Evaluating progressive-rendering algorithms in appearance design tasks.

PubMed

Jiawei Ou; Karlik, Ondrej; Křivánek, Jaroslav; Pellacini, Fabio

2013-01-01

Progressive rendering is becoming a popular alternative to precomputational approaches to appearance design. However, progressive algorithms create images exhibiting visual artifacts at early stages. A user study investigated these artifacts' effects on user performance in appearance design tasks. Novice and expert subjects performed lighting and material editing tasks with four algorithms: random path tracing, quasirandom path tracing, progressive photon mapping, and virtual-point-light rendering. Both the novices and experts strongly preferred path tracing to progressive photon mapping and virtual-point-light rendering. None of the participants preferred random path tracing to quasirandom path tracing or vice versa; the same situation held between progressive photon mapping and virtual-point-light rendering. The user workflow didn’t differ significantly with the four algorithms. The Web Extras include a video showing how four progressive-rendering algorithms converged (at http://youtu.be/ck-Gevl1e9s), the source code used, and other supplementary materials.

Accuracy of patient specific organ-dose estimates obtained using an automated image segmentation algorithm

NASA Astrophysics Data System (ADS)

Gilat-Schmidt, Taly; Wang, Adam; Coradi, Thomas; Haas, Benjamin; Star-Lack, Josh

2016-03-01

The overall goal of this work is to develop a rapid, accurate and fully automated software tool to estimate patient-specific organ doses from computed tomography (CT) scans using a deterministic Boltzmann Transport Equation solver and automated CT segmentation algorithms. This work quantified the accuracy of organ dose estimates obtained by an automated segmentation algorithm. The investigated algorithm uses a combination of feature-based and atlas-based methods. A multiatlas approach was also investigated. We hypothesize that the auto-segmentation algorithm is sufficiently accurate to provide organ dose estimates since random errors at the organ boundaries will average out when computing the total organ dose. To test this hypothesis, twenty head-neck CT scans were expertly segmented into nine regions. A leave-one-out validation study was performed, where every case was automatically segmented with each of the remaining cases used as the expert atlas, resulting in nineteen automated segmentations for each of the twenty datasets. The segmented regions were applied to gold-standard Monte Carlo dose maps to estimate mean and peak organ doses. The results demonstrated that the fully automated segmentation algorithm estimated the mean organ dose to within 10% of the expert segmentation for regions other than the spinal canal, with median error for each organ region below 2%. In the spinal canal region, the median error was 7% across all data sets and atlases, with a maximum error of 20%. The error in peak organ dose was below 10% for all regions, with a median error below 4% for all organ regions. The multiple-case atlas reduced the variation in the dose estimates and additional improvements may be possible with more robust multi-atlas approaches. Overall, the results support potential feasibility of an automated segmentation algorithm to provide accurate organ dose estimates.

Data mining for multiagent rules, strategies, and fuzzy decision tree structure

NASA Astrophysics Data System (ADS)

Smith, James F., III; Rhyne, Robert D., II; Fisher, Kristin

2002-03-01

A fuzzy logic based resource manager (RM) has been developed that automatically allocates electronic attack resources in real-time over many dissimilar platforms. Two different data mining algorithms have been developed to determine rules, strategies, and fuzzy decision tree structure. The first data mining algorithm uses a genetic algorithm as a data mining function and is called from an electronic game. The game allows a human expert to play against the resource manager in a simulated battlespace with each of the defending platforms being exclusively directed by the fuzzy resource manager and the attacking platforms being controlled by the human expert or operating autonomously under their own logic. This approach automates the data mining problem. The game automatically creates a database reflecting the domain expert's knowledge. It calls a data mining function, a genetic algorithm, for data mining of the database as required and allows easy evaluation of the information mined in the second step. The criterion for re- optimization is discussed as well as experimental results. Then a second data mining algorithm that uses a genetic program as a data mining function is introduced to automatically discover fuzzy decision tree structures. Finally, a fuzzy decision tree generated through this process is discussed.

Group prioritisation with unknown expert weights in incomplete linguistic context

NASA Astrophysics Data System (ADS)

Cheng, Dong; Cheng, Faxin; Zhou, Zhili; Wang, Juan

2017-09-01

In this paper, we study a group prioritisation problem in situations when the expert weights are completely unknown and their judgement preferences are linguistic and incomplete. Starting from the theory of relative entropy (RE) and multiplicative consistency, an optimisation model is provided for deriving an individual priority vector without estimating the missing value(s) of an incomplete linguistic preference relation. In order to address the unknown expert weights in the group aggregating process, we define two new kinds of expert weight indicators based on RE: proximity entropy weight and similarity entropy weight. Furthermore, a dynamic-adjusting algorithm (DAA) is proposed to obtain an objective expert weight vector and capture the dynamic properties involved in it. Unlike the extant literature of group prioritisation, the proposed RE approach does not require pre-allocation of expert weights and can solve incomplete preference relations. An interesting finding is that once all the experts express their preference relations, the final expert weight vector derived from the DAA is fixed irrespective of the initial settings of expert weights. Finally, an application example is conducted to validate the effectiveness and robustness of the RE approach.

Quantitative three-dimensional transrectal ultrasound (TRUS) for prostate imaging

NASA Astrophysics Data System (ADS)

Pathak, Sayan D.; Aarnink, Rene G.; de la Rosette, Jean J.; Chalana, Vikram; Wijkstra, Hessel; Haynor, David R.; Debruyne, Frans M. J.; Kim, Yongmin

1998-06-01

With the number of men seeking medical care for prostate diseases rising steadily, the need of a fast and accurate prostate boundary detection and volume estimation tool is being increasingly experienced by the clinicians. Currently, these measurements are made manually, which results in a large examination time. A possible solution is to improve the efficiency by automating the boundary detection and volume estimation process with minimal involvement from the human experts. In this paper, we present an algorithm based on SNAKES to detect the boundaries. Our approach is to selectively enhance the contrast along the edges using an algorithm called sticks and integrate it with a SNAKES model. This integrated algorithm requires an initial curve for each ultrasound image to initiate the boundary detection process. We have used different schemes to generate the curves with a varying degree of automation and evaluated its effects on the algorithm performance. After the boundaries are identified, the prostate volume is calculated using planimetric volumetry. We have tested our algorithm on 6 different prostate volumes and compared the performance against the volumes manually measured by 3 experts. With the increase in the user inputs, the algorithm performance improved as expected. The results demonstrate that given an initial contour reasonably close to the prostate boundaries, the algorithm successfully delineates the prostate boundaries in an image, and the resulting volume measurements are in close agreement with those made by the human experts.

Active Learning with Irrelevant Examples

NASA Technical Reports Server (NTRS)

Wagstaff, Kiri; Mazzoni, Dominic

2009-01-01

An improved active learning method has been devised for training data classifiers. One example of a data classifier is the algorithm used by the United States Postal Service since the 1960s to recognize scans of handwritten digits for processing zip codes. Active learning algorithms enable rapid training with minimal investment of time on the part of human experts to provide training examples consisting of correctly classified (labeled) input data. They function by identifying which examples would be most profitable for a human expert to label. The goal is to maximize classifier accuracy while minimizing the number of examples the expert must label. Although there are several well-established methods for active learning, they may not operate well when irrelevant examples are present in the data set. That is, they may select an item for labeling that the expert simply cannot assign to any of the valid classes. In the context of classifying handwritten digits, the irrelevant items may include stray marks, smudges, and mis-scans. Querying the expert about these items results in wasted time or erroneous labels, if the expert is forced to assign the item to one of the valid classes. In contrast, the new algorithm provides a specific mechanism for avoiding querying the irrelevant items. This algorithm has two components: an active learner (which could be a conventional active learning algorithm) and a relevance classifier. The combination of these components yields a method, denoted Relevance Bias, that enables the active learner to avoid querying irrelevant data so as to increase its learning rate and efficiency when irrelevant items are present. The algorithm collects irrelevant data in a set of rejected examples, then trains the relevance classifier to distinguish between labeled (relevant) training examples and the rejected ones. The active learner combines its ranking of the items with the probability that they are relevant to yield a final decision about which item to present to the expert for labeling. Experiments on several data sets have demonstrated that the Relevance Bias approach significantly decreases the number of irrelevant items queried and also accelerates learning speed.

Evaluation of a Machine-Learning Algorithm for Treatment Planning in Prostate Low-Dose-Rate Brachytherapy

DOE Office of Scientific and Technical Information (OSTI.GOV)

Nicolae, Alexandru; Department of Medical Physics, Odette Cancer Center, Sunnybrook Health Sciences Centre, Toronto, Ontario; Morton, Gerard

Purpose: This work presents the application of a machine learning (ML) algorithm to automatically generate high-quality, prostate low-dose-rate (LDR) brachytherapy treatment plans. The ML algorithm can mimic characteristics of preoperative treatment plans deemed clinically acceptable by brachytherapists. The planning efficiency, dosimetry, and quality (as assessed by experts) of preoperative plans generated with an ML planning approach was retrospectively evaluated in this study. Methods and Materials: Preimplantation and postimplantation treatment plans were extracted from 100 high-quality LDR treatments and stored within a training database. The ML training algorithm matches similar features from a new LDR case to those within the trainingmore » database to rapidly obtain an initial seed distribution; plans were then further fine-tuned using stochastic optimization. Preimplantation treatment plans generated by the ML algorithm were compared with brachytherapist (BT) treatment plans in terms of planning time (Wilcoxon rank sum, α = 0.05) and dosimetry (1-way analysis of variance, α = 0.05). Qualitative preimplantation plan quality was evaluated by expert LDR radiation oncologists using a Likert scale questionnaire. Results: The average planning time for the ML approach was 0.84 ± 0.57 minutes, compared with 17.88 ± 8.76 minutes for the expert planner (P=.020). Preimplantation plans were dosimetrically equivalent to the BT plans; the average prostate V150% was 4% lower for ML plans (P=.002), although the difference was not clinically significant. Respondents ranked the ML-generated plans as equivalent to expert BT treatment plans in terms of target coverage, normal tissue avoidance, implant confidence, and the need for plan modifications. Respondents had difficulty differentiating between plans generated by a human or those generated by the ML algorithm. Conclusions: Prostate LDR preimplantation treatment plans that have equivalent quality to plans created by brachytherapists can be rapidly generated using ML. The adoption of ML in the brachytherapy workflow is expected to improve LDR treatment plan uniformity while reducing planning time and resources.« less

A relational learning approach to Structure-Activity Relationships in drug design toxicity studies.

PubMed

Camacho, Rui; Pereira, Max; Costa, Vítor Santos; Fonseca, Nuno A; Adriano, Carlos; Simões, Carlos J V; Brito, Rui M M

2011-09-16

It has been recognized that the development of new therapeutic drugs is a complex and expensive process. A large number of factors affect the activity in vivo of putative candidate molecules and the propensity for causing adverse and toxic effects is recognized as one of the major hurdles behind the current "target-rich, lead-poor" scenario. Structure-Activity Relationship (SAR) studies, using relational Machine Learning (ML) algorithms, have already been shown to be very useful in the complex process of rational drug design. Despite the ML successes, human expertise is still of the utmost importance in the drug development process. An iterative process and tight integration between the models developed by ML algorithms and the know-how of medicinal chemistry experts would be a very useful symbiotic approach. In this paper we describe a software tool that achieves that goal--iLogCHEM. The tool allows the use of Relational Learners in the task of identifying molecules or molecular fragments with potential to produce toxic effects, and thus help in stream-lining drug design in silico. It also allows the expert to guide the search for useful molecules without the need to know the details of the algorithms used. The models produced by the algorithms may be visualized using a graphical interface, that is of common use amongst researchers in structural biology and medicinal chemistry. The graphical interface enables the expert to provide feedback to the learning system. The developed tool has also facilities to handle the similarity bias typical of large chemical databases. For that purpose the user can filter out similar compounds when assembling a data set. Additionally, we propose ways of providing background knowledge for Relational Learners using the results of Graph Mining algorithms. Copyright 2011 The Author(s). Published by Journal of Integrative Bioinformatics.

AutoNR: an automated system that measures ECAP thresholds with the Nucleus Freedom cochlear implant via machine intelligence.

PubMed

Botros, Andrew; van Dijk, Bas; Killian, Matthijs

2007-05-01

AutoNRT is an automated system that measures electrically evoked compound action potential (ECAP) thresholds from the auditory nerve with the Nucleus Freedom cochlear implant. ECAP thresholds along the electrode array are useful in objectively fitting cochlear implant systems for individual use. This paper provides the first detailed description of the AutoNRT algorithm and its expert systems, and reports the clinical success of AutoNRT to date. AutoNRT determines thresholds by visual detection, using two decision tree expert systems that automatically recognise ECAPs. The expert systems are guided by a dataset of 5393 neural response measurements. The algorithm approaches threshold from lower stimulus levels, ensuring recipient safety during postoperative measurements. Intraoperative measurements use the same algorithm but proceed faster by beginning at stimulus levels much closer to threshold. When searching for ECAPs, AutoNRT uses a highly specific expert system (specificity of 99% during training, 96% during testing; sensitivity of 91% during training, 89% during testing). Once ECAPs are established, AutoNRT uses an unbiased expert system to determine an accurate threshold. Throughout the execution of the algorithm, recording parameters (such as implant amplifier gain) are automatically optimised when needed. In a study that included 29 intraoperative and 29 postoperative subjects (a total of 418 electrodes), AutoNRT determined a threshold in 93% of cases where a human expert also determined a threshold. When compared to the median threshold of multiple human observers on 77 randomly selected electrodes, AutoNRT performed as accurately as the 'average' clinician. AutoNRT has demonstrated a high success rate and a level of performance that is comparable with human experts. It has been used in many clinics worldwide throughout the clinical trial and commercial launch of Nucleus Custom Sound Suite, significantly streamlining the clinical procedures associated with cochlear implant use.

A Knowledge Navigation Method for the Domain of Customers' Services of Mobile Communication Corporations in China

NASA Astrophysics Data System (ADS)

Wu, Jiangning; Wang, Xiaohuan

Rapidly increasing amount of mobile phone users and types of services leads to a great accumulation of complaining information. How to use this information to enhance the quality of customers' services is a big issue at present. To handle this kind of problem, the paper presents an approach to construct a domain knowledge map for navigating the explicit and tacit knowledge in two ways: building the Topic Map-based explicit knowledge navigation model, which includes domain TM construction, a semantic topic expansion algorithm and VSM-based similarity calculation; building Social Network Analysis-based tacit knowledge navigation model, which includes a multi-relational expert navigation algorithm and the criterions to evaluate the performance of expert networks. In doing so, both the customer managers and operators in call centers can find the appropriate knowledge and experts quickly and exactly. The experimental results show that the above method is very powerful for knowledge navigation.

Expert system constant false alarm rate processor

NASA Astrophysics Data System (ADS)

Baldygo, William J., Jr.; Wicks, Michael C.

1993-10-01

The requirements for high detection probability and low false alarm probability in modern wide area surveillance radars are rarely met due to spatial variations in clutter characteristics. Many filtering and CFAR detection algorithms have been developed to effectively deal with these variations; however, any single algorithm is likely to exhibit excessive false alarms and intolerably low detection probabilities in a dynamically changing environment. A great deal of research has led to advances in the state of the art in Artificial Intelligence (AI) and numerous areas have been identified for application to radar signal processing. The approach suggested here, discussed in a patent application submitted by the authors, is to intelligently select the filtering and CFAR detection algorithms being executed at any given time, based upon the observed characteristics of the interference environment. This approach requires sensing the environment, employing the most suitable algorithms, and applying an appropriate multiple algorithm fusion scheme or consensus algorithm to produce a global detection decision.

A mathematical framework for combining decisions of multiple experts toward accurate and remote diagnosis of malaria using tele-microscopy.

PubMed

Mavandadi, Sam; Feng, Steve; Yu, Frank; Dimitrov, Stoyan; Nielsen-Saines, Karin; Prescott, William R; Ozcan, Aydogan

2012-01-01

We propose a methodology for digitally fusing diagnostic decisions made by multiple medical experts in order to improve accuracy of diagnosis. Toward this goal, we report an experimental study involving nine experts, where each one was given more than 8,000 digital microscopic images of individual human red blood cells and asked to identify malaria infected cells. The results of this experiment reveal that even highly trained medical experts are not always self-consistent in their diagnostic decisions and that there exists a fair level of disagreement among experts, even for binary decisions (i.e., infected vs. uninfected). To tackle this general medical diagnosis problem, we propose a probabilistic algorithm to fuse the decisions made by trained medical experts to robustly achieve higher levels of accuracy when compared to individual experts making such decisions. By modelling the decisions of experts as a three component mixture model and solving for the underlying parameters using the Expectation Maximisation algorithm, we demonstrate the efficacy of our approach which significantly improves the overall diagnostic accuracy of malaria infected cells. Additionally, we present a mathematical framework for performing 'slide-level' diagnosis by using individual 'cell-level' diagnosis data, shedding more light on the statistical rules that should govern the routine practice in examination of e.g., thin blood smear samples. This framework could be generalized for various other tele-pathology needs, and can be used by trained experts within an efficient tele-medicine platform.

Advanced power system protection and incipient fault detection and protection of spaceborne power systems

NASA Technical Reports Server (NTRS)

Russell, B. Don

1989-01-01

This research concentrated on the application of advanced signal processing, expert system, and digital technologies for the detection and control of low grade, incipient faults on spaceborne power systems. The researchers have considerable experience in the application of advanced digital technologies and the protection of terrestrial power systems. This experience was used in the current contracts to develop new approaches for protecting the electrical distribution system in spaceborne applications. The project was divided into three distinct areas: (1) investigate the applicability of fault detection algorithms developed for terrestrial power systems to the detection of faults in spaceborne systems; (2) investigate the digital hardware and architectures required to monitor and control spaceborne power systems with full capability to implement new detection and diagnostic algorithms; and (3) develop a real-time expert operating system for implementing diagnostic and protection algorithms. Significant progress has been made in each of the above areas. Several terrestrial fault detection algorithms were modified to better adapt to spaceborne power system environments. Several digital architectures were developed and evaluated in light of the fault detection algorithms.

Extending Correlation Filter-Based Visual Tracking by Tree-Structured Ensemble and Spatial Windowing.

PubMed

Gundogdu, Erhan; Ozkan, Huseyin; Alatan, A Aydin

2017-11-01

Correlation filters have been successfully used in visual tracking due to their modeling power and computational efficiency. However, the state-of-the-art correlation filter-based (CFB) tracking algorithms tend to quickly discard the previous poses of the target, since they consider only a single filter in their models. On the contrary, our approach is to register multiple CFB trackers for previous poses and exploit the registered knowledge when an appearance change occurs. To this end, we propose a novel tracking algorithm [of complexity O(D) ] based on a large ensemble of CFB trackers. The ensemble [of size O(2 D ) ] is organized over a binary tree (depth D ), and learns the target appearance subspaces such that each constituent tracker becomes an expert of a certain appearance. During tracking, the proposed algorithm combines only the appearance-aware relevant experts to produce boosted tracking decisions. Additionally, we propose a versatile spatial windowing technique to enhance the individual expert trackers. For this purpose, spatial windows are learned for target objects as well as the correlation filters and then the windowed regions are processed for more robust correlations. In our extensive experiments on benchmark datasets, we achieve a substantial performance increase by using the proposed tracking algorithm together with the spatial windowing.

An Approach towards Ultrasound Kidney Cysts Detection using Vector Graphic Image Analysis

NASA Astrophysics Data System (ADS)

Mahmud, Wan Mahani Hafizah Wan; Supriyanto, Eko

2017-08-01

This study develops new approach towards detection of kidney ultrasound image for both with single cyst as well as multiple cysts. 50 single cyst images and 25 multiple cysts images were used to test the developed algorithm. Steps involved in developing this algorithm were vector graphic image formation and analysis, thresholding, binarization, filtering as well as roundness test. Performance evaluation to 50 single cyst images gave accuracy of 92%, while for multiple cysts images, the accuracy was about 86.89% when tested to 25 multiple cysts images. This developed algorithm may be used in developing a computerized system such as computer aided diagnosis system to help medical experts in diagnosis of kidney cysts.

A genetic algorithms approach for altering the membership functions in fuzzy logic controllers

NASA Technical Reports Server (NTRS)

Shehadeh, Hana; Lea, Robert N.

1992-01-01

Through previous work, a fuzzy control system was developed to perform translational and rotational control of a space vehicle. This problem was then re-examined to determine the effectiveness of genetic algorithms on fine tuning the controller. This paper explains the problems associated with the design of this fuzzy controller and offers a technique for tuning fuzzy logic controllers. A fuzzy logic controller is a rule-based system that uses fuzzy linguistic variables to model human rule-of-thumb approaches to control actions within a given system. This 'fuzzy expert system' features rules that direct the decision process and membership functions that convert the linguistic variables into the precise numeric values used for system control. Defining the fuzzy membership functions is the most time consuming aspect of the controller design. One single change in the membership functions could significantly alter the performance of the controller. This membership function definition can be accomplished by using a trial and error technique to alter the membership functions creating a highly tuned controller. This approach can be time consuming and requires a great deal of knowledge from human experts. In order to shorten development time, an iterative procedure for altering the membership functions to create a tuned set that used a minimal amount of fuel for velocity vector approach and station-keep maneuvers was developed. Genetic algorithms, search techniques used for optimization, were utilized to solve this problem.

Evidence-based algorithm for diagnosis and assessment in psoriatic arthritis: results by Italian DElphi in psoriatic Arthritis (IDEA).

PubMed

Lapadula, G; Marchesoni, A; Salaffi, F; Ramonda, R; Salvarani, C; Punzi, L; Costa, L; Caso, F; Simone, D; Baiocchi, G; Scioscia, C; Di Carlo, M; Scarpa, R; Ferraccioli, G

2016-12-16

Psoriatic arthritis (PsA) is a chronic inflammatory disease involving skin, peripheral joints, entheses, and axial skeleton. The disease is frequently associated with extrarticular manifestations (EAMs) and comorbidities. In order to create a protocol for PsA diagnosis and global assessment of patients with an algorithm based on anamnestic, clinical, laboratory and imaging procedures, we established a DElphi study on a national scale, named Italian DElphi in psoriatic Arthritis (IDEA). After a literature search, a Delphi poll, involving 52 rheumatologists, was performed. On the basis of the literature search, 202 potential items were identified. The steering committee planned at least two Delphi rounds. In the first Delphi round, the experts judged each of the 202 items using a score ranging from 1 to 9 based on its increasing clinical relevance. The questions posed to experts were How relevant is this procedure/observation/sign/symptom for assessment of a psoriatic arthritis patient? Proposals of additional items, not included in the questionnaire, were also encouraged. The results of the poll were discussed by the Steering Committee, which evaluated the necessity for removing selected procedures or adding additional ones, according to criteria of clinical appropriateness and sustainability. A total of 43 recommended diagnosis and assessment procedures, recognized as items, were derived by combination of the Delphi survey and two National Expert Meetings, and grouped in different areas. Favourable opinion was reached in 100% of cases for several aspects covering the following areas: medical (familial and personal) history, physical evaluation, imaging tool, second level laboratory tests, disease activity measurement and extrarticular manifestations. After performing PsA diagnosis, identification of specific disease activity scores and clinimetric approaches were suggested for assessing the different clinical subsets. Further, results showed the need for investigation on the presence of several EAMs and risk factors. In the context of any area, a rank was assigned for each item by Expert Committee members, in order to create the logical sequence of the algorithm. The final list of recommended diagnosis and assessment procedures, by the Delphi survey and the two National Expert Meetings, was also reported as an algorithm. This study shows results obtained by the combination of a DElphi survey of a group of Italian rheumatologists and two National Expert Meetings, created with the aim of establishing a clinical procedure and algorithm for the diagnosis and the assessment of PsA patients. In order to find accurate and practical diagnostic and assessment items in clinical practice, we have focused our attention on evaluating the different PsA domains. Hence, we conceived the IDEA algorithm in order to address PsA diagnosis and assessment in the context of daily clinical practice. The IDEA algorithm might eventually lead to a multidimensional approach and could represent a useful and practical tool for addressing diagnosis and for assessing the disease appropriately. However, the elaborated algorithm needs to be further investigated in daily practice, for evidencing and proving its eventual efficacy in detecting and staging PsA and its heterogeneous spectrum appropriately.

Automatic classification of scar tissue in late gadolinium enhancement cardiac MRI for the assessment of left-atrial wall injury after radiofrequency ablation

PubMed Central

Morris, Alan; Burgon, Nathan; McGann, Christopher; MacLeod, Robert; Cates, Joshua

2013-01-01

Radiofrequency ablation is a promising procedure for treating atrial fibrillation (AF) that relies on accurate lesion delivery in the left atrial (LA) wall for success. Late Gadolinium Enhancement MRI (LGE MRI) at three months post-ablation has proven effective for noninvasive assessment of the location and extent of scar formation, which are important factors for predicting patient outcome and planning of redo ablation procedures. We have developed an algorithm for automatic classification in LGE MRI of scar tissue in the LA wall and have evaluated accuracy and consistency compared to manual scar classifications by expert observers. Our approach clusters voxels based on normalized intensity and was chosen through a systematic comparison of the performance of multivariate clustering on many combinations of image texture. Algorithm performance was determined by overlap with ground truth, using multiple overlap measures, and the accuracy of the estimation of the total amount of scar in the LA. Ground truth was determined using the STAPLE algorithm, which produces a probabilistic estimate of the true scar classification from multiple expert manual segmentations. Evaluation of the ground truth data set was based on both inter- and intra-observer agreement, with variation among expert classifiers indicating the difficulty of scar classification for a given a dataset. Our proposed automatic scar classification algorithm performs well for both scar localization and estimation of scar volume: for ground truth datasets considered easy, variability from the ground truth was low; for those considered difficult, variability from ground truth was on par with the variability across experts. PMID:24236224

Automatic classification of scar tissue in late gadolinium enhancement cardiac MRI for the assessment of left-atrial wall injury after radiofrequency ablation

NASA Astrophysics Data System (ADS)

Perry, Daniel; Morris, Alan; Burgon, Nathan; McGann, Christopher; MacLeod, Robert; Cates, Joshua

2012-03-01

Radiofrequency ablation is a promising procedure for treating atrial fibrillation (AF) that relies on accurate lesion delivery in the left atrial (LA) wall for success. Late Gadolinium Enhancement MRI (LGE MRI) at three months post-ablation has proven effective for noninvasive assessment of the location and extent of scar formation, which are important factors for predicting patient outcome and planning of redo ablation procedures. We have developed an algorithm for automatic classification in LGE MRI of scar tissue in the LA wall and have evaluated accuracy and consistency compared to manual scar classifications by expert observers. Our approach clusters voxels based on normalized intensity and was chosen through a systematic comparison of the performance of multivariate clustering on many combinations of image texture. Algorithm performance was determined by overlap with ground truth, using multiple overlap measures, and the accuracy of the estimation of the total amount of scar in the LA. Ground truth was determined using the STAPLE algorithm, which produces a probabilistic estimate of the true scar classification from multiple expert manual segmentations. Evaluation of the ground truth data set was based on both inter- and intra-observer agreement, with variation among expert classifiers indicating the difficulty of scar classification for a given a dataset. Our proposed automatic scar classification algorithm performs well for both scar localization and estimation of scar volume: for ground truth datasets considered easy, variability from the ground truth was low; for those considered difficult, variability from ground truth was on par with the variability across experts.

Prognostic validation of a 17-segment score derived from a 20-segment score for myocardial perfusion SPECT interpretation.

PubMed

Berman, Daniel S; Abidov, Aiden; Kang, Xingping; Hayes, Sean W; Friedman, John D; Sciammarella, Maria G; Cohen, Ishac; Gerlach, James; Waechter, Parker B; Germano, Guido; Hachamovitch, Rory

2004-01-01

Recently, a 17-segment model of the left ventricle has been recommended as an optimally weighted approach for interpreting myocardial perfusion single photon emission computed tomography (SPECT). Methods to convert databases from previous 20- to new 17-segment data and criteria for abnormality for the 17-segment scores are needed. Initially, for derivation of the conversion algorithm, 65 patients were studied (algorithm population) (pilot group, n = 28; validation group, n = 37). Three conversion algorithms were derived: algorithm 1, which used mid, distal, and apical scores; algorithm 2, which used distal and apical scores alone; and algorithm 3, which used maximal scores of the distal septal, lateral, and apical segments in the 20-segment model for 3 corresponding segments of the 17-segment model. The prognosis population comprised 16,020 consecutive patients (mean age, 65 +/- 12 years; 41% women) who had exercise or vasodilator stress technetium 99m sestamibi myocardial perfusion SPECT and were followed up for 2.1 +/- 0.8 years. In this population, 17-segment scores were derived from 20-segment scores by use of algorithm 2, which demonstrated the best agreement with expert 17-segment reading in the algorithm population. The prognostic value of the 20- and 17-segment scores was compared by converting the respective summed scores into percent myocardium abnormal. Conversion algorithm 2 was found to be highly concordant with expert visual analysis by the 17-segment model (r = 0.982; kappa = 0.866) in the algorithm population. In the prognosis population, 456 cardiac deaths occurred during follow-up. When the conversion algorithm was applied, extent and severity of perfusion defects were nearly identical by 20- and derived 17-segment scores. The receiver operating characteristic curve areas by 20- and 17-segment perfusion scores were identical for predicting cardiac death (both 0.77 +/- 0.02, P = not significant). The optimal prognostic cutoff value for either 20- or derived 17-segment models was confirmed to be 5% myocardium abnormal, corresponding to a summed stress score greater than 3. Of note, the 17-segment model demonstrated a trend toward fewer mildly abnormal scans and more normal and severely abnormal scans. An algorithm for conversion of 20-segment perfusion scores to 17-segment scores has been developed that is highly concordant with expert visual analysis by the 17-segment model and provides nearly identical prognostic information. This conversion model may provide a mechanism for comparison of studies analyzed by the 17-segment system with previous studies analyzed by the 20-segment approach.

A smartphone-based pain management app for adolescents with cancer: establishing system requirements and a pain care algorithm based on literature review, interviews, and consensus.

PubMed

Jibb, Lindsay A; Stevens, Bonnie J; Nathan, Paul C; Seto, Emily; Cafazzo, Joseph A; Stinson, Jennifer N

2014-03-19

Pain that occurs both within and outside of the hospital setting is a common and distressing problem for adolescents with cancer. The use of smartphone technology may facilitate rapid, in-the-moment pain support for this population. To ensure the best possible pain management advice is given, evidence-based and expert-vetted care algorithms and system design features, which are designed using user-centered methods, are required. To develop the decision algorithm and system requirements that will inform the pain management advice provided by a real-time smartphone-based pain management app for adolescents with cancer. A systematic approach to algorithm development and system design was utilized. Initially, a comprehensive literature review was undertaken to understand the current body of knowledge pertaining to pediatric cancer pain management. A user-centered approach to development was used as the results of the review were disseminated to 15 international experts (clinicians, scientists, and a consumer) in pediatric pain, pediatric oncology and mHealth design, who participated in a 2-day consensus conference. This conference used nominal group technique to develop consensus on important pain inputs, pain management advice, and system design requirements. Using data generated at the conference, a prototype algorithm was developed. Iterative qualitative testing was conducted with adolescents with cancer, as well as pediatric oncology and pain health care providers to vet and refine the developed algorithm and system requirements for the real-time smartphone app. The systematic literature review established the current state of research related to nonpharmacological pediatric cancer pain management. The 2-day consensus conference established which clinically important pain inputs by adolescents would require action (pain management advice) from the app, the appropriate advice the app should provide to adolescents in pain, and the functional requirements of the app. These results were used to build a detailed prototype algorithm capable of providing adolescents with pain management support based on their individual pain. Analysis of qualitative interviews with 9 multidisciplinary health care professionals and 10 adolescents resulted in 4 themes that helped to adapt the algorithm and requirements to the needs of adolescents. Specifically, themes were overall endorsement of the system, the need for a clinical expert, the need to individualize the system, and changes to the algorithm to improve potential clinical effectiveness. This study used a phased and user-centered approach to develop a pain management algorithm for adolescents with cancer and the system requirements of an associated app. The smartphone software is currently being created and subsequent work will focus on the usability, feasibility, and effectiveness testing of the app for adolescents with cancer pain.

A Smartphone-Based Pain Management App for Adolescents With Cancer: Establishing System Requirements and a Pain Care Algorithm Based on Literature Review, Interviews, and Consensus

PubMed Central

Stevens, Bonnie J; Nathan, Paul C; Seto, Emily; Cafazzo, Joseph A; Stinson, Jennifer N

2014-01-01

Background Pain that occurs both within and outside of the hospital setting is a common and distressing problem for adolescents with cancer. The use of smartphone technology may facilitate rapid, in-the-moment pain support for this population. To ensure the best possible pain management advice is given, evidence-based and expert-vetted care algorithms and system design features, which are designed using user-centered methods, are required. Objective To develop the decision algorithm and system requirements that will inform the pain management advice provided by a real-time smartphone-based pain management app for adolescents with cancer. Methods A systematic approach to algorithm development and system design was utilized. Initially, a comprehensive literature review was undertaken to understand the current body of knowledge pertaining to pediatric cancer pain management. A user-centered approach to development was used as the results of the review were disseminated to 15 international experts (clinicians, scientists, and a consumer) in pediatric pain, pediatric oncology and mHealth design, who participated in a 2-day consensus conference. This conference used nominal group technique to develop consensus on important pain inputs, pain management advice, and system design requirements. Using data generated at the conference, a prototype algorithm was developed. Iterative qualitative testing was conducted with adolescents with cancer, as well as pediatric oncology and pain health care providers to vet and refine the developed algorithm and system requirements for the real-time smartphone app. Results The systematic literature review established the current state of research related to nonpharmacological pediatric cancer pain management. The 2-day consensus conference established which clinically important pain inputs by adolescents would require action (pain management advice) from the app, the appropriate advice the app should provide to adolescents in pain, and the functional requirements of the app. These results were used to build a detailed prototype algorithm capable of providing adolescents with pain management support based on their individual pain. Analysis of qualitative interviews with 9 multidisciplinary health care professionals and 10 adolescents resulted in 4 themes that helped to adapt the algorithm and requirements to the needs of adolescents. Specifically, themes were overall endorsement of the system, the need for a clinical expert, the need to individualize the system, and changes to the algorithm to improve potential clinical effectiveness. Conclusions This study used a phased and user-centered approach to develop a pain management algorithm for adolescents with cancer and the system requirements of an associated app. The smartphone software is currently being created and subsequent work will focus on the usability, feasibility, and effectiveness testing of the app for adolescents with cancer pain. PMID:24646454

Use of Management Pathways or Algorithms in Children With Chronic Cough: CHEST Guideline and Expert Panel Report.

PubMed

Chang, Anne B; Oppenheimer, John J; Weinberger, Miles M; Rubin, Bruce K; Weir, Kelly; Grant, Cameron C; Irwin, Richard S

2017-04-01

Using management algorithms or pathways potentially improves clinical outcomes. We undertook systematic reviews to examine various aspects in the generic approach (use of cough algorithms and tests) to the management of chronic cough in children (aged ≤ 14 years) based on key questions (KQs) using the Population, Intervention, Comparison, Outcome format. We used the CHEST Expert Cough Panel's protocol for the systematic reviews and the American College of Chest Physicians (CHEST) methodological guidelines and Grading of Recommendations Assessment, Development and Evaluation framework. Data from the systematic reviews in conjunction with patients' values and preferences and the clinical context were used to form recommendations. Delphi methodology was used to obtain the final grading. Combining data from systematic reviews addressing five KQs, we found high-quality evidence that a systematic approach to the management of chronic cough improves clinical outcomes. Although there was evidence from several pathways, the highest evidence was from the use of the CHEST approach. However, there was no or little evidence to address some of the KQs posed. Compared with the 2006 Cough Guidelines, there is now high-quality evidence that in children aged ≤ 14 years with chronic cough (> 4 weeks' duration), the use of cough management protocols (or algorithms) improves clinical outcomes, and cough management or testing algorithms should differ depending on the associated characteristics of the cough and clinical history. A chest radiograph and, when age appropriate, spirometry (pre- and post-β 2 agonist) should be undertaken. Other tests should not be routinely performed and undertaken in accordance with the clinical setting and the child's clinical symptoms and signs (eg, tests for tuberculosis when the child has been exposed). Copyright © 2017 American College of Chest Physicians. All rights reserved.

Difficulties in the diagnosis of vertebral fracture in men: agreement between doctors.

PubMed

Fechtenbaum, Jacques; Briot, Karine; Paternotte, Simon; Audran, Maurice; Breuil, Véronique; Cortet, Bernard; Debiais, Françoise; Grados, Franck; Guggenbuhl, Pascal; Laroche, Michel; Legrand, Erick; Lespessailles, Eric; Marcelli, Christian; Orcel, Philippe; Szulc, Pawel; Thomas, Thierry; Kolta, Sami; Roux, Christian

2014-03-01

The agreement for vertebral fracture (VF) diagnosis in men, between doctors is poor. To assess the agreement for VF diagnosis, in men, on standard radiographs, between experts, before and after consensual workshop and establishing an algorithm. The agreement between thirteen experimented rheumatologists has been calculated in thirty osteoporotic men. Then, the group discussed in a workshop and 28 other radiograph sets of osteoporotic men with follow-up radiographs and incident confirmed VF, have been reviewed. The experts identified and hierarchised 18 pathological features of vertebral deformation and established an algorithm of VF diagnosis. Eleven experts have realized a second reading of the first set of radiographs. We compared the agreement between the 2 readings without and with the algorithm. After consensus and the use of the algorithm the results are: number of fractured patients (with at least 1 VF) according to the experts varies from 13 to 26 patients out of 30 (13 to 28 during the first reading). The agreement between the experts at the patient level is 75% (70% at the first reading). Among the 390 vertebrae analyzed by the experts, the number of VF detected varies from 18 to 59 (18 to 98 at the first reading). The agreement between the experts at the vertebral level is 92% (89% at the first reading). The algorithm allows a good improvement of the agreement, especially for 8 of the 11 experts. Discrepancies for the VF diagnosis between experts exist. The algorithm improves the agreement. Copyright © 2013 Société française de rhumatologie. Published by Elsevier SAS. All rights reserved.

Leveraging the crowd for annotation of retinal images.

PubMed

Leifman, George; Swedish, Tristan; Roesch, Karin; Raskar, Ramesh

2015-01-01

Medical data presents a number of challenges. It tends to be unstructured, noisy and protected. To train algorithms to understand medical images, doctors can label the condition associated with a particular image, but obtaining enough labels can be difficult. We propose an annotation approach which starts with a small pool of expertly annotated images and uses their expertise to rate the performance of crowd-sourced annotations. In this paper we demonstrate how to apply our approach for annotation of large-scale datasets of retinal images. We introduce a novel data validation procedure which is designed to cope with noisy ground-truth data and with non-consistent input from both experts and crowd-workers.

Incorporating priors on expert performance parameters for segmentation validation and label fusion: a maximum a posteriori STAPLE

PubMed Central

Commowick, Olivier; Warfield, Simon K

2010-01-01

In order to evaluate the quality of segmentations of an image and assess intra- and inter-expert variability in segmentation performance, an Expectation Maximization (EM) algorithm for Simultaneous Truth And Performance Level Estimation (STAPLE) was recently developed. This algorithm, originally presented for segmentation validation, has since been used for many applications, such as atlas construction and decision fusion. However, the manual delineation of structures of interest is a very time consuming and burdensome task. Further, as the time required and burden of manual delineation increase, the accuracy of the delineation is decreased. Therefore, it may be desirable to ask the experts to delineate only a reduced number of structures or the segmentation of all structures by all experts may simply not be achieved. Fusion from data with some structures not segmented by each expert should be carried out in a manner that accounts for the missing information. In other applications, locally inconsistent segmentations may drive the STAPLE algorithm into an undesirable local optimum, leading to misclassifications or misleading experts performance parameters. We present a new algorithm that allows fusion with partial delineation and which can avoid convergence to undesirable local optima in the presence of strongly inconsistent segmentations. The algorithm extends STAPLE by incorporating prior probabilities for the expert performance parameters. This is achieved through a Maximum A Posteriori formulation, where the prior probabilities for the performance parameters are modeled by a beta distribution. We demonstrate that this new algorithm enables dramatically improved fusion from data with partial delineation by each expert in comparison to fusion with STAPLE. PMID:20879379

Incorporating priors on expert performance parameters for segmentation validation and label fusion: a maximum a posteriori STAPLE.

PubMed

Commowick, Olivier; Warfield, Simon K

2010-01-01

In order to evaluate the quality of segmentations of an image and assess intra- and inter-expert variability in segmentation performance, an Expectation Maximization (EM) algorithm for Simultaneous Truth And Performance Level Estimation (STAPLE) was recently developed. This algorithm, originally presented for segmentation validation, has since been used for many applications, such as atlas construction and decision fusion. However, the manual delineation of structures of interest is a very time consuming and burdensome task. Further, as the time required and burden of manual delineation increase, the accuracy of the delineation is decreased. Therefore, it may be desirable to ask the experts to delineate only a reduced number of structures or the segmentation of all structures by all experts may simply not be achieved. Fusion from data with some structures not segmented by each expert should be carried out in a manner that accounts for the missing information. In other applications, locally inconsistent segmentations may drive the STAPLE algorithm into an undesirable local optimum, leading to misclassifications or misleading experts performance parameters. We present a new algorithm that allows fusion with partial delineation and which can avoid convergence to undesirable local optima in the presence of strongly inconsistent segmentations. The algorithm extends STAPLE by incorporating prior probabilities for the expert performance parameters. This is achieved through a Maximum A Posteriori formulation, where the prior probabilities for the performance parameters are modeled by a beta distribution. We demonstrate that this new algorithm enables dramatically improved fusion from data with partial delineation by each expert in comparison to fusion with STAPLE.

Branch-pipe-routing approach for ships using improved genetic algorithm

NASA Astrophysics Data System (ADS)

Sui, Haiteng; Niu, Wentie

2016-09-01

Branch-pipe routing plays fundamental and critical roles in ship-pipe design. The branch-pipe-routing problem is a complex combinatorial optimization problem and is thus difficult to solve when depending only on human experts. A modified genetic-algorithm-based approach is proposed in this paper to solve this problem. The simplified layout space is first divided into threedimensional (3D) grids to build its mathematical model. Branch pipes in layout space are regarded as a combination of several two-point pipes, and the pipe route between two connection points is generated using an improved maze algorithm. The coding of branch pipes is then defined, and the genetic operators are devised, especially the complete crossover strategy that greatly accelerates the convergence speed. Finally, simulation tests demonstrate the performance of proposed method.

Rule groupings: A software engineering approach towards verification of expert systems

NASA Technical Reports Server (NTRS)

Mehrotra, Mala

1991-01-01

Currently, most expert system shells do not address software engineering issues for developing or maintaining expert systems. As a result, large expert systems tend to be incomprehensible, difficult to debug or modify and almost impossible to verify or validate. Partitioning rule based systems into rule groups which reflect the underlying subdomains of the problem should enhance the comprehensibility, maintainability, and reliability of expert system software. Attempts were made to semiautomatically structure a CLIPS rule base into groups of related rules that carry the same type of information. Different distance metrics that capture relevant information from the rules for grouping are discussed. Two clustering algorithms that partition the rule base into groups of related rules are given. Two independent evaluation criteria are developed to measure the effectiveness of the grouping strategies. Results of the experiment with three sample rule bases are presented.

Do we need annotation experts? A case study in celiac disease classification.

PubMed

Kwitt, Roland; Hegenbart, Sebastian; Rasiwasia, Nikhil; Vécsei, Andreas; Uhl, Andreas

2014-01-01

Inference of clinically-relevant findings from the visual appearance of images has become an essential part of processing pipelines for many problems in medical imaging. Typically, a sufficient amount labeled training data is assumed to be available, provided by domain experts. However, acquisition of this data is usually a time-consuming and expensive endeavor. In this work, we ask the question if, for certain problems, expert knowledge is actually required. In fact, we investigate the impact of letting non-expert volunteers annotate a database of endoscopy images which are then used to assess the absence/presence of celiac disease. Contrary to previous approaches, we are not interested in algorithms that can handle the label noise. Instead, we present compelling empirical evidence that label noise can be compensated by a sufficiently large corpus of training data, labeled by the non-experts.

Naïve Bayes Approach for Expert System Design of Children Skin Identification Based on Android

NASA Astrophysics Data System (ADS)

Hartatik; Purnomo, A.; Hartono, R.; Munawaroh, H.

2018-03-01

The development of technology gives some benefits to each person that we can use it properly and correctly. Technology has helped humans in every way. Such as the excess task of an expert in providing information or answers to a problem. Thus problem that often occurs is skin disease that affecting on child. That because the skin of children still vulnerable to the environment. The application was developed using the naïve Bayes algorithm. Through this application, users can consult with a system like an expert to know the symptoms that occur to the child and find the correct treatment to solve the problems.

Intelligent Visual Input: A Graphical Method for Rapid Entry of Patient-Specific Data

PubMed Central

Bergeron, Bryan P.; Greenes, Robert A.

1987-01-01

Intelligent Visual Input (IVI) provides a rapid, graphical method of data entry for both expert system interaction and medical record keeping purposes. Key components of IVI include: a high-resolution graphic display; an interface supportive of rapid selection, i.e., one utilizing a mouse or light pen; algorithm simplification modules; and intelligent graphic algorithm expansion modules. A prototype IVI system, designed to facilitate entry of physical exam findings, is used to illustrates the potential advantages of this approach.

Artificial intelligence within the chemical laboratory.

PubMed

Winkel, P

1994-01-01

Various techniques within the area of artificial intelligence such as expert systems and neural networks may play a role during the problem-solving processes within the clinical biochemical laboratory. Neural network analysis provides a non-algorithmic approach to information processing, which results in the ability of the computer to form associations and to recognize patterns or classes among data. It belongs to the machine learning techniques which also include probabilistic techniques such as discriminant function analysis and logistic regression and information theoretical techniques. These techniques may be used to extract knowledge from example patients to optimize decision limits and identify clinically important laboratory quantities. An expert system may be defined as a computer program that can give advice in a well-defined area of expertise and is able to explain its reasoning. Declarative knowledge consists of statements about logical or empirical relationships between things. Expert systems typically separate declarative knowledge residing in a knowledge base from the inference engine: an algorithm that dynamically directs and controls the system when it searches its knowledge base. A tool is an expert system without a knowledge base. The developer of an expert system uses a tool by entering knowledge into the system. Many, if not the majority of problems encountered at the laboratory level are procedural. A problem is procedural if it is possible to write up a step-by-step description of the expert's work or if it can be represented by a decision tree. To solve problems of this type only small expert system tools and/or conventional programming are required.(ABSTRACT TRUNCATED AT 250 WORDS)

Improving the efficiency of a user-driven learning system with reconfigurable hardware. Application to DNA splicing.

PubMed

Lemoine, E; Merceron, D; Sallantin, J; Nguifo, E M

1999-01-01

This paper describes a new approach to problem solving by splitting up problem component parts between software and hardware. Our main idea arises from the combination of two previously published works. The first one proposed a conceptual environment of concept modelling in which the machine and the human expert interact. The second one reported an algorithm based on reconfigurable hardware system which outperforms any kind of previously published genetic data base scanning hardware or algorithms. Here we show how efficient the interaction between the machine and the expert is when the concept modelling is based on reconfigurable hardware system. Their cooperation is thus achieved with an real time interaction speed. The designed system has been partially applied to the recognition of primate splice junctions sites in genetic sequences.

Classification of EEG abnormalities in partial epilepsy with simultaneous EEG-fMRI recordings.

PubMed

Pedreira, C; Vaudano, A E; Thornton, R C; Chaudhary, U J; Vulliemoz, S; Laufs, H; Rodionov, R; Carmichael, D W; Lhatoo, S D; Guye, M; Quian Quiroga, R; Lemieux, L

2014-10-01

Scalp EEG recordings and the classification of interictal epileptiform discharges (IED) in patients with epilepsy provide valuable information about the epileptogenic network, particularly by defining the boundaries of the "irritative zone" (IZ), and hence are helpful during pre-surgical evaluation of patients with severe refractory epilepsies. The current detection and classification of epileptiform signals essentially rely on expert observers. This is a very time-consuming procedure, which also leads to inter-observer variability. Here, we propose a novel approach to automatically classify epileptic activity and show how this method provides critical and reliable information related to the IZ localization beyond the one provided by previous approaches. We applied Wave_clus, an automatic spike sorting algorithm, for the classification of IED visually identified from pre-surgical simultaneous Electroencephalogram-functional Magnetic Resonance Imagining (EEG-fMRI) recordings in 8 patients affected by refractory partial epilepsy candidate for surgery. For each patient, two fMRI analyses were performed: one based on the visual classification and one based on the algorithmic sorting. This novel approach successfully identified a total of 29 IED classes (compared to 26 for visual identification). The general concordance between methods was good, providing a full match of EEG patterns in 2 cases, additional EEG information in 2 other cases and, in general, covering EEG patterns of the same areas as expert classification in 7 of the 8 cases. Most notably, evaluation of the method with EEG-fMRI data analysis showed hemodynamic maps related to the majority of IED classes representing improved performance than the visual IED classification-based analysis (72% versus 50%). Furthermore, the IED-related BOLD changes revealed by using the algorithm were localized within the presumed IZ for a larger number of IED classes (9) in a greater number of patients than the expert classification (7 and 5, respectively). In contrast, in only one case presented the new algorithm resulted in fewer classes and activation areas. We propose that the use of automated spike sorting algorithms to classify IED provides an efficient tool for mapping IED-related fMRI changes and increases the EEG-fMRI clinical value for the pre-surgical assessment of patients with severe epilepsy. Copyright © 2014 Elsevier Inc. All rights reserved.

An exploration of crowdsourcing citation screening for systematic reviews

PubMed Central

Mortensen, Michael L.; Adam, Gaelen P.; Trikalinos, Thomas A.; Kraska, Tim

2017-01-01

Systematic reviews are increasingly used to inform health care decisions, but are expensive to produce. We explore the use of crowdsourcing (distributing tasks to untrained workers via the web) to reduce the cost of screening citations. We used Amazon Mechanical Turk as our platform and 4 previously conducted systematic reviews as examples. For each citation, workers answered 4 or 5 questions that were equivalent to the eligibility criteria. We aggregated responses from multiple workers into an overall decision to include or exclude the citation using 1 of 9 algorithms and compared the performance of these algorithms to the corresponding decisions of trained experts. The most inclusive algorithm (designating a citation as relevant if any worker did) identified 95% to 99% of the citations that were ultimately included in the reviews while excluding 68% to 82% of irrelevant citations. Other algorithms increased the fraction of irrelevant articles excluded at some cost to the inclusion of relevant studies. Crowdworkers completed screening in 4 to 17 days, costing $460 to $2220, a cost reduction of up to 88% compared to trained experts. Crowdsourcing may represent a useful approach to reducing the cost of identifying literature for systematic reviews. PMID:28677322

On structure-exploiting trust-region regularized nonlinear least squares algorithms for neural-network learning.

PubMed

Mizutani, Eiji; Demmel, James W

2003-01-01

This paper briefly introduces our numerical linear algebra approaches for solving structured nonlinear least squares problems arising from 'multiple-output' neural-network (NN) models. Our algorithms feature trust-region regularization, and exploit sparsity of either the 'block-angular' residual Jacobian matrix or the 'block-arrow' Gauss-Newton Hessian (or Fisher information matrix in statistical sense) depending on problem scale so as to render a large class of NN-learning algorithms 'efficient' in both memory and operation costs. Using a relatively large real-world nonlinear regression application, we shall explain algorithmic strengths and weaknesses, analyzing simulation results obtained by both direct and iterative trust-region algorithms with two distinct NN models: 'multilayer perceptrons' (MLP) and 'complementary mixtures of MLP-experts' (or neuro-fuzzy modular networks).

Evolutionary Data Mining Approach to Creating Digital Logic

DTIC Science & Technology

2010-01-01

To deal with this problem a genetic program (GP) based data mining ( DM ) procedure has been invented (Smith 2005). A genetic program is an algorithm...that can operate on the variables. When a GP was used as a DM function in the past to automatically create fuzzy decision trees, the Report...rules represents an approach to the determining the effect of linguistic imprecision, i.e., the inability of experts to provide crisp rules. The

Spacecraft attitude control using a smart control system

NASA Technical Reports Server (NTRS)

Buckley, Brian; Wheatcraft, Louis

1992-01-01

Traditionally, spacecraft attitude control has been implemented using control loops written in native code for a space hardened processor. The Naval Research Lab has taken this approach during the development of the Attitude Control Electronics (ACE) package. After the system was developed and delivered, NRL decided to explore alternate technologies to accomplish this same task more efficiently. The approach taken by NRL was to implement the ACE control loops using systems technologies. The purpose of this effort was to: (1) research capabilities required of an expert system in processing a classic closed-loop control algorithm; (2) research the development environment required to design and test an embedded expert systems environment; (3) research the complexity of design and development of expert systems versus a conventional approach; and (4) test the resulting systems against the flight acceptance test software for both response and accuracy. Two expert systems were selected to implement the control loops. Criteria used for the selection of the expert systems included that they had to run in both embedded systems and ground based environments. Using two different expert systems allowed a comparison of the real-time capabilities, inferencing capabilities, and the ground-based development environment. The two expert systems chosen for the evaluation were Spacecraft Command Language (SCL), and NEXTPERT Object. SCL is a smart control system produced for the NRL by Interface and Control Systems (ICS). SCL was developed to be used for real-time command, control, and monitoring of a new generation of spacecraft. NEXPERT Object is a commercially available product developed by Neuron Data. Results of the effort were evaluated using the ACE test bed. The ACE test bed had been developed and used to test the original flight hardware and software using simulators and flight-like interfaces. The test bed was used for testing the expert systems in a 'near-flight' environment. The technical approach, the system architecture, the development environments, knowledge base development, and results of this effort are detailed.

A service relation model for web-based land cover change detection

NASA Astrophysics Data System (ADS)

Xing, Huaqiao; Chen, Jun; Wu, Hao; Zhang, Jun; Li, Songnian; Liu, Boyu

2017-10-01

Change detection with remotely sensed imagery is a critical step in land cover monitoring and updating. Although a variety of algorithms or models have been developed, none of them can be universal for all cases. The selection of appropriate algorithms and construction of processing workflows depend largely on the expertise of experts about the "algorithm-data" relations among change detection algorithms and the imagery data used. This paper presents a service relation model for land cover change detection by integrating the experts' knowledge about the "algorithm-data" relations into the web-based geo-processing. The "algorithm-data" relations are mapped into a set of web service relations with the analysis of functional and non-functional service semantics. These service relations are further classified into three different levels, i.e., interface, behavior and execution levels. A service relation model is then established using the Object and Relation Diagram (ORD) approach to represent the multi-granularity services and their relations for change detection. A set of semantic matching rules are built and used for deriving on-demand change detection service chains from the service relation model. A web-based prototype system is developed in .NET development environment, which encapsulates nine change detection and pre-processing algorithms and represents their service relations as an ORD. Three test areas from Shandong and Hebei provinces, China with different imagery conditions are selected for online change detection experiments, and the results indicate that on-demand service chains can be generated according to different users' demands.

Classifying BCI signals from novice users with extreme learning machine

NASA Astrophysics Data System (ADS)

Rodríguez-Bermúdez, Germán; Bueno-Crespo, Andrés; José Martinez-Albaladejo, F.

2017-07-01

Brain computer interface (BCI) allows to control external devices only with the electrical activity of the brain. In order to improve the system, several approaches have been proposed. However it is usual to test algorithms with standard BCI signals from experts users or from repositories available on Internet. In this work, extreme learning machine (ELM) has been tested with signals from 5 novel users to compare with standard classification algorithms. Experimental results show that ELM is a suitable method to classify electroencephalogram signals from novice users.

New directions for Artificial Intelligence (AI) methods in optimum design

NASA Technical Reports Server (NTRS)

Hajela, Prabhat

1989-01-01

Developments and applications of artificial intelligence (AI) methods in the design of structural systems is reviewed. Principal shortcomings in the current approach are emphasized, and the need for some degree of formalism in the development environment for such design tools is underscored. Emphasis is placed on efforts to integrate algorithmic computations in expert systems.

Deconstructing Chronic Low Back Pain in the Older Adult-Step by Step Evidence and Expert-Based Recommendations for Evaluation and Treatment. Part VI: Lumbar Spinal Stenosis.

PubMed

Fritz, Julie M; Rundell, Sean D; Dougherty, Paul; Gentili, Angela; Kochersberger, Gary; Morone, Natalia E; Naga Raja, Srinivasa; Rodriguez, Eric; Rossi, Michelle I; Shega, Joseph; Sowa, Gwendolyn; Weiner, Debra K

2016-03-01

. To present the sixth in a series of articles designed to deconstruct chronic low back pain (CLBP) in older adults. This article focuses on the evaluation and management of lumbar spinal stenosis (LSS), the most common condition for which older adults undergo spinal surgery. . The evaluation and treatment algorithm, a table articulating the rationale for the individual algorithm components, and stepped-care drug recommendations were developed using a modified Delphi approach. The Principal Investigator, a five-member content expert panel and a nine-member primary care panel were involved in the iterative development of these materials. The illustrative clinical case was taken from the clinical practice of a contributor's colleague (SR). . We present an algorithm and supportive materials to help guide the care of older adults with LSS, a condition that occurs not uncommonly in those with CLBP. The case illustrates the importance of function-focused management and a rational approach to conservative care. . Lumbar spinal stenosis exists not uncommonly in older adults with CLBP and management often can be accomplished without surgery. Treatment should address all conditions in addition to LSS contributing to pain and disability. © 2016 American Academy of Pain Medicine. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Expert AIV: Study and Prototyping of an Expert System, To Support the Conceptual AIV Phases Of Space Programs

NASA Astrophysics Data System (ADS)

Andrina, G.; Basso, V.; Saitta, L.

2004-08-01

The effort in optimising the AIV process has been mainly focused in the recent years on the standardisation of approaches and on the application of new methodologies. But the earlier the intervention, the greater the benefits in terms of cost and schedule. Early phases of AIV process relied up to now on standards that need to be tailored through company and personal expertise. A study has then been conducted in order to exploit the possibility to develop an expert system helping in making choices in the early, conceptual phase of Assembly, Integration and Verification, namely the Model Philosophy and the test definition. The work focused on a hybrid approach, allowing interaction between historical data and human expertise. The expert system that has been prototyped exploits both information elicited from domain experts and results of a Data Mining activity on the existent data bases of completed projects verification data. The Data Mining algorithms allow the extraction of past experience resident on ESA/ MATD data base, which contains information in the form of statistical summaries, costs, frequencies of on-ground and in flight failures. Finding non-trivial associations could then be utilised by the experts to manage new decisions in a controlled way (Standards driven) at the beginning or during the AIV Process Moreover, the Expert AIV could allow compilation of a set of feasible AIV schedules to support further programmatic-driven choices.

Western Trauma Association Critical Decisions in Trauma: Management of rib fractures.

PubMed

Brasel, Karen J; Moore, Ernest E; Albrecht, Roxie A; deMoya, Marc; Schreiber, Martin; Karmy-Jones, Riyad; Rowell, Susan; Namias, Nicholas; Cohen, Mitchell; Shatz, David V; Biffl, Walter L

2017-01-01

This is a recommended management algorithm from the Western Trauma Association addressing the management of adult patients with rib fractures. Because there is a paucity of published prospective randomized clinical trials that have generated Class I data, these recommendations are based primarily on published observational studies and expert opinion of Western Trauma Association members. The algorithm and accompanying comments represent a safe and sensible approach that can be followed at most trauma centers. We recognize that there will be patient, personnel, institutional, and situational factors that may warrant or require deviation from the recommended algorithm. We encourage institutions to use this as a guideline to develop their own local protocols.

An Algorithm Using Twelve Properties of Antibiotics to Find the Recommended Antibiotics, as in CPGs.

PubMed

Tsopra, R; Venot, A; Duclos, C

2014-01-01

Clinical Decision Support Systems (CDSS) incorporating justifications, updating and adjustable recommendations can considerably improve the quality of healthcare. We propose a new approach to the design of CDSS for empiric antibiotic prescription, based on implementation of the deeper medical reasoning used by experts in the development of clinical practice guidelines (CPGs), to deduce the recommended antibiotics. We investigated two methods ("exclusion" versus "scoring") for reproducing this reasoning based on antibiotic properties. The "exclusion" method reproduced expert reasoning the more accurately, retrieving the full list of recommended antibiotics for almost all clinical situations. This approach has several advantages: (i) it provides convincing explanations for physicians; (ii) updating could easily be incorporated into the CDSS; (iii) it can provide recommendations for clinical situations missing from CPGs.

Improving Automated Endmember Identification for Linear Unmixing of HyspIRI Spectral Data.

NASA Astrophysics Data System (ADS)

Gader, P.

2016-12-01

The size of data sets produced by imaging spectrometers is increasing rapidly. There is already a processing bottleneck. Part of the reason for this bottleneck is the need for expert input using interactive software tools. This process can be very time consuming and laborious but is currently crucial to ensuring the quality of the analysis. Automated algorithms can mitigate this problem. Although it is unlikely that processing systems can become completely automated, there is an urgent need to increase the level of automation. Spectral unmixing is a key component to processing HyspIRI data. Algorithms such as MESMA have been demonstrated to achieve results but require carefully, expert construction of endmember libraries. Unfortunately, many endmembers found by automated algorithms for finding endmembers are deemed unsuitable by experts because they are not physically reasonable. Unfortunately, endmembers that are not physically reasonable can achieve very low errors between the linear mixing model with those endmembers and the original data. Therefore, this error is not a reasonable way to resolve the problem on "non-physical" endmembers. There are many potential approaches for resolving these issues, including using Bayesian priors, but very little attention has been given to this problem. The study reported on here considers a modification of the Sparsity Promoting Iterated Constrained Endmember (SPICE) algorithm. SPICE finds endmembers and abundances and estimates the number of endmembers. The SPICE algorithm seeks to minimize a quadratic objective function with respect to endmembers E and fractions P. The modified SPICE algorithm, which we refer to as SPICED, is obtained by adding the term D to the objective function. The term D pressures the algorithm to minimize sum of the squared differences between each endmember and a weighted sum of the data. By appropriately modifying the, the endmembers are pushed towards a subset of the data with the potential for becoming exactly equal to the data points. The algorithm has been applied to spectral data and the differences between the endmembers resulting from ecorded. The results so far are that the endmembers found SPICED are approximately 25% closer to the data with indistinguishable reconstruction error compared to those found using SPICE.

Can masses of non-experts train highly accurate image classifiers? A crowdsourcing approach to instrument segmentation in laparoscopic images.

PubMed

Maier-Hein, Lena; Mersmann, Sven; Kondermann, Daniel; Bodenstedt, Sebastian; Sanchez, Alexandro; Stock, Christian; Kenngott, Hannes Gotz; Eisenmann, Mathias; Speidel, Stefanie

2014-01-01

Machine learning algorithms are gaining increasing interest in the context of computer-assisted interventions. One of the bottlenecks so far, however, has been the availability of training data, typically generated by medical experts with very limited resources. Crowdsourcing is a new trend that is based on outsourcing cognitive tasks to many anonymous untrained individuals from an online community. In this work, we investigate the potential of crowdsourcing for segmenting medical instruments in endoscopic image data. Our study suggests that (1) segmentations computed from annotations of multiple anonymous non-experts are comparable to those made by medical experts and (2) training data generated by the crowd is of the same quality as that annotated by medical experts. Given the speed of annotation, scalability and low costs, this implies that the scientific community might no longer need to rely on experts to generate reference or training data for certain applications. To trigger further research in endoscopic image processing, the data used in this study will be made publicly available.

AUI&GIV: Recommendation with Asymmetric User Influence and Global Importance Value.

PubMed

Zhao, Zhi-Lin; Wang, Chang-Dong; Lai, Jian-Huang

2016-01-01

The user-based collaborative filtering (CF) algorithm is one of the most popular approaches for making recommendation. Despite its success, the traditional user-based CF algorithm suffers one serious problem that it only measures the influence between two users based on their symmetric similarities calculated by their consumption histories. It means that, for a pair of users, the influences on each other are the same, which however may not be true. Intuitively, an expert may have an impact on a novice user but a novice user may not affect an expert at all. Besides, each user may possess a global importance factor that affects his/her influence to the remaining users. To this end, in this paper, we propose an asymmetric user influence model to measure the directed influence between two users and adopt the PageRank algorithm to calculate the global importance value of each user. And then the directed influence values and the global importance values are integrated to deduce the final influence values between two users. Finally, we use the final influence values to improve the performance of the traditional user-based CF algorithm. Extensive experiments have been conducted, the results of which have confirmed that both the asymmetric user influence model and global importance value play key roles in improving recommendation accuracy, and hence the proposed method significantly outperforms the existing recommendation algorithms, in particular the user-based CF algorithm on the datasets of high rating density.

AUI&GIV: Recommendation with Asymmetric User Influence and Global Importance Value

PubMed Central

Zhao, Zhi-Lin; Wang, Chang-Dong; Lai, Jian-Huang

2016-01-01

The user-based collaborative filtering (CF) algorithm is one of the most popular approaches for making recommendation. Despite its success, the traditional user-based CF algorithm suffers one serious problem that it only measures the influence between two users based on their symmetric similarities calculated by their consumption histories. It means that, for a pair of users, the influences on each other are the same, which however may not be true. Intuitively, an expert may have an impact on a novice user but a novice user may not affect an expert at all. Besides, each user may possess a global importance factor that affects his/her influence to the remaining users. To this end, in this paper, we propose an asymmetric user influence model to measure the directed influence between two users and adopt the PageRank algorithm to calculate the global importance value of each user. And then the directed influence values and the global importance values are integrated to deduce the final influence values between two users. Finally, we use the final influence values to improve the performance of the traditional user-based CF algorithm. Extensive experiments have been conducted, the results of which have confirmed that both the asymmetric user influence model and global importance value play key roles in improving recommendation accuracy, and hence the proposed method significantly outperforms the existing recommendation algorithms, in particular the user-based CF algorithm on the datasets of high rating density. PMID:26828803

Expert-guided evolutionary algorithm for layout design of complex space stations

NASA Astrophysics Data System (ADS)

Qian, Zhiqin; Bi, Zhuming; Cao, Qun; Ju, Weiguo; Teng, Hongfei; Zheng, Yang; Zheng, Siyu

2017-08-01

The layout of a space station should be designed in such a way that different equipment and instruments are placed for the station as a whole to achieve the best overall performance. The station layout design is a typical nondeterministic polynomial problem. In particular, how to manage the design complexity to achieve an acceptable solution within a reasonable timeframe poses a great challenge. In this article, a new evolutionary algorithm has been proposed to meet such a challenge. It is called as the expert-guided evolutionary algorithm with a tree-like structure decomposition (EGEA-TSD). Two innovations in EGEA-TSD are (i) to deal with the design complexity, the entire design space is divided into subspaces with a tree-like structure; it reduces the computation and facilitates experts' involvement in the solving process. (ii) A human-intervention interface is developed to allow experts' involvement in avoiding local optimums and accelerating convergence. To validate the proposed algorithm, the layout design of one-space station is formulated as a multi-disciplinary design problem, the developed algorithm is programmed and executed, and the result is compared with those from other two algorithms; it has illustrated the superior performance of the proposed EGEA-TSD.

Automatic computation of 2D cardiac measurements from B-mode echocardiography

NASA Astrophysics Data System (ADS)

Park, JinHyeong; Feng, Shaolei; Zhou, S. Kevin

2012-03-01

We propose a robust and fully automatic algorithm which computes the 2D echocardiography measurements recommended by America Society of Echocardiography. The algorithm employs knowledge-based imaging technologies which can learn the expert's knowledge from the training images and expert's annotation. Based on the models constructed from the learning stage, the algorithm searches initial location of the landmark points for the measurements by utilizing heart structure of left ventricle including mitral valve aortic valve. It employs the pseudo anatomic M-mode image generated by accumulating the line images in 2D parasternal long axis view along the time to refine the measurement landmark points. The experiment results with large volume of data show that the algorithm runs fast and is robust comparable to expert.

Towards a Framework for Evaluating and Comparing Diagnosis Algorithms

NASA Technical Reports Server (NTRS)

Kurtoglu, Tolga; Narasimhan, Sriram; Poll, Scott; Garcia,David; Kuhn, Lukas; deKleer, Johan; vanGemund, Arjan; Feldman, Alexander

2009-01-01

Diagnostic inference involves the detection of anomalous system behavior and the identification of its cause, possibly down to a failed unit or to a parameter of a failed unit. Traditional approaches to solving this problem include expert/rule-based, model-based, and data-driven methods. Each approach (and various techniques within each approach) use different representations of the knowledge required to perform the diagnosis. The sensor data is expected to be combined with these internal representations to produce the diagnosis result. In spite of the availability of various diagnosis technologies, there have been only minimal efforts to develop a standardized software framework to run, evaluate, and compare different diagnosis technologies on the same system. This paper presents a framework that defines a standardized representation of the system knowledge, the sensor data, and the form of the diagnosis results and provides a run-time architecture that can execute diagnosis algorithms, send sensor data to the algorithms at appropriate time steps from a variety of sources (including the actual physical system), and collect resulting diagnoses. We also define a set of metrics that can be used to evaluate and compare the performance of the algorithms, and provide software to calculate the metrics.

A new approach of data clustering using a flock of agents.

PubMed

Picarougne, Fabien; Azzag, Hanene; Venturini, Gilles; Guinot, Christiane

2007-01-01

This paper presents a new bio-inspired algorithm (FClust) that dynamically creates and visualizes groups of data. This algorithm uses the concepts of a flock of agents that move together in a complex manner with simple local rules. Each agent represents one data. The agents move together in a 2D environment with the aim of creating homogeneous groups of data. These groups are visualized in real time, and help the domain expert to understand the underlying structure of the data set, like for example a realistic number of classes, clusters of similar data, isolated data. We also present several extensions of this algorithm, which reduce its computational cost, and make use of a 3D display. This algorithm is then tested on artificial and real-world data, and a heuristic algorithm is used to evaluate the relevance of the obtained partitioning.

Machine Learning for Biological Trajectory Classification Applications

NASA Technical Reports Server (NTRS)

Sbalzarini, Ivo F.; Theriot, Julie; Koumoutsakos, Petros

2002-01-01

Machine-learning techniques, including clustering algorithms, support vector machines and hidden Markov models, are applied to the task of classifying trajectories of moving keratocyte cells. The different algorithms axe compared to each other as well as to expert and non-expert test persons, using concepts from signal-detection theory. The algorithms performed very well as compared to humans, suggesting a robust tool for trajectory classification in biological applications.

Expert networks in CLIPS

NASA Technical Reports Server (NTRS)

Hruska, S. I.; Dalke, A.; Ferguson, J. J.; Lacher, R. C.

1991-01-01

Rule-based expert systems may be structurally and functionally mapped onto a special class of neural networks called expert networks. This mapping lends itself to adaptation of connectionist learning strategies for the expert networks. A parsing algorithm to translate C Language Integrated Production System (CLIPS) rules into a network of interconnected assertion and operation nodes has been developed. The translation of CLIPS rules to an expert network and back again is illustrated. Measures of uncertainty similar to those rules in MYCIN-like systems are introduced into the CLIPS system and techniques for combining and hiring nodes in the network based on rule-firing with these certainty factors in the expert system are presented. Several learning algorithms are under study which automate the process of attaching certainty factors to rules.

Two New Tools for Glycopeptide Analysis Researchers: A Glycopeptide Decoy Generator and a Large Data Set of Assigned CID Spectra of Glycopeptides.

PubMed

Lakbub, Jude C; Su, Xiaomeng; Zhu, Zhikai; Patabandige, Milani W; Hua, David; Go, Eden P; Desaire, Heather

2017-08-04

The glycopeptide analysis field is tightly constrained by a lack of effective tools that translate mass spectrometry data into meaningful chemical information, and perhaps the most challenging aspect of building effective glycopeptide analysis software is designing an accurate scoring algorithm for MS/MS data. We provide the glycoproteomics community with two tools to address this challenge. The first tool, a curated set of 100 expert-assigned CID spectra of glycopeptides, contains a diverse set of spectra from a variety of glycan types; the second tool, Glycopeptide Decoy Generator, is a new software application that generates glycopeptide decoys de novo. We developed these tools so that emerging methods of assigning glycopeptides' CID spectra could be rigorously tested. Software developers or those interested in developing skills in expert (manual) analysis can use these tools to facilitate their work. We demonstrate the tools' utility in assessing the quality of one particular glycopeptide software package, GlycoPep Grader, which assigns glycopeptides to CID spectra. We first acquired the set of 100 expert assigned CID spectra; then, we used the Decoy Generator (described herein) to generate 20 decoys per target glycopeptide. The assigned spectra and decoys were used to test the accuracy of GlycoPep Grader's scoring algorithm; new strengths and weaknesses were identified in the algorithm using this approach. Both newly developed tools are freely available. The software can be downloaded at http://glycopro.chem.ku.edu/GPJ.jar.

An Algorithm Using Twelve Properties of Antibiotics to Find the Recommended Antibiotics, as in CPGs

PubMed Central

Tsopra, R.; Venot, A.; Duclos, C.

2014-01-01

Background Clinical Decision Support Systems (CDSS) incorporating justifications, updating and adjustable recommendations can considerably improve the quality of healthcare. We propose a new approach to the design of CDSS for empiric antibiotic prescription, based on implementation of the deeper medical reasoning used by experts in the development of clinical practice guidelines (CPGs), to deduce the recommended antibiotics. Methods We investigated two methods (“exclusion” versus “scoring”) for reproducing this reasoning based on antibiotic properties. Results The “exclusion” method reproduced expert reasoning the more accurately, retrieving the full list of recommended antibiotics for almost all clinical situations. Discussion This approach has several advantages: (i) it provides convincing explanations for physicians; (ii) updating could easily be incorporated into the CDSS; (iii) it can provide recommendations for clinical situations missing from CPGs. PMID:25954422

A comparison of neural network and fuzzy clustering techniques in segmenting magnetic resonance images of the brain.

PubMed

Hall, L O; Bensaid, A M; Clarke, L P; Velthuizen, R P; Silbiger, M S; Bezdek, J C

1992-01-01

Magnetic resonance (MR) brain section images are segmented and then synthetically colored to give visual representations of the original data with three approaches: the literal and approximate fuzzy c-means unsupervised clustering algorithms, and a supervised computational neural network. Initial clinical results are presented on normal volunteers and selected patients with brain tumors surrounded by edema. Supervised and unsupervised segmentation techniques provide broadly similar results. Unsupervised fuzzy algorithms were visually observed to show better segmentation when compared with raw image data for volunteer studies. For a more complex segmentation problem with tumor/edema or cerebrospinal fluid boundary, where the tissues have similar MR relaxation behavior, inconsistency in rating among experts was observed, with fuzz-c-means approaches being slightly preferred over feedforward cascade correlation results. Various facets of both approaches, such as supervised versus unsupervised learning, time complexity, and utility for the diagnostic process, are compared.

Bayesian Networks for Modeling Dredging Decisions

DTIC Science & Technology

2011-10-01

change scenarios. Arctic Expert elicitation Netica Bacon et al . 2002 Identify factors that might lead to a change in land use from farming to...tree) algorithms developed by Lauritzen and Spiegelhalter (1988) and Jensen et al . (1990). Statistical inference is simply the process of...causality when constructing a Bayesian network (Kjaerulff and Madsen 2008, Darwiche 2009, Marcot et al . 2006). A knowledge representation approach is the

Guidance, navigation, and control subsystem equipment selection algorithm using expert system methods

NASA Technical Reports Server (NTRS)

Allen, Cheryl L.

1991-01-01

Enhanced engineering tools can be obtained through the integration of expert system methodologies and existing design software. The application of these methodologies to the spacecraft design and cost model (SDCM) software provides an improved technique for the selection of hardware for unmanned spacecraft subsystem design. The knowledge engineering system (KES) expert system development tool was used to implement a smarter equipment section algorithm than that which is currently achievable through the use of a standard data base system. The guidance, navigation, and control subsystems of the SDCM software was chosen as the initial subsystem for implementation. The portions of the SDCM code which compute the selection criteria and constraints remain intact, and the expert system equipment selection algorithm is embedded within this existing code. The architecture of this new methodology is described and its implementation is reported. The project background and a brief overview of the expert system is described, and once the details of the design are characterized, an example of its implementation is demonstrated.

Quantitative CT based radiomics as predictor of resectability of pancreatic adenocarcinoma

NASA Astrophysics Data System (ADS)

van der Putten, Joost; Zinger, Svitlana; van der Sommen, Fons; de With, Peter H. N.; Prokop, Mathias; Hermans, John

2018-02-01

In current clinical practice, the resectability of pancreatic ductal adenocarcinoma (PDA) is determined subjec- tively by a physician, which is an error-prone procedure. In this paper, we present a method for automated determination of resectability of PDA from a routine abdominal CT, to reduce such decision errors. The tumor features are extracted from a group of patients with both hypo- and iso-attenuating tumors, of which 29 were resectable and 21 were not. The tumor contours are supplied by a medical expert. We present an approach that uses intensity, shape, and texture features to determine tumor resectability. The best classification results are obtained with fine Gaussian SVM and the L0 Feature Selection algorithms. Compared to expert predictions made on the same dataset, our method achieves better classification results. We obtain significantly better results on correctly predicting non-resectability (+17%) compared to a expert, which is essential for patient treatment (negative prediction value). Moreover, our predictions of resectability exceed expert predictions by approximately 3% (positive prediction value).

GREIT: a unified approach to 2D linear EIT reconstruction of lung images.

PubMed

Adler, Andy; Arnold, John H; Bayford, Richard; Borsic, Andrea; Brown, Brian; Dixon, Paul; Faes, Theo J C; Frerichs, Inéz; Gagnon, Hervé; Gärber, Yvo; Grychtol, Bartłomiej; Hahn, Günter; Lionheart, William R B; Malik, Anjum; Patterson, Robert P; Stocks, Janet; Tizzard, Andrew; Weiler, Norbert; Wolf, Gerhard K

2009-06-01

Electrical impedance tomography (EIT) is an attractive method for clinically monitoring patients during mechanical ventilation, because it can provide a non-invasive continuous image of pulmonary impedance which indicates the distribution of ventilation. However, most clinical and physiological research in lung EIT is done using older and proprietary algorithms; this is an obstacle to interpretation of EIT images because the reconstructed images are not well characterized. To address this issue, we develop a consensus linear reconstruction algorithm for lung EIT, called GREIT (Graz consensus Reconstruction algorithm for EIT). This paper describes the unified approach to linear image reconstruction developed for GREIT. The framework for the linear reconstruction algorithm consists of (1) detailed finite element models of a representative adult and neonatal thorax, (2) consensus on the performance figures of merit for EIT image reconstruction and (3) a systematic approach to optimize a linear reconstruction matrix to desired performance measures. Consensus figures of merit, in order of importance, are (a) uniform amplitude response, (b) small and uniform position error, (c) small ringing artefacts, (d) uniform resolution, (e) limited shape deformation and (f) high resolution. Such figures of merit must be attained while maintaining small noise amplification and small sensitivity to electrode and boundary movement. This approach represents the consensus of a large and representative group of experts in EIT algorithm design and clinical applications for pulmonary monitoring. All software and data to implement and test the algorithm have been made available under an open source license which allows free research and commercial use.

Stable Atlas-based Mapped Prior (STAMP) machine-learning segmentation for multicenter large-scale MRI data.

PubMed

Kim, Eun Young; Magnotta, Vincent A; Liu, Dawei; Johnson, Hans J

2014-09-01

Machine learning (ML)-based segmentation methods are a common technique in the medical image processing field. In spite of numerous research groups that have investigated ML-based segmentation frameworks, there remains unanswered aspects of performance variability for the choice of two key components: ML algorithm and intensity normalization. This investigation reveals that the choice of those elements plays a major part in determining segmentation accuracy and generalizability. The approach we have used in this study aims to evaluate relative benefits of the two elements within a subcortical MRI segmentation framework. Experiments were conducted to contrast eight machine-learning algorithm configurations and 11 normalization strategies for our brain MR segmentation framework. For the intensity normalization, a Stable Atlas-based Mapped Prior (STAMP) was utilized to take better account of contrast along boundaries of structures. Comparing eight machine learning algorithms on down-sampled segmentation MR data, it was obvious that a significant improvement was obtained using ensemble-based ML algorithms (i.e., random forest) or ANN algorithms. Further investigation between these two algorithms also revealed that the random forest results provided exceptionally good agreement with manual delineations by experts. Additional experiments showed that the effect of STAMP-based intensity normalization also improved the robustness of segmentation for multicenter data sets. The constructed framework obtained good multicenter reliability and was successfully applied on a large multicenter MR data set (n>3000). Less than 10% of automated segmentations were recommended for minimal expert intervention. These results demonstrate the feasibility of using the ML-based segmentation tools for processing large amount of multicenter MR images. We demonstrated dramatically different result profiles in segmentation accuracy according to the choice of ML algorithm and intensity normalization chosen. Copyright © 2014 Elsevier Inc. All rights reserved.

[Artificial intelligence in sleep analysis (ARTISANA)--modelling visual processes in sleep classification].

PubMed

Schwaibold, M; Schöller, B; Penzel, T; Bolz, A

2001-05-01

We describe a novel approach to the problem of automated sleep stage recognition. The ARTISANA algorithm mimics the behaviour of a human expert visually scoring sleep stages (Rechtschaffen and Kales classification). It comprises a number of interacting components that imitate the stepwise approach of the human expert, and artificial intelligence components. On the basis of parameters extracted at 1-s intervals from the signal curves, artificial neural networks recognize the incidence of typical patterns, e.g. delta activity or K complexes. This is followed by a rule interpretation stage that identifies the sleep stage with the aid of a neuro-fuzzy system while taking account of the context. Validation studies based on the records of 8 patients with obstructive sleep apnoea have confirmed the potential of this approach. Further features of the system include the transparency of the decision-taking process, and the flexibility of the option for expanding the system to cover new patterns and criteria.

Can human experts predict solubility better than computers?

PubMed

Boobier, Samuel; Osbourn, Anne; Mitchell, John B O

2017-12-13

In this study, we design and carry out a survey, asking human experts to predict the aqueous solubility of druglike organic compounds. We investigate whether these experts, drawn largely from the pharmaceutical industry and academia, can match or exceed the predictive power of algorithms. Alongside this, we implement 10 typical machine learning algorithms on the same dataset. The best algorithm, a variety of neural network known as a multi-layer perceptron, gave an RMSE of 0.985 log S units and an R 2 of 0.706. We would not have predicted the relative success of this particular algorithm in advance. We found that the best individual human predictor generated an almost identical prediction quality with an RMSE of 0.942 log S units and an R 2 of 0.723. The collection of algorithms contained a higher proportion of reasonably good predictors, nine out of ten compared with around half of the humans. We found that, for either humans or algorithms, combining individual predictions into a consensus predictor by taking their median generated excellent predictivity. While our consensus human predictor achieved very slightly better headline figures on various statistical measures, the difference between it and the consensus machine learning predictor was both small and statistically insignificant. We conclude that human experts can predict the aqueous solubility of druglike molecules essentially equally well as machine learning algorithms. We find that, for either humans or algorithms, combining individual predictions into a consensus predictor by taking their median is a powerful way of benefitting from the wisdom of crowds.

New insights into the classification and nomenclature of cortical GABAergic interneurons.

PubMed

DeFelipe, Javier; López-Cruz, Pedro L; Benavides-Piccione, Ruth; Bielza, Concha; Larrañaga, Pedro; Anderson, Stewart; Burkhalter, Andreas; Cauli, Bruno; Fairén, Alfonso; Feldmeyer, Dirk; Fishell, Gord; Fitzpatrick, David; Freund, Tamás F; González-Burgos, Guillermo; Hestrin, Shaul; Hill, Sean; Hof, Patrick R; Huang, Josh; Jones, Edward G; Kawaguchi, Yasuo; Kisvárday, Zoltán; Kubota, Yoshiyuki; Lewis, David A; Marín, Oscar; Markram, Henry; McBain, Chris J; Meyer, Hanno S; Monyer, Hannah; Nelson, Sacha B; Rockland, Kathleen; Rossier, Jean; Rubenstein, John L R; Rudy, Bernardo; Scanziani, Massimo; Shepherd, Gordon M; Sherwood, Chet C; Staiger, Jochen F; Tamás, Gábor; Thomson, Alex; Wang, Yun; Yuste, Rafael; Ascoli, Giorgio A

2013-03-01

A systematic classification and accepted nomenclature of neuron types is much needed but is currently lacking. This article describes a possible taxonomical solution for classifying GABAergic interneurons of the cerebral cortex based on a novel, web-based interactive system that allows experts to classify neurons with pre-determined criteria. Using Bayesian analysis and clustering algorithms on the resulting data, we investigated the suitability of several anatomical terms and neuron names for cortical GABAergic interneurons. Moreover, we show that supervised classification models could automatically categorize interneurons in agreement with experts' assignments. These results demonstrate a practical and objective approach to the naming, characterization and classification of neurons based on community consensus.

Joint Segmentation and Deformable Registration of Brain Scans Guided by a Tumor Growth Model

PubMed Central

Gooya, Ali; Pohl, Kilian M.; Bilello, Michel; Biros, George; Davatzikos, Christos

2011-01-01

This paper presents an approach for joint segmentation and deformable registration of brain scans of glioma patients to a normal atlas. The proposed method is based on the Expectation Maximization (EM) algorithm that incorporates a glioma growth model for atlas seeding, a process which modifies the normal atlas into one with a tumor and edema. The modified atlas is registered into the patient space and utilized for the posterior probability estimation of various tissue labels. EM iteratively refines the estimates of the registration parameters, the posterior probabilities of tissue labels and the tumor growth model parameters. We have applied this approach to 10 glioma scans acquired with four Magnetic Resonance (MR) modalities (T1, T1-CE, T2 and FLAIR ) and validated the result by comparing them to manual segmentations by clinical experts. The resulting segmentations look promising and quantitatively match well with the expert provided ground truth. PMID:21995070

Joint segmentation and deformable registration of brain scans guided by a tumor growth model.

PubMed

Gooya, Ali; Pohl, Kilian M; Bilello, Michel; Biros, George; Davatzikos, Christos

2011-01-01

This paper presents an approach for joint segmentation and deformable registration of brain scans of glioma patients to a normal atlas. The proposed method is based on the Expectation Maximization (EM) algorithm that incorporates a glioma growth model for atlas seeding, a process which modifies the normal atlas into one with a tumor and edema. The modified atlas is registered into the patient space and utilized for the posterior probability estimation of various tissue labels. EM iteratively refines the estimates of the registration parameters, the posterior probabilities of tissue labels and the tumor growth model parameters. We have applied this approach to 10 glioma scans acquired with four Magnetic Resonance (MR) modalities (T1, T1-CE, T2 and FLAIR) and validated the result by comparing them to manual segmentations by clinical experts. The resulting segmentations look promising and quantitatively match well with the expert provided ground truth.

Artificial immune system via Euclidean Distance Minimization for anomaly detection in bearings

NASA Astrophysics Data System (ADS)

Montechiesi, L.; Cocconcelli, M.; Rubini, R.

2016-08-01

In recent years new diagnostics methodologies have emerged, with particular interest into machinery operating in non-stationary conditions. In fact continuous speed changes and variable loads make non-trivial the spectrum analysis. A variable speed means a variable characteristic fault frequency related to the damage that is no more recognizable in the spectrum. To overcome this problem the scientific community proposed different approaches listed in two main categories: model-based approaches and expert systems. In this context the paper aims to present a simple expert system derived from the mechanisms of the immune system called Euclidean Distance Minimization, and its application in a real case of bearing faults recognition. The proposed method is a simplification of the original process, adapted by the class of Artificial Immune Systems, which proved to be useful and promising in different application fields. Comparative results are provided, with a complete explanation of the algorithm and its functioning aspects.

A Generalized Mixture Framework for Multi-label Classification

PubMed Central

Hong, Charmgil; Batal, Iyad; Hauskrecht, Milos

2015-01-01

We develop a novel probabilistic ensemble framework for multi-label classification that is based on the mixtures-of-experts architecture. In this framework, we combine multi-label classification models in the classifier chains family that decompose the class posterior distribution P(Y1, …, Yd|X) using a product of posterior distributions over components of the output space. Our approach captures different input–output and output–output relations that tend to change across data. As a result, we can recover a rich set of dependency relations among inputs and outputs that a single multi-label classification model cannot capture due to its modeling simplifications. We develop and present algorithms for learning the mixtures-of-experts models from data and for performing multi-label predictions on unseen data instances. Experiments on multiple benchmark datasets demonstrate that our approach achieves highly competitive results and outperforms the existing state-of-the-art multi-label classification methods. PMID:26613069

Discriminatory power of handwriting

NASA Astrophysics Data System (ADS)

Srihari, Sargur N.; Cha, Sung-Hyuk; Lee, Sangjik

2001-12-01

A study was undertaken to determine the power of handwriting to distinguish between individuals. Handwriting samples of one thousand five hundred individuals, representative of the US population with respect to gender, age, ethnic groups, etc., were obtained. Analyzing differences in handwriting was done by using computer algorithms for extracting features from scanned images of handwriting. Attributes characteristic of the handwriting were obtained, e.g., line separation, slant, character shapes, etc. These attributes, which are a subset of attributes used by expert document examiners, were used to quantitatively establish individuality by using machine learning approaches. Using global attributes of hadwriting and very few characters in the writing, the ability to determine the writer with a high degree of confidence was established. The work is a step towards providing scientific support for admitting handwriting evidence in court. The mathematical approach and the resulting software also have the promise of aiding the expert document examiner.

Automatic measurement of epidermal thickness from optical coherence tomography images using a new algorithm.

PubMed

Josse, G; George, J; Black, D

2011-08-01

Optical coherence tomography (OCT) is an imaging system that enables in vivo epidermal thickness (ET) measurement. In order to use OCT in large-scale clinical studies, automatic algorithm detection of the dermo-epidermal junction (DEJ) is needed. This may be difficult due to image noise from optical speckle, which requires specific image treatment procedures to reduce this. In the present work, a description of the position of the DEJ is given, and an algorithm for boundary detection is presented. Twenty-nine images were taken from the skin of normal healthy subjects, from five different body sites. Seven expert assessors were asked to trace the DEJ for ET measurement on each of the images. The variability between experts was compared with a new image processing method. Between-expert variability was relatively low with a mean standard deviation of 3.4 μm. However, local positioning of the DEJ between experts was often different. The described algorithm performed adequately on all images. ET was automatically measured with a precision of < 5 μm compared with the experts on all sites studied except that of the back. Moreover, the local algorithm positioning was verified. The new image processing method for measuring ET from OCT images significantly reduces calculation time for this parameter, and avoids user intervention. The main advantages of this are that data can be analyzed more rapidly and reproducibly in clinical trials. © 2011 John Wiley & Sons A/S.

A knowledge-based framework for image enhancement in aviation security.

PubMed

Singh, Maneesha; Singh, Sameer; Partridge, Derek

2004-12-01

The main aim of this paper is to present a knowledge-based framework for automatically selecting the best image enhancement algorithm from several available on a per image basis in the context of X-ray images of airport luggage. The approach detailed involves a system that learns to map image features that represent its viewability to one or more chosen enhancement algorithms. Viewability measures have been developed to provide an automatic check on the quality of the enhanced image, i.e., is it really enhanced? The choice is based on ground-truth information generated by human X-ray screening experts. Such a system, for a new image, predicts the best-suited enhancement algorithm. Our research details the various characteristics of the knowledge-based system and shows extensive results on real images.

A comparison of neural network and fuzzy clustering techniques in segmenting magnetic resonance images of the brain

NASA Technical Reports Server (NTRS)

Hall, Lawrence O.; Bensaid, Amine M.; Clarke, Laurence P.; Velthuizen, Robert P.; Silbiger, Martin S.; Bezdek, James C.

1992-01-01

Magnetic resonance (MR) brain section images are segmented and then synthetically colored to give visual representations of the original data with three approaches: the literal and approximate fuzzy c-means unsupervised clustering algorithms and a supervised computational neural network, a dynamic multilayered perception trained with the cascade correlation learning algorithm. Initial clinical results are presented on both normal volunteers and selected patients with brain tumors surrounded by edema. Supervised and unsupervised segmentation techniques provide broadly similar results. Unsupervised fuzzy algorithms were visually observed to show better segmentation when compared with raw image data for volunteer studies. However, for a more complex segmentation problem with tumor/edema or cerebrospinal fluid boundary, where the tissues have similar MR relaxation behavior, inconsistency in rating among experts was observed.

Hybrid expert system for decision supporting in the medical area: complexity and cognitive computing.

PubMed

Brasil, L M; de Azevedo, F M; Barreto, J M

2001-09-01

This paper proposes a hybrid expert system (HES) to minimise some complexity problems pervasive to the artificial intelligence such as: the knowledge elicitation process, known as the bottleneck of expert systems; the model choice for knowledge representation to code human reasoning; the number of neurons in the hidden layer and the topology used in the connectionist approach; the difficulty to obtain the explanation on how the network arrived to a conclusion. Two algorithms applied to developing of HES are also suggested. One of them is used to train the fuzzy neural network and the other to obtain explanations on how the fuzzy neural network attained a conclusion. To overcome these difficulties the cognitive computing was integrated to the developed system. A case study is presented (e.g. epileptic crisis) with the problem definition and simulations. Results are also discussed.

Syndromic Algorithms for Detection of Gambiense Human African Trypanosomiasis in South Sudan

PubMed Central

Palmer, Jennifer J.; Surur, Elizeous I.; Goch, Garang W.; Mayen, Mangar A.; Lindner, Andreas K.; Pittet, Anne; Kasparian, Serena; Checchi, Francesco; Whitty, Christopher J. M.

2013-01-01

Background Active screening by mobile teams is considered the best method for detecting human African trypanosomiasis (HAT) caused by Trypanosoma brucei gambiense but the current funding context in many post-conflict countries limits this approach. As an alternative, non-specialist health care workers (HCWs) in peripheral health facilities could be trained to identify potential cases who need testing based on their symptoms. We explored the predictive value of syndromic referral algorithms to identify symptomatic cases of HAT among a treatment-seeking population in Nimule, South Sudan. Methodology/Principal Findings Symptom data from 462 patients (27 cases) presenting for a HAT test via passive screening over a 7 month period were collected to construct and evaluate over 14,000 four item syndromic algorithms considered simple enough to be used by peripheral HCWs. For comparison, algorithms developed in other settings were also tested on our data, and a panel of expert HAT clinicians were asked to make referral decisions based on the symptom dataset. The best performing algorithms consisted of three core symptoms (sleep problems, neurological problems and weight loss), with or without a history of oedema, cervical adenopathy or proximity to livestock. They had a sensitivity of 88.9–92.6%, a negative predictive value of up to 98.8% and a positive predictive value in this context of 8.4–8.7%. In terms of sensitivity, these out-performed more complex algorithms identified in other studies, as well as the expert panel. The best-performing algorithm is predicted to identify about 9/10 treatment-seeking HAT cases, though only 1/10 patients referred would test positive. Conclusions/Significance In the absence of regular active screening, improving referrals of HAT patients through other means is essential. Systematic use of syndromic algorithms by peripheral HCWs has the potential to increase case detection and would increase their participation in HAT programmes. The algorithms proposed here, though promising, should be validated elsewhere. PMID:23350005

Natural and Artificial Intelligence in Neurosurgery: A Systematic Review.

PubMed

Senders, Joeky T; Arnaout, Omar; Karhade, Aditya V; Dasenbrock, Hormuzdiyar H; Gormley, William B; Broekman, Marike L; Smith, Timothy R

2017-09-07

Machine learning (ML) is a domain of artificial intelligence that allows computer algorithms to learn from experience without being explicitly programmed. To summarize neurosurgical applications of ML where it has been compared to clinical expertise, here referred to as "natural intelligence." A systematic search was performed in the PubMed and Embase databases as of August 2016 to review all studies comparing the performance of various ML approaches with that of clinical experts in neurosurgical literature. Twenty-three studies were identified that used ML algorithms for diagnosis, presurgical planning, or outcome prediction in neurosurgical patients. Compared to clinical experts, ML models demonstrated a median absolute improvement in accuracy and area under the receiver operating curve of 13% (interquartile range 4-21%) and 0.14 (interquartile range 0.07-0.21), respectively. In 29 (58%) of the 50 outcome measures for which a P -value was provided or calculated, ML models outperformed clinical experts ( P < .05). In 18 of 50 (36%), no difference was seen between ML and expert performance ( P > .05), while in 3 of 50 (6%) clinical experts outperformed ML models ( P < .05). All 4 studies that compared clinicians assisted by ML models vs clinicians alone demonstrated a better performance in the first group. We conclude that ML models have the potential to augment the decision-making capacity of clinicians in neurosurgical applications; however, significant hurdles remain associated with creating, validating, and deploying ML models in the clinical setting. Shifting from the preconceptions of a human-vs-machine to a human-and-machine paradigm could be essential to overcome these hurdles. Published by Oxford University Press on behalf of Congress of Neurological Surgeons 2017.

Second-line treatment for metastatic clear cell renal cell cancer: experts' consensus algorithms.

PubMed

Rothermundt, C; von Rappard, J; Eisen, T; Escudier, B; Grünwald, V; Larkin, J; McDermott, D; Oldenburg, J; Porta, C; Rini, B; Schmidinger, M; Sternberg, C N; Putora, P M

2017-04-01

Second-line systemic treatment options for metastatic clear cell renal cell cancer (mccRCC) are diverse and treatment strategies are variable among experts. Our aim was to investigate the approach for the second-line treatment after first-line therapy with a tyrosine kinase inhibitor (TKI). Recently two phase III trials have demonstrated a potential role for nivolumab (NIV) and cabozantinib (CAB) in this setting. We aimed to estimate the impact of these trials on clinical decision making. Eleven international experts were asked to provide their treatment strategies for second-line systemic therapy for mccRCC in the current setting and once NIV and CAB will be approved and available. The treatment strategies were analyzed with the objective consensus approach. The analysis of the decision trees revealed everolimus (EVE), axitinib (AXI), NIV and TKI switch (sTKI) as therapeutic options after first-line TKI therapy in the current situation and mostly NIV and CAB in the future setting. The most commonly used criteria for treatment decisions were duration of response, TKI tolerance and zugzwang a composite of several related criteria. In contrast to the first-line setting, recommendations for second-line systemic treatment of mccRCC among experts were not as heterogeneous. The agents mostly used after disease progression on a first-line TKI included: EVE, AXI, NIV and sTKI. In the future setting of NIV and CAB availability, NIV was the most commonly chosen drug, whereas several experts identified situations where CAB would be preferred.

Evaluating and Improving Automatic Sleep Spindle Detection by Using Multi-Objective Evolutionary Algorithms

PubMed Central

Liu, Min-Yin; Huang, Adam; Huang, Norden E.

2017-01-01

Sleep spindles are brief bursts of brain activity in the sigma frequency range (11–16 Hz) measured by electroencephalography (EEG) mostly during non-rapid eye movement (NREM) stage 2 sleep. These oscillations are of great biological and clinical interests because they potentially play an important role in identifying and characterizing the processes of various neurological disorders. Conventionally, sleep spindles are identified by expert sleep clinicians via visual inspection of EEG signals. The process is laborious and the results are inconsistent among different experts. To resolve the problem, numerous computerized methods have been developed to automate the process of sleep spindle identification. Still, the performance of these automated sleep spindle detection methods varies inconsistently from study to study. There are two reasons: (1) the lack of common benchmark databases, and (2) the lack of commonly accepted evaluation metrics. In this study, we focus on tackling the second problem by proposing to evaluate the performance of a spindle detector in a multi-objective optimization context and hypothesize that using the resultant Pareto fronts for deriving evaluation metrics will improve automatic sleep spindle detection. We use a popular multi-objective evolutionary algorithm (MOEA), the Strength Pareto Evolutionary Algorithm (SPEA2), to optimize six existing frequency-based sleep spindle detection algorithms. They include three Fourier, one continuous wavelet transform (CWT), and two Hilbert-Huang transform (HHT) based algorithms. We also explore three hybrid approaches. Trained and tested on open-access DREAMS and MASS databases, two new hybrid methods of combining Fourier with HHT algorithms show significant performance improvement with F1-scores of 0.726–0.737. PMID:28572762

Development and clinical application of an evidence-based pharmaceutical care service algorithm in acute coronary syndrome.

PubMed

Kang, J E; Yu, J M; Choi, J H; Chung, I-M; Pyun, W B; Kim, S A; Lee, E K; Han, N Y; Yoon, J-H; Oh, J M; Rhie, S J

2018-06-01

Drug therapies are critical for preventing secondary complications in acute coronary syndrome (ACS). The purpose of this study was to develop and apply a pharmaceutical care service (PCS) algorithm for ACS and confirm that it is applicable through a prospective clinical trial. The ACS-PCS algorithm was developed according to extant evidence-based treatment and pharmaceutical care guidelines. Quality assurance was conducted through two methods: literature comparison and expert panel evaluation. The literature comparison was used to compare the content of the algorithm with the referenced guidelines. Expert evaluations were conducted by nine experts for 75 questionnaire items. A trial was conducted to confirm its effectiveness. Seventy-nine patients were assigned to either the pharmacist-included multidisciplinary team care (MTC) group or the usual care (UC) group. The endpoints of the trial were the prescription rate of two important drugs, readmission, emergency room (ER) visit and mortality. The main frame of the algorithm was structured with three tasks: medication reconciliation, medication optimization and transition of care. The contents and context of the algorithm were compliant with class I recommendations and the main service items from the evidence-based guidelines. Opinions from the expert panel were mostly positive. There were significant differences in beta-blocker prescription rates in the overall period (P = .013) and ER visits (four cases, 9.76%, P = .016) in the MTC group compared to the UC group, respectively. We developed a PCS algorithm for ACS based on the contents of evidence-based drug therapy and the core concept of pharmacist services. © 2018 John Wiley & Sons Ltd.

Regionalization by fuzzy expert system based approach optimized by genetic algorithm

NASA Astrophysics Data System (ADS)

Chavoshi, Sattar; Azmin Sulaiman, Wan Nor; Saghafian, Bahram; Bin Sulaiman, Md. Nasir; Manaf, Latifah Abd

2013-04-01

SummaryIn recent years soft computing methods are being increasingly used to model complex hydrologic processes. These methods can simulate the real life processes without prior knowledge of the exact relationship between their components. The principal aim of this paper is perform hydrological regionalization based on soft computing concepts in the southern strip of the Caspian Sea basin, north of Iran. The basin with an area of 42,400 sq. km has been affected by severe floods in recent years that caused damages to human life and properties. Although some 61 hydrometric stations and 31 weather stations with 44 years of observed data (1961-2005) are operated in the study area, previous flood studies in this region have been hampered by insufficient and/or reliable observed rainfall-runoff records. In order to investigate the homogeneity (h) of catchments and overcome incompatibility that may occur on boundaries of cluster groups, a fuzzy expert system (FES) approach is used which incorporates physical and climatic characteristics, as well as flood seasonality and geographic location. Genetic algorithm (GA) was employed to adjust parameters of FES and optimize the system. In order to achieve the objective, a MATLAB programming code was developed which considers the heterogeneity criteria of less than 1 (H < 1) as the satisfying criteria. The adopted approach was found superior to the conventional hydrologic regionalization methods in the region because it employs greater number of homogeneity parameters and produces lower values of heterogeneity criteria.

Focusing attention on objects of interest using multiple matched filters.

PubMed

Stough, T M; Brodley, C E

2001-01-01

In order to be of use to scientists, large image databases need to be analyzed to create a catalog of the objects of interest. One approach is to apply a multiple tiered search algorithm that uses reduction techniques of increasing computational complexity to select the desired objects from the database. The first tier of this type of algorithm, often called a focus of attention (FOA) algorithm, selects candidate regions from the image data and passes them to the next tier of the algorithm. In this paper we present a new approach to FOA that employs multiple matched filters (MMF), one for each object prototype, to detect the regions of interest. The MMFs are formed using k-means clustering on a set of image patches identified by domain experts as positive examples of objects of interest. An innovation of the approach is to radically reduce the dimensionality of the feature space, used by the k-means algorithm, by taking block averages (spoiling) the sample image patches. The process of spoiling is analyzed and its applicability to other domains is discussed. The combination of the output of the MMFs is achieved through the projection of the detections back into an empty image and then thresholding. This research was motivated by the need to detect small volcanos in the Magellan probe data from Venus. An empirical evaluation of the approach illustrates that a combination of the MMF plus the average filter results in a higher likelihood of 100% detection of the objects of interest at a lower false positive rate than a single matched filter alone.

Extracting BI-RADS Features from Portuguese Clinical Texts.

PubMed

Nassif, Houssam; Cunha, Filipe; Moreira, Inês C; Cruz-Correia, Ricardo; Sousa, Eliana; Page, David; Burnside, Elizabeth; Dutra, Inês

2012-01-01

In this work we build the first BI-RADS parser for Portuguese free texts, modeled after existing approaches to extract BI-RADS features from English medical records. Our concept finder uses a semantic grammar based on the BIRADS lexicon and on iterative transferred expert knowledge. We compare the performance of our algorithm to manual annotation by a specialist in mammography. Our results show that our parser's performance is comparable to the manual method.

Sensitive and specific peak detection for SELDI-TOF mass spectrometry using a wavelet/neural-network based approach.

PubMed

Emanuele, Vincent A; Panicker, Gitika; Gurbaxani, Brian M; Lin, Jin-Mann S; Unger, Elizabeth R

2012-01-01

SELDI-TOF mass spectrometer's compact size and automated, high throughput design have been attractive to clinical researchers, and the platform has seen steady-use in biomarker studies. Despite new algorithms and preprocessing pipelines that have been developed to address reproducibility issues, visual inspection of the results of SELDI spectra preprocessing by the best algorithms still shows miscalled peaks and systematic sources of error. This suggests that there continues to be problems with SELDI preprocessing. In this work, we study the preprocessing of SELDI in detail and introduce improvements. While many algorithms, including the vendor supplied software, can identify peak clusters of specific mass (or m/z) in groups of spectra with high specificity and low false discover rate (FDR), the algorithms tend to underperform estimating the exact prevalence and intensity of peaks in those clusters. Thus group differences that at first appear very strong are shown, after careful and laborious hand inspection of the spectra, to be less than significant. Here we introduce a wavelet/neural network based algorithm which mimics what a team of expert, human users would call for peaks in each of several hundred spectra in a typical SELDI clinical study. The wavelet denoising part of the algorithm optimally smoothes the signal in each spectrum according to an improved suite of signal processing algorithms previously reported (the LibSELDI toolbox under development). The neural network part of the algorithm combines those results with the raw signal and a training dataset of expertly called peaks, to call peaks in a test set of spectra with approximately 95% accuracy. The new method was applied to data collected from a study of cervical mucus for the early detection of cervical cancer in HPV infected women. The method shows promise in addressing the ongoing SELDI reproducibility issues.

An Automatic Image Processing System for Glaucoma Screening

PubMed Central

Alodhayb, Sami; Lakshminarayanan, Vasudevan

2017-01-01

Horizontal and vertical cup to disc ratios are the most crucial parameters used clinically to detect glaucoma or monitor its progress and are manually evaluated from retinal fundus images of the optic nerve head. Due to the rarity of the glaucoma experts as well as the increasing in glaucoma's population, an automatically calculated horizontal and vertical cup to disc ratios (HCDR and VCDR, resp.) can be useful for glaucoma screening. We report on two algorithms to calculate the HCDR and VCDR. In the algorithms, level set and inpainting techniques were developed for segmenting the disc, while thresholding using Type-II fuzzy approach was developed for segmenting the cup. The results from the algorithms were verified using the manual markings of images from a dataset of glaucomatous images (retinal fundus images for glaucoma analysis (RIGA dataset)) by six ophthalmologists. The algorithm's accuracy for HCDR and VCDR combined was 74.2%. Only the accuracy of manual markings by one ophthalmologist was higher than the algorithm's accuracy. The algorithm's best agreement was with markings by ophthalmologist number 1 in 230 images (41.8%) of the total tested images. PMID:28947898

Multisensor data fusion for IED threat detection

NASA Astrophysics Data System (ADS)

Mees, Wim; Heremans, Roel

2012-10-01

In this paper we present the multi-sensor registration and fusion algorithms that were developed for a force protection research project in order to detect threats against military patrol vehicles. The fusion is performed at object level, using a hierarchical evidence aggregation approach. It first uses expert domain knowledge about the features used to characterize the detected threats, that is implemented in the form of a fuzzy expert system. The next level consists in fusing intra-sensor and inter-sensor information. Here an ordered weighted averaging operator is used. The object level fusion between candidate threats that are detected asynchronously on a moving vehicle by sensors with different imaging geometries, requires an accurate sensor to world coordinate transformation. This image registration will also be discussed in this paper.

A unified approach for debugging is-a structure and mappings in networked taxonomies

PubMed Central

2013-01-01

Background With the increased use of ontologies and ontology mappings in semantically-enabled applications such as ontology-based search and data integration, the issue of detecting and repairing defects in ontologies and ontology mappings has become increasingly important. These defects can lead to wrong or incomplete results for the applications. Results We propose a unified framework for debugging the is-a structure of and mappings between taxonomies, the most used kind of ontologies. We present theory and algorithms as well as an implemented system RepOSE, that supports a domain expert in detecting and repairing missing and wrong is-a relations and mappings. We also discuss two experiments performed by domain experts: an experiment on the Anatomy ontologies from the Ontology Alignment Evaluation Initiative, and a debugging session for the Swedish National Food Agency. Conclusions Semantically-enabled applications need high quality ontologies and ontology mappings. One key aspect is the detection and removal of defects in the ontologies and ontology mappings. Our system RepOSE provides an environment that supports domain experts to deal with this issue. We have shown the usefulness of the approach in two experiments by detecting and repairing circa 200 and 30 defects, respectively. PMID:23548155

Developing collaborative classifiers using an expert-based model

USGS Publications Warehouse

Mountrakis, G.; Watts, R.; Luo, L.; Wang, Jingyuan

2009-01-01

This paper presents a hierarchical, multi-stage adaptive strategy for image classification. We iteratively apply various classification methods (e.g., decision trees, neural networks), identify regions of parametric and geographic space where accuracy is low, and in these regions, test and apply alternate methods repeating the process until the entire image is classified. Currently, classifiers are evaluated through human input using an expert-based system; therefore, this paper acts as the proof of concept for collaborative classifiers. Because we decompose the problem into smaller, more manageable sub-tasks, our classification exhibits increased flexibility compared to existing methods since classification methods are tailored to the idiosyncrasies of specific regions. A major benefit of our approach is its scalability and collaborative support since selected low-accuracy classifiers can be easily replaced with others without affecting classification accuracy in high accuracy areas. At each stage, we develop spatially explicit accuracy metrics that provide straightforward assessment of results by non-experts and point to areas that need algorithmic improvement or ancillary data. Our approach is demonstrated in the task of detecting impervious surface areas, an important indicator for human-induced alterations to the environment, using a 2001 Landsat scene from Las Vegas, Nevada. ?? 2009 American Society for Photogrammetry and Remote Sensing.

Image analysis approach for development of a decision support system for detection of malaria parasites in thin blood smear images.

PubMed

Prasad, Keerthana; Winter, Jan; Bhat, Udayakrishna M; Acharya, Raviraja V; Prabhu, Gopalakrishna K

2012-08-01

This paper describes development of a decision support system for diagnosis of malaria using color image analysis. A hematologist has to study around 100 to 300 microscopic views of Giemsa-stained thin blood smear images to detect malaria parasites, evaluate the extent of infection and to identify the species of the parasite. The proposed algorithm picks up the suspicious regions and detects the parasites in images of all the views. The subimages representing all these parasites are put together to form a composite image which can be sent over a communication channel to obtain the opinion of a remote expert for accurate diagnosis and treatment. We demonstrate the use of the proposed technique for use as a decision support system by developing an android application which facilitates the communication with a remote expert for the final confirmation on the decision for treatment of malaria. Our algorithm detects around 96% of the parasites with a false positive rate of 20%. The Spearman correlation r was 0.88 with a confidence interval of 0.838 to 0.923, p<0.0001.

Reusable rocket engine turbopump health monitoring system, part 3

NASA Technical Reports Server (NTRS)

Perry, John G.

1989-01-01

Degradation mechanisms and sensor identification/selection resulted in a list of degradation modes and a list of sensors that are utilized in the diagnosis of these degradation modes. The sensor list is divided into primary and secondary indicators of the corresponding degradation modes. The signal conditioning requirements are discussed, describing the methods of producing the Space Shuttle Main Engine (SSME) post-hot-fire test data to be utilized by the Health Monitoring System. Development of the diagnostic logic and algorithms is also presented. The knowledge engineering approach, as utilized, includes the knowledge acquisition effort, characterization of the expert's problem solving strategy, conceptually defining the form of the applicable knowledge base, and rule base, and identifying an appropriate inferencing mechanism for the problem domain. The resulting logic flow graphs detail the diagnosis/prognosis procedure as followed by the experts. The nature and content of required support data and databases is also presented. The distinction between deep and shallow types of knowledge is identified. Computer coding of the Health Monitoring System is shown to follow the logical inferencing of the logic flow graphs/algorithms.

Towards a Computable Data Corpus of Temporal Correlations between Drug Administration and Lab Value Changes

PubMed Central

Newe, Axel; Wimmer, Stefan; Neubert, Antje; Becker, Linda; Prokosch, Hans-Ulrich; Ganslandt, Thomas

2015-01-01

Background The analysis of electronic health records for an automated detection of adverse drug reactions is an approach to solve the problems that arise from traditional methods like spontaneous reporting or manual chart review. Algorithms addressing this task should be modeled on the criteria for a standardized case causality assessment defined by the World Health Organization. One of these criteria is the temporal relationship between drug intake and the occurrence of a reaction or a laboratory test abnormality. Appropriate data that would allow for developing or validating related algorithms is not publicly available, though. Methods In order to provide such data, retrospective routine data of drug administrations and temporally corresponding laboratory observations from a university clinic were extracted, transformed and evaluated by experts in terms of a reasonable time relationship between drug administration and lab value alteration. Result The result is a data corpus of 400 episodes of normalized laboratory parameter values in temporal context with drug administrations. Each episode has been manually classified whether it contains data that might indicate a temporal correlation between the drug administration and the change of the lab value course, whether such a change is not observable or whether a decision between those two options is not possible due to the data. In addition, each episode has been assigned a concordance value which indicates how difficult it is to assess. This is the first open data corpus of a computable ground truth of temporal correlations between drug administration and lab value alterations. Discussion The main purpose of this data corpus is the provision of data for further research and the provision of a ground truth which allows for comparing the outcome of other assessments of this data with the outcome of assessments made by human experts. It can serve as a contribution towards systematic, computerized ADR detection in retrospective data. With this lab value curve data as a basis, algorithms for detecting temporal relationships can be developed, and with the classification made by human experts, these algorithms can immediately be validated. Due to the normalization of the lab value data, it allows for a generic approach rather than for specific or solitary drug/lab value combinations. PMID:26301507

Fuel management optimization using genetic algorithms and expert knowledge

DOE Office of Scientific and Technical Information (OSTI.GOV)

DeChaine, M.D.; Feltus, M.A.

1996-09-01

The CIGARO fuel management optimization code based on genetic algorithms is described and tested. The test problem optimized the core lifetime for a pressurized water reactor with a penalty function constraint on the peak normalized power. A bit-string genotype encoded the loading patterns, and genotype bias was reduced with additional bits. Expert knowledge about fuel management was incorporated into the genetic algorithm. Regional crossover exchanged physically adjacent fuel assemblies and improved the optimization slightly. Biasing the initial population toward a known priority table significantly improved the optimization.

Evaluating performance of a user-trained MR lung tumor autocontouring algorithm in the context of intra- and interobserver variations.

PubMed

Yip, Eugene; Yun, Jihyun; Gabos, Zsolt; Baker, Sarah; Yee, Don; Wachowicz, Keith; Rathee, Satyapal; Fallone, B Gino

2018-01-01

Real-time tracking of lung tumors using magnetic resonance imaging (MRI) has been proposed as a potential strategy to mitigate the ill-effects of breathing motion in radiation therapy. Several autocontouring methods have been evaluated against a "gold standard" of a single human expert user. However, contours drawn by experts have inherent intra- and interobserver variations. In this study, we aim to evaluate our user-trained autocontouring algorithm with manually drawn contours from multiple expert users, and to contextualize the accuracy of these autocontours within intra- and interobserver variations. Six nonsmall cell lung cancer patients were recruited, with institutional ethics approval. Patients were imaged with a clinical 3 T Philips MR scanner using a dynamic 2D balanced SSFP sequence under free breathing. Three radiation oncology experts, each in two separate sessions, contoured 130 dynamic images for each patient. For autocontouring, the first 30 images were used for algorithm training, and the remaining 100 images were autocontoured and evaluated. Autocontours were compared against manual contours in terms of Dice's coefficient (DC) and Hausdorff distances (d H ). Intra- and interobserver variations of the manual contours were also evaluated. When compared with the manual contours of the expert user who trained it, the algorithm generates autocontours whose evaluation metrics (same session: DC = 0.90(0.03), d H  = 3.8(1.6) mm; different session DC = 0.88(0.04), d H  = 4.3(1.5) mm) are similar to or better than intraobserver variations (DC = 0.88(0.04), and d H  = 4.3(1.7) mm) between two sessions. The algorithm's autocontours are also compared to the manual contours from different expert users with evaluation metrics (DC = 0.87(0.04), d H  = 4.8(1.7) mm) similar to interobserver variations (DC = 0.87(0.04), d H  = 4.7(1.6) mm). Our autocontouring algorithm delineates tumor contours (<20 ms per contour), in dynamic MRI of lung, that are comparable to multiple human experts (several seconds per contour), but at a much faster speed. At the same time, the agreement between autocontours and manual contours is comparable to the intra- and interobserver variations. This algorithm may be a key component of the real time tumor tracking workflow for our hybrid Linac-MR device in the future. © 2017 American Association of Physicists in Medicine.

Seizures in the elderly: development and validation of a diagnostic algorithm.

PubMed

Dupont, Sophie; Verny, Marc; Harston, Sandrine; Cartz-Piver, Leslie; Schück, Stéphane; Martin, Jennifer; Puisieux, François; Alecu, Cosmin; Vespignani, Hervé; Marchal, Cécile; Derambure, Philippe

2010-05-01

Seizures are frequent in the elderly, but their diagnosis can be challenging. The objective of this work was to develop and validate an expert-based algorithm for the diagnosis of seizures in elderly people. A multidisciplinary group of neurologists and geriatricians developed a diagnostic algorithm using a combination of selected clinical, electroencephalographical and radiological criteria. The algorithm was validated by multicentre retrospective analysis of data of patients referred for specific symptoms and classified by the experts as epileptic patients or not. The algorithm was applied to all the patients, and the diagnosis provided by the algorithm was compared to the clinical diagnosis of the experts. Twenty-nine clinical, electroencephalographical and radiological criteria were selected for the algorithm. According to criteria combination, seizures were classified in four levels of diagnosis: certain, highly probable, possible or improbable. To validate the algorithm, the medical records of 269 elderly patients were analyzed (138 with epileptic seizures, 131 with non-epileptic manifestations). Patients were mainly referred for a transient focal deficit (40%), confusion (38%), unconsciousness (27%). The algorithm best classified certain and probable seizures versus possible and improbable seizures, with 86.2% sensitivity and 67.2% specificity. Using logistical regression, 2 simplified models were developed, the first with 13 criteria (Se 85.5%, Sp 90.1%), and the second with 7 criteria only (Se 84.8%, Sp 88.6%). In conclusion, the present study validated the use of a revised diagnostic algorithm to help diagnosis epileptic seizures in the elderly. A prospective study is planned to further validate this algorithm. Copyright 2010 Elsevier B.V. All rights reserved.

Developing a new national approach to surveillance for ventilator-associated events: executive summary.

PubMed

Magill, Shelley S; Klompas, Michael; Balk, Robert; Burns, Suzanne M; Deutschman, Clifford S; Diekema, Daniel; Fridkin, Scott; Greene, Linda; Guh, Alice; Gutterman, David; Hammer, Beth; Henderson, David; Hess, Dean R; Hill, Nicholas S; Horan, Teresa; Kollef, Marin; Levy, Mitchell; Septimus, Edward; VanAntwerpen, Carole; Wright, Don; Lipsett, Pamela

2013-11-01

In September 2011, the Centers for Disease Control and Prevention (CDC) convened a Ventilator-Associated Pneumonia (VAP) Surveillance Definition Working Group to organize a formal process for leaders and experts of key stakeholder organizations to discuss the challenges of VAP surveillance definitions and to propose new approaches to VAP surveillance in adult patients (Table 1). The charges to the Working Group were to (1) critically review a draft, streamlined VAP surveillance definition developed for use in adult patients; (2) suggest modifications to enhance the reliability and credibility of the surveillance definition within the critical care and infection prevention communities; and (3) propose a final adult surveillance definition algorithm to be implemented in the CDC's National Healthcare Safety Network (NHSN), taking into consideration the potential future use of the definition algorithm in public reporting, interfacility comparisons, and pay-for-reporting and pay-for-performance programs. Published by Mosby, Inc.

Analysis of methods of processing of expert information by optimization of administrative decisions

NASA Astrophysics Data System (ADS)

Churakov, D. Y.; Tsarkova, E. G.; Marchenko, N. D.; Grechishnikov, E. V.

2018-03-01

In the real operation the measure definition methodology in case of expert estimation of quality and reliability of application-oriented software products is offered. In operation methods of aggregation of expert estimates on the example of a collective choice of an instrumental control projects in case of software development of a special purpose for needs of institutions are described. Results of operation of dialogue decision making support system are given an algorithm of the decision of the task of a choice on the basis of a method of the analysis of hierarchies and also. The developed algorithm can be applied by development of expert systems to the solution of a wide class of the tasks anyway connected to a multicriteria choice.

A machine learning-based framework to identify type 2 diabetes through electronic health records

PubMed Central

Zheng, Tao; Xie, Wei; Xu, Liling; He, Xiaoying; Zhang, Ya; You, Mingrong; Yang, Gong; Chen, You

2016-01-01

Objective To discover diverse genotype-phenotype associations affiliated with Type 2 Diabetes Mellitus (T2DM) via genome-wide association study (GWAS) and phenome-wide association study (PheWAS), more cases (T2DM subjects) and controls (subjects without T2DM) are required to be identified (e.g., via Electronic Health Records (EHR)). However, existing expert based identification algorithms often suffer in a low recall rate and could miss a large number of valuable samples under conservative filtering standards. The goal of this work is to develop a semi-automated framework based on machine learning as a pilot study to liberalize filtering criteria to improve recall rate with a keeping of low false positive rate. Materials and methods We propose a data informed framework for identifying subjects with and without T2DM from EHR via feature engineering and machine learning. We evaluate and contrast the identification performance of widely-used machine learning models within our framework, including k-Nearest-Neighbors, Naïve Bayes, Decision Tree, Random Forest, Support Vector Machine and Logistic Regression. Our framework was conducted on 300 patient samples (161 cases, 60 controls and 79 unconfirmed subjects), randomly selected from 23,281 diabetes related cohort retrieved from a regional distributed EHR repository ranging from 2012 to 2014. Results We apply top-performing machine learning algorithms on the engineered features. We benchmark and contrast the accuracy, precision, AUC, sensitivity and specificity of classification models against the state-of-the-art expert algorithm for identification of T2DM subjects. Our results indicate that the framework achieved high identification performances (∼0.98 in average AUC), which are much higher than the state-of-the-art algorithm (0.71 in AUC). Discussion Expert algorithm-based identification of T2DM subjects from EHR is often hampered by the high missing rates due to their conservative selection criteria. Our framework leverages machine learning and feature engineering to loosen such selection criteria to achieve a high identification rate of cases and controls. Conclusions Our proposed framework demonstrates a more accurate and efficient approach for identifying subjects with and without T2DM from EHR. PMID:27919371

A machine learning-based framework to identify type 2 diabetes through electronic health records.

PubMed

Zheng, Tao; Xie, Wei; Xu, Liling; He, Xiaoying; Zhang, Ya; You, Mingrong; Yang, Gong; Chen, You

2017-01-01

To discover diverse genotype-phenotype associations affiliated with Type 2 Diabetes Mellitus (T2DM) via genome-wide association study (GWAS) and phenome-wide association study (PheWAS), more cases (T2DM subjects) and controls (subjects without T2DM) are required to be identified (e.g., via Electronic Health Records (EHR)). However, existing expert based identification algorithms often suffer in a low recall rate and could miss a large number of valuable samples under conservative filtering standards. The goal of this work is to develop a semi-automated framework based on machine learning as a pilot study to liberalize filtering criteria to improve recall rate with a keeping of low false positive rate. We propose a data informed framework for identifying subjects with and without T2DM from EHR via feature engineering and machine learning. We evaluate and contrast the identification performance of widely-used machine learning models within our framework, including k-Nearest-Neighbors, Naïve Bayes, Decision Tree, Random Forest, Support Vector Machine and Logistic Regression. Our framework was conducted on 300 patient samples (161 cases, 60 controls and 79 unconfirmed subjects), randomly selected from 23,281 diabetes related cohort retrieved from a regional distributed EHR repository ranging from 2012 to 2014. We apply top-performing machine learning algorithms on the engineered features. We benchmark and contrast the accuracy, precision, AUC, sensitivity and specificity of classification models against the state-of-the-art expert algorithm for identification of T2DM subjects. Our results indicate that the framework achieved high identification performances (∼0.98 in average AUC), which are much higher than the state-of-the-art algorithm (0.71 in AUC). Expert algorithm-based identification of T2DM subjects from EHR is often hampered by the high missing rates due to their conservative selection criteria. Our framework leverages machine learning and feature engineering to loosen such selection criteria to achieve a high identification rate of cases and controls. Our proposed framework demonstrates a more accurate and efficient approach for identifying subjects with and without T2DM from EHR. Copyright © 2016 Elsevier Ireland Ltd. All rights reserved.

Mutually Augmented Cognition

NASA Astrophysics Data System (ADS)

Friesdorf, Florian; Pangercic, Dejan; Bubb, Heiner; Beetz, Michael

In mac, an ergonomic dialog-system and algorithms will be developed that enable human experts and companions to be integrated into knowledge gathering and decision making processes of highly complex cognitive systems (e.g. Assistive Household as manifested further in the paper). In this event we propose to join algorithms and methodologies coming from Ergonomics and Artificial Intelligence that: a) make cognitive systems more congenial for non-expert humans, b) facilitate their comprehension by utilizing a high-level expandable control code for human experts and c) augment representation of such cognitive system into “deep representation” obtained through an interaction with human companions.

Side-locked headaches: an algorithm-based approach.

PubMed

Prakash, Sanjay; Rathore, Chaturbhuj

2016-12-01

The differential diagnosis of strictly unilateral hemicranial pain includes a large number of primary and secondary headaches and cranial neuropathies. It may arise from both intracranial and extracranial structures such as cranium, neck, vessels, eyes, ears, nose, sinuses, teeth, mouth, and the other facial or cervical structure. Available data suggest that about two-third patients with side-locked headache visiting neurology or headache clinics have primary headaches. Other one-third will have either secondary headaches or neuralgias. Many of these hemicranial pain syndromes have overlapping presentations. Primary headache disorders may spread to involve the face and / or neck. Even various intracranial and extracranial pathologies may have similar overlapping presentations. Patients may present to a variety of clinicians, including headache experts, dentists, otolaryngologists, ophthalmologist, psychiatrists, and physiotherapists. Unfortunately, there is not uniform approach for such patients and diagnostic ambiguity is frequently encountered in clinical practice.Herein, we review the differential diagnoses of side-locked headaches and provide an algorithm based approach for patients presenting with side-locked headaches. Side-locked headache is itself a red flag. So, the first priority should be to rule out secondary headaches. A comprehensive history and thorough examinations will help one to formulate an algorithm to rule out or confirm secondary side-locked headaches. The diagnoses of most secondary side-locked headaches are largely investigations dependent. Therefore, each suspected secondary headache should be subjected for appropriate investigations or referral. The diagnostic approach of primary side-locked headache starts once one rule out all the possible secondary headaches. We have discussed an algorithmic approach for both secondary and primary side-locked headaches.

New approach to gallbladder ultrasonic images analysis and lesions recognition.

PubMed

Bodzioch, Sławomir; Ogiela, Marek R

2009-03-01

This paper presents a new approach to gallbladder ultrasonic image processing and analysis towards detection of disease symptoms on processed images. First, in this paper, there is presented a new method of filtering gallbladder contours from USG images. A major stage in this filtration is to segment and section off areas occupied by the said organ. In most cases this procedure is based on filtration that plays a key role in the process of diagnosing pathological changes. Unfortunately ultrasound images present among the most troublesome methods of analysis owing to the echogenic inconsistency of structures under observation. This paper provides for an inventive algorithm for the holistic extraction of gallbladder image contours. The algorithm is based on rank filtration, as well as on the analysis of histogram sections on tested organs. The second part concerns detecting lesion symptoms of the gallbladder. Automating a process of diagnosis always comes down to developing algorithms used to analyze the object of such diagnosis and verify the occurrence of symptoms related to given affection. Usually the final stage is to make a diagnosis based on the detected symptoms. This last stage can be carried out through either dedicated expert systems or more classic pattern analysis approach like using rules to determine illness basing on detected symptoms. This paper discusses the pattern analysis algorithms for gallbladder image interpretation towards classification of the most frequent illness symptoms of this organ.

Applied algorithm in the liner inspection of solid rocket motors

NASA Astrophysics Data System (ADS)

Hoffmann, Luiz Felipe Simões; Bizarria, Francisco Carlos Parquet; Bizarria, José Walter Parquet

2018-03-01

In rocket motors, the bonding between the solid propellant and thermal insulation is accomplished by a thin adhesive layer, known as liner. The liner application method involves a complex sequence of tasks, which includes in its final stage, the surface integrity inspection. Nowadays in Brazil, an expert carries out a thorough visual inspection to detect defects on the liner surface that may compromise the propellant interface bonding. Therefore, this paper proposes an algorithm that uses the photometric stereo technique and the K-nearest neighbor (KNN) classifier to assist the expert in the surface inspection. Photometric stereo allows the surface information recovery of the test images, while the KNN method enables image pixels classification into two classes: non-defect and defect. Tests performed on a computer vision based prototype validate the algorithm. The positive results suggest that the algorithm is feasible and when implemented in a real scenario, will be able to help the expert in detecting defective areas on the liner surface.

Correlation between automatic detection of malaria on thin film and experts' parasitaemia scores

NASA Astrophysics Data System (ADS)

Sunarko, Budi; Williams, Simon; Prescott, William R.; Byker, Scott M.; Bottema, Murk J.

2017-03-01

An algorithm was developed to diagnose the presence of malaria and to estimate the depth of infection by automatically counting individual normal and infected erythrocytes in images of thin blood smears. During the training stage, the parameters of the algorithm were optimized to maximize correlation with estimates of parasitaemia from expert human observers. The correlation was tested on a set of 1590 images from seven thin film blood smears. The correlation between the results from the algorithm and expert human readers was r = 0.836. Results indicate that reliable estimates of parasitaemia may be achieved by computational image analysis methods applied to images of thin film smears. Meanwhile, compared to biological experiments, the algorithm fitted well the three high parasitaemia slides and a mid-level parasitaemia slide, and overestimated the three low parasitaemia slides. To improve the parasitaemia estimation, the sources of the overestimation were identified. Emphasis is laid on the importance of further research in order to identify parasites independently of their erythrocyte hosts

Diagnosis and sensor validation through knowledge of structure and function

NASA Technical Reports Server (NTRS)

Scarl, Ethan A.; Jamieson, John R.; Delaune, Carl I.

1987-01-01

The liquid oxygen expert system 'LES' is proposed as the first capable of diagnostic reasoning from sensor data, using model-based knowledge of structure and function to find the expected state of all system objects, including sensors. The approach is generally algorithmic rather than heuristic, and represents uncertainties as sets of possibilities. Functional relationships are inverted to determine hypothetical values for potentially faulty objects, and may include conditional functions not normally considered to have inverses.

Extracting BI-RADS Features from Portuguese Clinical Texts

PubMed Central

Nassif, Houssam; Cunha, Filipe; Moreira, Inês C.; Cruz-Correia, Ricardo; Sousa, Eliana; Page, David; Burnside, Elizabeth; Dutra, Inês

2013-01-01

In this work we build the first BI-RADS parser for Portuguese free texts, modeled after existing approaches to extract BI-RADS features from English medical records. Our concept finder uses a semantic grammar based on the BIRADS lexicon and on iterative transferred expert knowledge. We compare the performance of our algorithm to manual annotation by a specialist in mammography. Our results show that our parser’s performance is comparable to the manual method. PMID:23797461

SU-C-207B-02: Maximal Noise Reduction Filter with Anatomical Structures Preservation

DOE Office of Scientific and Technical Information (OSTI.GOV)

Maitree, R; Guzman, G; Chundury, A

Purpose: All medical images contain noise, which can result in an undesirable appearance and can reduce the visibility of anatomical details. There are varieties of techniques utilized to reduce noise such as increasing the image acquisition time and using post-processing noise reduction algorithms. However, these techniques are increasing the imaging time and cost or reducing tissue contrast and effective spatial resolution which are useful diagnosis information. The three main focuses in this study are: 1) to develop a novel approach that can adaptively and maximally reduce noise while preserving valuable details of anatomical structures, 2) to evaluate the effectiveness ofmore » available noise reduction algorithms in comparison to the proposed algorithm, and 3) to demonstrate that the proposed noise reduction approach can be used clinically. Methods: To achieve a maximal noise reduction without destroying the anatomical details, the proposed approach automatically estimated the local image noise strength levels and detected the anatomical structures, i.e. tissue boundaries. Such information was used to adaptively adjust strength of the noise reduction filter. The proposed algorithm was tested on 34 repeating swine head datasets and 54 patients MRI and CT images. The performance was quantitatively evaluated by image quality metrics and manually validated for clinical usages by two radiation oncologists and one radiologist. Results: Qualitative measurements on repeated swine head images demonstrated that the proposed algorithm efficiently removed noise while preserving the structures and tissues boundaries. In comparisons, the proposed algorithm obtained competitive noise reduction performance and outperformed other filters in preserving anatomical structures. Assessments from the manual validation indicate that the proposed noise reduction algorithm is quite adequate for some clinical usages. Conclusion: According to both clinical evaluation (human expert ranking) and qualitative assessment, the proposed approach has superior noise reduction and anatomical structures preservation capabilities over existing noise removal methods. Senior Author Dr. Deshan Yang received research funding form ViewRay and Varian.« less

Topological mappings of video and audio data.

PubMed

Fyfe, Colin; Barbakh, Wesam; Ooi, Wei Chuan; Ko, Hanseok

2008-12-01

We review a new form of self-organizing map which is based on a nonlinear projection of latent points into data space, identical to that performed in the Generative Topographic Mapping (GTM).(1) But whereas the GTM is an extension of a mixture of experts, this model is an extension of a product of experts.(2) We show visualisation and clustering results on a data set composed of video data of lips uttering 5 Korean vowels. Finally we note that we may dispense with the probabilistic underpinnings of the product of experts and derive the same algorithm as a minimisation of mean squared error between the prototypes and the data. This leads us to suggest a new algorithm which incorporates local and global information in the clustering. Both ot the new algorithms achieve better results than the standard Self-Organizing Map.

Automated Diagnosis of Plus Disease in Retinopathy of Prematurity Using Deep Convolutional Neural Networks.

PubMed

Brown, James M; Campbell, J Peter; Beers, Andrew; Chang, Ken; Ostmo, Susan; Chan, R V Paul; Dy, Jennifer; Erdogmus, Deniz; Ioannidis, Stratis; Kalpathy-Cramer, Jayashree; Chiang, Michael F

2018-05-02

Retinopathy of prematurity (ROP) is a leading cause of childhood blindness worldwide. The decision to treat is primarily based on the presence of plus disease, defined as dilation and tortuosity of retinal vessels. However, clinical diagnosis of plus disease is highly subjective and variable. To implement and validate an algorithm based on deep learning to automatically diagnose plus disease from retinal photographs. A deep convolutional neural network was trained using a data set of 5511 retinal photographs. Each image was previously assigned a reference standard diagnosis (RSD) based on consensus of image grading by 3 experts and clinical diagnosis by 1 expert (ie, normal, pre-plus disease, or plus disease). The algorithm was evaluated by 5-fold cross-validation and tested on an independent set of 100 images. Images were collected from 8 academic institutions participating in the Imaging and Informatics in ROP (i-ROP) cohort study. The deep learning algorithm was tested against 8 ROP experts, each of whom had more than 10 years of clinical experience and more than 5 peer-reviewed publications about ROP. Data were collected from July 2011 to December 2016. Data were analyzed from December 2016 to September 2017. A deep learning algorithm trained on retinal photographs. Receiver operating characteristic analysis was performed to evaluate performance of the algorithm against the RSD. Quadratic-weighted κ coefficients were calculated for ternary classification (ie, normal, pre-plus disease, and plus disease) to measure agreement with the RSD and 8 independent experts. Of the 5511 included retinal photographs, 4535 (82.3%) were graded as normal, 805 (14.6%) as pre-plus disease, and 172 (3.1%) as plus disease, based on the RSD. Mean (SD) area under the receiver operating characteristic curve statistics were 0.94 (0.01) for the diagnosis of normal (vs pre-plus disease or plus disease) and 0.98 (0.01) for the diagnosis of plus disease (vs normal or pre-plus disease). For diagnosis of plus disease in an independent test set of 100 retinal images, the algorithm achieved a sensitivity of 93% with 94% specificity. For detection of pre-plus disease or worse, the sensitivity and specificity were 100% and 94%, respectively. On the same test set, the algorithm achieved a quadratic-weighted κ coefficient of 0.92 compared with the RSD, outperforming 6 of 8 ROP experts. This fully automated algorithm diagnosed plus disease in ROP with comparable or better accuracy than human experts. This has potential applications in disease detection, monitoring, and prognosis in infants at risk of ROP.

Digital imaging biomarkers feed machine learning for melanoma screening.

PubMed

Gareau, Daniel S; Correa da Rosa, Joel; Yagerman, Sarah; Carucci, John A; Gulati, Nicholas; Hueto, Ferran; DeFazio, Jennifer L; Suárez-Fariñas, Mayte; Marghoob, Ashfaq; Krueger, James G

2017-07-01

We developed an automated approach for generating quantitative image analysis metrics (imaging biomarkers) that are then analysed with a set of 13 machine learning algorithms to generate an overall risk score that is called a Q-score. These methods were applied to a set of 120 "difficult" dermoscopy images of dysplastic nevi and melanomas that were subsequently excised/classified. This approach yielded 98% sensitivity and 36% specificity for melanoma detection, approaching sensitivity/specificity of expert lesion evaluation. Importantly, we found strong spectral dependence of many imaging biomarkers in blue or red colour channels, suggesting the need to optimize spectral evaluation of pigmented lesions. © 2016 The Authors. Experimental Dermatology Published by John Wiley & Sons Ltd.

Data-driven approach for creating synthetic electronic medical records.

PubMed

Buczak, Anna L; Babin, Steven; Moniz, Linda

2010-10-14

New algorithms for disease outbreak detection are being developed to take advantage of full electronic medical records (EMRs) that contain a wealth of patient information. However, due to privacy concerns, even anonymized EMRs cannot be shared among researchers, resulting in great difficulty in comparing the effectiveness of these algorithms. To bridge the gap between novel bio-surveillance algorithms operating on full EMRs and the lack of non-identifiable EMR data, a method for generating complete and synthetic EMRs was developed. This paper describes a novel methodology for generating complete synthetic EMRs both for an outbreak illness of interest (tularemia) and for background records. The method developed has three major steps: 1) synthetic patient identity and basic information generation; 2) identification of care patterns that the synthetic patients would receive based on the information present in real EMR data for similar health problems; 3) adaptation of these care patterns to the synthetic patient population. We generated EMRs, including visit records, clinical activity, laboratory orders/results and radiology orders/results for 203 synthetic tularemia outbreak patients. Validation of the records by a medical expert revealed problems in 19% of the records; these were subsequently corrected. We also generated background EMRs for over 3000 patients in the 4-11 yr age group. Validation of those records by a medical expert revealed problems in fewer than 3% of these background patient EMRs and the errors were subsequently rectified. A data-driven method was developed for generating fully synthetic EMRs. The method is general and can be applied to any data set that has similar data elements (such as laboratory and radiology orders and results, clinical activity, prescription orders). The pilot synthetic outbreak records were for tularemia but our approach may be adapted to other infectious diseases. The pilot synthetic background records were in the 4-11 year old age group. The adaptations that must be made to the algorithms to produce synthetic background EMRs for other age groups are indicated.

Feature Selection for Classification of Polar Regions Using a Fuzzy Expert System

NASA Technical Reports Server (NTRS)

Penaloza, Mauel A.; Welch, Ronald M.

1996-01-01

Labeling, feature selection, and the choice of classifier are critical elements for classification of scenes and for image understanding. This study examines several methods for feature selection in polar regions, including the list, of a fuzzy logic-based expert system for further refinement of a set of selected features. Six Advanced Very High Resolution Radiometer (AVHRR) Local Area Coverage (LAC) arctic scenes are classified into nine classes: water, snow / ice, ice cloud, land, thin stratus, stratus over water, cumulus over water, textured snow over water, and snow-covered mountains. Sixty-seven spectral and textural features are computed and analyzed by the feature selection algorithms. The divergence, histogram analysis, and discriminant analysis approaches are intercompared for their effectiveness in feature selection. The fuzzy expert system method is used not only to determine the effectiveness of each approach in classifying polar scenes, but also to further reduce the features into a more optimal set. For each selection method,features are ranked from best to worst, and the best half of the features are selected. Then, rules using these selected features are defined. The results of running the fuzzy expert system with these rules show that the divergence method produces the best set features, not only does it produce the highest classification accuracy, but also it has the lowest computation requirements. A reduction of the set of features produced by the divergence method using the fuzzy expert system results in an overall classification accuracy of over 95 %. However, this increase of accuracy has a high computation cost.

Multivariate statistical model for 3D image segmentation with application to medical images.

PubMed

John, Nigel M; Kabuka, Mansur R; Ibrahim, Mohamed O

2003-12-01

In this article we describe a statistical model that was developed to segment brain magnetic resonance images. The statistical segmentation algorithm was applied after a pre-processing stage involving the use of a 3D anisotropic filter along with histogram equalization techniques. The segmentation algorithm makes use of prior knowledge and a probability-based multivariate model designed to semi-automate the process of segmentation. The algorithm was applied to images obtained from the Center for Morphometric Analysis at Massachusetts General Hospital as part of the Internet Brain Segmentation Repository (IBSR). The developed algorithm showed improved accuracy over the k-means, adaptive Maximum Apriori Probability (MAP), biased MAP, and other algorithms. Experimental results showing the segmentation and the results of comparisons with other algorithms are provided. Results are based on an overlap criterion against expertly segmented images from the IBSR. The algorithm produced average results of approximately 80% overlap with the expertly segmented images (compared with 85% for manual segmentation and 55% for other algorithms).

Meta-knowledge for charity: Learnings from a non-market-based prediction platform for climate scientists

NASA Astrophysics Data System (ADS)

Kriss, P.

2016-12-01

We present the results from Vision Prize, an online platform for capturing expert opinion on climate risks and solutions. Expert panelists provided their own opinions, and also predicted the views of their scientific colleagues. This approach (Prelec, 2004 Nature) gives new insight into the level of scientific consensus on various issues, which in some cases may be just as important as knowing the majority view. Questions ranged from causal attributions of past events to predictions about future risks and beliefs about possible solutions. Across all topics, we find that our expert participants agree with each other more than they think they do, often in surprising ways. Data collection for the Vision Prize project was run in collaboration with Environmental Research Web (IOP). Panelists were pre-screened to ensure they have relevant expertise and independently collected h-scores provide an approximate measure of research impact of those giving each response. Charity prizes were awarded for exceptional meta-knowledge, according to Prelec's algorithm.

Real Time Data System (RTDS)

NASA Technical Reports Server (NTRS)

Muratore, John F.

1991-01-01

Lessons learned from operational real time expert systems are examined. The basic system architecture is discussed. An expert system is any software that performs tasks to a standard that would normally require a human expert. An expert system implies knowledge contained in data rather than code. And an expert system implies the use of heuristics as well as algorithms. The 15 top lessons learned by the operation of a real time data system are presented.

High throughput light absorber discovery, Part 1: An algorithm for automated tauc analysis

DOE PAGES

Suram, Santosh K.; Newhouse, Paul F.; Gregoire, John M.

2016-09-23

High-throughput experimentation provides efficient mapping of composition-property relationships, and its implementation for the discovery of optical materials enables advancements in solar energy and other technologies. In a high throughput pipeline, automated data processing algorithms are often required to match experimental throughput, and we present an automated Tauc analysis algorithm for estimating band gap energies from optical spectroscopy data. The algorithm mimics the judgment of an expert scientist, which is demonstrated through its application to a variety of high throughput spectroscopy data, including the identification of indirect or direct band gaps in Fe 2O 3, Cu 2V 2O 7, and BiVOmore » 4. Here, the applicability of the algorithm to estimate a range of band gap energies for various materials is demonstrated by a comparison of direct-allowed band gaps estimated by expert scientists and by automated algorithm for 60 optical spectra.« less

Customization of biomedical terminologies.

PubMed

Homo, Julien; Dupuch, Laëtitia; Benbrahim, Allel; Grabar, Natalia; Dupuch, Marie

2012-01-01

Within the biomedical area over one hundred terminologies exist and are merged in the Unified Medical Language System Metathesaurus, which gives over 1 million concepts. When such huge terminological resources are available, the users must deal with them and specifically they must deal with irrelevant parts of these terminologies. We propose to exploit seed terms and semantic distance algorithms in order to customize the terminologies and to limit within them a semantically homogeneous space. An evaluation performed by a medical expert indicates that the proposed approach is relevant for the customization of terminologies and that the extracted terms are mostly relevant to the seeds. It also indicates that different algorithms provide with similar or identical results within a given terminology. The difference is due to the terminologies exploited. A special attention must be paid to the definition of optimal association between the semantic similarity algorithms and the thresholds specific to a given terminology.

Interpretable Categorization of Heterogeneous Time Series Data

NASA Technical Reports Server (NTRS)

Lee, Ritchie; Kochenderfer, Mykel J.; Mengshoel, Ole J.; Silbermann, Joshua

2017-01-01

We analyze data from simulated aircraft encounters to validate and inform the development of a prototype aircraft collision avoidance system. The high-dimensional and heterogeneous time series dataset is analyzed to discover properties of near mid-air collisions (NMACs) and categorize the NMAC encounters. Domain experts use these properties to better organize and understand NMAC occurrences. Existing solutions either are not capable of handling high-dimensional and heterogeneous time series datasets or do not provide explanations that are interpretable by a domain expert. The latter is critical to the acceptance and deployment of safety-critical systems. To address this gap, we propose grammar-based decision trees along with a learning algorithm. Our approach extends decision trees with a grammar framework for classifying heterogeneous time series data. A context-free grammar is used to derive decision expressions that are interpretable, application-specific, and support heterogeneous data types. In addition to classification, we show how grammar-based decision trees can also be used for categorization, which is a combination of clustering and generating interpretable explanations for each cluster. We apply grammar-based decision trees to a simulated aircraft encounter dataset and evaluate the performance of four variants of our learning algorithm. The best algorithm is used to analyze and categorize near mid-air collisions in the aircraft encounter dataset. We describe each discovered category in detail and discuss its relevance to aircraft collision avoidance.

Fetal heart rate deceleration detection using a discrete cosine transform implementation of singular spectrum analysis.

PubMed

Warrick, P A; Precup, D; Hamilton, E F; Kearney, R E

2007-01-01

To develop a singular-spectrum analysis (SSA) based change-point detection algorithm applicable to fetal heart rate (FHR) monitoring to improve the detection of deceleration events. We present a method for decomposing a signal into near-orthogonal components via the discrete cosine transform (DCT) and apply this in a novel online manner to change-point detection based on SSA. The SSA technique forms models of the underlying signal that can be compared over time; models that are sufficiently different indicate signal change points. To adapt the algorithm to deceleration detection where many successive similar change events can occur, we modify the standard SSA algorithm to hold the reference model constant under such conditions, an approach that we term "base-hold SSA". The algorithm is applied to a database of 15 FHR tracings that have been preprocessed to locate candidate decelerations and is compared to the markings of an expert obstetrician. Of the 528 true and 1285 false decelerations presented to the algorithm, the base-hold approach improved on standard SSA, reducing the number of missed decelerations from 64 to 49 (21.9%) while maintaining the same reduction in false-positives (278). The standard SSA assumption that changes are infrequent does not apply to FHR analysis where decelerations can occur successively and in close proximity; our base-hold SSA modification improves detection of these types of event series.

Comparison of manual and automatic segmentation methods for brain structures in the presence of space-occupying lesions: a multi-expert study

PubMed Central

Deeley, M A; Chen, A; Datteri, R; Noble, J; Cmelak, A; Donnelly, E; Malcolm, A; Moretti, L; Jaboin, J; Niermann, K; Yang, Eddy S; Yu, David S; Yei, F; Koyama, T; Ding, G X; Dawant, B M

2011-01-01

The purpose of this work was to characterize expert variation in segmentation of intracranial structures pertinent to radiation therapy, and to assess a registration-driven atlas-based segmentation algorithm in that context. Eight experts were recruited to segment the brainstem, optic chiasm, optic nerves, and eyes, of 20 patients who underwent therapy for large space-occupying tumors. Performance variability was assessed through three geometric measures: volume, Dice similarity coefficient, and Euclidean distance. In addition, two simulated ground truth segmentations were calculated via the simultaneous truth and performance level estimation (STAPLE) algorithm and a novel application of probability maps. The experts and automatic system were found to generate structures of similar volume, though the experts exhibited higher variation with respect to tubular structures. No difference was found between the mean Dice coefficient (DSC) of the automatic and expert delineations as a group at a 5% significance level over all cases and organs. The larger structures of the brainstem and eyes exhibited mean DSC of approximately 0.8–0.9, whereas the tubular chiasm and nerves were lower, approximately 0.4–0.5. Similarly low DSC have been reported previously without the context of several experts and patient volumes. This study, however, provides evidence that experts are similarly challenged. The average maximum distances (maximum inside, maximum outside) from a simulated ground truth ranged from (−4.3, +5.4) mm for the automatic system to (−3.9, +7.5) mm for the experts considered as a group. Over all the structures in a rank of true positive rates at a 2 mm threshold from the simulated ground truth, the automatic system ranked second of the nine raters. This work underscores the need for large scale studies utilizing statistically robust numbers of patients and experts in evaluating quality of automatic algorithms. PMID:21725140

Comparison of manual and automatic segmentation methods for brain structures in the presence of space-occupying lesions: a multi-expert study

NASA Astrophysics Data System (ADS)

Deeley, M. A.; Chen, A.; Datteri, R.; Noble, J. H.; Cmelak, A. J.; Donnelly, E. F.; Malcolm, A. W.; Moretti, L.; Jaboin, J.; Niermann, K.; Yang, Eddy S.; Yu, David S.; Yei, F.; Koyama, T.; Ding, G. X.; Dawant, B. M.

2011-07-01

The purpose of this work was to characterize expert variation in segmentation of intracranial structures pertinent to radiation therapy, and to assess a registration-driven atlas-based segmentation algorithm in that context. Eight experts were recruited to segment the brainstem, optic chiasm, optic nerves, and eyes, of 20 patients who underwent therapy for large space-occupying tumors. Performance variability was assessed through three geometric measures: volume, Dice similarity coefficient, and Euclidean distance. In addition, two simulated ground truth segmentations were calculated via the simultaneous truth and performance level estimation algorithm and a novel application of probability maps. The experts and automatic system were found to generate structures of similar volume, though the experts exhibited higher variation with respect to tubular structures. No difference was found between the mean Dice similarity coefficient (DSC) of the automatic and expert delineations as a group at a 5% significance level over all cases and organs. The larger structures of the brainstem and eyes exhibited mean DSC of approximately 0.8-0.9, whereas the tubular chiasm and nerves were lower, approximately 0.4-0.5. Similarly low DSCs have been reported previously without the context of several experts and patient volumes. This study, however, provides evidence that experts are similarly challenged. The average maximum distances (maximum inside, maximum outside) from a simulated ground truth ranged from (-4.3, +5.4) mm for the automatic system to (-3.9, +7.5) mm for the experts considered as a group. Over all the structures in a rank of true positive rates at a 2 mm threshold from the simulated ground truth, the automatic system ranked second of the nine raters. This work underscores the need for large scale studies utilizing statistically robust numbers of patients and experts in evaluating quality of automatic algorithms.

An interactive ontology-driven information system for simulating background radiation and generating scenarios for testing special nuclear materials detection algorithms

DOE PAGES

Sorokine, Alexandre; Schlicher, Bob G.; Ward, Richard C.; ...

2015-05-22

This paper describes an original approach to generating scenarios for the purpose of testing the algorithms used to detect special nuclear materials (SNM) that incorporates the use of ontologies. Separating the signal of SNM from the background requires sophisticated algorithms. To assist in developing such algorithms, there is a need for scenarios that capture a very wide range of variables affecting the detection process, depending on the type of detector being used. To provide such a cpability, we developed an ontology-driven information system (ODIS) for generating scenarios that can be used in creating scenarios for testing of algorithms for SNMmore » detection. The ontology-driven scenario generator (ODSG) is an ODIS based on information supplied by subject matter experts and other documentation. The details of the creation of the ontology, the development of the ontology-driven information system, and the design of the web user interface (UI) are presented along with specific examples of scenarios generated using the ODSG. We demonstrate that the paradigm behind the ODSG is capable of addressing the problem of semantic complexity at both the user and developer levels. Compared to traditional approaches, an ODIS provides benefits such as faithful representation of the users' domain conceptualization, simplified management of very large and semantically diverse datasets, and the ability to handle frequent changes to the application and the UI. Furthermore, the approach makes possible the generation of a much larger number of specific scenarios based on limited user-supplied information« less

Assessing system reliability and allocating resources: a bayesian approach that integrates multi-level data

DOE Office of Scientific and Technical Information (OSTI.GOV)

Graves, Todd L; Hamada, Michael S

2008-01-01

Good estimates of the reliability of a system make use of test data and expert knowledge at all available levels. Furthermore, by integrating all these information sources, one can determine how best to allocate scarce testing resources to reduce uncertainty. Both of these goals are facilitated by modern Bayesian computational methods. We apply these tools to examples that were previously solvable only through the use of ingenious approximations, and use genetic algorithms to guide resource allocation.

ANALYTiC: An Active Learning System for Trajectory Classification.

PubMed

Soares Junior, Amilcar; Renso, Chiara; Matwin, Stan

2017-01-01

The increasing availability and use of positioning devices has resulted in large volumes of trajectory data. However, semantic annotations for such data are typically added by domain experts, which is a time-consuming task. Machine-learning algorithms can help infer semantic annotations from trajectory data by learning from sets of labeled data. Specifically, active learning approaches can minimize the set of trajectories to be annotated while preserving good performance measures. The ANALYTiC web-based interactive tool visually guides users through this annotation process.

Automated electric power management and control for Space Station Freedom

NASA Technical Reports Server (NTRS)

Dolce, James L.; Mellor, Pamela A.; Kish, James A.

1990-01-01

A comprehensive automation design is being developed for Space Station Freedom's electric power system. It strives to increase station productivity by applying expert systems and conventional algorithms to automate power system operation. An integrated approach to the power system command and control problem is defined and used to direct technology development in: diagnosis, security monitoring and analysis, battery management, and cooperative problem-solving for resource allocation. The prototype automated power system is developed using simulations and test-beds.

Potential slab avalanche release area identification from estimated winter terrain: a multi-scale, fuzzy logic approach

NASA Astrophysics Data System (ADS)

Veitinger, Jochen; Purves, Ross Stuart; Sovilla, Betty

2016-10-01

Avalanche hazard assessment requires a very precise estimation of the release area, which still depends, to a large extent, on expert judgement of avalanche specialists. Therefore, a new algorithm for automated identification of potential avalanche release areas was developed. It overcomes some of the limitations of previous tools, which are currently not often applied in hazard mitigation practice. By introducing a multi-scale roughness parameter, fine-scale topography and its attenuation under snow influence is captured. This allows the assessment of snow influence on terrain morphology and, consequently, potential release area size and location. The integration of a wind shelter index enables the user to define release area scenarios as a function of the prevailing wind direction or single storm events. A case study illustrates the practical usefulness of this approach for the definition of release area scenarios under varying snow cover and wind conditions. A validation with historical data demonstrated an improved estimation of avalanche release areas. Our method outperforms a slope-based approach, in particular for more frequent avalanches; however, the application of the algorithm as a forecasting tool remains limited, as snowpack stability is not integrated. Future research activity should therefore focus on the coupling of the algorithm with snowpack conditions.

Atlas-based automatic measurements of the morphology of the tibiofemoral joint

NASA Astrophysics Data System (ADS)

Brehler, M.; Thawait, G.; Shyr, W.; Ramsay, J.; Siewerdsen, J. H.; Zbijewski, W.

2017-03-01

Purpose: Anatomical metrics of the tibiofemoral joint support assessment of joint stability and surgical planning. We propose an automated, atlas-based algorithm to streamline the measurements in 3D images of the joint and reduce userdependence of the metrics arising from manual identification of the anatomical landmarks. Methods: The method is initialized with coarse registrations of a set of atlas images to the fixed input image. The initial registrations are then refined separately for the tibia and femur and the best matching atlas is selected. Finally, the anatomical landmarks of the best matching atlas are transformed onto the input image by deforming a surface model of the atlas to fit the shape of the tibial plateau in the input image (a mesh-to-volume registration). We apply the method to weight-bearing volumetric images of the knee obtained from 23 subjects using an extremity cone-beam CT system. Results of the automated algorithm were compared to an expert radiologist for measurements of Static Alignment (SA), Medial Tibial Slope (MTS) and Lateral Tibial Slope (LTS). Results: Intra-reader variability as high as 10% for LTS and 7% for MTS (ratio of standard deviation to the mean in repeated measurements) was found for expert radiologist, illustrating the potential benefits of an automated approach in improving the precision of the metrics. The proposed method achieved excellent registration of the atlas mesh to the input volumes. The resulting automated measurements yielded high correlations with expert radiologist, as indicated by correlation coefficients of 0.72 for MTS, 0.8 for LTS, and 0.89 for SA. Conclusions: The automated method for measurement of anatomical metrics of the tibiofemoral joint achieves high correlation with expert radiologist without the need for time consuming and error prone manual selection of landmarks.

Atlas-based automatic measurements of the morphology of the tibiofemoral joint.

PubMed

Brehler, M; Thawait, G; Shyr, W; Ramsay, J; Siewerdsen, J H; Zbijewski, W

2017-02-11

Anatomical metrics of the tibiofemoral joint support assessment of joint stability and surgical planning. We propose an automated, atlas-based algorithm to streamline the measurements in 3D images of the joint and reduce user-dependence of the metrics arising from manual identification of the anatomical landmarks. The method is initialized with coarse registrations of a set of atlas images to the fixed input image. The initial registrations are then refined separately for the tibia and femur and the best matching atlas is selected. Finally, the anatomical landmarks of the best matching atlas are transformed onto the input image by deforming a surface model of the atlas to fit the shape of the tibial plateau in the input image (a mesh-to-volume registration). We apply the method to weight-bearing volumetric images of the knee obtained from 23 subjects using an extremity cone-beam CT system. Results of the automated algorithm were compared to an expert radiologist for measurements of Static Alignment (SA), Medial Tibial Slope (MTS) and Lateral Tibial Slope (LTS). Intra-reader variability as high as ~10% for LTS and 7% for MTS (ratio of standard deviation to the mean in repeated measurements) was found for expert radiologist, illustrating the potential benefits of an automated approach in improving the precision of the metrics. The proposed method achieved excellent registration of the atlas mesh to the input volumes. The resulting automated measurements yielded high correlations with expert radiologist, as indicated by correlation coefficients of 0.72 for MTS, 0.8 for LTS, and 0.89 for SA. The automated method for measurement of anatomical metrics of the tibiofemoral joint achieves high correlation with expert radiologist without the need for time consuming and error prone manual selection of landmarks.

Estimation and tracking of AP-diameter of the inferior vena cava in ultrasound images using a novel active circle algorithm.

PubMed

Karami, Ebrahim; Shehata, Mohamed S; Smith, Andrew

2018-05-04

Medical research suggests that the anterior-posterior (AP)-diameter of the inferior vena cava (IVC) and its associated temporal variation as imaged by bedside ultrasound is useful in guiding fluid resuscitation of the critically-ill patient. Unfortunately, indistinct edges and gaps in vessel walls are frequently present which impede accurate estimation of the IVC AP-diameter for both human operators and segmentation algorithms. The majority of research involving use of the IVC to guide fluid resuscitation involves manual measurement of the maximum and minimum AP-diameter as it varies over time. This effort proposes using a time-varying circle fitted inside the typically ellipsoid IVC as an efficient, consistent and novel approach to tracking and approximating the AP-diameter even in the context of poor image quality. In this active-circle algorithm, a novel evolution functional is proposed and shown to be a useful tool for ultrasound image processing. The proposed algorithm is compared with an expert manual measurement, and state-of-the-art relevant algorithms. It is shown that the algorithm outperforms other techniques and performs very close to manual measurement. Copyright © 2018 Elsevier Ltd. All rights reserved.

Towards automated sleep classification in infants using symbolic and subsymbolic approaches.

PubMed

Kubat, M; Flotzinger, D; Pfurtscheller, G

1993-04-01

The paper addresses the problem of automatic sleep classification. A special effort is made to find a method of extracting reasonable descriptions of the individual sleep stages from sample measurements of EGG, EMG, EOG, etc., and from a classification of these measurements provided by an expert. The method should satisfy three requirements: classification accuracy, interpretability of the results, and the ability to select the relevant and discard the irrelevant variables. The solution suggested in this paper consists of a combination of the subsymbolic algorithm LVQ with the symbolic decision tree generator ID3. Results demonstrating the feasibility and utility of our approach are also presented.

Practical interpretation of CYP2D6 haplotypes: Comparison and integration of automated and expert calling.

PubMed

Ruaño, Gualberto; Kocherla, Mohan; Graydon, James S; Holford, Theodore R; Makowski, Gregory S; Goethe, John W

2016-05-01

We describe a population genetic approach to compare samples interpreted with expert calling (EC) versus automated calling (AC) for CYP2D6 haplotyping. The analysis represents 4812 haplotype calls based on signal data generated by the Luminex xMap analyzers from 2406 patients referred to a high-complexity molecular diagnostics laboratory for CYP450 testing. DNA was extracted from buccal swabs. We compared the results of expert calls (EC) and automated calls (AC) with regard to haplotype number and frequency. The ratio of EC to AC was 1:3. Haplotype frequencies from EC and AC samples were convergent across haplotypes, and their distribution was not statistically different between the groups. Most duplications required EC, as only expansions with homozygous or hemizygous haplotypes could be automatedly called. High-complexity laboratories can offer equivalent interpretation to automated calling for non-expanded CYP2D6 loci, and superior interpretation for duplications. We have validated scientific expert calling specified by scoring rules as standard operating procedure integrated with an automated calling algorithm. The integration of EC with AC is a practical strategy for CYP2D6 clinical haplotyping. Copyright © 2016 Elsevier B.V. All rights reserved.

Validating hierarchical verbal autopsy expert algorithms in a large data set with known causes of death.

PubMed

Kalter, Henry D; Perin, Jamie; Black, Robert E

2016-06-01

Physician assessment historically has been the most common method of analyzing verbal autopsy (VA) data. Recently, the World Health Organization endorsed two automated methods, Tariff 2.0 and InterVA-4, which promise greater objectivity and lower cost. A disadvantage of the Tariff method is that it requires a training data set from a prior validation study, while InterVA relies on clinically specified conditional probabilities. We undertook to validate the hierarchical expert algorithm analysis of VA data, an automated, intuitive, deterministic method that does not require a training data set. Using Population Health Metrics Research Consortium study hospital source data, we compared the primary causes of 1629 neonatal and 1456 1-59 month-old child deaths from VA expert algorithms arranged in a hierarchy to their reference standard causes. The expert algorithms were held constant, while five prior and one new "compromise" neonatal hierarchy, and three former child hierarchies were tested. For each comparison, the reference standard data were resampled 1000 times within the range of cause-specific mortality fractions (CSMF) for one of three approximated community scenarios in the 2013 WHO global causes of death, plus one random mortality cause proportions scenario. We utilized CSMF accuracy to assess overall population-level validity, and the absolute difference between VA and reference standard CSMFs to examine particular causes. Chance-corrected concordance (CCC) and Cohen's kappa were used to evaluate individual-level cause assignment. Overall CSMF accuracy for the best-performing expert algorithm hierarchy was 0.80 (range 0.57-0.96) for neonatal deaths and 0.76 (0.50-0.97) for child deaths. Performance for particular causes of death varied, with fairly flat estimated CSMF over a range of reference values for several causes. Performance at the individual diagnosis level was also less favorable than that for overall CSMF (neonatal: best CCC = 0.23, range 0.16-0.33; best kappa = 0.29, 0.23-0.35; child: best CCC = 0.40, 0.19-0.45; best kappa = 0.29, 0.07-0.35). Expert algorithms in a hierarchy offer an accessible, automated method for assigning VA causes of death. Overall population-level accuracy is similar to that of more complex machine learning methods, but without need for a training data set from a prior validation study.

Analysis of Rules for Islamic Inheritance Law in Indonesia Using Hybrid Rule Based Learning

NASA Astrophysics Data System (ADS)

Khosyi'ah, S.; Irfan, M.; Maylawati, D. S.; Mukhlas, O. S.

2018-01-01

Along with the development of human civilization in Indonesia, the changes and reform of Islamic inheritance law so as to conform to the conditions and culture cannot be denied. The distribution of inheritance in Indonesia can be done automatically by storing the rule of Islamic inheritance law in the expert system. In this study, we analyze the knowledge of experts in Islamic inheritance in Indonesia and represent it in the form of rules using rule-based Forward Chaining (FC) and Davis-Putman-Logemann-Loveland (DPLL) algorithms. By hybridizing FC and DPLL algorithms, the rules of Islamic inheritance law in Indonesia are clearly defined and measured. The rules were conceptually validated by some experts in Islamic laws and informatics. The results revealed that generally all rules were ready for use in an expert system.

Knowledge-Based Object Detection in Laser Scanning Point Clouds

NASA Astrophysics Data System (ADS)

Boochs, F.; Karmacharya, A.; Marbs, A.

2012-07-01

Object identification and object processing in 3D point clouds have always posed challenges in terms of effectiveness and efficiency. In practice, this process is highly dependent on human interpretation of the scene represented by the point cloud data, as well as the set of modeling tools available for use. Such modeling algorithms are data-driven and concentrate on specific features of the objects, being accessible to numerical models. We present an approach that brings the human expert knowledge about the scene, the objects inside, and their representation by the data and the behavior of algorithms to the machine. This "understanding" enables the machine to assist human interpretation of the scene inside the point cloud. Furthermore, it allows the machine to understand possibilities and limitations of algorithms and to take this into account within the processing chain. This not only assists the researchers in defining optimal processing steps, but also provides suggestions when certain changes or new details emerge from the point cloud. Our approach benefits from the advancement in knowledge technologies within the Semantic Web framework. This advancement has provided a strong base for applications based on knowledge management. In the article we will present and describe the knowledge technologies used for our approach such as Web Ontology Language (OWL), used for formulating the knowledge base and the Semantic Web Rule Language (SWRL) with 3D processing and topologic built-ins, aiming to combine geometrical analysis of 3D point clouds, and specialists' knowledge of the scene and algorithmic processing.

Interaction mining and skill-dependent recommendations for multi-objective team composition

PubMed Central

Dorn, Christoph; Skopik, Florian; Schall, Daniel; Dustdar, Schahram

2011-01-01

Web-based collaboration and virtual environments supported by various Web 2.0 concepts enable the application of numerous monitoring, mining and analysis tools to study human interactions and team formation processes. The composition of an effective team requires a balance between adequate skill fulfillment and sufficient team connectivity. The underlying interaction structure reflects social behavior and relations of individuals and determines to a large degree how well people can be expected to collaborate. In this paper we address an extended team formation problem that does not only require direct interactions to determine team connectivity but additionally uses implicit recommendations of collaboration partners to support even sparsely connected networks. We provide two heuristics based on Genetic Algorithms and Simulated Annealing for discovering efficient team configurations that yield the best trade-off between skill coverage and team connectivity. Our self-adjusting mechanism aims to discover the best combination of direct interactions and recommendations when deriving connectivity. We evaluate our approach based on multiple configurations of a simulated collaboration network that features close resemblance to real world expert networks. We demonstrate that our algorithm successfully identifies efficient team configurations even when removing up to 40% of experts from various social network configurations. PMID:22298939

Relationship auditing of the FMA ontology

PubMed Central

Gu, Huanying (Helen); Wei, Duo; Mejino, Jose L.V.; Elhanan, Gai

2010-01-01

The Foundational Model of Anatomy (FMA) ontology is a domain reference ontology based on a disciplined modeling approach. Due to its large size, semantic complexity and manual data entry process, errors and inconsistencies are unavoidable and might remain within the FMA structure without detection. In this paper, we present computable methods to highlight candidate concepts for various relationship assignment errors. The process starts with locating structures formed by transitive structural relationships (part_of, tributary_of, branch_of) and examine their assignments in the context of the IS-A hierarchy. The algorithms were designed to detect five major categories of possible incorrect relationship assignments: circular, mutually exclusive, redundant, inconsistent, and missed entries. A domain expert reviewed samples of these presumptive errors to confirm the findings. Seven thousand and fifty-two presumptive errors were detected, the largest proportion related to part_of relationship assignments. The results highlight the fact that errors are unavoidable in complex ontologies and that well designed algorithms can help domain experts to focus on concepts with high likelihood of errors and maximize their effort to ensure consistency and reliability. In the future similar methods might be integrated with data entry processes to offer real-time error detection. PMID:19475727

Automatic recognition of surface landmarks of anatomical structures of back and posture

NASA Astrophysics Data System (ADS)

Michoński, Jakub; Glinkowski, Wojciech; Witkowski, Marcin; Sitnik, Robert

2012-05-01

Faulty postures, scoliosis and sagittal plane deformities should be detected as early as possible to apply preventive and treatment measures against major clinical consequences. To support documentation of the severity of deformity and diminish x-ray exposures, several solutions utilizing analysis of back surface topography data were introduced. A novel approach to automatic recognition and localization of anatomical landmarks of the human back is presented that may provide more repeatable results and speed up the whole procedure. The algorithm was designed as a two-step process involving a statistical model built upon expert knowledge and analysis of three-dimensional back surface shape data. Voronoi diagram is used to connect mean geometric relations, which provide a first approximation of the positions, with surface curvature distribution, which further guides the recognition process and gives final locations of landmarks. Positions obtained using the developed algorithms are validated with respect to accuracy of manual landmark indication by experts. Preliminary validation proved that the landmarks were localized correctly, with accuracy depending mostly on the characteristics of a given structure. It was concluded that recognition should mainly take into account the shape of the back surface, putting as little emphasis on the statistical approximation as possible.

Metric Ranking of Invariant Networks with Belief Propagation

DOE Office of Scientific and Technical Information (OSTI.GOV)

Tao, Changxia; Ge, Yong; Song, Qinbao

The management of large-scale distributed information systems relies on the effective use and modeling of monitoring data collected at various points in the distributed information systems. A promising approach is to discover invariant relationships among the monitoring data and generate invariant networks, where a node is a monitoring data source (metric) and a link indicates an invariant relationship between two monitoring data. Such an invariant network representation can help system experts to localize and diagnose the system faults by examining those broken invariant relationships and their related metrics, because system faults usually propagate among the monitoring data and eventually leadmore » to some broken invariant relationships. However, at one time, there are usually a lot of broken links (invariant relationships) within an invariant network. Without proper guidance, it is difficult for system experts to manually inspect this large number of broken links. Thus, a critical challenge is how to effectively and efficiently rank metrics (nodes) of invariant networks according to the anomaly levels of metrics. The ranked list of metrics will provide system experts with useful guidance for them to localize and diagnose the system faults. To this end, we propose to model the nodes and the broken links as a Markov Random Field (MRF), and develop an iteration algorithm to infer the anomaly of each node based on belief propagation (BP). Finally, we validate the proposed algorithm on both realworld and synthetic data sets to illustrate its effectiveness.« less

A comparison of the fractal and JPEG algorithms

NASA Technical Reports Server (NTRS)

Cheung, K.-M.; Shahshahani, M.

1991-01-01

A proprietary fractal image compression algorithm and the Joint Photographic Experts Group (JPEG) industry standard algorithm for image compression are compared. In every case, the JPEG algorithm was superior to the fractal method at a given compression ratio according to a root mean square criterion and a peak signal to noise criterion.

Content validation of a standardized algorithm for ostomy care.

PubMed

Beitz, Janice; Gerlach, Mary; Ginsburg, Pat; Ho, Marianne; McCann, Eileen; Schafer, Vickie; Scott, Vera; Stallings, Bobbie; Turnbull, Gwen

2010-10-01

The number of ostomy care clinician experts is limited and the majority of ostomy care is provided by non-specialized clinicians or unskilled caregivers and family. The purpose of this study was to obtain content validation data for a new standardized algorithm for ostomy care developed by expert wound ostomy continence nurse (WOCN) clinicians. After face validity was established using overall review and suggestions from WOCN experts, 166 WOCNs self-identified as having expertise in ostomy care were surveyed online for 6 weeks in 2009. Using a cross-sectional, mixed methods study design and a 30-item instrument with a 4-point Likert-type scale, the participants were asked to quantify the degree of validity of the Ostomy Algorithm's decisions and components. Participants' open-ended comments also were thematically analyzed. Using a scale of 1 to 4, the mean score of the entire algorithm was 3.8 (4 = relevant/very relevant). The algorithm's content validity index (CVI) was 0.95 (out of 1.0). Individual component mean scores ranged from 3.59 to 3.91. Individual CVIs ranged from 0.90 to 0.98. Qualitative data analysis revealed themes of difficulty associated with algorithm formatting, especially orientation and use of the Studio Alterazioni Cutanee Stomali (Study on Peristomal Skin Lesions [SACS™ Instrument]) and the inability of algorithms to capture all individual patient attributes affecting ostomy care. Positive themes included content thoroughness and the helpful clinical photos. Suggestions were offered for algorithm improvement. Study results support the strong content validity of the algorithm and research to ascertain its construct validity and effect on care outcomes is warranted.

DALMATIAN: An Algorithm for Automatic Cell Detection and Counting in 3D.

PubMed

Shuvaev, Sergey A; Lazutkin, Alexander A; Kedrov, Alexander V; Anokhin, Konstantin V; Enikolopov, Grigori N; Koulakov, Alexei A

2017-01-01

Current 3D imaging methods, including optical projection tomography, light-sheet microscopy, block-face imaging, and serial two photon tomography enable visualization of large samples of biological tissue. Large volumes of data obtained at high resolution require development of automatic image processing techniques, such as algorithms for automatic cell detection or, more generally, point-like object detection. Current approaches to automated cell detection suffer from difficulties originating from detection of particular cell types, cell populations of different brightness, non-uniformly stained, and overlapping cells. In this study, we present a set of algorithms for robust automatic cell detection in 3D. Our algorithms are suitable for, but not limited to, whole brain regions and individual brain sections. We used watershed procedure to split regional maxima representing overlapping cells. We developed a bootstrap Gaussian fit procedure to evaluate the statistical significance of detected cells. We compared cell detection quality of our algorithm and other software using 42 samples, representing 6 staining and imaging techniques. The results provided by our algorithm matched manual expert quantification with signal-to-noise dependent confidence, including samples with cells of different brightness, non-uniformly stained, and overlapping cells for whole brain regions and individual tissue sections. Our algorithm provided the best cell detection quality among tested free and commercial software.

Java-Based Diabetes Type 2 Prediction Tool for Better Diagnosis

PubMed Central

Odedra, Devang; Mallick, Medhavi; Shukla, Prateek; Samanta, Subir; Vidyarthi, Ambarish S.

2012-01-01

Abstract Background The concept of classification of clinical data can be utilized in the development of an effective diagnosis system by taking the advantage of computational intelligence. Diabetes disease diagnosis via proper interpretation of the diabetes data is an important problem in neural networks. Unfortunately, although several classification studies have been carried out with significant performance, many of the current methods often fail to reach out to patients. Graphical user interface-enabled tools need to be developed through which medical practitioners can simply enter the health profiles of their patients and receive an instant diabetes prediction with an acceptable degree of confidence. Methods In this study, the neural network approach was used for a dataset of 768 persons from a Pima Indian population living near Phoenix, AZ. A neural network mixture of experts model was trained with these data using the expectation-minimization algorithm. Results The mixture of experts method was used to train the algorithm with 97% accuracy. A graphical user interface was developed that would work in conjunction with the trained network to provide the output in a presentable format. Conclusions This study provides a machine-implementable approach that can be used by physicians and patients to minimize the extent of error in diagnosis. The authors are hopeful that replication of results of this study in other populations may lead to improved diagnosis. Physicians can simply enter the health profile of patients and get the diagnosis for diabetes type 2. PMID:22059431

Segmentation and tracking in echocardiographic sequences: active contours guided by optical flow estimates

NASA Technical Reports Server (NTRS)

Mikic, I.; Krucinski, S.; Thomas, J. D.

1998-01-01

This paper presents a method for segmentation and tracking of cardiac structures in ultrasound image sequences. The developed algorithm is based on the active contour framework. This approach requires initial placement of the contour close to the desired position in the image, usually an object outline. Best contour shape and position are then calculated, assuming that at this configuration a global energy function, associated with a contour, attains its minimum. Active contours can be used for tracking by selecting a solution from a previous frame as an initial position in a present frame. Such an approach, however, fails for large displacements of the object of interest. This paper presents a technique that incorporates the information on pixel velocities (optical flow) into the estimate of initial contour to enable tracking of fast-moving objects. The algorithm was tested on several ultrasound image sequences, each covering one complete cardiac cycle. The contour successfully tracked boundaries of mitral valve leaflets, aortic root and endocardial borders of the left ventricle. The algorithm-generated outlines were compared against manual tracings by expert physicians. The automated method resulted in contours that were within the boundaries of intraobserver variability.

Perspective on intelligent avionics

DOE Office of Scientific and Technical Information (OSTI.GOV)

Jones, H.L.

1987-01-01

Technical issues which could potentially limit the capability and acceptibility of expert systems decision-making for avionics applications are addressed. These issues are: real-time AI, mission-critical software, conventional algorithms, pilot interface, knowledge acquisition, and distributed expert systems. Examples from on-going expert system development programs are presented to illustrate likely architectures and applications of future intelligent avionic systems. 13 references.

Fast Fuzzy Arithmetic Operations

NASA Technical Reports Server (NTRS)

Hampton, Michael; Kosheleva, Olga

1997-01-01

In engineering applications of fuzzy logic, the main goal is not to simulate the way the experts really think, but to come up with a good engineering solution that would (ideally) be better than the expert's control, In such applications, it makes perfect sense to restrict ourselves to simplified approximate expressions for membership functions. If we need to perform arithmetic operations with the resulting fuzzy numbers, then we can use simple and fast algorithms that are known for operations with simple membership functions. In other applications, especially the ones that are related to humanities, simulating experts is one of the main goals. In such applications, we must use membership functions that capture every nuance of the expert's opinion; these functions are therefore complicated, and fuzzy arithmetic operations with the corresponding fuzzy numbers become a computational problem. In this paper, we design a new algorithm for performing such operations. This algorithm is applicable in the case when negative logarithms - log(u(x)) of membership functions u(x) are convex, and reduces computation time from O(n(exp 2))to O(n log(n)) (where n is the number of points x at which we know the membership functions u(x)).

Virtual commissioning of automated micro-optical assembly

NASA Astrophysics Data System (ADS)

Schlette, Christian; Losch, Daniel; Haag, Sebastian; Zontar, Daniel; Roßmann, Jürgen; Brecher, Christian

2015-02-01

In this contribution, we present a novel approach to enable virtual commissioning for process developers in micro-optical assembly. Our approach aims at supporting micro-optics experts to effectively develop assisted or fully automated assembly solutions without detailed prior experience in programming while at the same time enabling them to easily implement their own libraries of expert schemes and algorithms for handling optical components. Virtual commissioning is enabled by a 3D simulation and visualization system in which the functionalities and properties of automated systems are modeled, simulated and controlled based on multi-agent systems. For process development, our approach supports event-, state- and time-based visual programming techniques for the agents and allows for their kinematic motion simulation in combination with looped-in simulation results for the optical components. First results have been achieved for simply switching the agents to command the real hardware setup after successful process implementation and validation in the virtual environment. We evaluated and adapted our system to meet the requirements set by industrial partners-- laser manufacturers as well as hardware suppliers of assembly platforms. The concept is applied to the automated assembly of optical components for optically pumped semiconductor lasers and positioning of optical components for beam-shaping

Game playing.

PubMed

Rosin, Christopher D

2014-03-01

Game playing has been a core domain of artificial intelligence research since the beginnings of the field. Game playing provides clearly defined arenas within which computational approaches can be readily compared to human expertise through head-to-head competition and other benchmarks. Game playing research has identified several simple core algorithms that provide successful foundations, with development focused on the challenges of defeating human experts in specific games. Key developments include minimax search in chess, machine learning from self-play in backgammon, and Monte Carlo tree search in Go. These approaches have generalized successfully to additional games. While computers have surpassed human expertise in a wide variety of games, open challenges remain and research focuses on identifying and developing new successful algorithmic foundations. WIREs Cogn Sci 2014, 5:193-205. doi: 10.1002/wcs.1278 CONFLICT OF INTEREST: The author has declared no conflicts of interest for this article. For further resources related to this article, please visit the WIREs website. © 2014 John Wiley & Sons, Ltd.

Relational machine learning for electronic health record-driven phenotyping.

PubMed

Peissig, Peggy L; Santos Costa, Vitor; Caldwell, Michael D; Rottscheit, Carla; Berg, Richard L; Mendonca, Eneida A; Page, David

2014-12-01

Electronic health records (EHR) offer medical and pharmacogenomics research unprecedented opportunities to identify and classify patients at risk. EHRs are collections of highly inter-dependent records that include biological, anatomical, physiological, and behavioral observations. They comprise a patient's clinical phenome, where each patient has thousands of date-stamped records distributed across many relational tables. Development of EHR computer-based phenotyping algorithms require time and medical insight from clinical experts, who most often can only review a small patient subset representative of the total EHR records, to identify phenotype features. In this research we evaluate whether relational machine learning (ML) using inductive logic programming (ILP) can contribute to addressing these issues as a viable approach for EHR-based phenotyping. Two relational learning ILP approaches and three well-known WEKA (Waikato Environment for Knowledge Analysis) implementations of non-relational approaches (PART, J48, and JRIP) were used to develop models for nine phenotypes. International Classification of Diseases, Ninth Revision (ICD-9) coded EHR data were used to select training cohorts for the development of each phenotypic model. Accuracy, precision, recall, F-Measure, and Area Under the Receiver Operating Characteristic (AUROC) curve statistics were measured for each phenotypic model based on independent manually verified test cohorts. A two-sided binomial distribution test (sign test) compared the five ML approaches across phenotypes for statistical significance. We developed an approach to automatically label training examples using ICD-9 diagnosis codes for the ML approaches being evaluated. Nine phenotypic models for each ML approach were evaluated, resulting in better overall model performance in AUROC using ILP when compared to PART (p=0.039), J48 (p=0.003) and JRIP (p=0.003). ILP has the potential to improve phenotyping by independently delivering clinically expert interpretable rules for phenotype definitions, or intuitive phenotypes to assist experts. Relational learning using ILP offers a viable approach to EHR-driven phenotyping. Copyright © 2014 Elsevier Inc. All rights reserved.

Realizing parameterless automatic classification of remote sensing imagery using ontology engineering and cyberinfrastructure techniques

NASA Astrophysics Data System (ADS)

Sun, Ziheng; Fang, Hui; Di, Liping; Yue, Peng

2016-09-01

It was an untouchable dream for remote sensing experts to realize total automatic image classification without inputting any parameter values. Experts usually spend hours and hours on tuning the input parameters of classification algorithms in order to obtain the best results. With the rapid development of knowledge engineering and cyberinfrastructure, a lot of data processing and knowledge reasoning capabilities become online accessible, shareable and interoperable. Based on these recent improvements, this paper presents an idea of parameterless automatic classification which only requires an image and automatically outputs a labeled vector. No parameters and operations are needed from endpoint consumers. An approach is proposed to realize the idea. It adopts an ontology database to store the experiences of tuning values for classifiers. A sample database is used to record training samples of image segments. Geoprocessing Web services are used as functionality blocks to finish basic classification steps. Workflow technology is involved to turn the overall image classification into a total automatic process. A Web-based prototypical system named PACS (Parameterless Automatic Classification System) is implemented. A number of images are fed into the system for evaluation purposes. The results show that the approach could automatically classify remote sensing images and have a fairly good average accuracy. It is indicated that the classified results will be more accurate if the two databases have higher quality. Once the experiences and samples in the databases are accumulated as many as an expert has, the approach should be able to get the results with similar quality to that a human expert can get. Since the approach is total automatic and parameterless, it can not only relieve remote sensing workers from the heavy and time-consuming parameter tuning work, but also significantly shorten the waiting time for consumers and facilitate them to engage in image classification activities. Currently, the approach is used only on high resolution optical three-band remote sensing imagery. The feasibility using the approach on other kinds of remote sensing images or involving additional bands in classification will be studied in future.

Automated Item Generation with Recurrent Neural Networks.

PubMed

von Davier, Matthias

2018-03-12

Utilizing technology for automated item generation is not a new idea. However, test items used in commercial testing programs or in research are still predominantly written by humans, in most cases by content experts or professional item writers. Human experts are a limited resource and testing agencies incur high costs in the process of continuous renewal of item banks to sustain testing programs. Using algorithms instead holds the promise of providing unlimited resources for this crucial part of assessment development. The approach presented here deviates in several ways from previous attempts to solve this problem. In the past, automatic item generation relied either on generating clones of narrowly defined item types such as those found in language free intelligence tests (e.g., Raven's progressive matrices) or on an extensive analysis of task components and derivation of schemata to produce items with pre-specified variability that are hoped to have predictable levels of difficulty. It is somewhat unlikely that researchers utilizing these previous approaches would look at the proposed approach with favor; however, recent applications of machine learning show success in solving tasks that seemed impossible for machines not too long ago. The proposed approach uses deep learning to implement probabilistic language models, not unlike what Google brain and Amazon Alexa use for language processing and generation.

Consensus Guidelines on Evaluation and Management of the Febrile Child Presenting to the Emergency Department in India.

PubMed

Mahajan, Prashant; Batra, Prerna; Thakur, Neha; Patel, Reena; Rai, Narendra; Trivedi, Nitin; Fassl, Bernhard; Shah, Binita; Lozon, Marie; Oteng, Rockerfeller A; Saha, Abhijeet; Shah, Dheeraj; Galwankar, Sagar

2017-08-15

India, home to almost 1.5 billion people, is in need of a country-specific, evidence-based, consensus approach for the emergency department (ED) evaluation and management of the febrile child. We held two consensus meetings, performed an exhaustive literature review, and held ongoing web-based discussions to arrive at a formal consensus on the proposed evaluation and management algorithm. The first meeting was held in Delhi in October 2015, under the auspices of Pediatric Emergency Medicine (PEM) Section of Academic College of Emergency Experts in India (ACEE-INDIA); and the second meeting was conducted at Pune during Emergency Medical Pediatrics and Recent Trends (EMPART 2016) in March 2016. The second meeting was followed with futher e-mail-based discussions to arrive at a formal consensus on the proposed algorithm. To develop an algorithmic approach for the evaluation and management of the febrile child that can be easily applied in the context of emergency care and modified based on local epidemiology and practice standards. We created an algorithm that can assist the clinician in the evaluation and management of the febrile child presenting to the ED, contextualized to health care in India. This guideline includes the following key components: triage and the timely assessment; evaluation; and patient disposition from the ED. We urge the development and creation of a robust data repository of minimal standard data elements. This would provide a systematic measurement of the care processes and patient outcomes, and a better understanding of various etiologies of febrile illnesses in India; both of which can be used to further modify the proposed approach and algorithm.

Active machine learning for rapid landslide inventory mapping with VHR satellite images (Invited)

NASA Astrophysics Data System (ADS)

Stumpf, A.; Lachiche, N.; Malet, J.; Kerle, N.; Puissant, A.

2013-12-01

VHR satellite images have become a primary source for landslide inventory mapping after major triggering events such as earthquakes and heavy rainfalls. Visual image interpretation is still the prevailing standard method for operational purposes but is time-consuming and not well suited to fully exploit the increasingly better supply of remote sensing data. Recent studies have addressed the development of more automated image analysis workflows for landslide inventory mapping. In particular object-oriented approaches that account for spatial and textural image information have been demonstrated to be more adequate than pixel-based classification but manually elaborated rule-based classifiers are difficult to adapt under changing scene characteristics. Machine learning algorithm allow learning classification rules for complex image patterns from labelled examples and can be adapted straightforwardly with available training data. In order to reduce the amount of costly training data active learning (AL) has evolved as a key concept to guide the sampling for many applications. The underlying idea of AL is to initialize a machine learning model with a small training set, and to subsequently exploit the model state and data structure to iteratively select the most valuable samples that should be labelled by the user. With relatively few queries and labelled samples, an AL strategy yields higher accuracies than an equivalent classifier trained with many randomly selected samples. This study addressed the development of an AL method for landslide mapping from VHR remote sensing images with special consideration of the spatial distribution of the samples. Our approach [1] is based on the Random Forest algorithm and considers the classifier uncertainty as well as the variance of potential sampling regions to guide the user towards the most valuable sampling areas. The algorithm explicitly searches for compact regions and thereby avoids a spatially disperse sampling pattern inherent to most other AL methods. The accuracy, the sampling time and the computational runtime of the algorithm were evaluated on multiple satellite images capturing recent large scale landslide events. Sampling between 1-4% of the study areas the accuracies between 74% and 80% were achieved, whereas standard sampling schemes yielded only accuracies between 28% and 50% with equal sampling costs. Compared to commonly used point-wise AL algorithm the proposed approach significantly reduces the number of iterations and hence the computational runtime. Since the user can focus on relatively few compact areas (rather than on hundreds of distributed points) the overall labeling time is reduced by more than 50% compared to point-wise queries. An experimental evaluation of multiple expert mappings demonstrated strong relationships between the uncertainties of the experts and the machine learning model. It revealed that the achieved accuracies are within the range of the inter-expert disagreement and that it will be indispensable to consider ground truth uncertainties to truly achieve further enhancements in the future. The proposed method is generally applicable to a wide range of optical satellite images and landslide types. [1] A. Stumpf, N. Lachiche, J.-P. Malet, N. Kerle, and A. Puissant, Active learning in the spatial domain for remote sensing image classification, IEEE Transactions on Geosciece and Remote Sensing. 2013, DOI 10.1109/TGRS.2013.2262052.

Unsupervised learning of structure in spectroscopic cubes

NASA Astrophysics Data System (ADS)

Araya, M.; Mendoza, M.; Solar, M.; Mardones, D.; Bayo, A.

2018-07-01

We consider the problem of analyzing the structure of spectroscopic cubes using unsupervised machine learning techniques. We propose representing the target's signal as a homogeneous set of volumes through an iterative algorithm that separates the structured emission from the background while not overestimating the flux. Besides verifying some basic theoretical properties, the algorithm is designed to be tuned by domain experts, because its parameters have meaningful values in the astronomical context. Nevertheless, we propose a heuristic to automatically estimate the signal-to-noise ratio parameter of the algorithm directly from data. The resulting light-weighted set of samples (≤ 1% compared to the original data) offer several advantages. For instance, it is statistically correct and computationally inexpensive to apply well-established techniques of the pattern recognition and machine learning domains; such as clustering and dimensionality reduction algorithms. We use ALMA science verification data to validate our method, and present examples of the operations that can be performed by using the proposed representation. Even though this approach is focused on providing faster and better analysis tools for the end-user astronomer, it also opens the possibility of content-aware data discovery by applying our algorithm to big data.

A Clustering Algorithm for Ecological Stream Segment Identification from Spatially Extensive Digital Databases

NASA Astrophysics Data System (ADS)

Brenden, T. O.; Clark, R. D.; Wiley, M. J.; Seelbach, P. W.; Wang, L.

2005-05-01

Remote sensing and geographic information systems have made it possible to attribute variables for streams at increasingly detailed resolutions (e.g., individual river reaches). Nevertheless, management decisions still must be made at large scales because land and stream managers typically lack sufficient resources to manage on an individual reach basis. Managers thus require a method for identifying stream management units that are ecologically similar and that can be expected to respond similarly to management decisions. We have developed a spatially-constrained clustering algorithm that can merge neighboring river reaches with similar ecological characteristics into larger management units. The clustering algorithm is based on the Cluster Affinity Search Technique (CAST), which was developed for clustering gene expression data. Inputs to the clustering algorithm are the neighbor relationships of the reaches that comprise the digital river network, the ecological attributes of the reaches, and an affinity value, which identifies the minimum similarity for merging river reaches. In this presentation, we describe the clustering algorithm in greater detail and contrast its use with other methods (expert opinion, classification approach, regular clustering) for identifying management units using several Michigan watersheds as a backdrop.

[Effect of economic factors of perestroĭka on speeding-up the development of forensic medical services].

PubMed

Plaksin, V O; Dontsov, V G; Bakhmet'ev, V I

1989-01-01

The issues of intensification of medicolegal expert work on the basis of financial stimulation of competition between experts are considered. A new system of work distribution between experts, control and financial stimulation is developed. Three groups of objective indices are singled out and algorithms for different kinds of investigations are worked out. The system is approved in Medicolegal Bureau in Voronezh. Its use enlarged the volume of expert activities, increased the quality of expert investigations and shortened time of investigation performance.

Optical Coherence Tomography (OCT) Device Independent Intraretinal Layer Segmentation

PubMed Central

Ehnes, Alexander; Wenner, Yaroslava; Friedburg, Christoph; Preising, Markus N.; Bowl, Wadim; Sekundo, Walter; zu Bexten, Erdmuthe Meyer; Stieger, Knut; Lorenz, Birgit

2014-01-01

Purpose To develop and test an algorithm to segment intraretinal layers irrespectively of the actual Optical Coherence Tomography (OCT) device used. Methods The developed algorithm is based on the graph theory optimization. The algorithm's performance was evaluated against that of three expert graders for unsigned boundary position difference and thickness measurement of a retinal layer group in 50 and 41 B-scans, respectively. Reproducibility of the algorithm was tested in 30 C-scans of 10 healthy subjects each with the Spectralis and the Stratus OCT. Comparability between different devices was evaluated in 84 C-scans (volume or radial scans) obtained from 21 healthy subjects, two scans per subject with the Spectralis OCT, and one scan per subject each with the Stratus OCT and the RTVue-100 OCT. Each C-scan was segmented and the mean thickness for each retinal layer in sections of the early treatment of diabetic retinopathy study (ETDRS) grid was measured. Results The algorithm was able to segment up to 11 intraretinal layers. Measurements with the algorithm were within the 95% confidence interval of a single grader and the difference was smaller than the interindividual difference between the expert graders themselves. The cross-device examination of ETDRS-grid related layer thicknesses highly agreed between the three OCT devices. The algorithm correctly segmented a C-scan of a patient with X-linked retinitis pigmentosa. Conclusions The segmentation software provides device-independent, reliable, and reproducible analysis of intraretinal layers, similar to what is obtained from expert graders. Translational Relevance Potential application of the software includes routine clinical practice and multicenter clinical trials. PMID:24820053

FOREWORD: IV International Time-Scale Algorithms Symposium, BIPM, Sèvres, 18-19 March 2002

NASA Astrophysics Data System (ADS)

Leschiutta, Sigfrido

2003-06-01

Time-scale formation, along with atomic time/frequency standards and time comparison techniques, is one of the three basic ingredients of Time Metrology. Before summarizing this Symposium and the relevant outcomes, let me make a couple of very general remarks. Clocks and comparison methods have today reached a very high level of accuracy: the nanosecond level. Some applications in the real word are now challenging the capacity of the National Metrological Laboratories. It is therefore essential that the algorithms dealing with clocks and comparison techniques should be such as to make the most of existing technologies. The comfortable margin of accuracy we were used to, between Laboratories and the Field, is gone forever. While clock makers and time-comparison experts meet regularly (FCS, PTTI, EFTF, CPEM, URSI, UIT, etc), the somewhat secluded community of experts in time-scale formation lacks a similar point of contact, with the exception of the CCTF meeting. This venue must consequently be welcomed. Let me recall some highlights from this Symposium: there were about 60 attendees from 15 nations, plus international institutions, such as the host BIPM, and a supranational one, ESA. About 30 papers, prepared in some 20 laboratories, were received: among these papers, four tutorials were offered; descriptions of local time scales including the local algorithms were presented; four papers considered the algorithms applied to the results of time-comparison methods; and six papers covered the special requirements of some specialized time-scale 'users'. The four basic ingredients of time-scale formation: models, noise, filtering and steering, received attention and were also discussed, not just during the sessions. The most demanding applications for time scales now come from Global Navigation Satellite systems; in six papers the progress of some programmes was described and the present and future needs were presented and documented. The lively discussion on future navigation systems led to the following four points: an overall accuracy in timing of one nanosecond is a must; the combined 'clock and orbit' effects on the knowledge of satellite position should be less than one metre; a combined solution for positioning and timing should be pursued; a 'new' time window (2 h to 4 h) emerged, in which the accuracy and stability parameters of the clocks forming a time scale for space application are to be optimized. That interval is linked to some criteria and methods for on-board clock corrections. A revival of interest in the time-proven Kalman filter was noted; in the course of a tutorial on past experience, a number of new approaches were discussed. Some further research is in order, but one should heed the comment: 'do not ask too much of a filter'. The Kalman approach is indeed powerful in combining sets of different data, provided that the possible problems of convergence are suitably addressed. Attention was also focused on the possibility of becoming victims of ever-present 'hidden' correlations. The TAI algorithm, ALGOS, is about 30 years old and the fundamental approach remains unchanged and unchallenged. A number of small refinements, all justified, were introduced in the 'constants' and parameters, but the general philosophy holds. In so far as the BIPM Time Section and the CCTF Working Group on Algorithms are concerned, on the basis of the outcome of this Symposium it is clear that they should follow the evolution of TAI and suggest any appropriate action to the CCTF. This Symposium, which gathered the world experts on T/F algorithms in Paris for two days, offered a wonderful opportunity for cross-fertilization between researchers operating in different and interdependent communities that are loosely connected. Thanks are due to Felicitas Arias, Demetrios Matsakis and Patrizia Tavella and their host organizations for having provided the community with this learning experience. One last comment: please do not wait another 14 years for the next Time Scale Algorithm Symposium.

Examining applying high performance genetic data feature selection and classification algorithms for colon cancer diagnosis.

PubMed

Al-Rajab, Murad; Lu, Joan; Xu, Qiang

2017-07-01

This paper examines the accuracy and efficiency (time complexity) of high performance genetic data feature selection and classification algorithms for colon cancer diagnosis. The need for this research derives from the urgent and increasing need for accurate and efficient algorithms. Colon cancer is a leading cause of death worldwide, hence it is vitally important for the cancer tissues to be expertly identified and classified in a rapid and timely manner, to assure both a fast detection of the disease and to expedite the drug discovery process. In this research, a three-phase approach was proposed and implemented: Phases One and Two examined the feature selection algorithms and classification algorithms employed separately, and Phase Three examined the performance of the combination of these. It was found from Phase One that the Particle Swarm Optimization (PSO) algorithm performed best with the colon dataset as a feature selection (29 genes selected) and from Phase Two that the Support Vector Machine (SVM) algorithm outperformed other classifications, with an accuracy of almost 86%. It was also found from Phase Three that the combined use of PSO and SVM surpassed other algorithms in accuracy and performance, and was faster in terms of time analysis (94%). It is concluded that applying feature selection algorithms prior to classification algorithms results in better accuracy than when the latter are applied alone. This conclusion is important and significant to industry and society. Copyright © 2017 Elsevier B.V. All rights reserved.

Gold-standard for computer-assisted morphological sperm analysis.

PubMed

Chang, Violeta; Garcia, Alejandra; Hitschfeld, Nancy; Härtel, Steffen

2017-04-01

Published algorithms for classification of human sperm heads are based on relatively small image databases that are not open to the public, and thus no direct comparison is available for competing methods. We describe a gold-standard for morphological sperm analysis (SCIAN-MorphoSpermGS), a dataset of sperm head images with expert-classification labels in one of the following classes: normal, tapered, pyriform, small or amorphous. This gold-standard is for evaluating and comparing known techniques and future improvements to present approaches for classification of human sperm heads for semen analysis. Although this paper does not provide a computational tool for morphological sperm analysis, we present a set of experiments for comparing sperm head description and classification common techniques. This classification base-line is aimed to be used as a reference for future improvements to present approaches for human sperm head classification. The gold-standard provides a label for each sperm head, which is achieved by majority voting among experts. The classification base-line compares four supervised learning methods (1- Nearest Neighbor, naive Bayes, decision trees and Support Vector Machine (SVM)) and three shape-based descriptors (Hu moments, Zernike moments and Fourier descriptors), reporting the accuracy and the true positive rate for each experiment. We used Fleiss' Kappa Coefficient to evaluate the inter-expert agreement and Fisher's exact test for inter-expert variability and statistical significant differences between descriptors and learning techniques. Our results confirm the high degree of inter-expert variability in the morphological sperm analysis. Regarding the classification base line, we show that none of the standard descriptors or classification approaches is best suitable for tackling the problem of sperm head classification. We discovered that the correct classification rate was highly variable when trying to discriminate among non-normal sperm heads. By using the Fourier descriptor and SVM, we achieved the best mean correct classification: only 49%. We conclude that the SCIAN-MorphoSpermGS will provide a standard tool for evaluation of characterization and classification approaches for human sperm heads. Indeed, there is a clear need for a specific shape-based descriptor for human sperm heads and a specific classification approach to tackle the problem of high variability within subcategories of abnormal sperm cells. Copyright © 2017 Elsevier Ltd. All rights reserved.

Consensus in Gestational Diabetes MELLITUS: Looking for the Holy Grail.

PubMed

Agarwal, Mukesh M

2018-05-28

The world's pre-eminent diabetes, obstetric, endocrine, and health organizations advocate a plethora of diverse algorithms for the screening, diagnosis, management, and follow-up of gestational diabetes mellitus (GDM). Additionally, there are regional recommendations of local health societies. Several of these proposals for GDM are contentious because some of them were developed from unscientific studies, based on expert-opinion, catered to preserve resources, and subjectively modified for convenience. Due to the wide variety of choices available, the approach to GDM can be extremely diverse even within the same hospital. This lack of consensus creates major problems in addressing prevalence, complications, efficacy of treatment, and follow-up of GDM. Moreover, it becomes nearly impossible to compare the numerous studies. Furthermore, the lack of consensus confuses the health care providers of obstetric health who look to the experts for guidance. Therefore, a clear, objective, "evidence-based" global approach, which is simple, easy to follow, and validated by corroborative research, is crucial. We contend that, despite decades of research, a single acceptable global guideline is not yet on the horizon.

Multidisciplinary approach to the diagnosis and management of patients with peripheral arterial disease

PubMed Central

Walker, Craig M; Bunch, Frank T; Cavros, Nick G; Dippel, Eric J

2015-01-01

Peripheral arterial disease (PAD) is frequently diagnosed after permanent damage has occurred, resulting in a high rate of morbidity, amputation, and loss of life. Early and ongoing diagnosis and treatment is required for this progressive disease. Lifestyle modifications can prevent or delay disease progression and improve symptoms. Limb-sparing endovascular interventions can restore circulation based on appropriate diagnostic testing to pinpoint vascular targets, and intervention must occur as early as possible to ensure optimal clinical outcomes. An algorithm for the diagnosis and management of PAD was developed to enable a collaborative approach between the family practice and primary care physician or internist and various specialists that may include a diabetologist, endocrinologist, smoking cessation expert, hypertension and lipid specialist, endovascular interventionalist, vascular surgeon, orthopedist, neurologist, nurse practitioner, podiatrist, wound healing expert, and/or others. A multidisciplinary team working together has the greatest chance of providing optimal care for the patient with PAD and ensuring ongoing surveillance of the patient’s overall health, ultimately resulting in better quality of life and increased longevity for patients with PAD. PMID:26203234

Mathematical modelling of risk reduction in reinsurance

NASA Astrophysics Data System (ADS)

Balashov, R. B.; Kryanev, A. V.; Sliva, D. E.

2017-01-01

The paper presents a mathematical model of efficient portfolio formation in the reinsurance markets. The presented approach provides the optimal ratio between the expected value of return and the risk of yield values below a certain level. The uncertainty in the return values is conditioned by use of expert evaluations and preliminary calculations, which result in expected return values and the corresponding risk levels. The proposed method allows for implementation of computationally simple schemes and algorithms for numerical calculation of the numerical structure of the efficient portfolios of reinsurance contracts of a given insurance company.

Fast and robust estimation of ophthalmic wavefront aberrations

NASA Astrophysics Data System (ADS)

Dillon, Keith

2016-12-01

Rapidly rising levels of myopia, particularly in the developing world, have led to an increased need for inexpensive and automated approaches to optometry. A simple and robust technique is provided for estimating major ophthalmic aberrations using a gradient-based wavefront sensor. The approach is based on the use of numerical calculations to produce diverse combinations of phase components, followed by Fourier transforms to calculate the coefficients. The approach does not utilize phase unwrapping nor iterative solution of inverse problems. This makes the method very fast and tolerant to image artifacts, which do not need to be detected and masked or interpolated as is needed in other techniques. These features make it a promising algorithm on which to base low-cost devices for applications that may have limited access to expert maintenance and operation.

Quantitative investigation of inappropriate regression model construction and the importance of medical statistics experts in observational medical research: a cross-sectional study.

PubMed

Nojima, Masanori; Tokunaga, Mutsumi; Nagamura, Fumitaka

2018-05-05

To investigate under what circumstances inappropriate use of 'multivariate analysis' is likely to occur and to identify the population that needs more support with medical statistics. The frequency of inappropriate regression model construction in multivariate analysis and related factors were investigated in observational medical research publications. The inappropriate algorithm of using only variables that were significant in univariate analysis was estimated to occur at 6.4% (95% CI 4.8% to 8.5%). This was observed in 1.1% of the publications with a medical statistics expert (hereinafter 'expert') as the first author, 3.5% if an expert was included as coauthor and in 12.2% if experts were not involved. In the publications where the number of cases was 50 or less and the study did not include experts, inappropriate algorithm usage was observed with a high proportion of 20.2%. The OR of the involvement of experts for this outcome was 0.28 (95% CI 0.15 to 0.53). A further, nation-level, analysis showed that the involvement of experts and the implementation of unfavourable multivariate analysis are associated at the nation-level analysis (R=-0.652). Based on the results of this study, the benefit of participation of medical statistics experts is obvious. Experts should be involved for proper confounding adjustment and interpretation of statistical models. © Article author(s) (or their employer(s) unless otherwise stated in the text of the article) 2018. All rights reserved. No commercial use is permitted unless otherwise expressly granted.

High-grade video compression of echocardiographic studies: a multicenter validation study of selected motion pictures expert groups (MPEG)-4 algorithms.

PubMed

Barbier, Paolo; Alimento, Marina; Berna, Giovanni; Celeste, Fabrizio; Gentile, Francesco; Mantero, Antonio; Montericcio, Vincenzo; Muratori, Manuela

2007-05-01

Large files produced by standard compression algorithms slow down spread of digital and tele-echocardiography. We validated echocardiographic video high-grade compression with the new Motion Pictures Expert Groups (MPEG)-4 algorithms with a multicenter study. Seven expert cardiologists blindly scored (5-point scale) 165 uncompressed and compressed 2-dimensional and color Doppler video clips, based on combined diagnostic content and image quality (uncompressed files as references). One digital video and 3 MPEG-4 algorithms (WM9, MV2, and DivX) were used, the latter at 3 compression levels (0%, 35%, and 60%). Compressed file sizes decreased from 12 to 83 MB to 0.03 to 2.3 MB (1:1051-1:26 reduction ratios). Mean SD of differences was 0.81 for intraobserver variability (uncompressed and digital video files). Compared with uncompressed files, only the DivX mean score at 35% (P = .04) and 60% (P = .001) compression was significantly reduced. At subcategory analysis, these differences were still significant for gray-scale and fundamental imaging but not for color or second harmonic tissue imaging. Original image quality, session sequence, compression grade, and bitrate were all independent determinants of mean score. Our study supports use of MPEG-4 algorithms to greatly reduce echocardiographic file sizes, thus facilitating archiving and transmission. Quality evaluation studies should account for the many independent variables that affect image quality grading.

Utilization of Expert Knowledge in a Multi-Objective Hydrologic Model Automatic Calibration Process

NASA Astrophysics Data System (ADS)

Quebbeman, J.; Park, G. H.; Carney, S.; Day, G. N.; Micheletty, P. D.

2016-12-01

Spatially distributed continuous simulation hydrologic models have a large number of parameters for potential adjustment during the calibration process. Traditional manual calibration approaches of such a modeling system is extremely laborious, which has historically motivated the use of automatic calibration procedures. With a large selection of model parameters, achieving high degrees of objective space fitness - measured with typical metrics such as Nash-Sutcliffe, Kling-Gupta, RMSE, etc. - can easily be achieved using a range of evolutionary algorithms. A concern with this approach is the high degree of compensatory calibration, with many similarly performing solutions, and yet grossly varying parameter set solutions. To help alleviate this concern, and mimic manual calibration processes, expert knowledge is proposed for inclusion within the multi-objective functions, which evaluates the parameter decision space. As a result, Pareto solutions are identified with high degrees of fitness, but also create parameter sets that maintain and utilize available expert knowledge resulting in more realistic and consistent solutions. This process was tested using the joint SNOW-17 and Sacramento Soil Moisture Accounting method (SAC-SMA) within the Animas River basin in Colorado. Three different elevation zones, each with a range of parameters, resulted in over 35 model parameters simultaneously calibrated. As a result, high degrees of fitness were achieved, in addition to the development of more realistic and consistent parameter sets such as those typically achieved during manual calibration procedures.

Improved Vote Aggregation Techniques for the Geo-Wiki Cropland Capture Crowdsourcing Game

NASA Astrophysics Data System (ADS)

Baklanov, Artem; Fritz, Steffen; Khachay, Michael; Nurmukhametov, Oleg; Salk, Carl; See, Linda; Shchepashchenko, Dmitry

2016-04-01

Crowdsourcing is a new approach for solving data processing problems for which conventional methods appear to be inaccurate, expensive, or time-consuming. Nowadays, the development of new crowdsourcing techniques is mostly motivated by so called Big Data problems, including problems of assessment and clustering for large datasets obtained in aerospace imaging, remote sensing, and even in social network analysis. By involving volunteers from all over the world, the Geo-Wiki project tackles problems of environmental monitoring with applications to flood resilience, biomass data analysis and classification of land cover. For example, the Cropland Capture Game, which is a gamified version of Geo-Wiki, was developed to aid in the mapping of cultivated land, and was used to gather 4.5 million image classifications from the Earth's surface. More recently, the Picture Pile game, which is a more generalized version of Cropland Capture, aims to identify tree loss over time from pairs of very high resolution satellite images. Despite recent progress in image analysis, the solution to these problems is hard to automate since human experts still outperform the majority of machine learning algorithms and artificial systems in this field on certain image recognition tasks. The replacement of rare and expensive experts by a team of distributed volunteers seems to be promising, but this approach leads to challenging questions such as: how can individual opinions be aggregated optimally, how can confidence bounds be obtained, and how can the unreliability of volunteers be dealt with? In this paper, on the basis of several known machine learning techniques, we propose a technical approach to improve the overall performance of the majority voting decision rule used in the Cropland Capture Game. The proposed approach increases the estimated consistency with expert opinion from 77% to 86%.

Evidence reasoning method for constructing conditional probability tables in a Bayesian network of multimorbidity.

PubMed

Du, Yuanwei; Guo, Yubin

2015-01-01

The intrinsic mechanism of multimorbidity is difficult to recognize and prediction and diagnosis are difficult to carry out accordingly. Bayesian networks can help to diagnose multimorbidity in health care, but it is difficult to obtain the conditional probability table (CPT) because of the lack of clinically statistical data. Today, expert knowledge and experience are increasingly used in training Bayesian networks in order to help predict or diagnose diseases, but the CPT in Bayesian networks is usually irrational or ineffective for ignoring realistic constraints especially in multimorbidity. In order to solve these problems, an evidence reasoning (ER) approach is employed to extract and fuse inference data from experts using a belief distribution and recursive ER algorithm, based on which evidence reasoning method for constructing conditional probability tables in Bayesian network of multimorbidity is presented step by step. A multimorbidity numerical example is used to demonstrate the method and prove its feasibility and application. Bayesian network can be determined as long as the inference assessment is inferred by each expert according to his/her knowledge or experience. Our method is more effective than existing methods for extracting expert inference data accurately and is fused effectively for constructing CPTs in a Bayesian network of multimorbidity.

An active learning approach for rapid characterization of endothelial cells in human tumors.

PubMed

Padmanabhan, Raghav K; Somasundar, Vinay H; Griffith, Sandra D; Zhu, Jianliang; Samoyedny, Drew; Tan, Kay See; Hu, Jiahao; Liao, Xuejun; Carin, Lawrence; Yoon, Sam S; Flaherty, Keith T; Dipaola, Robert S; Heitjan, Daniel F; Lal, Priti; Feldman, Michael D; Roysam, Badrinath; Lee, William M F

2014-01-01

Currently, no available pathological or molecular measures of tumor angiogenesis predict response to antiangiogenic therapies used in clinical practice. Recognizing that tumor endothelial cells (EC) and EC activation and survival signaling are the direct targets of these therapies, we sought to develop an automated platform for quantifying activity of critical signaling pathways and other biological events in EC of patient tumors by histopathology. Computer image analysis of EC in highly heterogeneous human tumors by a statistical classifier trained using examples selected by human experts performed poorly due to subjectivity and selection bias. We hypothesized that the analysis can be optimized by a more active process to aid experts in identifying informative training examples. To test this hypothesis, we incorporated a novel active learning (AL) algorithm into FARSIGHT image analysis software that aids the expert by seeking out informative examples for the operator to label. The resulting FARSIGHT-AL system identified EC with specificity and sensitivity consistently greater than 0.9 and outperformed traditional supervised classification algorithms. The system modeled individual operator preferences and generated reproducible results. Using the results of EC classification, we also quantified proliferation (Ki67) and activity in important signal transduction pathways (MAP kinase, STAT3) in immunostained human clear cell renal cell carcinoma and other tumors. FARSIGHT-AL enables characterization of EC in conventionally preserved human tumors in a more automated process suitable for testing and validating in clinical trials. The results of our study support a unique opportunity for quantifying angiogenesis in a manner that can now be tested for its ability to identify novel predictive and response biomarkers.

ABCluster: the artificial bee colony algorithm for cluster global optimization.

PubMed

Zhang, Jun; Dolg, Michael

2015-10-07

Global optimization of cluster geometries is of fundamental importance in chemistry and an interesting problem in applied mathematics. In this work, we introduce a relatively new swarm intelligence algorithm, i.e. the artificial bee colony (ABC) algorithm proposed in 2005, to this field. It is inspired by the foraging behavior of a bee colony, and only three parameters are needed to control it. We applied it to several potential functions of quite different nature, i.e., the Coulomb-Born-Mayer, Lennard-Jones, Morse, Z and Gupta potentials. The benchmarks reveal that for long-ranged potentials the ABC algorithm is very efficient in locating the global minimum, while for short-ranged ones it is sometimes trapped into a local minimum funnel on a potential energy surface of large clusters. We have released an efficient, user-friendly, and free program "ABCluster" to realize the ABC algorithm. It is a black-box program for non-experts as well as experts and might become a useful tool for chemists to study clusters.

Diversified models for portfolio selection based on uncertain semivariance

NASA Astrophysics Data System (ADS)

Chen, Lin; Peng, Jin; Zhang, Bo; Rosyida, Isnaini

2017-02-01

Since the financial markets are complex, sometimes the future security returns are represented mainly based on experts' estimations due to lack of historical data. This paper proposes a semivariance method for diversified portfolio selection, in which the security returns are given subjective to experts' estimations and depicted as uncertain variables. In the paper, three properties of the semivariance of uncertain variables are verified. Based on the concept of semivariance of uncertain variables, two types of mean-semivariance diversified models for uncertain portfolio selection are proposed. Since the models are complex, a hybrid intelligent algorithm which is based on 99-method and genetic algorithm is designed to solve the models. In this hybrid intelligent algorithm, 99-method is applied to compute the expected value and semivariance of uncertain variables, and genetic algorithm is employed to seek the best allocation plan for portfolio selection. At last, several numerical examples are presented to illustrate the modelling idea and the effectiveness of the algorithm.

Process Mining for Individualized Behavior Modeling Using Wireless Tracking in Nursing Homes

PubMed Central

Fernández-Llatas, Carlos; Benedi, José-Miguel; García-Gómez, Juan M.; Traver, Vicente

2013-01-01

The analysis of human behavior patterns is increasingly used for several research fields. The individualized modeling of behavior using classical techniques requires too much time and resources to be effective. A possible solution would be the use of pattern recognition techniques to automatically infer models to allow experts to understand individual behavior. However, traditional pattern recognition algorithms infer models that are not readily understood by human experts. This limits the capacity to benefit from the inferred models. Process mining technologies can infer models as workflows, specifically designed to be understood by experts, enabling them to detect specific behavior patterns in users. In this paper, the eMotiva process mining algorithms are presented. These algorithms filter, infer and visualize workflows. The workflows are inferred from the samples produced by an indoor location system that stores the location of a resident in a nursing home. The visualization tool is able to compare and highlight behavior patterns in order to facilitate expert understanding of human behavior. This tool was tested with nine real users that were monitored for a 25-week period. The results achieved suggest that the behavior of users is continuously evolving and changing and that this change can be measured, allowing for behavioral change detection. PMID:24225907

The ranking algorithm of the Coach browser for the UMLS metathesaurus.

PubMed Central

Harbourt, A. M.; Syed, E. J.; Hole, W. T.; Kingsland, L. C.

1993-01-01

This paper presents the novel ranking algorithm of the Coach Metathesaurus browser which is a major module of the Coach expert search refinement program. An example shows how the ranking algorithm can assist in creating a list of candidate terms useful in augmenting a suboptimal Grateful Med search of MEDLINE. PMID:8130570

Reaching consensus on the physiotherapeutic management of patients following upper abdominal surgery: a pragmatic approach to interpret equivocal evidence.

PubMed

Hanekom, Susan D; Brooks, Dina; Denehy, Linda; Fagevik-Olsén, Monika; Hardcastle, Timothy C; Manie, Shamila; Louw, Quinette

2012-02-06

Postoperative pulmonary complications remain the most significant cause of morbidity following open upper abdominal surgery despite advances in perioperative care. However, due to the poor quality primary research uncertainty surrounding the value of prophylactic physiotherapy intervention in the management of patients following abdominal surgery persists. The Delphi process has been proposed as a pragmatic methodology to guide clinical practice when evidence is equivocal. The objective was to develop a clinical management algorithm for the post operative management of abdominal surgery patients. Eleven draft algorithm statements extracted from the extant literature by the primary research team were verified and rated by scientist clinicians (n=5) in an electronic three round Delphi process. Algorithm statements which reached a priori defined consensus-semi-interquartile range (SIQR)<0.5-were collated into the algorithm. The five panelists allocated to the abdominal surgery Delphi panel were from Australia, Canada, Sweden, and South Africa. The 11 draft algorithm statements were edited and 5 additional statements were formulated. The panel reached consensus on the rating of all statements. Four statements were rated essential. An expert Delphi panel interpreted the equivocal evidence for the physiotherapeutic management of patients following upper abdominal surgery. Through a process of consensus a clinical management algorithm was formulated. This algorithm can now be used by clinicians to guide clinical practice in this population.

MMKG: An approach to generate metallic materials knowledge graph based on DBpedia and Wikipedia

NASA Astrophysics Data System (ADS)

Zhang, Xiaoming; Liu, Xin; Li, Xin; Pan, Dongyu

2017-02-01

The research and development of metallic materials are playing an important role in today's society, and in the meanwhile lots of metallic materials knowledge is generated and available on the Web (e.g., Wikipedia) for materials experts. However, due to the diversity and complexity of metallic materials knowledge, the knowledge utilization may encounter much inconvenience. The idea of knowledge graph (e.g., DBpedia) provides a good way to organize the knowledge into a comprehensive entity network. Therefore, the motivation of our work is to generate a metallic materials knowledge graph (MMKG) using available knowledge on the Web. In this paper, an approach is proposed to build MMKG based on DBpedia and Wikipedia. First, we use an algorithm based on directly linked sub-graph semantic distance (DLSSD) to preliminarily extract metallic materials entities from DBpedia according to some predefined seed entities; then based on the results of the preliminary extraction, we use an algorithm, which considers both semantic distance and string similarity (SDSS), to achieve the further extraction. Second, due to the absence of materials properties in DBpedia, we use an ontology-based method to extract properties knowledge from the HTML tables of corresponding Wikipedia Web pages for enriching MMKG. Materials ontology is used to locate materials properties tables as well as to identify the structure of the tables. The proposed approach is evaluated by precision, recall, F1 and time performance, and meanwhile the appropriate thresholds for the algorithms in our approach are determined through experiments. The experimental results show that our approach returns expected performance. A tool prototype is also designed to facilitate the process of building the MMKG as well as to demonstrate the effectiveness of our approach.

Classification of microscopy images of Langerhans islets

NASA Astrophysics Data System (ADS)

Å vihlík, Jan; Kybic, Jan; Habart, David; Berková, Zuzana; Girman, Peter; Kříž, Jan; Zacharovová, Klára

2014-03-01

Evaluation of images of Langerhans islets is a crucial procedure for planning an islet transplantation, which is a promising diabetes treatment. This paper deals with segmentation of microscopy images of Langerhans islets and evaluation of islet parameters such as area, diameter, or volume (IE). For all the available images, the ground truth and the islet parameters were independently evaluated by four medical experts. We use a pixelwise linear classifier (perceptron algorithm) and SVM (support vector machine) for image segmentation. The volume is estimated based on circle or ellipse fitting to individual islets. The segmentations were compared with the corresponding ground truth. Quantitative islet parameters were also evaluated and compared with parameters given by medical experts. We can conclude that accuracy of the presented fully automatic algorithm is fully comparable with medical experts.

A bioinformatics expert system linking functional data to anatomical outcomes in limb regeneration

PubMed Central

Lobo, Daniel; Feldman, Erica B.; Shah, Michelle; Malone, Taylor J.

2014-01-01

Abstract Amphibians and molting arthropods have the remarkable capacity to regenerate amputated limbs, as described by an extensive literature of experimental cuts, amputations, grafts, and molecular techniques. Despite a rich history of experimental effort, no comprehensive mechanistic model exists that can account for the pattern regulation observed in these experiments. While bioinformatics algorithms have revolutionized the study of signaling pathways, no such tools have heretofore been available to assist scientists in formulating testable models of large‐scale morphogenesis that match published data in the limb regeneration field. Major barriers to preventing an algorithmic approach are the lack of formal descriptions for experimental regenerative information and a repository to centralize storage and mining of functional data on limb regeneration. Establishing a new bioinformatics of shape would significantly accelerate the discovery of key insights into the mechanisms that implement complex regeneration. Here, we describe a novel mathematical ontology for limb regeneration to unambiguously encode phenotype, manipulation, and experiment data. Based on this formalism, we present the first centralized formal database of published limb regeneration experiments together with a user‐friendly expert system tool to facilitate its access and mining. These resources are freely available for the community and will assist both human biologists and artificial intelligence systems to discover testable, mechanistic models of limb regeneration. PMID:25729585

Geodesic Distance Algorithm for Extracting the Ascending Aorta from 3D CT Images

PubMed Central

Jang, Yeonggul; Jung, Ho Yub; Hong, Youngtaek; Cho, Iksung; Shim, Hackjoon; Chang, Hyuk-Jae

2016-01-01

This paper presents a method for the automatic 3D segmentation of the ascending aorta from coronary computed tomography angiography (CCTA). The segmentation is performed in three steps. First, the initial seed points are selected by minimizing a newly proposed energy function across the Hough circles. Second, the ascending aorta is segmented by geodesic distance transformation. Third, the seed points are effectively transferred through the next axial slice by a novel transfer function. Experiments are performed using a database composed of 10 patients' CCTA images. For the experiment, the ground truths are annotated manually on the axial image slices by a medical expert. A comparative evaluation with state-of-the-art commercial aorta segmentation algorithms shows that our approach is computationally more efficient and accurate under the DSC (Dice Similarity Coefficient) measurements. PMID:26904151

A fall prevention guideline for older adults living in long-term care facilities.

PubMed

Jung, D; Shin, S; Kim, H

2014-12-01

Falls are among the most frequent critical health problems for older adults over 65 years of age and often result in consequential injuries. This study developed a guideline covering risk factors and interventions for falls in order to prevent them from occurring in long-term care facilities. This study was grounded in the methodological approach of the Scottish Intercollegiate Guideline Network for establishing evidence-based guidelines: (1) establishment of the target population and scope of the guideline, (2) systematic literature review and critical analysis, (3) determination of the recommendation grade, (4) development of a draft nursing intervention guideline and algorithm, (5) expert evaluation of the draft nursing intervention guideline, and (6) confirmation of the final intervention guideline and completion of the algorithm. The resulting evidence-based fall prevention guideline consists of a three-step factor assessment and a three-step intervention approach. The resulting guideline was based on the literature and clinical experts. Further research is required to test the guideline's feasibility in across long term care facilities. This guideline can be used by nurses to screen patients who are at a high risk of falling to provide patient interventions to help prevent falls. Considering the high rate of falls at long-term care facilities and the absence of evidence-based guidelines to prevent them, additional studies on falls at long-term care facilities are necessary. Meanwhile, given prior research that indicates the importance of human resources in the application of such guidelines, continuous investigations are needed as to whether the research outcomes are actually conveyed to nurses. © 2014 International Council of Nurses.

Evaluation of a research diagnostic algorithm for DSM-5 neurocognitive disorders in a population-based cohort of older adults.

PubMed

Eramudugolla, Ranmalee; Mortby, Moyra E; Sachdev, Perminder; Meslin, Chantal; Kumar, Rajeev; Anstey, Kaarin J

2017-03-04

There is little information on the application and impact of revised criteria for diagnosing dementia and mild cognitive impairment (MCI), now termed major and mild neurocognitive disorders (NCDs) in the DSM-5. We evaluate a psychometric algorithm for diagnosing DSM-5 NCDs in a community-dwelling sample, and characterize the neuropsychological and functional profile of expert-diagnosed DSM-5 NCDs relative to DSM-IV dementia and International Working Group criteria for MCI. A population-based sample of 1644 adults aged 72-78 years was assessed. Algorithmic diagnostic criteria used detailed neuropsychological data, medical history, longitudinal cognitive performance, and informant interview. Those meeting all criteria for at least one diagnosis had data reviewed by a neurologist (expert diagnosis) who achieved consensus with a psychiatrist for complex cases. The algorithm accurately classified DSM-5 major NCD (area under the curve (AUC) = 0.95, 95% confidence interval (CI) 0.92-0.97), DSM-IV dementia (AUC = 0.91, 95% CI 0.85-0.97), DSM-5 mild NCD (AUC = 0.75, 95% CI 0.70-0.80), and MCI (AUC = 0.76, 95% CI 0.72-0.81) when compared to expert diagnosis. Expert diagnosis of dementia using DSM-5 criteria overlapped with 90% of DSM-IV dementia cases, but resulted in a 127% increase in diagnosis relative to DSM-IV. Additional cases had less severe memory, language impairment, and instrumental activities of daily living (IADL) impairments compared to cases meeting DSM-IV criteria for dementia. DSM-5 mild NCD overlapped with 83% of MCI cases and resulted in a 19% increase in diagnosis. These additional cases had a subtly different neurocognitive profile to MCI cases, including poorer social cognition. DSM-5 NCD criteria can be operationalized in a psychometric algorithm in a population setting. Expert diagnosis using DSM-5 NCD criteria captured most cases with DSM-IV dementia and MCI in our sample, but included many additional cases suggesting that DSM-5 criteria are broader in their categorization.

Performance optimization of an online retailer by a unique online resilience engineering algorithm

NASA Astrophysics Data System (ADS)

Azadeh, A.; Salehi, V.; Salehi, R.; Hassani, S. M.

2018-03-01

Online shopping has become more attractive and competitive in electronic markets. Resilience engineering (RE) can help such systems divert to the normal state in case of encountering unexpected events. This study presents a unique online resilience engineering (ORE) approach for online shopping systems and customer service performance. Moreover, this study presents a new ORE algorithm for the performance optimisation of an actual online shopping system. The data are collected by standard questionnaires from both expert employees and customers. The problem is then formulated mathematically using data envelopment analysis (DEA). The results show that the design process which is based on ORE is more efficient than the conventional design approach. Moreover, on-time delivery is the most important factor from the personnel's perspective. In addition, according to customers' view, trust, security and good quality assurance are the most effective factors during transactions. This is the first study that introduces ORE for electronic markets. Second, it investigates impact of RE on online shopping through DEA and statistical methods. Third, a practical approach is employed in this study and it may be used for similar online shops. Fourth, the results are verified and validated through complete sensitivity analysis.

Microcomputer-based classification of environmental data in municipal areas

NASA Astrophysics Data System (ADS)

Thiergärtner, H.

1995-10-01

Multivariate data-processing methods used in mineral resource identification can be used to classify urban regions. Using elements of expert systems, geographical information systems, as well as known classification and prognosis systems, it is possible to outline a single model that consists of resistant and of temporary parts of a knowledge base including graphical input and output treatment and of resistant and temporary elements of a bank of methods and algorithms. Whereas decision rules created by experts will be stored in expert systems directly, powerful classification rules in form of resistant but latent (implicit) decision algorithms may be implemented in the suggested model. The latent functions will be transformed into temporary explicit decision rules by learning processes depending on the actual task(s), parameter set(s), pixels selection(s), and expert control(s). This takes place both at supervised and nonsupervised classification of multivariately described pixel sets representing municipal subareas. The model is outlined briefly and illustrated by results obtained in a target area covering a part of the city of Berlin (Germany).

Use of a machine learning algorithm to classify expertise: analysis of hand motion patterns during a simulated surgical task.

PubMed

Watson, Robert A

2014-08-01

To test the hypothesis that machine learning algorithms increase the predictive power to classify surgical expertise using surgeons' hand motion patterns. In 2012 at the University of North Carolina at Chapel Hill, 14 surgical attendings and 10 first- and second-year surgical residents each performed two bench model venous anastomoses. During the simulated tasks, the participants wore an inertial measurement unit on the dorsum of their dominant (right) hand to capture their hand motion patterns. The pattern from each bench model task performed was preprocessed into a symbolic time series and labeled as expert (attending) or novice (resident). The labeled hand motion patterns were processed and used to train a Support Vector Machine (SVM) classification algorithm. The trained algorithm was then tested for discriminative/predictive power against unlabeled (blinded) hand motion patterns from tasks not used in the training. The Lempel-Ziv (LZ) complexity metric was also measured from each hand motion pattern, with an optimal threshold calculated to separately classify the patterns. The LZ metric classified unlabeled (blinded) hand motion patterns into expert and novice groups with an accuracy of 70% (sensitivity 64%, specificity 80%). The SVM algorithm had an accuracy of 83% (sensitivity 86%, specificity 80%). The results confirmed the hypothesis. The SVM algorithm increased the predictive power to classify blinded surgical hand motion patterns into expert versus novice groups. With further development, the system used in this study could become a viable tool for low-cost, objective assessment of procedural proficiency in a competency-based curriculum.

Retinopathy of Prematurity-assist: Novel Software for Detecting Plus Disease

PubMed Central

Pour, Elias Khalili; Pourreza, Hamidreza; Zamani, Kambiz Ameli; Mahmoudi, Alireza; Sadeghi, Arash Mir Mohammad; Shadravan, Mahla; Karkhaneh, Reza; Pour, Ramak Rouhi

2017-01-01

Purpose To design software with a novel algorithm, which analyzes the tortuosity and vascular dilatation in fundal images of retinopathy of prematurity (ROP) patients with an acceptable accuracy for detecting plus disease. Methods Eighty-seven well-focused fundal images taken with RetCam were classified to three groups of plus, non-plus, and pre-plus by agreement between three ROP experts. Automated algorithms in this study were designed based on two methods: the curvature measure and distance transform for assessment of tortuosity and vascular dilatation, respectively as two major parameters of plus disease detection. Results Thirty-eight plus, 12 pre-plus, and 37 non-plus images, which were classified by three experts, were tested by an automated algorithm and software evaluated the correct grouping of images in comparison to expert voting with three different classifiers, k-nearest neighbor, support vector machine and multilayer perceptron network. The plus, pre-plus, and non-plus images were analyzed with 72.3%, 83.7%, and 84.4% accuracy, respectively. Conclusions The new automated algorithm used in this pilot scheme for diagnosis and screening of patients with plus ROP has acceptable accuracy. With more improvements, it may become particularly useful, especially in centers without a skilled person in the ROP field. PMID:29022295

Comparison of Decision Assist and Clinical Judgment of Experts for Prediction of Lifesaving Interventions

DTIC Science & Technology

2015-03-01

min of pulse oximeter photopletysmograph waveforms and extracted features to predict LSIs. We compared this with clinical judgment of LSIs by...Curve (AUROC). We obtained clinical judgment of need for LSI from 405 expert clinicians in135 trauma patients. The pulse oximeter algorithm...15 min of pulse oximeter waveforms predicts the need for LSIs during initial trauma resuscitation as accurately as judgment of expert trauma

A Novel Approach to ECG Classification Based upon Two-Layered HMMs in Body Sensor Networks

PubMed Central

Liang, Wei; Zhang, Yinlong; Tan, Jindong; Li, Yang

2014-01-01

This paper presents a novel approach to ECG signal filtering and classification. Unlike the traditional techniques which aim at collecting and processing the ECG signals with the patient being still, lying in bed in hospitals, our proposed algorithm is intentionally designed for monitoring and classifying the patient's ECG signals in the free-living environment. The patients are equipped with wearable ambulatory devices the whole day, which facilitates the real-time heart attack detection. In ECG preprocessing, an integral-coefficient-band-stop (ICBS) filter is applied, which omits time-consuming floating-point computations. In addition, two-layered Hidden Markov Models (HMMs) are applied to achieve ECG feature extraction and classification. The periodic ECG waveforms are segmented into ISO intervals, P subwave, QRS complex and T subwave respectively in the first HMM layer where expert-annotation assisted Baum-Welch algorithm is utilized in HMM modeling. Then the corresponding interval features are selected and applied to categorize the ECG into normal type or abnormal type (PVC, APC) in the second HMM layer. For verifying the effectiveness of our algorithm on abnormal signal detection, we have developed an ECG body sensor network (BSN) platform, whereby real-time ECG signals are collected, transmitted, displayed and the corresponding classification outcomes are deduced and shown on the BSN screen. PMID:24681668

Tuberculosis-Diagnostic Expert System: an architecture for translating patients information from the web for use in tuberculosis diagnosis.

PubMed

Osamor, Victor C; Azeta, Ambrose A; Ajulo, Oluseyi O

2014-12-01

Over 1.5-2 million tuberculosis deaths occur annually. Medical professionals are faced with a lot of challenges in delivering good health-care with unassisted automation in hospitals where there are several patients who need the doctor's attention. To automate the pre-laboratory screening process against tuberculosis infection to aid diagnosis and make it fast and accessible to the public via the Internet. The expert system we have built is designed to also take care of people who do not have access to medical experts, but would want to check their medical status. A rule-based approach has been used, and unified modeling language and the client-server architecture technique were applied to model the system and to develop it as a web-based expert system for tuberculosis diagnosis. Algorithmic rules in the Tuberculosis-Diagnosis Expert System necessitate decision coverage where tuberculosis is either suspected or not suspected. The architecture consists of a rule base, knowledge base, and patient database. These units interact with the inference engine, which receives patient' data through the Internet via a user interface. We present the architecture of the Tuberculosis-Diagnosis Expert System and its implementation. We evaluated it for usability to determine the level of effectiveness, efficiency and user satisfaction. The result of the usability evaluation reveals that the system has a usability of 4.08 out of a scale of 5. This is an indication of a more-than-average system performance. Several existing expert systems have been developed for the purpose of supporting different medical diagnoses, but none is designed to translate tuberculosis patients' symptomatic data for online pre-laboratory screening. Our Tuberculosis-Diagnosis Expert System is an effective solution for the implementation of the needed web-based expert system diagnosis. © The Author(s) 2013.

Imaging spectroscopy: Earth and planetary remote sensing with the USGS Tetracorder and expert systems

USGS Publications Warehouse

Clark, Roger N.; Swayze, Gregg A.; Livo, K. Eric; Kokaly, Raymond F.; Sutley, Steve J.; Dalton, J. Brad; McDougal, Robert R.; Gent, Carol A.

2003-01-01

Imaging spectroscopy is a tool that can be used to spectrally identify and spatially map materials based on their specific chemical bonds. Spectroscopic analysis requires significantly more sophistication than has been employed in conventional broadband remote sensing analysis. We describe a new system that is effective at material identification and mapping: a set of algorithms within an expert system decision‐making framework that we call Tetracorder. The expertise in the system has been derived from scientific knowledge of spectral identification. The expert system rules are implemented in a decision tree where multiple algorithms are applied to spectral analysis, additional expert rules and algorithms can be applied based on initial results, and more decisions are made until spectral analysis is complete. Because certain spectral features are indicative of specific chemical bonds in materials, the system can accurately identify and map those materials. In this paper we describe the framework of the decision making process used for spectral identification, describe specific spectral feature analysis algorithms, and give examples of what analyses and types of maps are possible with imaging spectroscopy data. We also present the expert system rules that describe which diagnostic spectral features are used in the decision making process for a set of spectra of minerals and other common materials. We demonstrate the applications of Tetracorder to identify and map surface minerals, to detect sources of acid rock drainage, and to map vegetation species, ice, melting snow, water, and water pollution, all with one set of expert system rules. Mineral mapping can aid in geologic mapping and fault detection and can provide a better understanding of weathering, mineralization, hydrothermal alteration, and other geologic processes. Environmental site assessment, such as mapping source areas of acid mine drainage, has resulted in the acceleration of site cleanup, saving millions of dollars and years in cleanup time. Imaging spectroscopy data and Tetracorder analysis can be used to study both terrestrial and planetary science problems. Imaging spectroscopy can be used to probe planetary systems, including their atmospheres, oceans, and land surfaces.

Constructing Benchmark Databases and Protocols for Medical Image Analysis: Diabetic Retinopathy

PubMed Central

Kauppi, Tomi; Kämäräinen, Joni-Kristian; Kalesnykiene, Valentina; Sorri, Iiris; Uusitalo, Hannu; Kälviäinen, Heikki

2013-01-01

We address the performance evaluation practices for developing medical image analysis methods, in particular, how to establish and share databases of medical images with verified ground truth and solid evaluation protocols. Such databases support the development of better algorithms, execution of profound method comparisons, and, consequently, technology transfer from research laboratories to clinical practice. For this purpose, we propose a framework consisting of reusable methods and tools for the laborious task of constructing a benchmark database. We provide a software tool for medical image annotation helping to collect class label, spatial span, and expert's confidence on lesions and a method to appropriately combine the manual segmentations from multiple experts. The tool and all necessary functionality for method evaluation are provided as public software packages. As a case study, we utilized the framework and tools to establish the DiaRetDB1 V2.1 database for benchmarking diabetic retinopathy detection algorithms. The database contains a set of retinal images, ground truth based on information from multiple experts, and a baseline algorithm for the detection of retinopathy lesions. PMID:23956787

Psychophysiological Approach To Support Of Cosmonauts Performance Reliability

NASA Astrophysics Data System (ADS)

Nechaev, A. P.

Space flight factors may effect negatively on psychophysiological state (PPS) and work capacity of crewmembers, lead to errors in tasks performance. Such errors cannot be entirely prevented during "crew-spacecraft" system designing and/or crewmembers training as they are consequence of a human "psychophysiological troubles". In the present investigation we paid special attention to work and rest schedule (WRS) intensity due to sleep-wake rhythm disturbances and crewmembers overload may cause PPS aggravation. By methods of the correlation analysis of the data collected during 9 "Mir" missions (18 Russian cosmonauts, 226 flight weeks) the significant (p<0.05) interrelations between WRS intensity and cosmonauts PPS as well as between cosmonauts PPS and crewmembers errors (CE) frequency have been established. It allows to consider WRS intensity as "the controlling factor" in relation to crewmember's PPS. Quantitative characteristics of these interrelations have been also determined. This findings have been used for development of algorithm of crewmember's PPS management with the purpose of decrease in CE frequency. The algorithm can be briefly described in the following kind. On the basis of the analysis of crew forthcoming work experts should assess WRS intensity, possible crewmembers PPS alterations, and expected CE frequency. If CE frequency is allowable the management may be limited to recommendations that operative vigilance increase. If CE frequency is higher allowable it is necessary to carry out the measures on crewmembers PPS normalization by means of decrease in WRS intensity. The algorithm is intended for use in the on-ground expert system of psychophysiological support of cosmonauts performance.

French national consensus clinical guidelines for the management of ulcerative colitis.

PubMed

Peyrin-Biroulet, Laurent; Bouhnik, Yoram; Roblin, Xavier; Bonnaud, Guillaume; Hagège, Hervé; Hébuterne, Xavier

2016-07-01

Ulcerative colitis (UC) is a chronic inflammatory bowel disease of multifactorial etiology that primarily affects the colonic mucosa. The disease progresses over time, and clinical management guidelines should reflect its dynamic nature. There is limited evidence supporting UC management in specific clinical situations, thus precluding an evidence-based approach. To use a formal consensus method - the nominal group technique (NGT) - to develop a clinical practice expert opinion to outline simple algorithms and practices, optimize UC management, and assist clinicians in making treatment decisions. The consensus was developed by an expert panel of 37 gastroenterologists from various professional organizations with experience in UC management using the qualitative and iterative NGT, incorporating deliberations based on the European Crohn's and Colitis Organisation recommendations, recent reviews of scientific literature, and pertinent discussion topics developed by a steering committee. Examples of clinical cases for which there are limited evidence-based data from clinical trials were used. Two working groups proposed and voted on treatment algorithms that were then discussed and voted for by the nominal group as a whole, in order to reach a consensus. A clinical practice guideline covering management of the following clinical situations was developed: (i) moderate and severe UC; (ii) acute severe UC; (iii) pouchitis; (iv) refractory proctitis, in the form of treatment algorithms. Given the limited available evidence-based data, a formal consensus methodology was used to develop simple treatment guidelines for UC management in different clinical situations that is now accessible via an online application. Copyright © 2016 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved.

Big Data GPU-Driven Parallel Processing Spatial and Spatio-Temporal Clustering Algorithms

NASA Astrophysics Data System (ADS)

Konstantaras, Antonios; Skounakis, Emmanouil; Kilty, James-Alexander; Frantzeskakis, Theofanis; Maravelakis, Emmanuel

2016-04-01

Advances in graphics processing units' technology towards encompassing parallel architectures [1], comprised of thousands of cores and multiples of parallel threads, provide the foundation in terms of hardware for the rapid processing of various parallel applications regarding seismic big data analysis. Seismic data are normally stored as collections of vectors in massive matrices, growing rapidly in size as wider areas are covered, denser recording networks are being established and decades of data are being compiled together [2]. Yet, many processes regarding seismic data analysis are performed on each seismic event independently or as distinct tiles [3] of specific grouped seismic events within a much larger data set. Such processes, independent of one another can be performed in parallel narrowing down processing times drastically [1,3]. This research work presents the development and implementation of three parallel processing algorithms using Cuda C [4] for the investigation of potentially distinct seismic regions [5,6] present in the vicinity of the southern Hellenic seismic arc. The algorithms, programmed and executed in parallel comparatively, are the: fuzzy k-means clustering with expert knowledge [7] in assigning overall clusters' number; density-based clustering [8]; and a selves-developed spatio-temporal clustering algorithm encompassing expert [9] and empirical knowledge [10] for the specific area under investigation. Indexing terms: GPU parallel programming, Cuda C, heterogeneous processing, distinct seismic regions, parallel clustering algorithms, spatio-temporal clustering References [1] Kirk, D. and Hwu, W.: 'Programming massively parallel processors - A hands-on approach', 2nd Edition, Morgan Kaufman Publisher, 2013 [2] Konstantaras, A., Valianatos, F., Varley, M.R. and Makris, J.P.: 'Soft-Computing Modelling of Seismicity in the Southern Hellenic Arc', Geoscience and Remote Sensing Letters, vol. 5 (3), pp. 323-327, 2008 [3] Papadakis, S. and Diamantaras, K.: 'Programming and architecture of parallel processing systems', 1st Edition, Eds. Kleidarithmos, 2011 [4] NVIDIA.: 'NVidia CUDA C Programming Guide', version 5.0, NVidia (reference book) [5] Konstantaras, A.: 'Classification of Distinct Seismic Regions and Regional Temporal Modelling of Seismicity in the Vicinity of the Hellenic Seismic Arc', IEEE Selected Topics in Applied Earth Observations and Remote Sensing, vol. 6 (4), pp. 1857-1863, 2013 [6] Konstantaras, A. Varley, M.R.,. Valianatos, F., Collins, G. and Holifield, P.: 'Recognition of electric earthquake precursors using neuro-fuzzy models: methodology and simulation results', Proc. IASTED International Conference on Signal Processing Pattern Recognition and Applications (SPPRA 2002), Crete, Greece, 2002, pp 303-308, 2002 [7] Konstantaras, A., Katsifarakis, E., Maravelakis, E., Skounakis, E., Kokkinos, E. and Karapidakis, E.: 'Intelligent Spatial-Clustering of Seismicity in the Vicinity of the Hellenic Seismic Arc', Earth Science Research, vol. 1 (2), pp. 1-10, 2012 [8] Georgoulas, G., Konstantaras, A., Katsifarakis, E., Stylios, C.D., Maravelakis, E. and Vachtsevanos, G.: '"Seismic-Mass" Density-based Algorithm for Spatio-Temporal Clustering', Expert Systems with Applications, vol. 40 (10), pp. 4183-4189, 2013 [9] Konstantaras, A. J.: 'Expert knowledge-based algorithm for the dynamic discrimination of interactive natural clusters', Earth Science Informatics, 2015 (In Press, see: www.scopus.com) [10] Drakatos, G. and Latoussakis, J.: 'A catalog of aftershock sequences in Greece (1971-1997): Their spatial and temporal characteristics', Journal of Seismology, vol. 5, pp. 137-145, 2001

Assimilation of a knowledge base and physical models to reduce errors in passive-microwave classifications of sea ice

NASA Technical Reports Server (NTRS)

Maslanik, J. A.; Key, J.

1992-01-01

An expert system framework has been developed to classify sea ice types using satellite passive microwave data, an operational classification algorithm, spatial and temporal information, ice types estimated from a dynamic-thermodynamic model, output from a neural network that detects the onset of melt, and knowledge about season and region. The rule base imposes boundary conditions upon the ice classification, modifies parameters in the ice algorithm, determines a `confidence' measure for the classified data, and under certain conditions, replaces the algorithm output with model output. Results demonstrate the potential power of such a system for minimizing overall error in the classification and for providing non-expert data users with a means of assessing the usefulness of the classification results for their applications.

A new automatic algorithm for quantification of myocardial infarction imaged by late gadolinium enhancement cardiovascular magnetic resonance: experimental validation and comparison to expert delineations in multi-center, multi-vendor patient data.

PubMed

Engblom, Henrik; Tufvesson, Jane; Jablonowski, Robert; Carlsson, Marcus; Aletras, Anthony H; Hoffmann, Pavel; Jacquier, Alexis; Kober, Frank; Metzler, Bernhard; Erlinge, David; Atar, Dan; Arheden, Håkan; Heiberg, Einar

2016-05-04

Late gadolinium enhancement (LGE) cardiovascular magnetic resonance (CMR) using magnitude inversion recovery (IR) or phase sensitive inversion recovery (PSIR) has become clinical standard for assessment of myocardial infarction (MI). However, there is no clinical standard for quantification of MI even though multiple methods have been proposed. Simple thresholds have yielded varying results and advanced algorithms have only been validated in single center studies. Therefore, the aim of this study was to develop an automatic algorithm for MI quantification in IR and PSIR LGE images and to validate the new algorithm experimentally and compare it to expert delineations in multi-center, multi-vendor patient data. The new automatic algorithm, EWA (Expectation Maximization, weighted intensity, a priori information), was implemented using an intensity threshold by Expectation Maximization (EM) and a weighted summation to account for partial volume effects. The EWA algorithm was validated in-vivo against triphenyltetrazolium-chloride (TTC) staining (n = 7 pigs with paired IR and PSIR images) and against ex-vivo high resolution T1-weighted images (n = 23 IR and n = 13 PSIR images). The EWA algorithm was also compared to expert delineation in 124 patients from multi-center, multi-vendor clinical trials 2-6 days following first time ST-elevation myocardial infarction (STEMI) treated with percutaneous coronary intervention (PCI) (n = 124 IR and n = 49 PSIR images). Infarct size by the EWA algorithm in vivo in pigs showed a bias to ex-vivo TTC of -1 ± 4%LVM (R = 0.84) in IR and -2 ± 3%LVM (R = 0.92) in PSIR images and a bias to ex-vivo T1-weighted images of 0 ± 4%LVM (R = 0.94) in IR and 0 ± 5%LVM (R = 0.79) in PSIR images. In multi-center patient studies, infarct size by the EWA algorithm showed a bias to expert delineation of -2 ± 6 %LVM (R = 0.81) in IR images (n = 124) and 0 ± 5%LVM (R = 0.89) in PSIR images (n = 49). The EWA algorithm was validated experimentally and in patient data with a low bias in both IR and PSIR LGE images. Thus, the use of EM and a weighted intensity as in the EWA algorithm, may serve as a clinical standard for the quantification of myocardial infarction in LGE CMR images. CHILL-MI: NCT01379261 . NCT01374321 .

Towards an Online Seizure Advisory System-An Adaptive Seizure Prediction Framework Using Active Learning Heuristics.

PubMed

Karuppiah Ramachandran, Vignesh Raja; Alblas, Huibert J; Le, Duc V; Meratnia, Nirvana

2018-05-24

In the last decade, seizure prediction systems have gained a lot of attention because of their enormous potential to largely improve the quality-of-life of the epileptic patients. The accuracy of the prediction algorithms to detect seizure in real-world applications is largely limited because the brain signals are inherently uncertain and affected by various factors, such as environment, age, drug intake, etc., in addition to the internal artefacts that occur during the process of recording the brain signals. To deal with such ambiguity, researchers transitionally use active learning, which selects the ambiguous data to be annotated by an expert and updates the classification model dynamically. However, selecting the particular data from a pool of large ambiguous datasets to be labelled by an expert is still a challenging problem. In this paper, we propose an active learning-based prediction framework that aims to improve the accuracy of the prediction with a minimum number of labelled data. The core technique of our framework is employing the Bernoulli-Gaussian Mixture model (BGMM) to determine the feature samples that have the most ambiguity to be annotated by an expert. By doing so, our approach facilitates expert intervention as well as increasing medical reliability. We evaluate seven different classifiers in terms of the classification time and memory required. An active learning framework built on top of the best performing classifier is evaluated in terms of required annotation effort to achieve a high level of prediction accuracy. The results show that our approach can achieve the same accuracy as a Support Vector Machine (SVM) classifier using only 20 % of the labelled data and also improve the prediction accuracy even under the noisy condition.

Probabilistic Structural Analysis Methods (PSAM) for select space propulsion system components

NASA Technical Reports Server (NTRS)

1991-01-01

This annual report summarizes the work completed during the third year of technical effort on the referenced contract. Principal developments continue to focus on the Probabilistic Finite Element Method (PFEM) which has been under development for three years. Essentially all of the linear capabilities within the PFEM code are in place. Major progress in the application or verifications phase was achieved. An EXPERT module architecture was designed and partially implemented. EXPERT is a user interface module which incorporates an expert system shell for the implementation of a rule-based interface utilizing the experience and expertise of the user community. The Fast Probability Integration (FPI) Algorithm continues to demonstrate outstanding performance characteristics for the integration of probability density functions for multiple variables. Additionally, an enhanced Monte Carlo simulation algorithm was developed and demonstrated for a variety of numerical strategies.

Automated EEG artifact elimination by applying machine learning algorithms to ICA-based features

NASA Astrophysics Data System (ADS)

Radüntz, Thea; Scouten, Jon; Hochmuth, Olaf; Meffert, Beate

2017-08-01

Objective. Biological and non-biological artifacts cause severe problems when dealing with electroencephalogram (EEG) recordings. Independent component analysis (ICA) is a widely used method for eliminating various artifacts from recordings. However, evaluating and classifying the calculated independent components (IC) as artifact or EEG is not fully automated at present. Approach. In this study, we propose a new approach for automated artifact elimination, which applies machine learning algorithms to ICA-based features. Main results. We compared the performance of our classifiers with the visual classification results given by experts. The best result with an accuracy rate of 95% was achieved using features obtained by range filtering of the topoplots and IC power spectra combined with an artificial neural network. Significance. Compared with the existing automated solutions, our proposed method is not limited to specific types of artifacts, electrode configurations, or number of EEG channels. The main advantages of the proposed method is that it provides an automatic, reliable, real-time capable, and practical tool, which avoids the need for the time-consuming manual selection of ICs during artifact removal.

Drought Management Activities of the National Drought Mitigation Center (NDMC): Contributions Toward a Global Drought Early Warning System (GDEWS)

NASA Astrophysics Data System (ADS)

Stumpf, A.; Lachiche, N.; Malet, J.; Kerle, N.; Puissant, A.

2011-12-01

VHR satellite images have become a primary source for landslide inventory mapping after major triggering events such as earthquakes and heavy rainfalls. Visual image interpretation is still the prevailing standard method for operational purposes but is time-consuming and not well suited to fully exploit the increasingly better supply of remote sensing data. Recent studies have addressed the development of more automated image analysis workflows for landslide inventory mapping. In particular object-oriented approaches that account for spatial and textural image information have been demonstrated to be more adequate than pixel-based classification but manually elaborated rule-based classifiers are difficult to adapt under changing scene characteristics. Machine learning algorithm allow learning classification rules for complex image patterns from labelled examples and can be adapted straightforwardly with available training data. In order to reduce the amount of costly training data active learning (AL) has evolved as a key concept to guide the sampling for many applications. The underlying idea of AL is to initialize a machine learning model with a small training set, and to subsequently exploit the model state and data structure to iteratively select the most valuable samples that should be labelled by the user. With relatively few queries and labelled samples, an AL strategy yields higher accuracies than an equivalent classifier trained with many randomly selected samples. This study addressed the development of an AL method for landslide mapping from VHR remote sensing images with special consideration of the spatial distribution of the samples. Our approach [1] is based on the Random Forest algorithm and considers the classifier uncertainty as well as the variance of potential sampling regions to guide the user towards the most valuable sampling areas. The algorithm explicitly searches for compact regions and thereby avoids a spatially disperse sampling pattern inherent to most other AL methods. The accuracy, the sampling time and the computational runtime of the algorithm were evaluated on multiple satellite images capturing recent large scale landslide events. Sampling between 1-4% of the study areas the accuracies between 74% and 80% were achieved, whereas standard sampling schemes yielded only accuracies between 28% and 50% with equal sampling costs. Compared to commonly used point-wise AL algorithm the proposed approach significantly reduces the number of iterations and hence the computational runtime. Since the user can focus on relatively few compact areas (rather than on hundreds of distributed points) the overall labeling time is reduced by more than 50% compared to point-wise queries. An experimental evaluation of multiple expert mappings demonstrated strong relationships between the uncertainties of the experts and the machine learning model. It revealed that the achieved accuracies are within the range of the inter-expert disagreement and that it will be indispensable to consider ground truth uncertainties to truly achieve further enhancements in the future. The proposed method is generally applicable to a wide range of optical satellite images and landslide types. [1] A. Stumpf, N. Lachiche, J.-P. Malet, N. Kerle, and A. Puissant, Active learning in the spatial domain for remote sensing image classification, IEEE Transactions on Geosciece and Remote Sensing. 2013, DOI 10.1109/TGRS.2013.2262052.

Detection of subtle nocturnal motor activity from 3-D accelerometry recordings in epilepsy patients.

PubMed

Nijsen, Tamara M E; Cluitmans, Pierre J M; Arends, Johan B A M; Griep, Paul A M

2007-11-01

This paper presents a first step towards reliable detection of nocturnal epileptic seizures based on 3-D accelerometry (ACM) recordings. The main goal is to distinguish between data with and without subtle nocturnal motor activity, thus reducing the amount of data that needs further (more complex) analysis for seizure detection. From 15 ACM signals (measured on five positions on the body), two features are computed, the variance and the jerk. In the resulting 2-D feature space, a linear threshold function is used for classification. For training and testing, the algorithm ACM data along with video data is used from nocturnal registrations in seven mentally retarded patients with severe epilepsy. Per patient, the algorithm detected 100% of the periods of motor activity that are marked in video recordings and the ACM signals by experts. From all the detections, 43%-89% was correct (mean =65%). We were able to reduce the amount of data that need to be analyzed considerably. The results show that our approach can be used for detection of subtle nocturnal motor activity. Furthermore, our results indicate that our algorithm is robust for fluctuations across patients. Consequently, there is no need for training the algorithm for each new patient.

The Method of Immersion the Problem of Comparing Technical Objects in an Expert Shell in the Class of Artificial Intelligence Algorithms

NASA Astrophysics Data System (ADS)

Sergey Vasilievich, Buharin; Aleksandr Vladimirovich, Melnikov; Svetlana Nikolaevna, Chernyaeva; Lyudmila Anatolievna, Korobova

2017-08-01

The method of dip of the underlying computational problem of comparing technical object in an expert shell in the class of data mining methods is examined. An example of using the proposed method is given.

METEOR - an artificial intelligence system for convective storm forecasting

DOE Office of Scientific and Technical Information (OSTI.GOV)

Elio, R.; De haan, J.; Strong, G.S.

1987-03-01

An AI system called METEOR, which uses the meteorologist's heuristics, strategies, and statistical tools to forecast severe hailstorms in Alberta, is described, emphasizing the information and knowledge that METEOR uses to mimic the forecasting procedure of an expert meteorologist. METEOR is then discussed as an AI system, emphasizing the ways in which it is qualitatively different from algorithmic or statistical approaches to prediction. Some features of METEOR's design and the AI techniques for representing meteorological knowledge and for reasoning and inference are presented. Finally, some observations on designing and implementing intelligent consultants for meteorological applications are made. 7 references.

Combining watershed and graph cuts methods to segment organs at risk in radiotherapy

NASA Astrophysics Data System (ADS)

Dolz, Jose; Kirisli, Hortense A.; Viard, Romain; Massoptier, Laurent

2014-03-01

Computer-aided segmentation of anatomical structures in medical images is a valuable tool for efficient radiation therapy planning (RTP). As delineation errors highly affect the radiation oncology treatment, it is crucial to delineate geometric structures accurately. In this paper, a semi-automatic segmentation approach for computed tomography (CT) images, based on watershed and graph-cuts methods, is presented. The watershed pre-segmentation groups small areas of similar intensities in homogeneous labels, which are subsequently used as input for the graph-cuts algorithm. This methodology does not require of prior knowledge of the structure to be segmented; even so, it performs well with complex shapes and low intensity. The presented method also allows the user to add foreground and background strokes in any of the three standard orthogonal views - axial, sagittal or coronal - making the interaction with the algorithm easy and fast. Hence, the segmentation information is propagated within the whole volume, providing a spatially coherent result. The proposed algorithm has been evaluated using 9 CT volumes, by comparing its segmentation performance over several organs - lungs, liver, spleen, heart and aorta - to those of manual delineation from experts. A Dicés coefficient higher than 0.89 was achieved in every case. That demonstrates that the proposed approach works well for all the anatomical structures analyzed. Due to the quality of the results, the introduction of the proposed approach in the RTP process will be a helpful tool for organs at risk (OARs) segmentation.

Hospital-based expert model for health technology procurement planning in hospitals.

PubMed

Miniati, R; Cecconi, G; Frosini, F; Dori, F; Regolini, J; Iadanza, E; Biffi Gentili, G

2014-01-01

Although in the last years technology innovation in healthcare brought big improvements in care level and patient quality of life, hospital complexity and management cost became higher. For this reason, necessity of planning for medical equipment procurement within hospitals is getting more and more important in order to sustainable provide appropriate technology for both routine activity and innovative procedures. In order to support hospital decision makers for technology procurement planning, an expert model was designed as reported in the following paper. It combines the most widely used approaches for technology evaluation by taking into consideration Health Technology Assessment (HTA) and Medical Equipment Replacement Model (MERM). The designing phases include a first definition of prioritization algorithms, then the weighting process through experts' interviews and a final step for the model validation that included both statistical testing and comparison with real decisions. In conclusion, the designed model was able to provide a semi-automated tool that through the use of multidisciplinary information is able to prioritize different requests of technology acquisition in hospitals. Validation outcomes improved the model accuracy and created different "user profiles" according to the specific needs of decision makers.

Stage-independent, single lead EEG sleep spindle detection using the continuous wavelet transform and local weighted smoothing.

PubMed

Tsanas, Athanasios; Clifford, Gari D

2015-01-01

Sleep spindles are critical in characterizing sleep and have been associated with cognitive function and pathophysiological assessment. Typically, their detection relies on the subjective and time-consuming visual examination of electroencephalogram (EEG) signal(s) by experts, and has led to large inter-rater variability as a result of poor definition of sleep spindle characteristics. Hitherto, many algorithmic spindle detectors inherently make signal stationarity assumptions (e.g., Fourier transform-based approaches) which are inappropriate for EEG signals, and frequently rely on additional information which may not be readily available in many practical settings (e.g., more than one EEG channels, or prior hypnogram assessment). This study proposes a novel signal processing methodology relying solely on a single EEG channel, and provides objective, accurate means toward probabilistically assessing the presence of sleep spindles in EEG signals. We use the intuitively appealing continuous wavelet transform (CWT) with a Morlet basis function, identifying regions of interest where the power of the CWT coefficients corresponding to the frequencies of spindles (11-16 Hz) is large. The potential for assessing the signal segment as a spindle is refined using local weighted smoothing techniques. We evaluate our findings on two databases: the MASS database comprising 19 healthy controls and the DREAMS sleep spindle database comprising eight participants diagnosed with various sleep pathologies. We demonstrate that we can replicate the experts' sleep spindles assessment accurately in both databases (MASS database: sensitivity: 84%, specificity: 90%, false discovery rate 83%, DREAMS database: sensitivity: 76%, specificity: 92%, false discovery rate: 67%), outperforming six competing automatic sleep spindle detection algorithms in terms of correctly replicating the experts' assessment of detected spindles.

Treatment Algorithms in Systemic Lupus Erythematosus.

PubMed

Muangchan, Chayawee; van Vollenhoven, Ronald F; Bernatsky, Sasha R; Smith, C Douglas; Hudson, Marie; Inanç, Murat; Rothfield, Naomi F; Nash, Peter T; Furie, Richard A; Senécal, Jean-Luc; Chandran, Vinod; Burgos-Vargas, Ruben; Ramsey-Goldman, Rosalind; Pope, Janet E

2015-09-01

To establish agreement on systemic lupus erythematosus (SLE) treatment. SLE experts (n = 69) were e-mailed scenarios and indicated preferred treatments. Algorithms were constructed and agreement determined (≥50% respondents indicating ≥70% agreement). Initially, 54% (n = 37) responded suggesting treatment for scenarios; 13 experts rated agreement with scenarios. Fourteen of 16 scenarios had agreement as follows: discoid lupus: first-line therapy was topical agents and hydroxychloroquine and/or glucocorticoids then azathioprine and subsequently mycophenolate (mofetil); uncomplicated cutaneous vasculitis: initial treatment was glucocorticoids ± hydroxychloroquine ± methotrexate, followed by azathioprine or mycophenolate and then cyclophosphamide; arthritis: initial therapy was hydroxychloroquine and/or glucocorticoids, then methotrexate and subsequently rituximab; pericarditis: first-line therapy was nonsteroidal antiinflammatory drugs, then glucocorticoids with/without hydroxychloroquine, then azathioprine, mycophenolate, or methotrexate and finally belimumab or rituximab, and/or a pericardial window; interstitial lung disease/alveolitis: induction was glucocorticoids and mycophenolate or cyclophosphamide, then rituximab or intravenous gamma globulin (IVIG), and maintenance followed with azathioprine or mycophenolate; pulmonary hypertension: glucocorticoids and mycophenolate or cyclophosphamide and an endothelin receptor antagonist were initial therapies, subsequent treatments were phosphodiesterase-5 inhibitors and then prostanoids and rituximab; antiphospholipid antibody syndrome: standard anticoagulation with/without hydroxychloroquine, then a thrombin inhibitor for venous thrombosis, versus adding aspirin or platelet inhibition drugs for arterial events; mononeuritis multiplex and central nervous system vasculitis: first-line therapy was glucocorticoids and cyclophosphamide followed by maintenance with azathioprine or mycophenolate, and then rituximab, IVIG, or plasmapheresis; and serious lupus nephritis: first-line therapy was glucocorticoids and mycophenolate, then cyclophosphamide then rituximab. We established variable agreement on treatment approaches. For some treatment decisions there was good agreement between experts even if no randomized controlled trial data were available. © 2015, American College of Rheumatology.

Automated Recognition of 3D Features in GPIR Images

NASA Technical Reports Server (NTRS)

Park, Han; Stough, Timothy; Fijany, Amir

2007-01-01

A method of automated recognition of three-dimensional (3D) features in images generated by ground-penetrating imaging radar (GPIR) is undergoing development. GPIR 3D images can be analyzed to detect and identify such subsurface features as pipes and other utility conduits. Until now, much of the analysis of GPIR images has been performed manually by expert operators who must visually identify and track each feature. The present method is intended to satisfy a need for more efficient and accurate analysis by means of algorithms that can automatically identify and track subsurface features, with minimal supervision by human operators. In this method, data from multiple sources (for example, data on different features extracted by different algorithms) are fused together for identifying subsurface objects. The algorithms of this method can be classified in several different ways. In one classification, the algorithms fall into three classes: (1) image-processing algorithms, (2) feature- extraction algorithms, and (3) a multiaxis data-fusion/pattern-recognition algorithm that includes a combination of machine-learning, pattern-recognition, and object-linking algorithms. The image-processing class includes preprocessing algorithms for reducing noise and enhancing target features for pattern recognition. The feature-extraction algorithms operate on preprocessed data to extract such specific features in images as two-dimensional (2D) slices of a pipe. Then the multiaxis data-fusion/ pattern-recognition algorithm identifies, classifies, and reconstructs 3D objects from the extracted features. In this process, multiple 2D features extracted by use of different algorithms and representing views along different directions are used to identify and reconstruct 3D objects. In object linking, which is an essential part of this process, features identified in successive 2D slices and located within a threshold radius of identical features in adjacent slices are linked in a directed-graph data structure. Relative to past approaches, this multiaxis approach offers the advantages of more reliable detections, better discrimination of objects, and provision of redundant information, which can be helpful in filling gaps in feature recognition by one of the component algorithms. The image-processing class also includes postprocessing algorithms that enhance identified features to prepare them for further scrutiny by human analysts (see figure). Enhancement of images as a postprocessing step is a significant departure from traditional practice, in which enhancement of images is a preprocessing step.

Improved Estimation of Cardiac Function Parameters Using a Combination of Independent Automated Segmentation Results in Cardiovascular Magnetic Resonance Imaging.

PubMed

Lebenberg, Jessica; Lalande, Alain; Clarysse, Patrick; Buvat, Irene; Casta, Christopher; Cochet, Alexandre; Constantinidès, Constantin; Cousty, Jean; de Cesare, Alain; Jehan-Besson, Stephanie; Lefort, Muriel; Najman, Laurent; Roullot, Elodie; Sarry, Laurent; Tilmant, Christophe; Frouin, Frederique; Garreau, Mireille

2015-01-01

This work aimed at combining different segmentation approaches to produce a robust and accurate segmentation result. Three to five segmentation results of the left ventricle were combined using the STAPLE algorithm and the reliability of the resulting segmentation was evaluated in comparison with the result of each individual segmentation method. This comparison was performed using a supervised approach based on a reference method. Then, we used an unsupervised statistical evaluation, the extended Regression Without Truth (eRWT) that ranks different methods according to their accuracy in estimating a specific biomarker in a population. The segmentation accuracy was evaluated by estimating six cardiac function parameters resulting from the left ventricle contour delineation using a public cardiac cine MRI database. Eight different segmentation methods, including three expert delineations and five automated methods, were considered, and sixteen combinations of the automated methods using STAPLE were investigated. The supervised and unsupervised evaluations demonstrated that in most cases, STAPLE results provided better estimates than individual automated segmentation methods. Overall, combining different automated segmentation methods improved the reliability of the segmentation result compared to that obtained using an individual method and could achieve the accuracy of an expert.

Improved Estimation of Cardiac Function Parameters Using a Combination of Independent Automated Segmentation Results in Cardiovascular Magnetic Resonance Imaging

PubMed Central

Lebenberg, Jessica; Lalande, Alain; Clarysse, Patrick; Buvat, Irene; Casta, Christopher; Cochet, Alexandre; Constantinidès, Constantin; Cousty, Jean; de Cesare, Alain; Jehan-Besson, Stephanie; Lefort, Muriel; Najman, Laurent; Roullot, Elodie; Sarry, Laurent; Tilmant, Christophe

2015-01-01

This work aimed at combining different segmentation approaches to produce a robust and accurate segmentation result. Three to five segmentation results of the left ventricle were combined using the STAPLE algorithm and the reliability of the resulting segmentation was evaluated in comparison with the result of each individual segmentation method. This comparison was performed using a supervised approach based on a reference method. Then, we used an unsupervised statistical evaluation, the extended Regression Without Truth (eRWT) that ranks different methods according to their accuracy in estimating a specific biomarker in a population. The segmentation accuracy was evaluated by estimating six cardiac function parameters resulting from the left ventricle contour delineation using a public cardiac cine MRI database. Eight different segmentation methods, including three expert delineations and five automated methods, were considered, and sixteen combinations of the automated methods using STAPLE were investigated. The supervised and unsupervised evaluations demonstrated that in most cases, STAPLE results provided better estimates than individual automated segmentation methods. Overall, combining different automated segmentation methods improved the reliability of the segmentation result compared to that obtained using an individual method and could achieve the accuracy of an expert. PMID:26287691

Sweet-spot training for early esophageal cancer detection

NASA Astrophysics Data System (ADS)

van der Sommen, Fons; Zinger, Svitlana; Schoon, Erik J.; de With, Peter H. N.

2016-03-01

Over the past decade, the imaging tools for endoscopists have improved drastically. This has enabled physicians to visually inspect the intestinal tissue for early signs of malignant lesions. Besides this, recent studies show the feasibility of supportive image analysis for endoscopists, but the analysis problem is typically approached as a segmentation task where binary ground truth is employed. In this study, we show that the detection of early cancerous tissue in the gastrointestinal tract cannot be approached as a binary segmentation problem and it is crucial and clinically relevant to involve multiple experts for annotating early lesions. By employing the so-called sweet spot for training purposes as a metric, a much better detection performance can be achieved. Furthermore, a multi-expert-based ground truth, i.e. a golden standard, enables an improved validation of the resulting delineations. For this purpose, besides the sweet spot we also propose another novel metric, the Jaccard Golden Standard (JIGS) that can handle multiple ground-truth annotations. Our experiments involving these new metrics and based on the golden standard show that the performance of a detection algorithm of early neoplastic lesions in Barrett's esophagus can be increased significantly, demonstrating a 10 percent point increase in the resulting F1 detection score.

Expert perspectives on hereditary angioedema: Key areas for advancements in care across the patient journey

PubMed Central

Baş, Murat; Bernstein, Jonathan A.; Boccon-Gibod, Isabelle; Bova, Maria; Dempster, John; Grumach, Anete Sevciovic; Magerl, Markus; Poarch, Kimberly; Ferreira, Manuel Branco

2016-01-01

Background: Published literature documents the substantial burden of hereditary angioedema (HAE) with C1 inhibitor deficiency on the quality of life and work productivity of patients. However, despite advances in the field and the availability of guidelines to advise health care providers (HCP) on the diagnosis and management of HAE, there are still many challenges to overcome. For example, delayed diagnosis and misdiagnosis are common, and treatment practices vary worldwide. Objective: An international expert panel was convened to consider opportunities for improvements that would benefit patients with HAE. Methods: Based on professional and personal experiences, the experts developed schematics to describe the journey of patients through the following stages: (1) onset of symptoms and initial evaluation; (2) referral/diagnosis; and (3) management of HAE. More importantly, the panel identified key areas in which it was possible to optimize the support provided to patients and HCPs along this journey. Results: Overall, this approach highlighted the need for wider dissemination of algorithms and scientific data to more effectively educate HCPs from multiple disciplines and the need for more research to inform appropriate treatment decisions. Furthermore, HAE awareness campaigns, accurate online information, and referral to patient advocacy groups were all considered helpful approaches to support patients. Conclusion: More detailed and widespread information on the diagnosis and management of HAE is needed and may lead to advancements in care throughout the journey of the patient with HAE. PMID:27661998

Toward high-throughput phenotyping: unbiased automated feature extraction and selection from knowledge sources.

PubMed

Yu, Sheng; Liao, Katherine P; Shaw, Stanley Y; Gainer, Vivian S; Churchill, Susanne E; Szolovits, Peter; Murphy, Shawn N; Kohane, Isaac S; Cai, Tianxi

2015-09-01

Analysis of narrative (text) data from electronic health records (EHRs) can improve population-scale phenotyping for clinical and genetic research. Currently, selection of text features for phenotyping algorithms is slow and laborious, requiring extensive and iterative involvement by domain experts. This paper introduces a method to develop phenotyping algorithms in an unbiased manner by automatically extracting and selecting informative features, which can be comparable to expert-curated ones in classification accuracy. Comprehensive medical concepts were collected from publicly available knowledge sources in an automated, unbiased fashion. Natural language processing (NLP) revealed the occurrence patterns of these concepts in EHR narrative notes, which enabled selection of informative features for phenotype classification. When combined with additional codified features, a penalized logistic regression model was trained to classify the target phenotype. The authors applied our method to develop algorithms to identify patients with rheumatoid arthritis and coronary artery disease cases among those with rheumatoid arthritis from a large multi-institutional EHR. The area under the receiver operating characteristic curves (AUC) for classifying RA and CAD using models trained with automated features were 0.951 and 0.929, respectively, compared to the AUCs of 0.938 and 0.929 by models trained with expert-curated features. Models trained with NLP text features selected through an unbiased, automated procedure achieved comparable or slightly higher accuracy than those trained with expert-curated features. The majority of the selected model features were interpretable. The proposed automated feature extraction method, generating highly accurate phenotyping algorithms with improved efficiency, is a significant step toward high-throughput phenotyping. © The Author 2015. Published by Oxford University Press on behalf of the American Medical Informatics Association. All rights reserved. For Permissions, please email: journals.permissions@oup.com.

An Ensemble Deep Convolutional Neural Network Model with Improved D-S Evidence Fusion for Bearing Fault Diagnosis.

PubMed

Li, Shaobo; Liu, Guokai; Tang, Xianghong; Lu, Jianguang; Hu, Jianjun

2017-07-28

Intelligent machine health monitoring and fault diagnosis are becoming increasingly important for modern manufacturing industries. Current fault diagnosis approaches mostly depend on expert-designed features for building prediction models. In this paper, we proposed IDSCNN, a novel bearing fault diagnosis algorithm based on ensemble deep convolutional neural networks and an improved Dempster-Shafer theory based evidence fusion. The convolutional neural networks take the root mean square (RMS) maps from the FFT (Fast Fourier Transformation) features of the vibration signals from two sensors as inputs. The improved D-S evidence theory is implemented via distance matrix from evidences and modified Gini Index. Extensive evaluations of the IDSCNN on the Case Western Reserve Dataset showed that our IDSCNN algorithm can achieve better fault diagnosis performance than existing machine learning methods by fusing complementary or conflicting evidences from different models and sensors and adapting to different load conditions.

An Ensemble Deep Convolutional Neural Network Model with Improved D-S Evidence Fusion for Bearing Fault Diagnosis

PubMed Central

Li, Shaobo; Liu, Guokai; Tang, Xianghong; Lu, Jianguang

2017-01-01

Intelligent machine health monitoring and fault diagnosis are becoming increasingly important for modern manufacturing industries. Current fault diagnosis approaches mostly depend on expert-designed features for building prediction models. In this paper, we proposed IDSCNN, a novel bearing fault diagnosis algorithm based on ensemble deep convolutional neural networks and an improved Dempster–Shafer theory based evidence fusion. The convolutional neural networks take the root mean square (RMS) maps from the FFT (Fast Fourier Transformation) features of the vibration signals from two sensors as inputs. The improved D-S evidence theory is implemented via distance matrix from evidences and modified Gini Index. Extensive evaluations of the IDSCNN on the Case Western Reserve Dataset showed that our IDSCNN algorithm can achieve better fault diagnosis performance than existing machine learning methods by fusing complementary or conflicting evidences from different models and sensors and adapting to different load conditions. PMID:28788099

Texture segmentation by genetic programming.

PubMed

Song, Andy; Ciesielski, Vic

2008-01-01

This paper describes a texture segmentation method using genetic programming (GP), which is one of the most powerful evolutionary computation algorithms. By choosing an appropriate representation texture, classifiers can be evolved without computing texture features. Due to the absence of time-consuming feature extraction, the evolved classifiers enable the development of the proposed texture segmentation algorithm. This GP based method can achieve a segmentation speed that is significantly higher than that of conventional methods. This method does not require a human expert to manually construct models for texture feature extraction. In an analysis of the evolved classifiers, it can be seen that these GP classifiers are not arbitrary. Certain textural regularities are captured by these classifiers to discriminate different textures. GP has been shown in this study as a feasible and a powerful approach for texture classification and segmentation, which are generally considered as complex vision tasks.

Robust Segmentation of Overlapping Cells in Histopathology Specimens Using Parallel Seed Detection and Repulsive Level Set

PubMed Central

Qi, Xin; Xing, Fuyong; Foran, David J.; Yang, Lin

2013-01-01

Automated image analysis of histopathology specimens could potentially provide support for early detection and improved characterization of breast cancer. Automated segmentation of the cells comprising imaged tissue microarrays (TMA) is a prerequisite for any subsequent quantitative analysis. Unfortunately, crowding and overlapping of cells present significant challenges for most traditional segmentation algorithms. In this paper, we propose a novel algorithm which can reliably separate touching cells in hematoxylin stained breast TMA specimens which have been acquired using a standard RGB camera. The algorithm is composed of two steps. It begins with a fast, reliable object center localization approach which utilizes single-path voting followed by mean-shift clustering. Next, the contour of each cell is obtained using a level set algorithm based on an interactive model. We compared the experimental results with those reported in the most current literature. Finally, performance was evaluated by comparing the pixel-wise accuracy provided by human experts with that produced by the new automated segmentation algorithm. The method was systematically tested on 234 image patches exhibiting dense overlap and containing more than 2200 cells. It was also tested on whole slide images including blood smears and tissue microarrays containing thousands of cells. Since the voting step of the seed detection algorithm is well suited for parallelization, a parallel version of the algorithm was implemented using graphic processing units (GPU) which resulted in significant speed-up over the C/C++ implementation. PMID:22167559

A Tool for Automatic Verification of Real-Time Expert Systems

NASA Technical Reports Server (NTRS)

Traylor, B.; Schwuttke, U.; Quan, A.

1994-01-01

The creation of an automated, user-driven tool for expert system development, validation, and verification is curretly onoging at NASA's Jet Propulsion Laboratory. In the new age of faster, better, cheaper missions, there is an increased willingness to utilize embedded expert systems for encapsulating and preserving mission expertise in systems which combine conventional algorithmic processing and artifical intelligence. The once-questioned role of automation in spacecraft monitoring is now becoming one of increasing importance.

A validation framework for brain tumor segmentation.

PubMed

Archip, Neculai; Jolesz, Ferenc A; Warfield, Simon K

2007-10-01

We introduce a validation framework for the segmentation of brain tumors from magnetic resonance (MR) images. A novel unsupervised semiautomatic brain tumor segmentation algorithm is also presented. The proposed framework consists of 1) T1-weighted MR images of patients with brain tumors, 2) segmentation of brain tumors performed by four independent experts, 3) segmentation of brain tumors generated by a semiautomatic algorithm, and 4) a software tool that estimates the performance of segmentation algorithms. We demonstrate the validation of the novel segmentation algorithm within the proposed framework. We show its performance and compare it with existent segmentation. The image datasets and software are available at http://www.brain-tumor-repository.org/. We present an Internet resource that provides access to MR brain tumor image data and segmentation that can be openly used by the research community. Its purpose is to encourage the development and evaluation of segmentation methods by providing raw test and image data, human expert segmentation results, and methods for comparing segmentation results.

Drug- and Herb-Induced Liver Injury in Clinical and Translational Hepatology: Causality Assessment Methods, Quo Vadis?

PubMed Central

Eickhoff, Axel; Schulze, Johannes

2013-01-01

Drug-induced liver injury (DILI) and herb-induced liver injury (HILI) are typical diseases of clinical and translational hepatology. Their diagnosis is complex and requires an experienced clinician to translate basic science into clinical judgment and identify a valid causality algorithm. To prospectively assess causality starting on the day DILI or HILI is suspected, the best approach for physicians is to use the Council for International Organizations of Medical Sciences (CIOMS) scale in its original or preferably its updated version. The CIOMS scale is validated, liver-specific, structured, and quantitative, providing final causality grades based on scores of specific items for individual patients. These items include latency period, decline in liver values after treatment cessation, risk factors, co-medication, alternative diagnoses, hepatotoxicity track record of the suspected product, and unintentional re-exposure. Provided causality is established as probable or highly probable, data of the CIOMS scale with all individual items, a short clinical report, and complete raw data should be transmitted to the regulatory agencies, manufacturers, expert panels, and possibly to the scientific community for further refinement of the causality evaluation in a setting of retrospective expert opinion. Good-quality case data combined with thorough CIOMS-based assessment as a standardized approach should avert subsequent necessity for other complex causality assessment methods that may have inter-rater problems because of poor-quality data. In the future, the CIOMS scale will continue to be the preferred tool to assess causality of DILI and HILI cases and should be used consistently, both prospectively by physicians, and retrospectively for subsequent expert opinion if needed. For comparability and international harmonization, all parties assessing causality in DILI and HILI cases should attempt this standardized approach using the updated CIOMS scale. PMID:26357608

Automated model-based quantitative analysis of phantoms with spherical inserts in FDG PET scans.

PubMed

Ulrich, Ethan J; Sunderland, John J; Smith, Brian J; Mohiuddin, Imran; Parkhurst, Jessica; Plichta, Kristin A; Buatti, John M; Beichel, Reinhard R

2018-01-01

Quality control plays an increasingly important role in quantitative PET imaging and is typically performed using phantoms. The purpose of this work was to develop and validate a fully automated analysis method for two common PET/CT quality assurance phantoms: the NEMA NU-2 IQ and SNMMI/CTN oncology phantom. The algorithm was designed to only utilize the PET scan to enable the analysis of phantoms with thin-walled inserts. We introduce a model-based method for automated analysis of phantoms with spherical inserts. Models are first constructed for each type of phantom to be analyzed. A robust insert detection algorithm uses the model to locate all inserts inside the phantom. First, candidates for inserts are detected using a scale-space detection approach. Second, candidates are given an initial label using a score-based optimization algorithm. Third, a robust model fitting step aligns the phantom model to the initial labeling and fixes incorrect labels. Finally, the detected insert locations are refined and measurements are taken for each insert and several background regions. In addition, an approach for automated selection of NEMA and CTN phantom models is presented. The method was evaluated on a diverse set of 15 NEMA and 20 CTN phantom PET/CT scans. NEMA phantoms were filled with radioactive tracer solution at 9.7:1 activity ratio over background, and CTN phantoms were filled with 4:1 and 2:1 activity ratio over background. For quantitative evaluation, an independent reference standard was generated by two experts using PET/CT scans of the phantoms. In addition, the automated approach was compared against manual analysis, which represents the current clinical standard approach, of the PET phantom scans by four experts. The automated analysis method successfully detected and measured all inserts in all test phantom scans. It is a deterministic algorithm (zero variability), and the insert detection RMS error (i.e., bias) was 0.97, 1.12, and 1.48 mm for phantom activity ratios 9.7:1, 4:1, and 2:1, respectively. For all phantoms and at all contrast ratios, the average RMS error was found to be significantly lower for the proposed automated method compared to the manual analysis of the phantom scans. The uptake measurements produced by the automated method showed high correlation with the independent reference standard (R 2 ≥ 0.9987). In addition, the average computing time for the automated method was 30.6 s and was found to be significantly lower (P ≪ 0.001) compared to manual analysis (mean: 247.8 s). The proposed automated approach was found to have less error when measured against the independent reference than the manual approach. It can be easily adapted to other phantoms with spherical inserts. In addition, it eliminates inter- and intraoperator variability in PET phantom analysis and is significantly more time efficient, and therefore, represents a promising approach to facilitate and simplify PET standardization and harmonization efforts. © 2017 American Association of Physicists in Medicine.

Quantitative knowledge acquisition for expert systems

NASA Technical Reports Server (NTRS)

Belkin, Brenda L.; Stengel, Robert F.

1991-01-01

A common problem in the design of expert systems is the definition of rules from data obtained in system operation or simulation. While it is relatively easy to collect data and to log the comments of human operators engaged in experiments, generalizing such information to a set of rules has not previously been a direct task. A statistical method is presented for generating rule bases from numerical data, motivated by an example based on aircraft navigation with multiple sensors. The specific objective is to design an expert system that selects a satisfactory suite of measurements from a dissimilar, redundant set, given an arbitrary navigation geometry and possible sensor failures. The systematic development is described of a Navigation Sensor Management (NSM) Expert System from Kalman Filter convariance data. The method invokes two statistical techniques: Analysis of Variance (ANOVA) and the ID3 Algorithm. The ANOVA technique indicates whether variations of problem parameters give statistically different covariance results, and the ID3 algorithms identifies the relationships between the problem parameters using probabilistic knowledge extracted from a simulation example set. Both are detailed.

Research Data Alliance: Understanding Big Data Analytics Applications in Earth Science

NASA Astrophysics Data System (ADS)

Riedel, Morris; Ramachandran, Rahul; Baumann, Peter

2014-05-01

The Research Data Alliance (RDA) enables data to be shared across barriers through focused working groups and interest groups, formed of experts from around the world - from academia, industry and government. Its Big Data Analytics (BDA) interest groups seeks to develop community based recommendations on feasible data analytics approaches to address scientific community needs of utilizing large quantities of data. BDA seeks to analyze different scientific domain applications (e.g. earth science use cases) and their potential use of various big data analytics techniques. These techniques reach from hardware deployment models up to various different algorithms (e.g. machine learning algorithms such as support vector machines for classification). A systematic classification of feasible combinations of analysis algorithms, analytical tools, data and resource characteristics and scientific queries will be covered in these recommendations. This contribution will outline initial parts of such a classification and recommendations in the specific context of the field of Earth Sciences. Given lessons learned and experiences are based on a survey of use cases and also providing insights in a few use cases in detail.

Research Data Alliance: Understanding Big Data Analytics Applications in Earth Science

NASA Technical Reports Server (NTRS)

Riedel, Morris; Ramachandran, Rahul; Baumann, Peter

2014-01-01

The Research Data Alliance (RDA) enables data to be shared across barriers through focused working groups and interest groups, formed of experts from around the world - from academia, industry and government. Its Big Data Analytics (BDA) interest groups seeks to develop community based recommendations on feasible data analytics approaches to address scientific community needs of utilizing large quantities of data. BDA seeks to analyze different scientific domain applications (e.g. earth science use cases) and their potential use of various big data analytics techniques. These techniques reach from hardware deployment models up to various different algorithms (e.g. machine learning algorithms such as support vector machines for classification). A systematic classification of feasible combinations of analysis algorithms, analytical tools, data and resource characteristics and scientific queries will be covered in these recommendations. This contribution will outline initial parts of such a classification and recommendations in the specific context of the field of Earth Sciences. Given lessons learned and experiences are based on a survey of use cases and also providing insights in a few use cases in detail.

Parallel definition of tear film maps on distributed-memory clusters for the support of dry eye diagnosis.

PubMed

González-Domínguez, Jorge; Remeseiro, Beatriz; Martín, María J

2017-02-01

The analysis of the interference patterns on the tear film lipid layer is a useful clinical test to diagnose dry eye syndrome. This task can be automated with a high degree of accuracy by means of the use of tear film maps. However, the time required by the existing applications to generate them prevents a wider acceptance of this method by medical experts. Multithreading has been previously successfully employed by the authors to accelerate the tear film map definition on multicore single-node machines. In this work, we propose a hybrid message-passing and multithreading parallel approach that further accelerates the generation of tear film maps by exploiting the computational capabilities of distributed-memory systems such as multicore clusters and supercomputers. The algorithm for drawing tear film maps is parallelized using Message Passing Interface (MPI) for inter-node communications and the multithreading support available in the C++11 standard for intra-node parallelization. The original algorithm is modified to reduce the communications and increase the scalability. The hybrid method has been tested on 32 nodes of an Intel cluster (with two 12-core Haswell 2680v3 processors per node) using 50 representative images. Results show that maximum runtime is reduced from almost two minutes using the previous only-multithreaded approach to less than ten seconds using the hybrid method. The hybrid MPI/multithreaded implementation can be used by medical experts to obtain tear film maps in only a few seconds, which will significantly accelerate and facilitate the diagnosis of the dry eye syndrome. Copyright © 2016 Elsevier Ireland Ltd. All rights reserved.

The hockey-stick method to estimate evening dim light melatonin onset (DLMO) in humans.

PubMed

Danilenko, Konstantin V; Verevkin, Evgeniy G; Antyufeev, Viktor S; Wirz-Justice, Anna; Cajochen, Christian

2014-04-01

The onset of melatonin secretion in the evening is the most reliable and most widely used index of circadian timing in humans. Saliva (or plasma) is usually sampled every 0.5-1 hours under dim-light conditions in the evening 5-6 hours before usual bedtime to assess the dim-light melatonin onset (DLMO). For many years, attempts have been made to find a reliable objective determination of melatonin onset time either by fixed or dynamic threshold approaches. The here-developed hockey-stick algorithm, used as an interactive computer-based approach, fits the evening melatonin profile by a piecewise linear-parabolic function represented as a straight line switching to the branch of a parabola. The switch point is considered to reliably estimate melatonin rise time. We applied the hockey-stick method to 109 half-hourly melatonin profiles to assess the DLMOs and compared these estimates to visual ratings from three experts in the field. The DLMOs of 103 profiles were considered to be clearly quantifiable. The hockey-stick DLMO estimates were on average 4 minutes earlier than the experts' estimates, with a range of -27 to +13 minutes; in 47% of the cases the difference fell within ±5 minutes, in 98% within -20 to +13 minutes. The raters' and hockey-stick estimates showed poor accordance with DLMOs defined by threshold methods. Thus, the hockey-stick algorithm is a reliable objective method to estimate melatonin rise time, which does not depend on a threshold value and is free from errors arising from differences in subjective circadian phase estimates. The method is available as a computerized program that can be easily used in research settings and clinical practice either for salivary or plasma melatonin values.

The integrated care pathway for post stroke patients (iCaPPS): a shared care approach between stakeholders in areas with limited access to specialist stroke care services.

PubMed

Abdul Aziz, Aznida Firzah; Mohd Nordin, Nor Azlin; Ali, Mohd Fairuz; Abd Aziz, Noor Azah; Sulong, Saperi; Aljunid, Syed Mohamed

2017-01-13

Lack of intersectoral collaboration within public health sectors compound efforts to promote effective multidisciplinary post stroke care after discharge following acute phase. A coordinated, primary care-led care pathway to manage post stroke patients residing at home in the community was designed by an expert panel of specialist stroke care providers to help overcome fragmented post stroke care in areas where access is limited or lacking. Expert panel discussions comprising Family Medicine Specialists, Neurologists, Rehabilitation Physicians and Therapists, and Nurse Managers from Ministry of Health and acadaemia were conducted. In Phase One, experts chartered current care processes in public healthcare facilities, from acute stroke till discharge and also patients who presented late with stroke symptoms to public primary care health centres. In Phase Two, modified Delphi technique was employed to obtain consensus on recommendations, based on current evidence and best care practices. Care algorithms were designed around existing work schedules at public health centres. Indication for patients eligible for monitoring by primary care at public health centres were identified. Gaps in transfer of care occurred either at post discharge from acute care or primary care patients diagnosed at or beyond subacute phase at health centres. Essential information required during transfer of care from tertiary care to primary care providers was identified. Care algorithms including appropriate tools were summarised to guide primary care teams to identify patients requiring further multidisciplinary interventions. Shared care approaches with Specialist Stroke care team were outlined. Components of the iCaPPS were developed simultaneously: (i) iCaPPS-Rehab© for rehabilitation of stroke patients at community level (ii) iCaPPS-Swallow© guided the primary care team to screen and manage stroke related swallowing problems. Coordinated post stroke care monitoring service for patients at community level is achievable using the iCaPPS and its components as a guide. The iCaPPS may be used for post stroke care monitoring of patients in similar fragmented healthcare delivery systems or areas with limited access to specialist stroke care services. No.: ACTRN12616001322426 (Registration Date: 21st September 2016).

Integrating expert- and algorithm-derived data to generate a hemispheric ice edge

NASA Astrophysics Data System (ADS)

Tsatsoulis, C.; Komp, E.

The Arctic ice edge is the area of the Arctic where sea ice concentration is less than 15%, and is considered navigable by most vessels. Experts at the National Ice Center generate a daily ice edge product that is available to the public. Data of preference is that of active, high resolution satellite sensors such as RADARSAT which yields all-weather images of 100m resolution, and a second source is OLS data with 550m resolution. Unfortunately, RADARSAT does not provide full, daily coverage of the Arctic and OLS can be obscured by clouds. The SSM/I sensor provides complete coverage of the Arctic at 25km resolution and is independent of cloud cover and solar illumination during the Arctic winter. SSM/I data is analyzed by the NASA Team algorithm to establish ice concentration. Our work integrates the ice edge created by experts using high resolution data with the ice edge generated out of the coarser SSM/I microwave data. The result is a product that combines human and algorithmic outputs, deals with gross differences in resolution of the underlying data sets, and results in a useful, operational product.

The Development and Delivery of On-Demand RADARSAT Constellation Mission Ground Deformation Products Based on Advanced Insar Technology

NASA Astrophysics Data System (ADS)

Samsonov, S. V.; Feng, W.

2017-12-01

InSAR-based mapping of surface deformation (displacement) has proven valuable to a variety of geoscience applications within NRCan. Conventional approaches to InSAR analysis require significant expert intervention to separate useful signal from noise and are not suited to the address the opportunities and challenges presented by the large multi-temporal SAR datasets provided by future radar constellations. The Canada Centre for Mapping and Earth Observation (CCMEO) develops, in support of NRCAN and Government of Canada priorities a framework for automatic generation of standard and advanced deformation products based on Interferometric Synthetic Aperture Radar (InSAR) technology from RADARSAT Constellation Mission (RCM) Synthetic Aperture Radar data. We utilize existing processing algorithms that are currently used for processing RADARSAT-2 data and adapt them to RCM specifications. In addition we develop novel advanced processing algorithms that address large data sets made possible by the satellites' rapid revisit cycle and expand InSAR functionality to regional and national scales across a wide range of time scales. Through automation the system makes it possible to extend the mapping of surface deformation to non-SAR experts. The architecture is scalable and expandable to serve large number of clients and simultaneously address multiple application areas including: natural and anthropogenic hazards, natural resource development, permafrost and glacier monitoring, coastal and environmental change and wetlands mapping.

Finding regions of interest in pathological images: an attentional model approach

NASA Astrophysics Data System (ADS)

Gómez, Francisco; Villalón, Julio; Gutierrez, Ricardo; Romero, Eduardo

2009-02-01

This paper introduces an automated method for finding diagnostic regions-of-interest (RoIs) in histopathological images. This method is based on the cognitive process of visual selective attention that arises during a pathologist's image examination. Specifically, it emulates the first examination phase, which consists in a coarse search for tissue structures at a "low zoom" to separate the image into relevant regions.1 The pathologist's cognitive performance depends on inherent image visual cues - bottom-up information - and on acquired clinical medicine knowledge - top-down mechanisms -. Our pathologist's visual attention model integrates the latter two components. The selected bottom-up information includes local low level features such as intensity, color, orientation and texture information. Top-down information is related to the anatomical and pathological structures known by the expert. A coarse approximation to these structures is achieved by an oversegmentation algorithm, inspired by psychological grouping theories. The algorithm parameters are learned from an expert pathologist's segmentation. Top-down and bottom-up integration is achieved by calculating a unique index for each of the low level characteristics inside the region. Relevancy is estimated as a simple average of these indexes. Finally, a binary decision rule defines whether or not a region is interesting. The method was evaluated on a set of 49 images using a perceptually-weighted evaluation criterion, finding a quality gain of 3dB when comparing to a classical bottom-up model of attention.

Direct estimates of national neonatal and child cause–specific mortality proportions in Niger by expert algorithm and physician–coded analysis of verbal autopsy interviews

PubMed Central

Kalter, Henry D.; Roubanatou, Abdoulaye–Mamadou; Koffi, Alain; Black, Robert E.

2015-01-01

Background This study was one of a set of verbal autopsy investigations undertaken by the WHO/UNCEF–supported Child Health Epidemiology Reference Group (CHERG) to derive direct estimates of the causes of neonatal and child deaths in high priority countries of sub–Saharan Africa. The objective of the study was to determine the cause distributions of neonatal (0–27 days) and child (1–59 months) mortality in Niger. Methods Verbal autopsy interviews were conducted of random samples of 453 neonatal deaths and 620 child deaths from 2007 to 2010 identified by the 2011 Niger National Mortality Survey. The cause of each death was assigned using two methods: computerized expert algorithms arranged in a hierarchy and physician completion of a death certificate for each child. The findings of the two methods were compared to each other, and plausibility checks were conducted to assess which is the preferred method. Comparison of some direct measures from this study with CHERG modeled cause of death estimates are discussed. Findings The cause distributions of neonatal deaths as determined by expert algorithms and the physician were similar, with the same top three causes by both methods and all but two other causes within one rank of each other. Although child causes of death differed more, the reasons often could be discerned by analyzing algorithmic criteria alongside the physician’s application of required minimal diagnostic criteria. Including all algorithmic (primary and co–morbid) and physician (direct, underlying and contributing) diagnoses in the comparison minimized the differences, with kappa coefficients greater than 0.40 for five of 11 neonatal diagnoses and nine of 13 child diagnoses. By algorithmic diagnosis, early onset neonatal infection was significantly associated (χ2 = 13.2, P < 0.001) with maternal infection, and the geographic distribution of child meningitis deaths closely corresponded with that for meningitis surveillance cases and deaths. Conclusions Verbal autopsy conducted in the context of a national mortality survey can provide useful estimates of the cause distributions of neonatal and child deaths. While the current study found reasonable agreement between the expert algorithm and physician analyses, it also demonstrated greater plausibility for two algorithmic diagnoses and validation work is needed to ascertain the findings. Direct, large–scale measurement of causes of death complement, can strengthen, and in some settings may be preferred over modeled estimates. PMID:25969734

Clean birth kits to improve birth practices: development and testing of a country level decision support tool.

PubMed

Hundley, Vanora A; Avan, Bilal I; Ahmed, Haris; Graham, Wendy J

2012-12-19

Clean birth practices can prevent sepsis, one of the leading causes of both maternal and newborn mortality. Evidence suggests that clean birth kits (CBKs), as part of package that includes education, are associated with a reduction in newborn mortality, omphalitis, and puerperal sepsis. However, questions remain about how best to approach the introduction of CBKs in country. We set out to develop a practical decision support tool for programme managers of public health systems who are considering the potential role of CBKs in their strategy for care at birth. Development and testing of the decision support tool was a three-stage process involving an international expert group and country level testing. Stage 1, the development of the tool was undertaken by the Birth Kit Working Group and involved a review of the evidence, a consensus meeting, drafting of the proposed tool and expert review. In Stage 2 the tool was tested with users through interviews (9) and a focus group, with federal and provincial level decision makers in Pakistan. In Stage 3 the findings from the country level testing were reviewed by the expert group. The decision support tool comprised three separate algorithms to guide the policy maker or programme manager through the specific steps required in making the country level decision about whether to use CBKs. The algorithms were supported by a series of questions (that could be administered by interview, focus group or questionnaire) to help the decision maker identify the information needed. The country level testing revealed that the decision support tool was easy to follow and helpful in making decisions about the potential role of CBKs. Minor modifications were made and the final algorithms are presented. Testing of the tool with users in Pakistan suggests that the tool facilitates discussion and aids decision making. However, testing in other countries is needed to determine whether these results can be replicated and to identify how the tool can be adapted to meet country specific needs.

Non-iterative double-frame 2D/3D particle tracking velocimetry

NASA Astrophysics Data System (ADS)

Fuchs, Thomas; Hain, Rainer; Kähler, Christian J.

2017-09-01

In recent years, the detection of individual particle images and their tracking over time to determine the local flow velocity has become quite popular for planar and volumetric measurements. Particle tracking velocimetry has strong advantages compared to the statistical analysis of an ensemble of particle images by means of cross-correlation approaches, such as particle image velocimetry. Tracking individual particles does not suffer from spatial averaging and therefore bias errors can be avoided. Furthermore, the spatial resolution can be increased up to the sub-pixel level for mean fields. A maximization of the spatial resolution for instantaneous measurements requires high seeding concentrations. However, it is still challenging to track particles at high seeding concentrations, if no time series is available. Tracking methods used under these conditions are typically very complex iterative algorithms, which require expert knowledge due to the large number of adjustable parameters. To overcome these drawbacks, a new non-iterative tracking approach is introduced in this letter, which automatically analyzes the motion of the neighboring particles without requiring to specify any parameters, except for the displacement limits. This makes the algorithm very user friendly and also offers unexperienced users to use and implement particle tracking. In addition, the algorithm enables measurements of high speed flows using standard double-pulse equipment and estimates the flow velocity reliably even at large particle image densities.

In search of patient characteristics that may guide empirically based treatment selection for personality disorder patients-a concept map approach.

PubMed

van Manen, J G; Kamphuis, J H; Goossensen, A; Timman, R; Busschbach, J J V; Verheul, R

2012-08-01

Using the concept map method, this study aimed to summarize and describe patient characteristics pertinent to treatment selection for patients with personality disorders (PDs). Initial patient characteristics were derived from the research literature and a survey among Dutch expert clinicians. Concept mapping is a formalized conceptualization procedure that describes the underlying cognitive structures people use in complex tasks, such as treatment allocation. Based on expert opinions of 29 Dutch clinicians, a concept map was generated that yielded eight domains of patient characteristics, i.e., Severity of symptoms, Severity of personality pathology, Ego-adaptive capacities, Motivation and working alliance, Social context, Social demographic characteristics, Trauma, and Treatment history and medical condition. These domains can be ordered along two bipolar axes, running from internal to external concepts and from vulnerability to strength concepts, respectively. Our findings may serve as input for the delineation of algorithms for patient-treatment matching research in PD.

ExSTraCS 2.0: Description and Evaluation of a Scalable Learning Classifier System.

PubMed

Urbanowicz, Ryan J; Moore, Jason H

2015-09-01

Algorithmic scalability is a major concern for any machine learning strategy in this age of 'big data'. A large number of potentially predictive attributes is emblematic of problems in bioinformatics, genetic epidemiology, and many other fields. Previously, ExS-TraCS was introduced as an extended Michigan-style supervised learning classifier system that combined a set of powerful heuristics to successfully tackle the challenges of classification, prediction, and knowledge discovery in complex, noisy, and heterogeneous problem domains. While Michigan-style learning classifier systems are powerful and flexible learners, they are not considered to be particularly scalable. For the first time, this paper presents a complete description of the ExS-TraCS algorithm and introduces an effective strategy to dramatically improve learning classifier system scalability. ExSTraCS 2.0 addresses scalability with (1) a rule specificity limit, (2) new approaches to expert knowledge guided covering and mutation mechanisms, and (3) the implementation and utilization of the TuRF algorithm for improving the quality of expert knowledge discovery in larger datasets. Performance over a complex spectrum of simulated genetic datasets demonstrated that these new mechanisms dramatically improve nearly every performance metric on datasets with 20 attributes and made it possible for ExSTraCS to reliably scale up to perform on related 200 and 2000-attribute datasets. ExSTraCS 2.0 was also able to reliably solve the 6, 11, 20, 37, 70, and 135 multiplexer problems, and did so in similar or fewer learning iterations than previously reported, with smaller finite training sets, and without using building blocks discovered from simpler multiplexer problems. Furthermore, ExS-TraCS usability was made simpler through the elimination of previously critical run parameters.

Learning to change taxonomies

NASA Astrophysics Data System (ADS)

Eneva, Elena; Petrushin, Valery A.

2002-03-01

Taxonomies are valuable tools for structuring and representing our knowledge about the world. They are widely used in many domains, where information about species, products, customers, publications, etc. needs to be organized. In the absence of standards, many taxonomies of the same entities can co-exist. A problem arises when data categorized in a particular taxonomy needs to be used by a procedure (methodology or algorithm) that uses a different taxonomy. Usually, a labor-intensive manual approach is used to solve this problem. This paper describes a machine learning approach which aids domain experts in changing taxonomies. It allows learning relationships between two taxonomies and mapping the data from one taxonomy into another. The proposed approach uses decision trees and bootstrapping for learning mappings of instances from the source to the target taxonomies. A C4.5 decision tree classifier is trained on a small manually labeled training set and applied to a randomly selected sample from the unlabeled data. The classification results are analyzed and the misclassified items are corrected and all items are added to the training set. This procedure is iterated until unlabeled data is available or an acceptable error rate is reached. In the latter case the last classifier is used to label all the remaining data. We test our approach on a database of products obtained from as grocery store chain and find that it performs well, reaching 92.6% accuracy while requiring the human expert to explicitly label only 18% of the entire data.

Introduction of a New Diagnostic Method for Breast Cancer Based on Fine Needle Aspiration (FNA) Test Data and Combining Intelligent Systems.

PubMed

Fiuzy, Mohammad; Haddadnia, Javad; Mollania, Nasrin; Hashemian, Maryam; Hassanpour, Kazem

2012-01-01

Accurate Diagnosis of Breast Cancer is of prime importance. Fine Needle Aspiration test or "FNA", which has been used for several years in Europe, is a simple, inexpensive, noninvasive and accurate technique for detecting breast cancer. Expending the suitable features of the Fine Needle Aspiration results is the most important diagnostic problem in early stages of breast cancer. In this study, we introduced a new algorithm that can detect breast cancer based on combining artificial intelligent system and Fine Needle Aspiration (FNA). We studied the Features of Wisconsin Data Base Cancer which contained about 569 FNA test samples (212 patient samples (malignant) and 357 healthy samples (benign)). In this research, we combined Artificial Intelligence Approaches, such as Evolutionary Algorithm (EA) with Genetic Algorithm (GA), and also used Exact Classifier Systems (here by Fuzzy C-Means (FCM)) to separate malignant from benign samples. Furthermore, we examined artificial Neural Networks (NN) to identify the model and structure. This research proposed a new algorithm for an accurate diagnosis of breast cancer. According to Wisconsin Data Base Cancer (WDBC) data base, 62.75% of samples were benign, and 37.25% were malignant. After applying the proposed algorithm, we achieved high detection accuracy of about "96.579%" on 205 patients who were diagnosed as having breast cancer. It was found that the method had 93% sensitivity, 73% specialty, 65% positive predictive value, and 95% negative predictive value, respectively. If done by experts, Fine Needle Aspiration (FNA) can be a reliable replacement for open biopsy in palpable breast masses. Evaluation of FNA samples during aspiration can decrease insufficient samples. FNA can be the first line of diagnosis in women with breast masses, at least in deprived regions, and may increase health standards and clinical supervision of patients. Such a smart, economical, non-invasive, rapid and accurate system can be introduced as a useful diagnostic system for comprehensive treatment of breast cancer. Another advantage of this method is the possibility of diagnosing breast abnormalities. If done by experts, FNA can be a reliable replacement for open biopsy in palpable breast masses. Evaluation of FNA samples during aspiration can decrease insufficient samples.

An algorithmic and information-theoretic approach to multimetric index construction

USGS Publications Warehouse

Schoolmaster, Donald R.; Grace, James B.; Schweiger, E. William; Guntenspergen, Glenn R.; Mitchell, Brian R.; Miller, Kathryn M.; Little, Amanda M.

2013-01-01

The use of multimetric indices (MMIs), such as the widely used index of biological integrity (IBI), to measure, track, summarize and infer the overall impact of human disturbance on biological communities has been steadily growing in recent years. Initially, MMIs were developed for aquatic communities using pre-selected biological metrics as indicators of system integrity. As interest in these bioassessment tools has grown, so have the types of biological systems to which they are applied. For many ecosystem types the appropriate biological metrics to use as measures of biological integrity are not known a priori. As a result, a variety of ad hoc protocols for selecting metrics empirically has developed. However, the assumptions made by proposed protocols have not be explicitly described or justified, causing many investigators to call for a clear, repeatable methodology for developing empirically derived metrics and indices that can be applied to any biological system. An issue of particular importance that has not been sufficiently addressed is the way that individual metrics combine to produce an MMI that is a sensitive composite indicator of human disturbance. In this paper, we present and demonstrate an algorithm for constructing MMIs given a set of candidate metrics and a measure of human disturbance. The algorithm uses each metric to inform a candidate MMI, and then uses information-theoretic principles to select MMIs that capture the information in the multidimensional system response from among possible MMIs. Such an approach can be used to create purely empirical (data-based) MMIs or can, optionally, be influenced by expert opinion or biological theory through the use of a weighting vector to create value-weighted MMIs. We demonstrate the algorithm with simulated data to demonstrate the predictive capacity of the final MMIs and with real data from wetlands from Acadia and Rocky Mountain National Parks. For the Acadia wetland data, the algorithm identified 4 metrics that combined to produce a -0.88 correlation with the human disturbance index. When compared to other methods, we find this algorithmic approach resulted in MMIs that were more predictive and comprise fewer metrics.

Quasi-Algorithm Methods and Techniques for Specifying Objective Job/Task Performance Requirements

DTIC Science & Technology

1978-07-01

succeeding experts. While "dottings of i’s and crossings of t’s" may still occur, these trivia no longer significantly affect the course of task...That is, as soon as a branch entered under the assunption that condi- tion A applied was completed, administrator and expert recycled to the

An expert systems approach to automated fault management in a regenerative life support subsystem

NASA Technical Reports Server (NTRS)

Malin, J. T.; Lance, N., Jr.

1986-01-01

This paper describes FIXER, a prototype expert system for automated fault management in a regenerative life support subsystem typical of Space Station applications. The development project provided an evaluation of the use of expert systems technology to enhance controller functions in space subsystems. The software development approach permitted evaluation of the effectiveness of direct involvement of the expert in design and development. The approach also permitted intensive observation of the knowledge and methods of the expert. This paper describes the development of the prototype expert system and presents results of the evaluation.

Toward accelerating landslide mapping with interactive machine learning techniques

NASA Astrophysics Data System (ADS)

Stumpf, André; Lachiche, Nicolas; Malet, Jean-Philippe; Kerle, Norman; Puissant, Anne

2013-04-01

Despite important advances in the development of more automated methods for landslide mapping from optical remote sensing images, the elaboration of inventory maps after major triggering events still remains a tedious task. Image classification with expert defined rules typically still requires significant manual labour for the elaboration and adaption of rule sets for each particular case. Machine learning algorithm, on the contrary, have the ability to learn and identify complex image patterns from labelled examples but may require relatively large amounts of training data. In order to reduce the amount of required training data active learning has evolved as key concept to guide the sampling for applications such as document classification, genetics and remote sensing. The general underlying idea of most active learning approaches is to initialize a machine learning model with a small training set, and to subsequently exploit the model state and/or the data structure to iteratively select the most valuable samples that should be labelled by the user and added in the training set. With relatively few queries and labelled samples, an active learning strategy should ideally yield at least the same accuracy than an equivalent classifier trained with many randomly selected samples. Our study was dedicated to the development of an active learning approach for landslide mapping from VHR remote sensing images with special consideration of the spatial distribution of the samples. The developed approach is a region-based query heuristic that enables to guide the user attention towards few compact spatial batches rather than distributed points resulting in time savings of 50% and more compared to standard active learning techniques. The approach was tested with multi-temporal and multi-sensor satellite images capturing recent large scale triggering events in Brazil and China and demonstrated balanced user's and producer's accuracies between 74% and 80%. The assessment also included an experimental evaluation of the uncertainties of manual mappings from multiple experts and demonstrated strong relationships between the uncertainty of the experts and the machine learning model.

Mixed-initiative control of intelligent systems

NASA Technical Reports Server (NTRS)

Borchardt, G. C.

1987-01-01

Mixed-initiative user interfaces provide a means by which a human operator and an intelligent system may collectively share the task of deciding what to do next. Such interfaces are important to the effective utilization of real-time expert systems as assistants in the execution of critical tasks. Presented here is the Incremental Inference algorithm, a symbolic reasoning mechanism based on propositional logic and suited to the construction of mixed-initiative interfaces. The algorithm is similar in some respects to the Truth Maintenance System, but replaces the notion of 'justifications' with a notion of recency, allowing newer values to override older values yet permitting various interested parties to refresh these values as they become older and thus more vulnerable to change. A simple example is given of the use of the Incremental Inference algorithm plus an overview of the integration of this mechanism within the SPECTRUM expert system for geological interpretation of imaging spectrometer data.

Denoising and 4D visualization of OCT images

PubMed Central

Gargesha, Madhusudhana; Jenkins, Michael W.; Rollins, Andrew M.; Wilson, David L.

2009-01-01

We are using Optical Coherence Tomography (OCT) to image structure and function of the developing embryonic heart in avian models. Fast OCT imaging produces very large 3D (2D + time) and 4D (3D volumes + time) data sets, which greatly challenge ones ability to visualize results. Noise in OCT images poses additional challenges. We created an algorithm with a quick, data set specific optimization for reduction of both shot and speckle noise and applied it to 3D visualization and image segmentation in OCT. When compared to baseline algorithms (median, Wiener, orthogonal wavelet, basic non-orthogonal wavelet), a panel of experts judged the new algorithm to give much improved volume renderings concerning both noise and 3D visualization. Specifically, the algorithm provided a better visualization of the myocardial and endocardial surfaces, and the interaction of the embryonic heart tube with surrounding tissue. Quantitative evaluation using an image quality figure of merit also indicated superiority of the new algorithm. Noise reduction aided semi-automatic 2D image segmentation, as quantitatively evaluated using a contour distance measure with respect to an expert segmented contour. In conclusion, the noise reduction algorithm should be quite useful for visualization and quantitative measurements (e.g., heart volume, stroke volume, contraction velocity, etc.) in OCT embryo images. With its semi-automatic, data set specific optimization, we believe that the algorithm can be applied to OCT images from other applications. PMID:18679509

Unified treatment algorithm for the management of crotaline snakebite in the United States: results of an evidence-informed consensus workshop

PubMed Central

2011-01-01

Background Envenomation by crotaline snakes (rattlesnake, cottonmouth, copperhead) is a complex, potentially lethal condition affecting thousands of people in the United States each year. Treatment of crotaline envenomation is not standardized, and significant variation in practice exists. Methods A geographically diverse panel of experts was convened for the purpose of deriving an evidence-informed unified treatment algorithm. Research staff analyzed the extant medical literature and performed targeted analyses of existing databases to inform specific clinical decisions. A trained external facilitator used modified Delphi and structured consensus methodology to achieve consensus on the final treatment algorithm. Results A unified treatment algorithm was produced and endorsed by all nine expert panel members. This algorithm provides guidance about clinical and laboratory observations, indications for and dosing of antivenom, adjunctive therapies, post-stabilization care, and management of complications from envenomation and therapy. Conclusions Clinical manifestations and ideal treatment of crotaline snakebite differ greatly, and can result in severe complications. Using a modified Delphi method, we provide evidence-informed treatment guidelines in an attempt to reduce variation in care and possibly improve clinical outcomes. PMID:21291549

Cyber security and data collection approaches for smartphone sensor systems

NASA Astrophysics Data System (ADS)

Turner, Hamilton; White, Jules

2012-06-01

In recent years the ubiquity and resources provided by smartphone devices have encouraged scientists to explore using these devices as remote sensing nodes. In addition, the United States Department of Defense has stated a mission of increasing persistent intelligence, surveillance, and reconnaissance capabilities or U.S. units. This paper presents a method of enabling large-scale, long-term smartphone-powered data collection. Key solutions discussed include the ability to directly allow domain experts to define and refine smartphone applications for data collection, technical advancements that allow rapid dissemination of a smartphone data collection application, and an algorithm for preserving the locational privacy of participating users.

Automated extraction of metadata from remotely sensed satellite imagery

NASA Technical Reports Server (NTRS)

Cromp, Robert F.

1991-01-01

The paper discusses research in the Intelligent Data Management project at the NASA/Goddard Space Flight Center, with emphasis on recent improvements in low-level feature detection algorithms for performing real-time characterization of images. Images, including MSS and TM data, are characterized using neural networks and the interpretation of the neural network output by an expert system for subsequent archiving in an object-oriented data base. The data show the applicability of this approach to different arrangements of low-level remote sensing channels. The technique works well when the neural network is trained on data similar to the data used for testing.

Setting objective thresholds for rare event detection in flow cytometry

PubMed Central

Richards, Adam J.; Staats, Janet; Enzor, Jennifer; McKinnon, Katherine; Frelinger, Jacob; Denny, Thomas N.; Weinhold, Kent J.; Chan, Cliburn

2014-01-01

The accurate identification of rare antigen-specific cytokine positive cells from peripheral blood mononuclear cells (PBMC) after antigenic stimulation in an intracellular staining (ICS) flow cytometry assay is challenging, as cytokine positive events may be fairly diffusely distributed and lack an obvious separation from the negative population. Traditionally, the approach by flow operators has been to manually set a positivity threshold to partition events into cytokine-positive and cytokine-negative. This approach suffers from subjectivity and inconsistency across different flow operators. The use of statistical clustering methods does not remove the need to find an objective threshold between between positive and negative events since consistent identification of rare event subsets is highly challenging for automated algorithms, especially when there is distributional overlap between the positive and negative events (“smear”). We present a new approach, based on the Fβ measure, that is similar to manual thresholding in providing a hard cutoff, but has the advantage of being determined objectively. The performance of this algorithm is compared with results obtained by expert visual gating. Several ICS data sets from the External Quality Assurance Program Oversight Laboratory (EQAPOL) proficiency program were used to make the comparisons. We first show that visually determined thresholds are difficult to reproduce and pose a problem when comparing results across operators or laboratories, as well as problems that occur with the use of commonly employed clustering algorithms. In contrast, a single parameterization for the Fβ method performs consistently across different centers, samples, and instruments because it optimizes the precision/recall tradeoff by using both negative and positive controls. PMID:24727143

Knowledge acquisition from natural language for expert systems based on classification problem-solving methods

NASA Technical Reports Server (NTRS)

Gomez, Fernando

1989-01-01

It is shown how certain kinds of domain independent expert systems based on classification problem-solving methods can be constructed directly from natural language descriptions by a human expert. The expert knowledge is not translated into production rules. Rather, it is mapped into conceptual structures which are integrated into long-term memory (LTM). The resulting system is one in which problem-solving, retrieval and memory organization are integrated processes. In other words, the same algorithm and knowledge representation structures are shared by these processes. As a result of this, the system can answer questions, solve problems or reorganize LTM.

Investigating the Language of Engineering Education

NASA Astrophysics Data System (ADS)

Variawa, Chirag

A significant part of professional communication development in engineering is the ability to learn and understand technical vocabulary. Mastering such vocabulary is often a desired learning outcome of engineering education. In promoting this goal, this research investigates the development of a tool that creates wordlists of characteristic discipline-specific vocabulary for a given course. These wordlists explicitly highlight requisite vocabulary learning and, when used as a teaching aid, can promote greater accessibility in the learning environment. Literature, including work in higher education, diversity and language learning, suggest that designing accessible learning environments can increase the quality of instruction and learning for all students. Studying the student/instructor interface using the framework of Universal Instructional Design identified vocabulary learning as an invisible barrier in engineering education. A preliminary investigation of this barrier suggested that students have difficulty assessing their understanding of technical vocabulary. Subsequently, computing word frequency on engineering course material was investigated as an approach for characterizing this barrier. However, it was concluded that a more nuanced method was necessary. This research program was built on previous work in the fields of linguistics and computer science, and lead to the design of an algorithm. The developed algorithm is based on a statistical technique called, Term Frequency-Inverse Document Frequency. Comparator sets of documents are used to hierarchically identify characteristic terms on a target document, such as course materials from a previous term of study. The approach draws on a standardized artifact of the engineering learning environment as its dataset; a repository of 2254 engineering final exams from the University of Toronto, to process the target material. After producing wordlists for ten courses, with the goal of highlighting characteristic discipline-specific terms, the effectiveness of the approach was evaluated by comparing the computed results to the judgment of subject-matter experts. The overall data show a good correlation between the program and the subject-matter experts. The results indicated a balance between accuracy and feasibility, and suggested that this approach could mimic subject-matter expertise to create a list discipline-specific vocabulary from course materials.

Evaluation of an Algorithm to Predict Menstrual-Cycle Phase at the Time of Injury.

PubMed

Tourville, Timothy W; Shultz, Sandra J; Vacek, Pamela M; Knudsen, Emily J; Bernstein, Ira M; Tourville, Kelly J; Hardy, Daniel M; Johnson, Robert J; Slauterbeck, James R; Beynnon, Bruce D

2016-01-01

Women are 2 to 8 times more likely to sustain an anterior cruciate ligament (ACL) injury than men, and previous studies indicated an increased risk for injury during the preovulatory phase of the menstrual cycle (MC). However, investigations of risk rely on retrospective classification of MC phase, and no tools for this have been validated. To evaluate the accuracy of an algorithm for retrospectively classifying MC phase at the time of a mock injury based on MC history and salivary progesterone (P4) concentration. Descriptive laboratory study. Research laboratory. Thirty-one healthy female collegiate athletes (age range, 18-24 years) provided serum or saliva (or both) samples at 8 visits over 1 complete MC. Self-reported MC information was obtained on a randomized date (1-45 days) after mock injury, which is the typical timeframe in which researchers have access to ACL-injured study participants. The MC phase was classified using the algorithm as applied in a stand-alone computational fashion and also by 4 clinical experts using the algorithm and additional subjective hormonal history information to help inform their decision. To assess algorithm accuracy, phase classifications were compared with the actual MC phase at the time of mock injury (ascertained using urinary luteinizing hormone tests and serial serum P4 samples). Clinical expert and computed classifications were compared using κ statistics. Fourteen participants (45%) experienced anovulatory cycles. The algorithm correctly classified MC phase for 23 participants (74%): 22 (76%) of 29 who were preovulatory/anovulatory and 1 (50%) of 2 who were postovulatory. Agreement between expert and algorithm classifications ranged from 80.6% (κ = 0.50) to 93% (κ = 0.83). Classifications based on same-day saliva sample and optimal P4 threshold were the same as those based on MC history alone (87.1% correct). Algorithm accuracy varied during the MC but at no time were both sensitivity and specificity levels acceptable. These findings raise concerns about the accuracy of previous retrospective MC-phase classification systems, particularly in a population with a high occurrence of anovulatory cycles.

Algorithm for automatic forced spirometry quality assessment: technological developments.

PubMed

Melia, Umberto; Burgos, Felip; Vallverdú, Montserrat; Velickovski, Filip; Lluch-Ariet, Magí; Roca, Josep; Caminal, Pere

2014-01-01

We hypothesized that the implementation of automatic real-time assessment of quality of forced spirometry (FS) may significantly enhance the potential for extensive deployment of a FS program in the community. Recent studies have demonstrated that the application of quality criteria defined by the ATS/ERS (American Thoracic Society/European Respiratory Society) in commercially available equipment with automatic quality assessment can be markedly improved. To this end, an algorithm for assessing quality of FS automatically was reported. The current research describes the mathematical developments of the algorithm. An innovative analysis of the shape of the spirometric curve, adding 23 new metrics to the traditional 4 recommended by ATS/ERS, was done. The algorithm was created through a two-step iterative process including: (1) an initial version using the standard FS curves recommended by the ATS; and, (2) a refined version using curves from patients. In each of these steps the results were assessed against one expert's opinion. Finally, an independent set of FS curves from 291 patients was used for validation purposes. The novel mathematical approach to characterize the FS curves led to appropriate FS classification with high specificity (95%) and sensitivity (96%). The results constitute the basis for a successful transfer of FS testing to non-specialized professionals in the community.

Priority target conditions for algorithms for monitoring children's growth: Interdisciplinary consensus.

PubMed

Scherdel, Pauline; Reynaud, Rachel; Pietrement, Christine; Salaün, Jean-François; Bellaïche, Marc; Arnould, Michel; Chevallier, Bertrand; Piloquet, Hugues; Jobez, Emmanuel; Cheymol, Jacques; Bichara, Emmanuelle; Heude, Barbara; Chalumeau, Martin

2017-01-01

Growth monitoring of apparently healthy children aims at early detection of serious conditions through the use of both clinical expertise and algorithms that define abnormal growth. Optimization of growth monitoring requires standardization of the definition of abnormal growth, and the selection of the priority target conditions is a prerequisite of such standardization. To obtain a consensus about the priority target conditions for algorithms monitoring children's growth. We applied a formal consensus method with a modified version of the RAND/UCLA method, based on three phases (preparatory, literature review, and rating), with the participation of expert advisory groups from the relevant professional medical societies (ranging from primary care providers to hospital subspecialists) as well as parent associations. We asked experts in the pilot (n = 11), reading (n = 8) and rating (n = 60) groups to complete the list of diagnostic classification of the European Society for Paediatric Endocrinology and then to select the conditions meeting the four predefined criteria of an ideal type of priority target condition. Strong agreement was obtained for the 8 conditions selected by the experts among the 133 possible: celiac disease, Crohn disease, craniopharyngioma, juvenile nephronophthisis, Turner syndrome, growth hormone deficiency with pituitary stalk interruption syndrome, infantile cystinosis, and hypothalamic-optochiasmatic astrocytoma (in decreasing order of agreement). This national consensus can be used to evaluate the algorithms currently suggested for growth monitoring. The method used for this national consensus could be re-used to obtain an international consensus.

Evaluation of prostate segmentation algorithms for MRI: the PROMISE12 challenge

PubMed Central

Litjens, Geert; Toth, Robert; van de Ven, Wendy; Hoeks, Caroline; Kerkstra, Sjoerd; van Ginneken, Bram; Vincent, Graham; Guillard, Gwenael; Birbeck, Neil; Zhang, Jindang; Strand, Robin; Malmberg, Filip; Ou, Yangming; Davatzikos, Christos; Kirschner, Matthias; Jung, Florian; Yuan, Jing; Qiu, Wu; Gao, Qinquan; Edwards, Philip “Eddie”; Maan, Bianca; van der Heijden, Ferdinand; Ghose, Soumya; Mitra, Jhimli; Dowling, Jason; Barratt, Dean; Huisman, Henkjan; Madabhushi, Anant

2014-01-01

Prostate MRI image segmentation has been an area of intense research due to the increased use of MRI as a modality for the clinical workup of prostate cancer. Segmentation is useful for various tasks, e.g. to accurately localize prostate boundaries for radiotherapy or to initialize multi-modal registration algorithms. In the past, it has been difficult for research groups to evaluate prostate segmentation algorithms on multi-center, multi-vendor and multi-protocol data. Especially because we are dealing with MR images, image appearance, resolution and the presence of artifacts are affected by differences in scanners and/or protocols, which in turn can have a large influence on algorithm accuracy. The Prostate MR Image Segmentation (PROMISE12) challenge was setup to allow a fair and meaningful comparison of segmentation methods on the basis of performance and robustness. In this work we will discuss the initial results of the online PROMISE12 challenge, and the results obtained in the live challenge workshop hosted by the MICCAI2012 conference. In the challenge, 100 prostate MR cases from 4 different centers were included, with differences in scanner manufacturer, field strength and protocol. A total of 11 teams from academic research groups and industry participated. Algorithms showed a wide variety in methods and implementation, including active appearance models, atlas registration and level sets. Evaluation was performed using boundary and volume based metrics which were combined into a single score relating the metrics to human expert performance. The winners of the challenge where the algorithms by teams Imorphics and ScrAutoProstate, with scores of 85.72 and 84.29 overall. Both algorithms where significantly better than all other algorithms in the challenge (p < 0.05) and had an efficient implementation with a run time of 8 minutes and 3 second per case respectively. Overall, active appearance model based approaches seemed to outperform other approaches like multi-atlas registration, both on accuracy and computation time. Although average algorithm performance was good to excellent and the Imorphics algorithm outperformed the second observer on average, we showed that algorithm combination might lead to further improvement, indicating that optimal performance for prostate segmentation is not yet obtained. All results are available online at http://promise12.grand-challenge.org/. PMID:24418598

An evolutionary computation based algorithm for calculating solar differential rotation by automatic tracking of coronal bright points

NASA Astrophysics Data System (ADS)

Shahamatnia, Ehsan; Dorotovič, Ivan; Fonseca, Jose M.; Ribeiro, Rita A.

2016-03-01

Developing specialized software tools is essential to support studies of solar activity evolution. With new space missions such as Solar Dynamics Observatory (SDO), solar images are being produced in unprecedented volumes. To capitalize on that huge data availability, the scientific community needs a new generation of software tools for automatic and efficient data processing. In this paper a prototype of a modular framework for solar feature detection, characterization, and tracking is presented. To develop an efficient system capable of automatic solar feature tracking and measuring, a hybrid approach combining specialized image processing, evolutionary optimization, and soft computing algorithms is being followed. The specialized hybrid algorithm for tracking solar features allows automatic feature tracking while gathering characterization details about the tracked features. The hybrid algorithm takes advantages of the snake model, a specialized image processing algorithm widely used in applications such as boundary delineation, image segmentation, and object tracking. Further, it exploits the flexibility and efficiency of Particle Swarm Optimization (PSO), a stochastic population based optimization algorithm. PSO has been used successfully in a wide range of applications including combinatorial optimization, control, clustering, robotics, scheduling, and image processing and video analysis applications. The proposed tool, denoted PSO-Snake model, was already successfully tested in other works for tracking sunspots and coronal bright points. In this work, we discuss the application of the PSO-Snake algorithm for calculating the sidereal rotational angular velocity of the solar corona. To validate the results we compare them with published manual results performed by an expert.

LOGISMOS-B for primates: primate cortical surface reconstruction and thickness measurement

NASA Astrophysics Data System (ADS)

Oguz, Ipek; Styner, Martin; Sanchez, Mar; Shi, Yundi; Sonka, Milan

2015-03-01

Cortical thickness and surface area are important morphological measures with implications for many psychiatric and neurological conditions. Automated segmentation and reconstruction of the cortical surface from 3D MRI scans is challenging due to the variable anatomy of the cortex and its highly complex geometry. While many methods exist for this task in the context of the human brain, these methods are typically not readily applicable to the primate brain. We propose an innovative approach based on our recently proposed human cortical reconstruction algorithm, LOGISMOS-B, and the Laplace-based thickness measurement method. Quantitative evaluation of our approach was performed based on a dataset of T1- and T2-weighted MRI scans from 12-month-old macaques where labeling by our anatomical experts was used as independent standard. In this dataset, LOGISMOS-B has an average signed surface error of 0.01 +/- 0.03mm and an unsigned surface error of 0.42 +/- 0.03mm over the whole brain. Excluding the rather problematic temporal pole region further improves unsigned surface distance to 0.34 +/- 0.03mm. This high level of accuracy reached by our algorithm even in this challenging developmental dataset illustrates its robustness and its potential for primate brain studies.

Co-evolutionary data mining for fuzzy rules: automatic fitness function creation phase space, and experiments

NASA Astrophysics Data System (ADS)

Smith, James F., III; Blank, Joseph A.

2003-03-01

An approach is being explored that involves embedding a fuzzy logic based resource manager in an electronic game environment. Game agents can function under their own autonomous logic or human control. This approach automates the data mining problem. The game automatically creates a cleansed database reflecting the domain expert's knowledge, it calls a data mining function, a genetic algorithm, for data mining of the data base as required and allows easy evaluation of the information extracted. The co-evolutionary fitness functions, chromosomes and stopping criteria for ending the game are discussed. Genetic algorithm and genetic program based data mining procedures are discussed that automatically discover new fuzzy rules and strategies. The strategy tree concept and its relationship to co-evolutionary data mining are examined as well as the associated phase space representation of fuzzy concepts. The overlap of fuzzy concepts in phase space reduces the effective strategies available to adversaries. Co-evolutionary data mining alters the geometric properties of the overlap region known as the admissible region of phase space significantly enhancing the performance of the resource manager. Procedures for validation of the information data mined are discussed and significant experimental results provided.

MR and CT data with multiobserver delineations of organs in the pelvic area-Part of the Gold Atlas project.

PubMed

Nyholm, Tufve; Svensson, Stina; Andersson, Sebastian; Jonsson, Joakim; Sohlin, Maja; Gustafsson, Christian; Kjellén, Elisabeth; Söderström, Karin; Albertsson, Per; Blomqvist, Lennart; Zackrisson, Björn; Olsson, Lars E; Gunnlaugsson, Adalsteinn

2018-03-01

We describe a public dataset with MR and CT images of patients performed in the same position with both multiobserver and expert consensus delineations of relevant organs in the male pelvic region. The purpose was to provide means for training and validation of segmentation algorithms and methods to convert MR to CT like data, i.e., so called synthetic CT (sCT). T1- and T2-weighted MR images as well as CT data were collected for 19 patients at three different departments. Five experts delineated nine organs for each patient based on the T2-weighted MR images. An automatic method was used to fuse the delineations. Starting from each fused delineation, a consensus delineation was agreed upon by the five experts for each organ and patient. Segmentation overlap between user delineations with respect to the consensus delineations was measured to describe the spread of the collected data. Finally, an open-source software was used to create deformation vector fields describing the relation between MR and CT images to further increase the usability of the dataset. The dataset has been made publically available to be used for academic purposes, and can be accessed from https://zenodo.org/record/583096. The dataset provides a useful source for training and validation of segmentation algorithms as well as methods to convert MR to CT-like data (sCT). To give some examples: The T2-weighted MR images with their consensus delineations can directly be used as a template in an existing atlas-based segmentation engine; the expert delineations are useful to validate the performance of a segmentation algorithm as they provide a way to measure variability among users which can be compared with the result of an automatic segmentation; and the pairwise deformably registered MR and CT images can be a source for an atlas-based sCT algorithm or for validation of sCT algorithm. © 2018 The Authors. Medical Physics published by Wiley Periodicals, Inc. on behalf of American Association of Physicists in Medicine.

Who Owns Educational Theory? Big Data, Algorithms and the Expert Power of Education Data Science

ERIC Educational Resources Information Center

Williamson, Ben

2017-01-01

"Education data science" is an emerging methodological field which possesses the algorithm-driven technologies required to generate insights and knowledge from educational big data. This article consists of an analysis of the Lytics Lab, Stanford University's laboratory for research and development in learning analytics, and the Center…

Supervised interpretation of echocardiograms with a psychological model of expert supervision

NASA Astrophysics Data System (ADS)

Revankar, Shriram V.; Sher, David B.; Shalin, Valerie L.; Ramamurthy, Maya

1993-07-01

We have developed a collaborative scheme that facilitates active human supervision of the binary segmentation of an echocardiogram. The scheme complements the reliability of a human expert with the precision of segmentation algorithms. In the developed system, an expert user compares the computer generated segmentation with the original image in a user friendly graphics environment, and interactively indicates the incorrectly classified regions either by pointing or by circling. The precise boundaries of the indicated regions are computed by studying original image properties at that region, and a human visual attention distribution map obtained from the published psychological and psychophysical research. We use the developed system to extract contours of heart chambers from a sequence of two dimensional echocardiograms. We are currently extending this method to incorporate a richer set of inputs from the human supervisor, to facilitate multi-classification of image regions depending on their functionality. We are integrating into our system the knowledge related constraints that cardiologists use, to improve the capabilities of our existing system. This extension involves developing a psychological model of expert reasoning, functional and relational models of typical views in echocardiograms, and corresponding interface modifications to map the suggested actions to image processing algorithms.

The Texas Medication Algorithm Project antipsychotic algorithm for schizophrenia: 2003 update.

PubMed

Miller, Alexander L; Hall, Catherine S; Buchanan, Robert W; Buckley, Peter F; Chiles, John A; Conley, Robert R; Crismon, M Lynn; Ereshefsky, Larry; Essock, Susan M; Finnerty, Molly; Marder, Stephen R; Miller, Del D; McEvoy, Joseph P; Rush, A John; Saeed, Sy A; Schooler, Nina R; Shon, Steven P; Stroup, Scott; Tarin-Godoy, Bernardo

2004-04-01

The Texas Medication Algorithm Project (TMAP) has been a public-academic collaboration in which guidelines for medication treatment of schizophrenia, bipolar disorder, and major depressive disorder were used in selected public outpatient clinics in Texas. Subsequently, these algorithms were implemented throughout Texas and are being used in other states. Guidelines require updating when significant new evidence emerges; the antipsychotic algorithm for schizophrenia was last updated in 1999. This article reports the recommendations developed in 2002 and 2003 by a group of experts, clinicians, and administrators. A conference in January 2002 began the update process. Before the conference, experts in the pharmacologic treatment of schizophrenia, clinicians, and administrators reviewed literature topics and prepared presentations. Topics included ziprasidone's inclusion in the algorithm, the number of antipsychotics tried before clozapine, and the role of first generation antipsychotics. Data were rated according to Agency for Healthcare Research and Quality criteria. After discussing the presentations, conference attendees arrived at consensus recommendations. Consideration of aripiprazole's inclusion was subsequently handled by electronic communications. The antipsychotic algorithm for schizophrenia was updated to include ziprasidone and aripiprazole among the first-line agents. Relative to the prior algorithm, the number of stages before clozapine was reduced. First generation antipsychotics were included but not as first-line choices. For patients refusing or not responding to clozapine and clozapine augmentation, preference was given to trying monotherapy with another antipsychotic before resorting to antipsychotic combinations. Consensus on algorithm revisions was achieved, but only further well-controlled research will answer many key questions about sequence and type of medication treatments of schizophrenia.

Aquifer Hydrogeologic Layer Zonation at the Hanford Site

DOE Office of Scientific and Technical Information (OSTI.GOV)

Savelieva-Trofimova, Elena A.; Kanevski, Mikhail; timonin, v.

2003-09-10

Sedimentary aquifer layers are characterized by spatial variability of hydraulic properties. Nevertheless, zones with similar values of hydraulic parameters (parameter zones) can be distinguished. This parameter zonation approach is an alternative to the analysis of spatial variation of the continuous hydraulic parameters. The parameter zonation approach is primarily motivated by the lack of measurements that would be needed for direct spatial modeling of the hydraulic properties. The current work is devoted to the problem of zonation of the Hanford formation, the uppermost sedimentary aquifer unit (U1) included in hydrogeologic models at the Hanford site. U1 is characterized by 5 zonesmore » with different hydraulic properties. Each sampled location is ascribed to a parameter zone by an expert. This initial classification is accompanied by a measure of quality (also indicated by an expert) that addresses the level of classification confidence. In the current study, the coneptual zonation map developed by an expert geologist was used as an a priori model. The parameter zonation problem was formulated as a multiclass classification task. Different geostatistical and machine learning algorithms were adapted and applied to solve this problem, including: indicator kriging, conditional simulations, neural networks of different architectures, and support vector machines. All methods were trained using additional soft information based on expert estimates. Regularization methods were used to overcome possible overfitting. The zonation problem was complicated because there were few samples for some zones (classes) and by the spatial non-stationarity of the data. Special approaches were developed to overcome these complications. The comparison of different methods was performed using qualitative and quantitative statistical methods and image analysis. We examined the correspondence of the results with the geologically based interpretation, including the reproduction of the spatial orientation of the different classes and the spatial correlation structure of the classes. The uncertainty of the classification task was examined using both probabilistic interpretation of the estimators and by examining the results of a set of stochastic realizations. Characterization of the classification uncertainty is the main advantage of the proposed methods.« less

Validation of an automated seizure detection algorithm for term neonates

PubMed Central

Mathieson, Sean R.; Stevenson, Nathan J.; Low, Evonne; Marnane, William P.; Rennie, Janet M.; Temko, Andrey; Lightbody, Gordon; Boylan, Geraldine B.

2016-01-01

Objective The objective of this study was to validate the performance of a seizure detection algorithm (SDA) developed by our group, on previously unseen, prolonged, unedited EEG recordings from 70 babies from 2 centres. Methods EEGs of 70 babies (35 seizure, 35 non-seizure) were annotated for seizures by experts as the gold standard. The SDA was tested on the EEGs at a range of sensitivity settings. Annotations from the expert and SDA were compared using event and epoch based metrics. The effect of seizure duration on SDA performance was also analysed. Results Between sensitivity settings of 0.5 and 0.3, the algorithm achieved seizure detection rates of 52.6–75.0%, with false detection (FD) rates of 0.04–0.36 FD/h for event based analysis, which was deemed to be acceptable in a clinical environment. Time based comparison of expert and SDA annotations using Cohen’s Kappa Index revealed a best performing SDA threshold of 0.4 (Kappa 0.630). The SDA showed improved detection performance with longer seizures. Conclusion The SDA achieved promising performance and warrants further testing in a live clinical evaluation. Significance The SDA has the potential to improve seizure detection and provide a robust tool for comparing treatment regimens. PMID:26055336

Development of an algorithm for assessing the risk to food safety posed by a new animal disease.

PubMed

Parker, E M; Jenson, I; Jordan, D; Ward, M P

2012-05-01

An algorithm was developed as a tool to rapidly assess the potential for a new or emerging disease of livestock to adversely affect humans via consumption or handling of meat product, so that the risks and uncertainties can be understood and appropriate risk management and communication implemented. An algorithm describing the sequence of events from occurrence of the disease in livestock, release of the causative agent from an infected animal, contamination of fresh meat and then possible adverse effects in humans following meat handling and consumption was created. A list of questions complements the algorithm to help the assessors address the issues of concern at each step of the decision pathway. The algorithm was refined and validated through consultation with a panel of experts and a review group of animal health and food safety policy advisors via five case studies of potential emerging diseases of cattle. Tasks for model validation included describing the path taken in the algorithm and stating an outcome. Twenty-nine per cent of the 62 experts commented on the model, and one-third of those responding also completed the tasks required for model validation. The feedback from the panel of experts and the review group was used to further develop the tool and remove redundancies and ambiguities. There was agreement in the pathways and assessments for diseases in which the causative agent was well understood (for example, bovine pneumonia due to Mycoplasma bovis). The stated pathways and assessments of other diseases (for example, bovine Johne's disease) were not as consistent. The framework helps to promote objectivity by requiring questions to be answered sequentially and providing the opportunity to record consensus or differences of opinion. Areas for discussion and future investigation are highlighted by the points of diversion on the pathway taken by different assessors. © 2011 Blackwell Verlag GmbH.

Expert-guided optimization for 3D printing of soft and liquid materials.

PubMed

Abdollahi, Sara; Davis, Alexander; Miller, John H; Feinberg, Adam W

2018-01-01

Additive manufacturing (AM) has rapidly emerged as a disruptive technology to build mechanical parts, enabling increased design complexity, low-cost customization and an ever-increasing range of materials. Yet these capabilities have also created an immense challenge in optimizing the large number of process parameters in order achieve a high-performance part. This is especially true for AM of soft, deformable materials and for liquid-like resins that require experimental printing methods. Here, we developed an expert-guided optimization (EGO) strategy to provide structure in exploring and improving the 3D printing of liquid polydimethylsiloxane (PDMS) elastomer resin. EGO uses three steps, starting first with expert screening to select the parameter space, factors, and factor levels. Second is a hill-climbing algorithm to search the parameter space defined by the expert for the best set of parameters. Third is expert decision making to try new factors or a new parameter space to improve on the best current solution. We applied the algorithm to two calibration objects, a hollow cylinder and a five-sided hollow cube that were evaluated based on a multi-factor scoring system. The optimum print settings were then used to print complex PDMS and epoxy 3D objects, including a twisted vase, water drop, toe, and ear, at a level of detail and fidelity previously not obtained.

Expert-guided optimization for 3D printing of soft and liquid materials

PubMed Central

Abdollahi, Sara; Davis, Alexander; Miller, John H.

2018-01-01

Additive manufacturing (AM) has rapidly emerged as a disruptive technology to build mechanical parts, enabling increased design complexity, low-cost customization and an ever-increasing range of materials. Yet these capabilities have also created an immense challenge in optimizing the large number of process parameters in order achieve a high-performance part. This is especially true for AM of soft, deformable materials and for liquid-like resins that require experimental printing methods. Here, we developed an expert-guided optimization (EGO) strategy to provide structure in exploring and improving the 3D printing of liquid polydimethylsiloxane (PDMS) elastomer resin. EGO uses three steps, starting first with expert screening to select the parameter space, factors, and factor levels. Second is a hill-climbing algorithm to search the parameter space defined by the expert for the best set of parameters. Third is expert decision making to try new factors or a new parameter space to improve on the best current solution. We applied the algorithm to two calibration objects, a hollow cylinder and a five-sided hollow cube that were evaluated based on a multi-factor scoring system. The optimum print settings were then used to print complex PDMS and epoxy 3D objects, including a twisted vase, water drop, toe, and ear, at a level of detail and fidelity previously not obtained. PMID:29621286

Systematic methods for knowledge acquisition and expert system development

NASA Technical Reports Server (NTRS)

Belkin, Brenda L.; Stengel, Robert F.

1991-01-01

Nine cooperating rule-based systems, collectively called AUTOCREW which were designed to automate functions and decisions associated with a combat aircraft's subsystems, are discussed. The organization of tasks within each system is described; performance metrics were developed to evaluate the workload of each rule base and to assess the cooperation between the rule bases. Simulation and comparative workload results for two mission scenarios are given. The scenarios are inbound surface-to-air-missile attack on the aircraft and pilot incapacitation. The methodology used to develop the AUTOCREW knowledge bases is summarized. Issues involved in designing the navigation sensor selection expert in AUTOCREW's NAVIGATOR knowledge base are discussed in detail. The performance of seven navigation systems aiding a medium-accuracy INS was investigated using Kalman filter covariance analyses. A navigation sensor management (NSM) expert system was formulated from covariance simulation data using the analysis of variance (ANOVA) method and the ID3 algorithm. ANOVA results show that statistically different position accuracies are obtained when different navaids are used, the number of navaids aiding the INS is varied, the aircraft's trajectory is varied, and the performance history is varied. The ID3 algorithm determines the NSM expert's classification rules in the form of decision trees. The performance of these decision trees was assessed on two arbitrary trajectories, and the results demonstrate that the NSM expert adapts to new situations and provides reasonable estimates of the expected hybrid performance.

Virtual Raters for Reproducible and Objective Assessments in Radiology

NASA Astrophysics Data System (ADS)

Kleesiek, Jens; Petersen, Jens; Döring, Markus; Maier-Hein, Klaus; Köthe, Ullrich; Wick, Wolfgang; Hamprecht, Fred A.; Bendszus, Martin; Biller, Armin

2016-04-01

Volumetric measurements in radiologic images are important for monitoring tumor growth and treatment response. To make these more reproducible and objective we introduce the concept of virtual raters (VRs). A virtual rater is obtained by combining knowledge of machine-learning algorithms trained with past annotations of multiple human raters with the instantaneous rating of one human expert. Thus, he is virtually guided by several experts. To evaluate the approach we perform experiments with multi-channel magnetic resonance imaging (MRI) data sets. Next to gross tumor volume (GTV) we also investigate subcategories like edema, contrast-enhancing and non-enhancing tumor. The first data set consists of N = 71 longitudinal follow-up scans of 15 patients suffering from glioblastoma (GB). The second data set comprises N = 30 scans of low- and high-grade gliomas. For comparison we computed Pearson Correlation, Intra-class Correlation Coefficient (ICC) and Dice score. Virtual raters always lead to an improvement w.r.t. inter- and intra-rater agreement. Comparing the 2D Response Assessment in Neuro-Oncology (RANO) measurements to the volumetric measurements of the virtual raters results in one-third of the cases in a deviating rating. Hence, we believe that our approach will have an impact on the evaluation of clinical studies as well as on routine imaging diagnostics.

Korean Medication Algorithm for Depressive Disorder: Comparisons with Other Treatment Guidelines

PubMed Central

Wang, Hee Ryung; Bahk, Won-Myong; Seo, Jeong Seok; Woo, Young Sup; Park, Young-Min; Jeong, Jong-Hyun; Kim, Won; Shim, Se-Hoon; Lee, Jung Goo; Jon, Duk-In; Min, Kyung Joon

2017-01-01

In this review, we compared recommendations from the Korean Medication Algorithm Project for Depressive Disorder 2017 (KMAP-DD 2017) to other global treatment guidelines for depression. Six global treatment guidelines were reviewed; among the six, 4 were evidence-based guidelines, 1 was an expert consensus-based guideline, and 1 was an amalgamation of both evidence and expert consensus-based recommendations. The recommendations in the KMAP-DD 2017 were generally similar to those in other global treatment guidelines, although there were some differences between the guidelines. The KMAP-DD 2017 appeared to reflect current changes in the psychopharmacology of depression quite well, like other recently published evidence-based guidelines. As an expert consensus-based guideline, the KMAP-DD 2017 had some limitations. However, considering there are situations in which clinical evidence cannot be drawn from planned clinical trials, the KMAP-DD 2017 may be helpful for Korean psychiatrists making decisions in the clinical settings by complementing previously published evidence-based guidelines. PMID:28783928

Pediatric cyanide poisoning by fire smoke inhalation: a European expert consensus. Toxicology Surveillance System of the Intoxications Working Group of the Spanish Society of Paediatric Emergencies.

PubMed

Mintegi, Santiago; Clerigue, Nuria; Tipo, Vincenzo; Ponticiello, Eduardo; Lonati, Davide; Burillo-Putze, Guillermo; Delvau, Nicolas; Anseeuw, Kurt

2013-11-01

Most fire-related deaths are attributable to smoke inhalation rather than burns. The inhalation of fire smoke, which contains not only carbon monoxide but also a complex mixture of gases, seems to be the major cause of morbidity and mortality in fire victims, mainly in enclosed spaces. Cyanide gas exposure is quite common during smoke inhalation, and cyanide is present in the blood of fire victims in most cases and may play an important role in death by smoke inhalation. Cyanide poisoning may, however, be difficult to diagnose and treat. In these children, hydrogen cyanide seems to be a major source of concern, and the rapid administration of the antidote, hydroxocobalamin, may be critical for these children.European experts recently met to formulate an algorithm for prehospital and hospital management of adult patients with acute cyanide poisoning. Subsequently, a group of European pediatric experts met to evaluate and adopt that algorithm for use in the pediatric population.

The EURO-FORTA (Fit fOR The Aged) List: International Consensus Validation of a Clinical Tool for Improved Drug Treatment in Older People.

PubMed

Pazan, Farhad; Weiss, Christel; Wehling, Martin

2018-01-01

Drug treatment of older people is still potentially inappropriate in many cases as multimorbidity and related polypharmacy are highly prevalent. To increase the quality of drug treatment in older people, the FORTA (Fit fOR The Aged) List (first version 2012) was developed in a Delphi consensus procedure and updated (FORTA2015) by 21 experts from Germany and Austria. It has been validated in a randomized, controlled, prospective trial demonstrating significant improvement in the quality of drug treatment and clinical endpoints (VALFORTA). Based on these results, Delphi consensus validations (two rounds) of country/region-specific FORTA Lists were conducted in the UK/Ireland, France, Poland, Italy, Spain, the Nordic countries and The Netherlands. An algorithm based on geriatric/pharmacologic expertise, publications and professional position was used to find experts in the field. Forty-seven experts agreed to participate in the Delphi process (return rate of 97.9%). For each country/region, the overall mean consensus coefficient (deviation from the initiator proposal) was > 0.9. FORTA Lists from six countries/regions with a minimum of four participating experts (excluding The Netherlands) plus the original FORTA List were collated into the EURO-FORTA List containing 264 items in 26 main indication groups. Two drugs had to be added to the proposed items, as proposed by at least four countries/regions; none had to be removed. This project produced seven new country/region-specific FORTA Lists, as well as the overarching EURO-FORTA List showing a high consensual level based on a broader expert base. EURO-FORTA should help to spread the FORTA approach and improve geriatric pharmacotherapy internationally.

A machine independent expert system for diagnosing environmentally induced spacecraft anomalies

NASA Technical Reports Server (NTRS)

Rolincik, Mark J.

1991-01-01

A new rule-based, machine independent analytical tool for diagnosing spacecraft anomalies, the EnviroNET expert system, was developed. Expert systems provide an effective method for storing knowledge, allow computers to sift through large amounts of data pinpointing significant parts, and most importantly, use heuristics in addition to algorithms which allow approximate reasoning and inference, and the ability to attack problems not rigidly defines. The EviroNET expert system knowledge base currently contains over two hundred rules, and links to databases which include past environmental data, satellite data, and previous known anomalies. The environmental causes considered are bulk charging, single event upsets (SEU), surface charging, and total radiation dose.

Classification of health webpages as expert and non expert with a reduced set of cross-language features.

PubMed

Grabar, Natalia; Krivine, Sonia; Jaulent, Marie-Christine

2007-10-11

Making the distinction between expert and non expert health documents can help users to select the information which is more suitable for them, according to whether they are familiar or not with medical terminology. This issue is particularly important for the information retrieval area. In our work we address this purpose through stylistic corpus analysis and the application of machine learning algorithms. Our hypothesis is that this distinction can be performed on the basis of a small number of features and that such features can be language and domain independent. The used features were acquired in source corpus (Russian language, diabetes topic) and then tested on target (French language, pneumology topic) and source corpora. These cross-language features show 90% precision and 93% recall with non expert documents in source language; and 85% precision and 74% recall with expert documents in target language.

Macromolecular target prediction by self-organizing feature maps.

PubMed

Schneider, Gisbert; Schneider, Petra

2017-03-01

Rational drug discovery would greatly benefit from a more nuanced appreciation of the activity of pharmacologically active compounds against a diverse panel of macromolecular targets. Already, computational target-prediction models assist medicinal chemists in library screening, de novo molecular design, optimization of active chemical agents, drug re-purposing, in the spotting of potential undesired off-target activities, and in the 'de-orphaning' of phenotypic screening hits. The self-organizing map (SOM) algorithm has been employed successfully for these and other purposes. Areas covered: The authors recapitulate contemporary artificial neural network methods for macromolecular target prediction, and present the basic SOM algorithm at a conceptual level. Specifically, they highlight consensus target-scoring by the employment of multiple SOMs, and discuss the opportunities and limitations of this technique. Expert opinion: Self-organizing feature maps represent a straightforward approach to ligand clustering and classification. Some of the appeal lies in their conceptual simplicity and broad applicability domain. Despite known algorithmic shortcomings, this computational target prediction concept has been proven to work in prospective settings with high success rates. It represents a prototypic technique for future advances in the in silico identification of the modes of action and macromolecular targets of bioactive molecules.

Semi-Automatic Terminology Generation for Information Extraction from German Chest X-Ray Reports.

PubMed

Krebs, Jonathan; Corovic, Hamo; Dietrich, Georg; Ertl, Max; Fette, Georg; Kaspar, Mathias; Krug, Markus; Stoerk, Stefan; Puppe, Frank

2017-01-01

Extraction of structured data from textual reports is an important subtask for building medical data warehouses for research and care. Many medical and most radiology reports are written in a telegraphic style with a concatenation of noun phrases describing the presence or absence of findings. Therefore a lexico-syntactical approach is promising, where key terms and their relations are recognized and mapped on a predefined standard terminology (ontology). We propose a two-phase algorithm for terminology matching: In the first pass, a local terminology for recognition is derived as close as possible to the terms used in the radiology reports. In the second pass, the local terminology is mapped to a standard terminology. In this paper, we report on an algorithm for the first step of semi-automatic generation of the local terminology and evaluate the algorithm with radiology reports of chest X-ray examinations from Würzburg university hospital. With an effort of about 20 hours work of a radiologist as domain expert and 10 hours for meetings, a local terminology with about 250 attributes and various value patterns was built. In an evaluation with 100 randomly chosen reports it achieved an F1-Score of about 95% for information extraction.

Development of expert systems for modeling of technological process of pressure casting on the basis of artificial intelligence

NASA Astrophysics Data System (ADS)

Gavarieva, K. N.; Simonova, L. A.; Pankratov, D. L.; Gavariev, R. V.

2017-09-01

In article the main component of expert system of process of casting under pressure which consists of algorithms, united in logical models is considered. The characteristics of system showing data on a condition of an object of management are described. A number of logically interconnected steps allowing to increase quality of the received castings is developed

Texas Children's Medication Algorithm Project: Update from Texas Consensus Conference Panel on Medication Treatment of Childhood Major Depressive Disorder

ERIC Educational Resources Information Center

Hughes, Carroll W.; Emslie, Graham J.; Crismon, M. Lynn; Posner, Kelly; Birmaher, Boris; Ryan, Neal; Jensen, Peter; Curry, John; Vitiello, Benedetto; Lopez, Molly; Shon, Steve P.; Pliszka, Steven R.; Trivedi, Madhukar H.

2007-01-01

Objective: To revise and update consensus guidelines for medication treatment algorithms for childhood major depressive disorder based on new scientific evidence and expert clinical consensus when evidence is lacking. Method: A consensus conference was held January 13-14, 2005, that included academic clinicians and researchers, practicing…

Imitating manual curation of text-mined facts in biomedicine.

PubMed

Rodriguez-Esteban, Raul; Iossifov, Ivan; Rzhetsky, Andrey

2006-09-08

Text-mining algorithms make mistakes in extracting facts from natural-language texts. In biomedical applications, which rely on use of text-mined data, it is critical to assess the quality (the probability that the message is correctly extracted) of individual facts--to resolve data conflicts and inconsistencies. Using a large set of almost 100,000 manually produced evaluations (most facts were independently reviewed more than once, producing independent evaluations), we implemented and tested a collection of algorithms that mimic human evaluation of facts provided by an automated information-extraction system. The performance of our best automated classifiers closely approached that of our human evaluators (ROC score close to 0.95). Our hypothesis is that, were we to use a larger number of human experts to evaluate any given sentence, we could implement an artificial-intelligence curator that would perform the classification job at least as accurately as an average individual human evaluator. We illustrated our analysis by visualizing the predicted accuracy of the text-mined relations involving the term cocaine.

Crowdsourcing the creation of image segmentation algorithms for connectomics.

PubMed

Arganda-Carreras, Ignacio; Turaga, Srinivas C; Berger, Daniel R; Cireşan, Dan; Giusti, Alessandro; Gambardella, Luca M; Schmidhuber, Jürgen; Laptev, Dmitry; Dwivedi, Sarvesh; Buhmann, Joachim M; Liu, Ting; Seyedhosseini, Mojtaba; Tasdizen, Tolga; Kamentsky, Lee; Burget, Radim; Uher, Vaclav; Tan, Xiao; Sun, Changming; Pham, Tuan D; Bas, Erhan; Uzunbas, Mustafa G; Cardona, Albert; Schindelin, Johannes; Seung, H Sebastian

2015-01-01

To stimulate progress in automating the reconstruction of neural circuits, we organized the first international challenge on 2D segmentation of electron microscopic (EM) images of the brain. Participants submitted boundary maps predicted for a test set of images, and were scored based on their agreement with a consensus of human expert annotations. The winning team had no prior experience with EM images, and employed a convolutional network. This "deep learning" approach has since become accepted as a standard for segmentation of EM images. The challenge has continued to accept submissions, and the best so far has resulted from cooperation between two teams. The challenge has probably saturated, as algorithms cannot progress beyond limits set by ambiguities inherent in 2D scoring and the size of the test dataset. Retrospective evaluation of the challenge scoring system reveals that it was not sufficiently robust to variations in the widths of neurite borders. We propose a solution to this problem, which should be useful for a future 3D segmentation challenge.

A Knowledge Discovery Approach to Diagnosing Intracranial Hematomas on Brain CT: Recognition, Measurement and Classification

NASA Astrophysics Data System (ADS)

Liao, Chun-Chih; Xiao, Furen; Wong, Jau-Min; Chiang, I.-Jen

Computed tomography (CT) of the brain is preferred study on neurological emergencies. Physicians use CT to diagnose various types of intracranial hematomas, including epidural, subdural and intracerebral hematomas according to their locations and shapes. We propose a novel method that can automatically diagnose intracranial hematomas by combining machine vision and knowledge discovery techniques. The skull on the CT slice is located and the depth of each intracranial pixel is labeled. After normalization of the pixel intensities by their depth, the hyperdense area of intracranial hematoma is segmented with multi-resolution thresholding and region-growing. We then apply C4.5 algorithm to construct a decision tree using the features of the segmented hematoma and the diagnoses made by physicians. The algorithm was evaluated on 48 pathological images treated in a single institute. The two discovered rules closely resemble those used by human experts, and are able to make correct diagnoses in all cases.

Hot-spot selection and evaluation methods for whole slice images of meningiomas and oligodendrogliomas.

PubMed

Swiderska, Zaneta; Markiewicz, Tomasz; Grala, Bartlomiej; Slodkowska, Janina

2015-01-01

The paper presents a combined method for an automatic hot-spot areas selection based on penalty factor in the whole slide images to support the pathomorphological diagnostic procedure. The studied slides represent the meningiomas and oligodendrogliomas tumor on the basis of the Ki-67/MIB-1 immunohistochemical reaction. It allows determining the tumor proliferation index as well as gives an indication to the medical treatment and prognosis. The combined method based on mathematical morphology, thresholding, texture analysis and classification is proposed and verified. The presented algorithm includes building a specimen map, elimination of hemorrhages from them, two methods for detection of hot-spot fields with respect to an introduced penalty factor. Furthermore, we propose localization concordance measure to evaluation localization of hot spot selection by the algorithms in respect to the expert's results. Thus, the results of the influence of the penalty factor are presented and discussed. It was found that the best results are obtained for 0.2 value of them. They confirm effectiveness of applied approach.

Clinical management algorithm of uveitis associated with juvenile idiopathic arthritis: interdisciplinary panel consensus.

PubMed

Bou, Rosa; Adán, Alfredo; Borrás, Fátima; Bravo, Beatriz; Calvo, Inmaculada; De Inocencio, Jaime; Díaz, Jesús; Escudero, Julia; Fonollosa, Alex; de Vicuña, Carmen García; Hernández, Victoria; Merino, Rosa; Peralta, Jesús; Rúa, María-Jesús; Tejada, Pilar; Antón, Jordi

2015-05-01

Uveitis associated with juvenile idiopathic arthritis (JIA) typically involves the anterior chamber segment, follows an indolent chronic course, and presents a high rate of uveitic complications and a worse outcome as compared to other aetiologies of uveitis. Disease assessment, treatment, and outcome measures have not been standardized. Collaboration between pediatric rheumatologists and ophthalmologists is critical for effective management and prevention of morbidity, impaired vision, and irreparable visual loss. Although the Standardization of Uveitis Nomenclature Working Group recommendations have been a great advance to help clinicians to improve consistency in grading and reporting data, difficulties arise at the time of deciding the best treatment approach in the individual patient in routine daily practice. For this reason, recommendations for a systematized control and treatment strategies according to clinical characteristics and disease severity in children with JIA-related uveitis were developed by a panel of experts with special interest in uveitis associated with JIA. A clinical management algorithm organized in a stepwise regimen is here presented.

Automatic lung lobe segmentation of COPD patients using iterative B-spline fitting

NASA Astrophysics Data System (ADS)

Shamonin, D. P.; Staring, M.; Bakker, M. E.; Xiao, C.; Stolk, J.; Reiber, J. H. C.; Stoel, B. C.

2012-02-01

We present an automatic lung lobe segmentation algorithm for COPD patients. The method enhances fissures, removes unlikely fissure candidates, after which a B-spline is fitted iteratively through the remaining candidate objects. The iterative fitting approach circumvents the need to classify each object as being part of the fissure or being noise, and allows the fissure to be detected in multiple disconnected parts. This property is beneficial for good performance in patient data, containing incomplete and disease-affected fissures. The proposed algorithm is tested on 22 COPD patients, resulting in accurate lobe-based densitometry, and a median overlap of the fissure (defined 3 voxels wide) with an expert ground truth of 0.65, 0.54 and 0.44 for the three main fissures. This compares to complete lobe overlaps of 0.99, 0.98, 0.98, 0.97 and 0.87 for the five main lobes, showing promise for lobe segmentation on data of patients with moderate to severe COPD.

Medical image segmentation based on SLIC superpixels model

NASA Astrophysics Data System (ADS)

Chen, Xiang-ting; Zhang, Fan; Zhang, Ruo-ya

2017-01-01

Medical imaging has been widely used in clinical practice. It is an important basis for medical experts to diagnose the disease. However, medical images have many unstable factors such as complex imaging mechanism, the target displacement will cause constructed defect and the partial volume effect will lead to error and equipment wear, which increases the complexity of subsequent image processing greatly. The segmentation algorithm which based on SLIC (Simple Linear Iterative Clustering, SLIC) superpixels is used to eliminate the influence of constructed defect and noise by means of the feature similarity in the preprocessing stage. At the same time, excellent clustering effect can reduce the complexity of the algorithm extremely, which provides an effective basis for the rapid diagnosis of experts.

Accurate lumen diameter measurement in curved vessels in carotid ultrasound: an iterative scale-space and spatial transformation approach.

PubMed

Krishna Kumar, P; Araki, Tadashi; Rajan, Jeny; Saba, Luca; Lavra, Francesco; Ikeda, Nobutaka; Sharma, Aditya M; Shafique, Shoaib; Nicolaides, Andrew; Laird, John R; Gupta, Ajay; Suri, Jasjit S

2017-08-01

Monitoring of cerebrovascular diseases via carotid ultrasound has started to become a routine. The measurement of image-based lumen diameter (LD) or inter-adventitial diameter (IAD) is a promising approach for quantification of the degree of stenosis. The manual measurements of LD/IAD are not reliable, subjective and slow. The curvature associated with the vessels along with non-uniformity in the plaque growth poses further challenges. This study uses a novel and generalized approach for automated LD and IAD measurement based on a combination of spatial transformation and scale-space. In this iterative procedure, the scale-space is first used to get the lumen axis which is then used with spatial image transformation paradigm to get a transformed image. The scale-space is then reapplied to retrieve the lumen region and boundary in the transformed framework. Then, inverse transformation is applied to display the results in original image framework. Two hundred and two patients' left and right common carotid artery (404 carotid images) B-mode ultrasound images were retrospectively analyzed. The validation of our algorithm has done against the two manual expert tracings. The coefficient of correlation between the two manual tracings for LD was 0.98 (p < 0.0001) and 0.99 (p < 0.0001), respectively. The precision of merit between the manual expert tracings and the automated system was 97.7 and 98.7%, respectively. The experimental analysis demonstrated superior performance of the proposed method over conventional approaches. Several statistical tests demonstrated the stability and reliability of the automated system.

Reliability and performance evaluation of systems containing embedded rule-based expert systems

NASA Technical Reports Server (NTRS)

Beaton, Robert M.; Adams, Milton B.; Harrison, James V. A.

1989-01-01

A method for evaluating the reliability of real-time systems containing embedded rule-based expert systems is proposed and investigated. It is a three stage technique that addresses the impact of knowledge-base uncertainties on the performance of expert systems. In the first stage, a Markov reliability model of the system is developed which identifies the key performance parameters of the expert system. In the second stage, the evaluation method is used to determine the values of the expert system's key performance parameters. The performance parameters can be evaluated directly by using a probabilistic model of uncertainties in the knowledge-base or by using sensitivity analyses. In the third and final state, the performance parameters of the expert system are combined with performance parameters for other system components and subsystems to evaluate the reliability and performance of the complete system. The evaluation method is demonstrated in the context of a simple expert system used to supervise the performances of an FDI algorithm associated with an aircraft longitudinal flight-control system.

Perceptual-cognitive expertise in sport: some considerations when applying the expert performance approach.

PubMed

Williams, A Mark; Ericsson, K Anders

2005-06-01

The number of researchers studying perceptual-cognitive expertise in sport is increasing. The intention in this paper is to review the currently accepted framework for studying expert performance and to consider implications for undertaking research work in the area of perceptual-cognitive expertise in sport. The expert performance approach presents a descriptive and inductive approach for the systematic study of expert performance. The nature of expert performance is initially captured in the laboratory using representative tasks that identify reliably superior performance. Process-tracing measures are employed to determine the mechanisms that mediate expert performance on the task. Finally, the specific types of activities that lead to the acquisition and development of these mediating mechanisms are identified. General principles and mechanisms may be discovered and then validated by more traditional experimental designs. The relevance of this approach to the study of perceptual-cognitive expertise in sport is discussed and suggestions for future work highlighted.

Vibration and acoustic frequency spectra for industrial process modeling using selective fusion multi-condition samples and multi-source features

NASA Astrophysics Data System (ADS)

Tang, Jian; Qiao, Junfei; Wu, ZhiWei; Chai, Tianyou; Zhang, Jian; Yu, Wen

2018-01-01

Frequency spectral data of mechanical vibration and acoustic signals relate to difficult-to-measure production quality and quantity parameters of complex industrial processes. A selective ensemble (SEN) algorithm can be used to build a soft sensor model of these process parameters by fusing valued information selectively from different perspectives. However, a combination of several optimized ensemble sub-models with SEN cannot guarantee the best prediction model. In this study, we use several techniques to construct mechanical vibration and acoustic frequency spectra of a data-driven industrial process parameter model based on selective fusion multi-condition samples and multi-source features. Multi-layer SEN (MLSEN) strategy is used to simulate the domain expert cognitive process. Genetic algorithm and kernel partial least squares are used to construct the inside-layer SEN sub-model based on each mechanical vibration and acoustic frequency spectral feature subset. Branch-and-bound and adaptive weighted fusion algorithms are integrated to select and combine outputs of the inside-layer SEN sub-models. Then, the outside-layer SEN is constructed. Thus, "sub-sampling training examples"-based and "manipulating input features"-based ensemble construction methods are integrated, thereby realizing the selective information fusion process based on multi-condition history samples and multi-source input features. This novel approach is applied to a laboratory-scale ball mill grinding process. A comparison with other methods indicates that the proposed MLSEN approach effectively models mechanical vibration and acoustic signals.

A study of metaheuristic algorithms for high dimensional feature selection on microarray data

NASA Astrophysics Data System (ADS)

Dankolo, Muhammad Nasiru; Radzi, Nor Haizan Mohamed; Sallehuddin, Roselina; Mustaffa, Noorfa Haszlinna

2017-11-01

Microarray systems enable experts to examine gene profile at molecular level using machine learning algorithms. It increases the potentials of classification and diagnosis of many diseases at gene expression level. Though, numerous difficulties may affect the efficiency of machine learning algorithms which includes vast number of genes features comprised in the original data. Many of these features may be unrelated to the intended analysis. Therefore, feature selection is necessary to be performed in the data pre-processing. Many feature selection algorithms are developed and applied on microarray which including the metaheuristic optimization algorithms. This paper discusses the application of the metaheuristics algorithms for feature selection in microarray dataset. This study reveals that, the algorithms have yield an interesting result with limited resources thereby saving computational expenses of machine learning algorithms.

EXSPRT: An Expert Systems Approach to Computer-Based Adaptive Testing.

ERIC Educational Resources Information Center

Frick, Theodore W.; And Others

Expert systems can be used to aid decision making. A computerized adaptive test (CAT) is one kind of expert system, although it is not commonly recognized as such. A new approach, termed EXSPRT, was devised that combines expert systems reasoning and sequential probability ratio test stopping rules. EXSPRT-R uses random selection of test items,…

Using the "Expert Performance Approach" as a Framework for Improving Understanding of Expert Learning

ERIC Educational Resources Information Center

Williams, A. Mark; Fawver, Bradley; Hodges, Nicola J.

2017-01-01

The expert performance approach, initially proposed by Ericsson and Smith (1991), is reviewed as a systematic framework for the study of "expert" learning. The need to develop representative tasks to capture learning is discussed, as is the need to employ process-tracing measures during acquisition to examine what actually changes during…

Spatially Invariant Vector Quantization: A pattern matching algorithm for multiple classes of image subject matter including pathology.

PubMed

Hipp, Jason D; Cheng, Jerome Y; Toner, Mehmet; Tompkins, Ronald G; Balis, Ulysses J

2011-02-26

HISTORICALLY, EFFECTIVE CLINICAL UTILIZATION OF IMAGE ANALYSIS AND PATTERN RECOGNITION ALGORITHMS IN PATHOLOGY HAS BEEN HAMPERED BY TWO CRITICAL LIMITATIONS: 1) the availability of digital whole slide imagery data sets and 2) a relative domain knowledge deficit in terms of application of such algorithms, on the part of practicing pathologists. With the advent of the recent and rapid adoption of whole slide imaging solutions, the former limitation has been largely resolved. However, with the expectation that it is unlikely for the general cohort of contemporary pathologists to gain advanced image analysis skills in the short term, the latter problem remains, thus underscoring the need for a class of algorithm that has the concurrent properties of image domain (or organ system) independence and extreme ease of use, without the need for specialized training or expertise. In this report, we present a novel, general case pattern recognition algorithm, Spatially Invariant Vector Quantization (SIVQ), that overcomes the aforementioned knowledge deficit. Fundamentally based on conventional Vector Quantization (VQ) pattern recognition approaches, SIVQ gains its superior performance and essentially zero-training workflow model from its use of ring vectors, which exhibit continuous symmetry, as opposed to square or rectangular vectors, which do not. By use of the stochastic matching properties inherent in continuous symmetry, a single ring vector can exhibit as much as a millionfold improvement in matching possibilities, as opposed to conventional VQ vectors. SIVQ was utilized to demonstrate rapid and highly precise pattern recognition capability in a broad range of gross and microscopic use-case settings. With the performance of SIVQ observed thus far, we find evidence that indeed there exist classes of image analysis/pattern recognition algorithms suitable for deployment in settings where pathologists alone can effectively incorporate their use into clinical workflow, as a turnkey solution. We anticipate that SIVQ, and other related class-independent pattern recognition algorithms, will become part of the overall armamentarium of digital image analysis approaches that are immediately available to practicing pathologists, without the need for the immediate availability of an image analysis expert.

Neuroprosthetic Decoder Training as Imitation Learning.

PubMed

Merel, Josh; Carlson, David; Paninski, Liam; Cunningham, John P

2016-05-01

Neuroprosthetic brain-computer interfaces function via an algorithm which decodes neural activity of the user into movements of an end effector, such as a cursor or robotic arm. In practice, the decoder is often learned by updating its parameters while the user performs a task. When the user's intention is not directly observable, recent methods have demonstrated value in training the decoder against a surrogate for the user's intended movement. Here we show that training a decoder in this way is a novel variant of an imitation learning problem, where an oracle or expert is employed for supervised training in lieu of direct observations, which are not available. Specifically, we describe how a generic imitation learning meta-algorithm, dataset aggregation (DAgger), can be adapted to train a generic brain-computer interface. By deriving existing learning algorithms for brain-computer interfaces in this framework, we provide a novel analysis of regret (an important metric of learning efficacy) for brain-computer interfaces. This analysis allows us to characterize the space of algorithmic variants and bounds on their regret rates. Existing approaches for decoder learning have been performed in the cursor control setting, but the available design principles for these decoders are such that it has been impossible to scale them to naturalistic settings. Leveraging our findings, we then offer an algorithm that combines imitation learning with optimal control, which should allow for training of arbitrary effectors for which optimal control can generate goal-oriented control. We demonstrate this novel and general BCI algorithm with simulated neuroprosthetic control of a 26 degree-of-freedom model of an arm, a sophisticated and realistic end effector.

Personalized management of hyperglycemia in type 2 diabetes: reflections from a Diabetes Care Editors' Expert Forum.

PubMed

Raz, Itamar; Riddle, Matthew C; Rosenstock, Julio; Buse, John B; Inzucchi, Silvio E; Home, Philip D; Del Prato, Stefano; Ferrannini, Ele; Chan, Juliana C N; Leiter, Lawrence A; Leroith, Derek; Defronzo, Ralph; Cefalu, William T

2013-06-01

In June 2012, 13 thought leaders convened in a Diabetes Care Editors' Expert Forum to discuss the concept of personalized medicine in the wake of a recently published American Diabetes Association/European Association for the Study of Diabetes position statement calling for a patient-centered approach to hyperglycemia management in type 2 diabetes. This article, an outgrowth of that forum, offers a clinical translation of the underlying issues that need to be considered for effectively personalizing diabetes care. The medical management of type 2 diabetes has become increasingly complex, and its complications remain a great burden to individual patients and the larger society. The burgeoning armamentarium of pharmacological agents for hyperglycemia management should aid clinicians in providing early treatment to delay or prevent these complications. However, trial evidence is limited for the optimal use of these agents, especially in dual or triple combinations. In the distant future, genotyping and testing for metabolomic markers may help us to better phenotype patients and predict their responses to antihyperglycemic drugs. For now, a personalized ("n of 1") approach in which drugs are tested in a trial-and-error manner in each patient may be the most practical strategy for achieving therapeutic targets. Patient-centered care and standardized algorithmic management are conflicting approaches, but they can be made more compatible by recognizing instances in which personalized A1C targets are warranted and clinical circumstances that may call for comanagement by primary care and specialty clinicians.

New insights into the classification and nomenclature of cortical GABAergic interneurons

PubMed Central

DeFelipe, Javier; López-Cruz, Pedro L.; Benavides-Piccione, Ruth; Bielza, Concha; Larrañaga, Pedro; Anderson, Stewart; Burkhalter, Andreas; Cauli, Bruno; Fairén, Alfonso; Feldmeyer, Dirk; Fishell, Gord; Fitzpatrick, David; Freund, Tamás F.; González-Burgos, Guillermo; Hestrin, Shaul; Hill, Sean; Hof, Patrick R.; Huang, Josh; Jones, Edward G.; Kawaguchi, Yasuo; Kisvárday, Zoltán; Kubota, Yoshiyuki; Lewis, David A.; Marín, Oscar; Markram, Henry; McBain, Chris J.; Meyer, Hanno S.; Monyer, Hannah; Nelson, Sacha B.; Rockland, Kathleen; Rossier, Jean; Rubenstein, John L. R.; Rudy, Bernardo; Scanziani, Massimo; Shepherd, Gordon M.; Sherwood, Chet C.; Staiger, Jochen F.; Tamás, Gábor; Thomson, Alex; Wang, Yun; Yuste, Rafael; Ascoli, Giorgio A.

2013-01-01

A systematic classification and accepted nomenclature of neuron types is much needed but is currently lacking. This article describes a possible taxonomical solution for classifying GABAergic interneurons of the cerebral cortex based on a novel, web-based interactive system that allows experts to classify neurons with pre-determined criteria. Using Bayesian analysis and clustering algorithms on the resulting data, we investigated the suitability of several anatomical terms and neuron names for cortical GABAergic interneurons. Moreover, we show that supervised classification models could automatically categorize interneurons in agreement with experts’ assignments. These results demonstrate a practical and objective approach to the naming, characterization and classification of neurons based on community consensus. PMID:23385869

The "expert patient" approach for non-communicable disease management in low and middle income settings: When the reality confronts the rhetoric.

PubMed

Xiao, Yue

2015-09-01

This paper seeks to explore the relevance between the Western "expert patient" rhetoric and the reality of non-communicable diseases (NCDs) control and management in low and middle income settings from the health sociological perspective. It firstly sets up a conceptual framework of the "expert patient" or the patient self-management approach, showing the rhetoric of the initiative in the developed countries. Then by examining the situation of NCDs control and management in low income settings, the paper tries to evaluate the possibilities of implementing the "expert patient" approach in these countries. Kober and Van Damme's study on the relevance of the "expert patient" for an HIV/AIDS program in low income settings is critically studied to show the relevance of the developed countries' rhetoric of the "expert patient" approach for the reality of developing countries. In addition, the MoPoTsyo diabetes peer educator program is analyzed to show the challenges faced by the low income countries in implementing patient self-management programs. Finally, applications of the expert patient approach in China are discussed as well, to remind us of the possible difficulties in introducing it into rural settings.

Task-Driven Dictionary Learning Based on Mutual Information for Medical Image Classification.

PubMed

Diamant, Idit; Klang, Eyal; Amitai, Michal; Konen, Eli; Goldberger, Jacob; Greenspan, Hayit

2017-06-01

We present a novel variant of the bag-of-visual-words (BoVW) method for automated medical image classification. Our approach improves the BoVW model by learning a task-driven dictionary of the most relevant visual words per task using a mutual information-based criterion. Additionally, we generate relevance maps to visualize and localize the decision of the automatic classification algorithm. These maps demonstrate how the algorithm works and show the spatial layout of the most relevant words. We applied our algorithm to three different tasks: chest x-ray pathology identification (of four pathologies: cardiomegaly, enlarged mediastinum, right consolidation, and left consolidation), liver lesion classification into four categories in computed tomography (CT) images and benign/malignant clusters of microcalcifications (MCs) classification in breast mammograms. Validation was conducted on three datasets: 443 chest x-rays, 118 portal phase CT images of liver lesions, and 260 mammography MCs. The proposed method improves the classical BoVW method for all tested applications. For chest x-ray, area under curve of 0.876 was obtained for enlarged mediastinum identification compared to 0.855 using classical BoVW (with p-value 0.01). For MC classification, a significant improvement of 4% was achieved using our new approach (with p-value = 0.03). For liver lesion classification, an improvement of 6% in sensitivity and 2% in specificity were obtained (with p-value 0.001). We demonstrated that classification based on informative selected set of words results in significant improvement. Our new BoVW approach shows promising results in clinically important domains. Additionally, it can discover relevant parts of images for the task at hand without explicit annotations for training data. This can provide computer-aided support for medical experts in challenging image analysis tasks.

Identifying marker genes in transcription profiling data using a mixture of feature relevance experts.

PubMed

Chow, M L; Moler, E J; Mian, I S

2001-03-08

Transcription profiling experiments permit the expression levels of many genes to be measured simultaneously. Given profiling data from two types of samples, genes that most distinguish the samples (marker genes) are good candidates for subsequent in-depth experimental studies and developing decision support systems for diagnosis, prognosis, and monitoring. This work proposes a mixture of feature relevance experts as a method for identifying marker genes and illustrates the idea using published data from samples labeled as acute lymphoblastic and myeloid leukemia (ALL, AML). A feature relevance expert implements an algorithm that calculates how well a gene distinguishes samples, reorders genes according to this relevance measure, and uses a supervised learning method [here, support vector machines (SVMs)] to determine the generalization performances of different nested gene subsets. The mixture of three feature relevance experts examined implement two existing and one novel feature relevance measures. For each expert, a gene subset consisting of the top 50 genes distinguished ALL from AML samples as completely as all 7,070 genes. The 125 genes at the union of the top 50s are plausible markers for a prototype decision support system. Chromosomal aberration and other data support the prediction that the three genes at the intersection of the top 50s, cystatin C, azurocidin, and adipsin, are good targets for investigating the basic biology of ALL/AML. The same data were employed to identify markers that distinguish samples based on their labels of T cell/B cell, peripheral blood/bone marrow, and male/female. Selenoprotein W may discriminate T cells from B cells. Results from analysis of transcription profiling data from tumor/nontumor colon adenocarcinoma samples support the general utility of the aforementioned approach. Theoretical issues such as choosing SVM kernels and their parameters, training and evaluating feature relevance experts, and the impact of potentially mislabeled samples on marker identification (feature selection) are discussed.

Comparison of exposure assessment methods in a lung cancer case-control study: performance of a lifelong task-based questionnaire for asbestos and PAHs.

PubMed

Bourgkard, Eve; Wild, Pascal; Gonzalez, Maria; Févotte, Joëlle; Penven, Emmanuelle; Paris, Christophe

2013-12-01

To describe the performance of a lifelong task-based questionnaire (TBQ) in estimating exposures compared with other approaches in the context of a case-control study. A sample of 93 subjects was randomly selected from a lung cancer case-control study corresponding to 497 jobs. For each job, exposure assessments for asbestos and polycyclic aromatic hydrocarbons (PAHs) were obtained by expertise (TBQ expertise) and by algorithm using the TBQ (TBQ algorithm) as well as by expert appraisals based on all available occupational data (REFERENCE expertise) considered to be the gold standard. Additionally, a Job Exposure Matrix (JEM)-based evaluation for asbestos was also obtained. On the 497 jobs, the various evaluations were contrasted using Cohen's κ coefficient of agreement. Additionally, on the total case-control population, the asbestos dose-response relationship based on the TBQ algorithm was compared with the JEM-based assessment. Regarding asbestos, the TBQ-exposure estimates agreed well with the REFERENCE estimate (TBQ expertise: level-weighted κ (lwk)=0.68; TBQ algorithm: lwk=0.61) but less so with the JEM estimate (TBQ expertise: lwk=0.31; TBQ algorithm: lwk=0.26). Regarding PAHs, the agreements between REFERENCE expertise and TBQ were less good (TBQ expertise: lwk=0.43; TBQ algorithm: lwk=0.36). In the case-control study analysis, the dose-response relationship between lung cancer and cumulative asbestos based on the JEM is less steep than with the TBQ-algorithm exposure assessment and statistically non-significant. Asbestos-exposure estimates based on the TBQ were consistent with the REFERENCE expertise and yielded a steeper dose-response relationship than the JEM. For PAHs, results were less clear.

[Problems in organization of medical criminological registration and personality identification for subjects occupationally exposed to life risk].

PubMed

Shcherbakov, V V

2000-01-01

The paper discusses problems in organization of identification studies under conditions of mass deaths as exemplified by forensic medical records of medical criminological identification studies of subjects killed during war conflict in Chechnya. The evolution of the organization model of identification studies is shown transformation of organization philosophy, formation of expert algorithms, formalization and technologic realization of expert solutions.

Data Mining for Anomaly Detection

NASA Technical Reports Server (NTRS)

Biswas, Gautam; Mack, Daniel; Mylaraswamy, Dinkar; Bharadwaj, Raj

2013-01-01

The Vehicle Integrated Prognostics Reasoner (VIPR) program describes methods for enhanced diagnostics as well as a prognostic extension to current state of art Aircraft Diagnostic and Maintenance System (ADMS). VIPR introduced a new anomaly detection function for discovering previously undetected and undocumented situations, where there are clear deviations from nominal behavior. Once a baseline (nominal model of operations) is established, the detection and analysis is split between on-aircraft outlier generation and off-aircraft expert analysis to characterize and classify events that may not have been anticipated by individual system providers. Offline expert analysis is supported by data curation and data mining algorithms that can be applied in the contexts of supervised learning methods and unsupervised learning. In this report, we discuss efficient methods to implement the Kolmogorov complexity measure using compression algorithms, and run a systematic empirical analysis to determine the best compression measure. Our experiments established that the combination of the DZIP compression algorithm and CiDM distance measure provides the best results for capturing relevant properties of time series data encountered in aircraft operations. This combination was used as the basis for developing an unsupervised learning algorithm to define "nominal" flight segments using historical flight segments.

Inductive learning of thyroid functional states using the ID3 algorithm. The effect of poor examples on the learning result.

PubMed

Forsström, J

1992-01-01

The ID3 algorithm for inductive learning was tested using preclassified material for patients suspected to have a thyroid illness. Classification followed a rule-based expert system for the diagnosis of thyroid function. Thus, the knowledge to be learned was limited to the rules existing in the knowledge base of that expert system. The learning capability of the ID3 algorithm was tested with an unselected learning material (with some inherent missing data) and with a selected learning material (no missing data). The selected learning material was a subgroup which formed a part of the unselected learning material. When the number of learning cases was increased, the accuracy of the program improved. When the learning material was large enough, an increase in the learning material did not improve the results further. A better learning result was achieved with the selected learning material not including missing data as compared to unselected learning material. With this material we demonstrate a weakness in the ID3 algorithm: it can not find available information from good example cases if we add poor examples to the data.

Where to search top-K biomedical ontologies?

PubMed

Oliveira, Daniela; Butt, Anila Sahar; Haller, Armin; Rebholz-Schuhmann, Dietrich; Sahay, Ratnesh

2018-03-20

Searching for precise terms and terminological definitions in the biomedical data space is problematic, as researchers find overlapping, closely related and even equivalent concepts in a single or multiple ontologies. Search engines that retrieve ontological resources often suggest an extensive list of search results for a given input term, which leads to the tedious task of selecting the best-fit ontological resource (class or property) for the input term and reduces user confidence in the retrieval engines. A systematic evaluation of these search engines is necessary to understand their strengths and weaknesses in different search requirements. We have implemented seven comparable Information Retrieval ranking algorithms to search through ontologies and compared them against four search engines for ontologies. Free-text queries have been performed, the outcomes have been judged by experts and the ranking algorithms and search engines have been evaluated against the expert-based ground truth (GT). In addition, we propose a probabilistic GT that is developed automatically to provide deeper insights and confidence to the expert-based GT as well as evaluating a broader range of search queries. The main outcome of this work is the identification of key search factors for biomedical ontologies together with search requirements and a set of recommendations that will help biomedical experts and ontology engineers to select the best-suited retrieval mechanism in their search scenarios. We expect that this evaluation will allow researchers and practitioners to apply the current search techniques more reliably and that it will help them to select the right solution for their daily work. The source code (of seven ranking algorithms), ground truths and experimental results are available at https://github.com/danielapoliveira/bioont-search-benchmark.

MRI Brain Tumor Segmentation and Necrosis Detection Using Adaptive Sobolev Snakes.

PubMed

Nakhmani, Arie; Kikinis, Ron; Tannenbaum, Allen

2014-03-21

Brain tumor segmentation in brain MRI volumes is used in neurosurgical planning and illness staging. It is important to explore the tumor shape and necrosis regions at different points of time to evaluate the disease progression. We propose an algorithm for semi-automatic tumor segmentation and necrosis detection. Our algorithm consists of three parts: conversion of MRI volume to a probability space based on the on-line learned model, tumor probability density estimation, and adaptive segmentation in the probability space. We use manually selected acceptance and rejection classes on a single MRI slice to learn the background and foreground statistical models. Then, we propagate this model to all MRI slices to compute the most probable regions of the tumor. Anisotropic 3D diffusion is used to estimate the probability density. Finally, the estimated density is segmented by the Sobolev active contour (snake) algorithm to select smoothed regions of the maximum tumor probability. The segmentation approach is robust to noise and not very sensitive to the manual initialization in the volumes tested. Also, it is appropriate for low contrast imagery. The irregular necrosis regions are detected by using the outliers of the probability distribution inside the segmented region. The necrosis regions of small width are removed due to a high probability of noisy measurements. The MRI volume segmentation results obtained by our algorithm are very similar to expert manual segmentation.

MRI brain tumor segmentation and necrosis detection using adaptive Sobolev snakes

NASA Astrophysics Data System (ADS)

Nakhmani, Arie; Kikinis, Ron; Tannenbaum, Allen

2014-03-01

Brain tumor segmentation in brain MRI volumes is used in neurosurgical planning and illness staging. It is important to explore the tumor shape and necrosis regions at di erent points of time to evaluate the disease progression. We propose an algorithm for semi-automatic tumor segmentation and necrosis detection. Our algorithm consists of three parts: conversion of MRI volume to a probability space based on the on-line learned model, tumor probability density estimation, and adaptive segmentation in the probability space. We use manually selected acceptance and rejection classes on a single MRI slice to learn the background and foreground statistical models. Then, we propagate this model to all MRI slices to compute the most probable regions of the tumor. Anisotropic 3D di usion is used to estimate the probability density. Finally, the estimated density is segmented by the Sobolev active contour (snake) algorithm to select smoothed regions of the maximum tumor probability. The segmentation approach is robust to noise and not very sensitive to the manual initialization in the volumes tested. Also, it is appropriate for low contrast imagery. The irregular necrosis regions are detected by using the outliers of the probability distribution inside the segmented region. The necrosis regions of small width are removed due to a high probability of noisy measurements. The MRI volume segmentation results obtained by our algorithm are very similar to expert manual segmentation.

A generalized approach for producing, quantifying, and validating citizen science data from wildlife images.

PubMed

Swanson, Alexandra; Kosmala, Margaret; Lintott, Chris; Packer, Craig

2016-06-01

Citizen science has the potential to expand the scope and scale of research in ecology and conservation, but many professional researchers remain skeptical of data produced by nonexperts. We devised an approach for producing accurate, reliable data from untrained, nonexpert volunteers. On the citizen science website www.snapshotserengeti.org, more than 28,000 volunteers classified 1.51 million images taken in a large-scale camera-trap survey in Serengeti National Park, Tanzania. Each image was circulated to, on average, 27 volunteers, and their classifications were aggregated using a simple plurality algorithm. We validated the aggregated answers against a data set of 3829 images verified by experts and calculated 3 certainty metrics-level of agreement among classifications (evenness), fraction of classifications supporting the aggregated answer (fraction support), and fraction of classifiers who reported "nothing here" for an image that was ultimately classified as containing an animal (fraction blank)-to measure confidence that an aggregated answer was correct. Overall, aggregated volunteer answers agreed with the expert-verified data on 98% of images, but accuracy differed by species commonness such that rare species had higher rates of false positives and false negatives. Easily calculated analysis of variance and post-hoc Tukey tests indicated that the certainty metrics were significant indicators of whether each image was correctly classified or classifiable. Thus, the certainty metrics can be used to identify images for expert review. Bootstrapping analyses further indicated that 90% of images were correctly classified with just 5 volunteers per image. Species classifications based on the plurality vote of multiple citizen scientists can provide a reliable foundation for large-scale monitoring of African wildlife. © 2016 The Authors. Conservation Biology published by Wiley Periodicals, Inc. on behalf of Society for Conservation Biology.

Assigning clinical codes with data-driven concept representation on Dutch clinical free text.

PubMed

Scheurwegs, Elyne; Luyckx, Kim; Luyten, Léon; Goethals, Bart; Daelemans, Walter

2017-05-01

Clinical codes are used for public reporting purposes, are fundamental to determining public financing for hospitals, and form the basis for reimbursement claims to insurance providers. They are assigned to a patient stay to reflect the diagnosis and performed procedures during that stay. This paper aims to enrich algorithms for automated clinical coding by taking a data-driven approach and by using unsupervised and semi-supervised techniques for the extraction of multi-word expressions that convey a generalisable medical meaning (referred to as concepts). Several methods for extracting concepts from text are compared, two of which are constructed from a large unannotated corpus of clinical free text. A distributional semantic model (i.c. the word2vec skip-gram model) is used to generalize over concepts and retrieve relations between them. These methods are validated on three sets of patient stay data, in the disease areas of urology, cardiology, and gastroenterology. The datasets are in Dutch, which introduces a limitation on available concept definitions from expert-based ontologies (e.g. UMLS). The results show that when expert-based knowledge in ontologies is unavailable, concepts derived from raw clinical texts are a reliable alternative. Both concepts derived from raw clinical texts perform and concepts derived from expert-created dictionaries outperform a bag-of-words approach in clinical code assignment. Adding features based on tokens that appear in a semantically similar context has a positive influence for predicting diagnostic codes. Furthermore, the experiments indicate that a distributional semantics model can find relations between semantically related concepts in texts but also introduces erroneous and redundant relations, which can undermine clinical coding performance. Copyright © 2017. Published by Elsevier Inc.

Rotation-invariant convolutional neural networks for galaxy morphology prediction

NASA Astrophysics Data System (ADS)

Dieleman, Sander; Willett, Kyle W.; Dambre, Joni

2015-06-01

Measuring the morphological parameters of galaxies is a key requirement for studying their formation and evolution. Surveys such as the Sloan Digital Sky Survey have resulted in the availability of very large collections of images, which have permitted population-wide analyses of galaxy morphology. Morphological analysis has traditionally been carried out mostly via visual inspection by trained experts, which is time consuming and does not scale to large (≳104) numbers of images. Although attempts have been made to build automated classification systems, these have not been able to achieve the desired level of accuracy. The Galaxy Zoo project successfully applied a crowdsourcing strategy, inviting online users to classify images by answering a series of questions. Unfortunately, even this approach does not scale well enough to keep up with the increasing availability of galaxy images. We present a deep neural network model for galaxy morphology classification which exploits translational and rotational symmetry. It was developed in the context of the Galaxy Challenge, an international competition to build the best model for morphology classification based on annotated images from the Galaxy Zoo project. For images with high agreement among the Galaxy Zoo participants, our model is able to reproduce their consensus with near-perfect accuracy (>99 per cent) for most questions. Confident model predictions are highly accurate, which makes the model suitable for filtering large collections of images and forwarding challenging images to experts for manual annotation. This approach greatly reduces the experts' workload without affecting accuracy. The application of these algorithms to larger sets of training data will be critical for analysing results from future surveys such as the Large Synoptic Survey Telescope.

Detecting ‘Wrong Blood in Tube’ Errors: Evaluation of a Bayesian Network Approach

PubMed Central

Doctor, Jason N.; Strylewicz, Greg

2010-01-01

Objective In an effort to address the problem of laboratory errors, we develop and evaluate a method to detect mismatched specimens from nationally collected blood laboratory data in two experiments. Methods In Experiment 1 and 2 using blood labs from National Health and Nutrition Examination Survey (NHANES) and values derived from the Diabetes Prevention Program (DPP) respectively, a proportion of glucose and HbA1c specimens were randomly mismatched. A Bayesian network that encoded probabilistic relationships among analytes was used to predict mismatches. In Experiment 1 the performance of the network was compared against existing error detection software. In Experiment 2 the network was compared against 11 human experts recruited from the American Academy of Clinical Chemists. Results were compared via area under the receiver-operating characteristics curves (AUCs) and with agreement statistics. Results In Experiment 1 the network was most predictive of mismatches that produced clinically significant discrepancies between true and mismatched scores ((AUC of 0.87 (±0.04) for HbA1c and 0.83 (±0.02) for glucose), performed well in identifying errors among those self-reporting diabetes (N = 329) (AUC = 0.79 (± 0.02)) and performed significantly better than the established approach it was tested against (in all cases p < .0.05). In Experiment 2 it performed better (and in no case worse) than 7 of the 11 human experts. Average percent agreement was 0.79. and Kappa (κ) was 0.59, between experts and the Bayesian network. Conclusions Bayesian network can accurately identify mismatched specimens. The algorithm is best at identifying mismatches that result in a clinically significant magnitude of error. PMID:20566275

Improving oncoplastic breast tumor bed localization for radiotherapy planning using image registration algorithms

NASA Astrophysics Data System (ADS)

Wodzinski, Marek; Skalski, Andrzej; Ciepiela, Izabela; Kuszewski, Tomasz; Kedzierawski, Piotr; Gajda, Janusz

2018-02-01

Knowledge about tumor bed localization and its shape analysis is a crucial factor for preventing irradiation of healthy tissues during supportive radiotherapy and as a result, cancer recurrence. The localization process is especially hard for tumors placed nearby soft tissues, which undergo complex, nonrigid deformations. Among them, breast cancer can be considered as the most representative example. A natural approach to improving tumor bed localization is the use of image registration algorithms. However, this involves two unusual aspects which are not common in typical medical image registration: the real deformation field is discontinuous, and there is no direct correspondence between the cancer and its bed in the source and the target 3D images respectively. The tumor no longer exists during radiotherapy planning. Therefore, a traditional evaluation approach based on known, smooth deformations and target registration error are not directly applicable. In this work, we propose alternative artificial deformations which model the tumor bed creation process. We perform a comprehensive evaluation of the most commonly used deformable registration algorithms: B-Splines free form deformations (B-Splines FFD), different variants of the Demons and TV-L1 optical flow. The evaluation procedure includes quantitative assessment of the dedicated artificial deformations, target registration error calculation, 3D contour propagation and medical experts visual judgment. The results demonstrate that the currently, practically applied image registration (rigid registration and B-Splines FFD) are not able to correctly reconstruct discontinuous deformation fields. We show that the symmetric Demons provide the most accurate soft tissues alignment in terms of the ability to reconstruct the deformation field, target registration error and relative tumor volume change, while B-Splines FFD and TV-L1 optical flow are not an appropriate choice for the breast tumor bed localization problem, even though the visual alignment seems to be better than for the Demons algorithm. However, no algorithm could recover the deformation field with sufficient accuracy in terms of vector length and rotation angle differences.

Towards guided data assimilation for operational hydrologic forecasting in the US Tennessee River basin

NASA Astrophysics Data System (ADS)

Weerts, A.; Wood, A. W.; Clark, M. P.; Carney, S.; Day, G. N.; Lemans, M.; Sumihar, J.; Newman, A. J.

2014-12-01

In the US, the forecasting approach used by the NWS River Forecast Centers and other regional organizations such as the Bonneville Power Administration (BPA) or Tennessee Valley Authority (TVA) has traditionally involved manual model input and state modifications made by forecasters in real-time. This process is time consuming and requires expert knowledge and experience. The benefits of automated data assimilation (DA) as a strategy for avoiding manual modification approaches have been demonstrated in research studies (eg. Seo et al., 2009). This study explores the usage of various ensemble DA algorithms within the operational platform used by TVA. The final goal is to identify a DA algorithm that will guide the manual modification process used by TVA forecasters and realize considerable time gains (without loss of quality or even enhance the quality) within the forecast process. We evaluate the usability of various popular algorithms for DA that have been applied on a limited basis for operational hydrology. To this end, Delft-FEWS was wrapped (via piwebservice) in OpenDA to enable execution of FEWS workflows (and the chained models within these workflows, including SACSMA, UNITHG and LAGK) in a DA framework. Within OpenDA, several filter methods are available. We considered 4 algorithms: particle filter (RRF), Ensemble Kalman Filter and Asynchronous Ensemble Kalman and Particle filter. Retrospective simulation results for one location and algorithm (AEnKF) are illustrated in Figure 1. The initial results are promising. We will present verification results for these methods (and possible more) for a variety of sub basins in the Tennessee River basin. Finally, we will offer recommendations for guided DA based on our results. References Seo, D.-J., L. Cajina, R. Corby and T. Howieson, 2009: Automatic State Updating for Operational Streamflow Forecasting via Variational Data Assimilation, 367, Journal of Hydrology, 255-275. Figure 1. Retrospectively simulated streamflow for the headwater basin above Powell River at Jonesville (red is observed flow, blue is simulated flow without DA, black is simulated flow with DA)

Learning classification with auxiliary probabilistic information

PubMed Central

Nguyen, Quang; Valizadegan, Hamed; Hauskrecht, Milos

2012-01-01

Finding ways of incorporating auxiliary information or auxiliary data into the learning process has been the topic of active data mining and machine learning research in recent years. In this work we study and develop a new framework for classification learning problem in which, in addition to class labels, the learner is provided with an auxiliary (probabilistic) information that reflects how strong the expert feels about the class label. This approach can be extremely useful for many practical classification tasks that rely on subjective label assessment and where the cost of acquiring additional auxiliary information is negligible when compared to the cost of the example analysis and labelling. We develop classification algorithms capable of using the auxiliary information to make the learning process more efficient in terms of the sample complexity. We demonstrate the benefit of the approach on a number of synthetic and real world data sets by comparing it to the learning with class labels only. PMID:25309141

A Machine Learning Approach for Business Intelligence Analysis using Commercial Shipping Transaction Data

DOE Office of Scientific and Technical Information (OSTI.GOV)

Bramer, Lisa M.; Chatterjee, Samrat; Holmes, Aimee E.

Business intelligence problems are particularly challenging due to the use of large volume and high velocity data in attempts to model and explain complex underlying phenomena. Incremental machine learning based approaches for summarizing trends and identifying anomalous behavior are often desirable in such conditions to assist domain experts in characterizing their data. The overall goal of this research is to develop a machine learning algorithm that enables predictive analysis on streaming data, detects changes and anomalies in the data, and can evolve based on the dynamic behavior of the data. Commercial shipping transaction data for the U.S. is used tomore » develop and test a Naïve Bayes model that classifies several companies into lines of businesses and demonstrates an ability to predict when the behavior of these companies changes by venturing into other lines of businesses.« less

DeepStack: Expert-level artificial intelligence in heads-up no-limit poker.

PubMed

Moravčík, Matej; Schmid, Martin; Burch, Neil; Lisý, Viliam; Morrill, Dustin; Bard, Nolan; Davis, Trevor; Waugh, Kevin; Johanson, Michael; Bowling, Michael

2017-05-05

Artificial intelligence has seen several breakthroughs in recent years, with games often serving as milestones. A common feature of these games is that players have perfect information. Poker, the quintessential game of imperfect information, is a long-standing challenge problem in artificial intelligence. We introduce DeepStack, an algorithm for imperfect-information settings. It combines recursive reasoning to handle information asymmetry, decomposition to focus computation on the relevant decision, and a form of intuition that is automatically learned from self-play using deep learning. In a study involving 44,000 hands of poker, DeepStack defeated, with statistical significance, professional poker players in heads-up no-limit Texas hold'em. The approach is theoretically sound and is shown to produce strategies that are more difficult to exploit than prior approaches. Copyright © 2017, American Association for the Advancement of Science.

Ground Truth Creation for Complex Clinical NLP Tasks - an Iterative Vetting Approach and Lessons Learned.

PubMed

Liang, Jennifer J; Tsou, Ching-Huei; Devarakonda, Murthy V

2017-01-01

Natural language processing (NLP) holds the promise of effectively analyzing patient record data to reduce cognitive load on physicians and clinicians in patient care, clinical research, and hospital operations management. A critical need in developing such methods is the "ground truth" dataset needed for training and testing the algorithms. Beyond localizable, relatively simple tasks, ground truth creation is a significant challenge because medical experts, just as physicians in patient care, have to assimilate vast amounts of data in EHR systems. To mitigate potential inaccuracies of the cognitive challenges, we present an iterative vetting approach for creating the ground truth for complex NLP tasks. In this paper, we present the methodology, and report on its use for an automated problem list generation task, its effect on the ground truth quality and system accuracy, and lessons learned from the effort.

Comparison of four morphometric definitions and a semiquantitative consensus reading for assessing prevalent vertebral fractures.

PubMed

Grados, F; Roux, C; de Vernejoul, M C; Utard, G; Sebert, J L; Fardellone, P

2001-01-01

The assessment of vertebral fracture in patients with osteoporosis by conventional radiography has been improved over the past 10 years using either the semiquantitative (SQ) method devised by Genant et al. or quantitative morphometry. However, there is still no internationally agreed definition for vertebral fracture and there have been few comparative studies between these different approaches. Our study assessed the reproducibility of the SQ method and of four commonly used morphometric algorithms (Melton's, Eastell's, Minne's and McCloskey's methods) for assessing prevalent vertebral fractures, and examined the agreement of each morphometric algorithm with a SQ consensus reading performed by three experts. With this consensus reading in place of a gold standard, we determined relative measures of sensitivity, specificity and optimal cutoff threshold for each morphometric algorithm. The study was conducted in 39 postmenopausal women who had at least one osteoporotic vertebral fracture. Normal values were derived from 84 healthy postmenopausal women with apparently normal vertebral bodies. Our results indicate that the concordance of SQ method was excellent (intraobserver agreement on serial radiographs = 96.4%, kappa = 0.91; agreement between individual readings and the consensus reading = 98%, kappa = 0.95). Three morphometric approaches demonstrated good intra- and interobserver concordance (Melton: intraobserver agreement on serial radiographs = 92.7%, kappa = 0.82, interobserver agreement = 91.1%, kappa = 0.79; Eastell: intraobserver agreement on serial radiographs = 87.6%, kappa = 0.66, interobserver agreement = 88.6%, kappa = 0.68; McCloskey: intraobserver agreement on serial radiographs = 91.5%, kappa = 0.72, interobserver agreement = 93.9%, kappa = 0.78). Except for McCloskey's method, the optimal cutoff thresholds defined in our study by highest kappa score or Youden index in comparison with the SQ consensus reading were near the cutoff thresholds that were arbitrarily fixed. The four morphometric algorithms provided a good agreement with the results of the SQ consensus reading, but the more complex algorithm did not provide better results and even if we adjusted the cutoff threshold, no morphometric algorithm agreed perfectly with the SQ consensus reading. We conclude that morphometric approaches currently used should not be employed alone to detect prevalent vertebral fractures in studies on osteoporosis, but should rather be used in combination with a visual assessment. The SQ approach that allows differential diagnosis of vertebral deformities and has demonstrated a better reproducibility can be employed alone when it is performed by experienced and well-trained readers.

Systems biology by the rules: hybrid intelligent systems for pathway modeling and discovery.

PubMed

Bosl, William J

2007-02-15

Expert knowledge in journal articles is an important source of data for reconstructing biological pathways and creating new hypotheses. An important need for medical research is to integrate this data with high throughput sources to build useful models that span several scales. Researchers traditionally use mental models of pathways to integrate information and development new hypotheses. Unfortunately, the amount of information is often overwhelming and these are inadequate for predicting the dynamic response of complex pathways. Hierarchical computational models that allow exploration of semi-quantitative dynamics are useful systems biology tools for theoreticians, experimentalists and clinicians and may provide a means for cross-communication. A novel approach for biological pathway modeling based on hybrid intelligent systems or soft computing technologies is presented here. Intelligent hybrid systems, which refers to several related computing methods such as fuzzy logic, neural nets, genetic algorithms, and statistical analysis, has become ubiquitous in engineering applications for complex control system modeling and design. Biological pathways may be considered to be complex control systems, which medicine tries to manipulate to achieve desired results. Thus, hybrid intelligent systems may provide a useful tool for modeling biological system dynamics and computational exploration of new drug targets. A new modeling approach based on these methods is presented in the context of hedgehog regulation of the cell cycle in granule cells. Code and input files can be found at the Bionet website: www.chip.ord/~wbosl/Software/Bionet. This paper presents the algorithmic methods needed for modeling complicated biochemical dynamics using rule-based models to represent expert knowledge in the context of cell cycle regulation and tumor growth. A notable feature of this modeling approach is that it allows biologists to build complex models from their knowledge base without the need to translate that knowledge into mathematical form. Dynamics on several levels, from molecular pathways to tissue growth, are seamlessly integrated. A number of common network motifs are examined and used to build a model of hedgehog regulation of the cell cycle in cerebellar neurons, which is believed to play a key role in the etiology of medulloblastoma, a devastating childhood brain cancer.

A precise and accurate acupoint location obtained on the face using consistency matrix pointwise fusion method.

PubMed

Yanq, Xuming; Ye, Yijun; Xia, Yong; Wei, Xuanzhong; Wang, Zheyu; Ni, Hongmei; Zhu, Ying; Xu, Lingyu

2015-02-01

To develop a more precise and accurate method, and identified a procedure to measure whether an acupoint had been correctly located. On the face, we used an acupoint location from different acupuncture experts and obtained the most precise and accurate values of acupoint location based on the consistency information fusion algorithm, through a virtual simulation of the facial orientation coordinate system. Because of inconsistencies in each acupuncture expert's original data, the system error the general weight calculation. First, we corrected each expert of acupoint location system error itself, to obtain a rational quantification for each expert of acupuncture and moxibustion acupoint location consistent support degree, to obtain pointwise variable precision fusion results, to put every expert's acupuncture acupoint location fusion error enhanced to pointwise variable precision. Then, we more effectively used the measured characteristics of different acupuncture expert's acupoint location, to improve the measurement information utilization efficiency and acupuncture acupoint location precision and accuracy. Based on using the consistency matrix pointwise fusion method on the acupuncture experts' acupoint location values, each expert's acupoint location information could be calculated, and the most precise and accurate values of each expert's acupoint location could be obtained.

Introduction of a New Diagnostic Method for Breast Cancer Based on Fine Needle Aspiration (FNA) Test Data and Combining Intelligent Systems

PubMed Central

Fiuzy, Mohammad; Haddadnia, Javad; Mollania, Nasrin; Hashemian, Maryam; Hassanpour, Kazem

2012-01-01

Background Accurate Diagnosis of Breast Cancer is of prime importance. Fine Needle Aspiration test or "FNA”, which has been used for several years in Europe, is a simple, inexpensive, noninvasive and accurate technique for detecting breast cancer. Expending the suitable features of the Fine Needle Aspiration results is the most important diagnostic problem in early stages of breast cancer. In this study, we introduced a new algorithm that can detect breast cancer based on combining artificial intelligent system and Fine Needle Aspiration (FNA). Methods We studied the Features of Wisconsin Data Base Cancer which contained about 569 FNA test samples (212 patient samples (malignant) and 357 healthy samples (benign)). In this research, we combined Artificial Intelligence Approaches, such as Evolutionary Algorithm (EA) with Genetic Algorithm (GA), and also used Exact Classifier Systems (here by Fuzzy C-Means (FCM)) to separate malignant from benign samples. Furthermore, we examined artificial Neural Networks (NN) to identify the model and structure. This research proposed a new algorithm for an accurate diagnosis of breast cancer. Results According to Wisconsin Data Base Cancer (WDBC) data base, 62.75% of samples were benign, and 37.25% were malignant. After applying the proposed algorithm, we achieved high detection accuracy of about "96.579%” on 205 patients who were diagnosed as having breast cancer. It was found that the method had 93% sensitivity, 73% specialty, 65% positive predictive value, and 95% negative predictive value, respectively. If done by experts, Fine Needle Aspiration (FNA) can be a reliable replacement for open biopsy in palpable breast masses. Evaluation of FNA samples during aspiration can decrease insufficient samples. FNA can be the first line of diagnosis in women with breast masses, at least in deprived regions, and may increase health standards and clinical supervision of patients. Conclusion Such a smart, economical, non-invasive, rapid and accurate system can be introduced as a useful diagnostic system for comprehensive treatment of breast cancer. Another advantage of this method is the possibility of diagnosing breast abnormalities. If done by experts, FNA can be a reliable replacement for open biopsy in palpable breast masses. Evaluation of FNA samples during aspiration can decrease insufficient samples. PMID:25352966

Improved operator agreement and efficiency using the minimum area contour change method for delineation of hyperintense multiple sclerosis lesions on FLAIR MRI

PubMed Central

2013-01-01

Background Activity of disease in patients with multiple sclerosis (MS) is monitored by detecting and delineating hyper-intense lesions on MRI scans. The Minimum Area Contour Change (MACC) algorithm has been created with two main goals: a) to improve inter-operator agreement on outlining regions of interest (ROIs) and b) to automatically propagate longitudinal ROIs from the baseline scan to a follow-up scan. Methods The MACC algorithm first identifies an outer bound for the solution path, forms a high number of iso-contour curves based on equally spaced contour values, and then selects the best contour value to outline the lesion. The MACC software was tested on a set of 17 FLAIR MRI images evaluated by a pair of human experts and a longitudinal dataset of 12 pairs of T2-weighted Fluid Attenuated Inversion Recovery (FLAIR) images that had lesion analysis ROIs drawn by a single expert operator. Results In the tests where two human experts evaluated the same MRI images, the MACC program demonstrated that it could markedly reduce inter-operator outline error. In the longitudinal part of the study, the MACC program created ROIs on follow-up scans that were in close agreement to the original expert’s ROIs. Finally, in a post-hoc analysis of 424 follow-up scans 91% of propagated MACC were accepted by an expert and only 9% of the final accepted ROIS had to be created or edited by the expert. Conclusion When used with an expert operator's verification of automatically created ROIs, MACC can be used to improve inter- operator agreement and decrease analysis time, which should improve data collected and analyzed in multicenter clinical trials. PMID:24004511

An expert fitness diagnosis system based on elastic cloud computing.

PubMed

Tseng, Kevin C; Wu, Chia-Chuan

2014-01-01

This paper presents an expert diagnosis system based on cloud computing. It classifies a user's fitness level based on supervised machine learning techniques. This system is able to learn and make customized diagnoses according to the user's physiological data, such as age, gender, and body mass index (BMI). In addition, an elastic algorithm based on Poisson distribution is presented to allocate computation resources dynamically. It predicts the required resources in the future according to the exponential moving average of past observations. The experimental results show that Naïve Bayes is the best classifier with the highest accuracy (90.8%) and that the elastic algorithm is able to capture tightly the trend of requests generated from the Internet and thus assign corresponding computation resources to ensure the quality of service.

Automated power management and control

NASA Technical Reports Server (NTRS)

Dolce, James L.

1991-01-01

A comprehensive automation design is being developed for Space Station Freedom's electric power system. A joint effort between NASA's Office of Aeronautics and Exploration Technology and NASA's Office of Space Station Freedom, it strives to increase station productivity by applying expert systems and conventional algorithms to automate power system operation. The initial station operation will use ground-based dispatches to perform the necessary command and control tasks. These tasks constitute planning and decision-making activities that strive to eliminate unplanned outages. We perceive an opportunity to help these dispatchers make fast and consistent on-line decisions by automating three key tasks: failure detection and diagnosis, resource scheduling, and security analysis. Expert systems will be used for the diagnostics and for the security analysis; conventional algorithms will be used for the resource scheduling.

Construct validation of an interactive digital algorithm for ostomy care.

PubMed

Beitz, Janice M; Gerlach, Mary A; Schafer, Vickie

2014-01-01

The purpose of this study was to evaluate construct validity for a previously face and content validated Ostomy Algorithm using digital real-life clinical scenarios. A cross-sectional, mixed-methods Web-based survey design study was conducted. Two hundred ninety-seven English-speaking RNs completed the study; participants practiced in both acute care and postacute settings, with 1 expert ostomy nurse (WOC nurse) and 2 nonexpert nurses. Following written consent, respondents answered demographic questions and completed a brief algorithm tutorial. Participants were then presented with 7 ostomy-related digital scenarios consisting of real-life photos and pertinent clinical information. Respondents used the 11 assessment components of the digital algorithm to choose management options. Participant written comments about the scenarios and the research process were collected. The mean overall percentage of correct responses was 84.23%. Mean percentage of correct responses for respondents with a self-reported basic ostomy knowledge was 87.7%; for those with a self-reported intermediate ostomy knowledge was 85.88% and those who were self-reported experts in ostomy care achieved 82.77% correct response rate. Five respondents reported having no prior ostomy care knowledge at screening and achieved an overall 45.71% correct response rate. No negative comments regarding the algorithm were recorded by participants. The new standardized Ostomy Algorithm remains the only face, content, and construct validated digital clinical decision instrument currently available. Further research on application at the bedside while tracking patient outcomes is warranted.

SpikeGUI: software for rapid interictal discharge annotation via template matching and online machine learning.

PubMed

Jing Jin; Dauwels, Justin; Cash, Sydney; Westover, M Brandon

2014-01-01

Detection of interictal discharges is a key element of interpreting EEGs during the diagnosis and management of epilepsy. Because interpretation of clinical EEG data is time-intensive and reliant on experts who are in short supply, there is a great need for automated spike detectors. However, attempts to develop general-purpose spike detectors have so far been severely limited by a lack of expert-annotated data. Huge databases of interictal discharges are therefore in great demand for the development of general-purpose detectors. Detailed manual annotation of interictal discharges is time consuming, which severely limits the willingness of experts to participate. To address such problems, a graphical user interface "SpikeGUI" was developed in our work for the purposes of EEG viewing and rapid interictal discharge annotation. "SpikeGUI" substantially speeds up the task of annotating interictal discharges using a custom-built algorithm based on a combination of template matching and online machine learning techniques. While the algorithm is currently tailored to annotation of interictal epileptiform discharges, it can easily be generalized to other waveforms and signal types.

SpikeGUI: Software for Rapid Interictal Discharge Annotation via Template Matching and Online Machine Learning

PubMed Central

Jin, Jing; Dauwels, Justin; Cash, Sydney; Westover, M. Brandon

2015-01-01

Detection of interictal discharges is a key element of interpreting EEGs during the diagnosis and management of epilepsy. Because interpretation of clinical EEG data is time-intensive and reliant on experts who are in short supply, there is a great need for automated spike detectors. However, attempts to develop general-purpose spike detectors have so far been severely limited by a lack of expert-annotated data. Huge databases of interictal discharges are therefore in great demand for the development of general-purpose detectors. Detailed manual annotation of interictal discharges is time consuming, which severely limits the willingness of experts to participate. To address such problems, a graphical user interface “SpikeGUI” was developed in our work for the purposes of EEG viewing and rapid interictal discharge annotation. “SpikeGUI” substantially speeds up the task of annotating interictal discharges using a custom-built algorithm based on a combination of template matching and online machine learning techniques. While the algorithm is currently tailored to annotation of interictal epileptiform discharges, it can easily be generalized to other waveforms and signal types. PMID:25570976

A photon recycling approach to the denoising of ultra-low dose X-ray sequences.

PubMed

Hariharan, Sai Gokul; Strobel, Norbert; Kaethner, Christian; Kowarschik, Markus; Demirci, Stefanie; Albarqouni, Shadi; Fahrig, Rebecca; Navab, Nassir

2018-06-01

Clinical procedures that make use of fluoroscopy may expose patients as well as the clinical staff (throughout their career) to non-negligible doses of radiation. The potential consequences of such exposures fall under two categories, namely stochastic (mostly cancer) and deterministic risks (skin injury). According to the "as low as reasonably achievable" principle, the radiation dose can be lowered only if the necessary image quality can be maintained. Our work improves upon the existing patch-based denoising algorithms by utilizing a more sophisticated noise model to exploit non-local self-similarity better and this in turn improves the performance of low-rank approximation. The novelty of the proposed approach lies in its properly designed and parameterized noise model and the elimination of initial estimates. This reduces the computational cost significantly. The algorithm has been evaluated on 500 clinical images (7 patients, 20 sequences, 3 clinical sites), taken at ultra-low dose levels, i.e. 50% of the standard low dose level, during electrophysiology procedures. An average improvement in the contrast-to-noise ratio (CNR) by a factor of around 3.5 has been found. This is associated with an image quality achieved at around 12 (square of 3.5) times the ultra-low dose level. Qualitative evaluation by X-ray image quality experts suggests that the method produces denoised images that comply with the required image quality criteria. The results are consistent with the number of patches used, and they demonstrate that it is possible to use motion estimation techniques and "recycle" photons from previous frames to improve the image quality of the current frame. Our results are comparable in terms of CNR to Video Block Matching 3D-a state-of-the-art denoising method. But qualitative analysis by experts confirms that the denoised ultra-low dose X-ray images obtained using our method are more realistic with respect to appearance.

AlexSys: a knowledge-based expert system for multiple sequence alignment construction and analysis

PubMed Central

Aniba, Mohamed Radhouene; Poch, Olivier; Marchler-Bauer, Aron; Thompson, Julie Dawn

2010-01-01

Multiple sequence alignment (MSA) is a cornerstone of modern molecular biology and represents a unique means of investigating the patterns of conservation and diversity in complex biological systems. Many different algorithms have been developed to construct MSAs, but previous studies have shown that no single aligner consistently outperforms the rest. This has led to the development of a number of ‘meta-methods’ that systematically run several aligners and merge the output into one single solution. Although these methods generally produce more accurate alignments, they are inefficient because all the aligners need to be run first and the choice of the best solution is made a posteriori. Here, we describe the development of a new expert system, AlexSys, for the multiple alignment of protein sequences. AlexSys incorporates an intelligent inference engine to automatically select an appropriate aligner a priori, depending only on the nature of the input sequences. The inference engine was trained on a large set of reference multiple alignments, using a novel machine learning approach. Applying AlexSys to a test set of 178 alignments, we show that the expert system represents a good compromise between alignment quality and running time, making it suitable for high throughput projects. AlexSys is freely available from http://alnitak.u-strasbg.fr/∼aniba/alexsys. PMID:20530533

Model of experts for decision support in the diagnosis of leukemia patients.

PubMed

Corchado, Juan M; De Paz, Juan F; Rodríguez, Sara; Bajo, Javier

2009-07-01

Recent advances in the field of biomedicine, specifically in the field of genomics, have led to an increase in the information available for conducting expression analysis. Expression analysis is a technique used in transcriptomics, a branch of genomics that deals with the study of messenger ribonucleic acid (mRNA) and the extraction of information contained in the genes. This increase in information is reflected in the exon arrays, which require the use of new techniques in order to extract the information. The purpose of this study is to provide a tool based on a mixture of experts model that allows the analysis of the information contained in the exon arrays, from which automatic classifications for decision support in diagnoses of leukemia patients can be made. The proposed model integrates several cooperative algorithms characterized for their efficiency for data processing, filtering, classification and knowledge extraction. The Cancer Institute of the University of Salamanca is making an effort to develop tools to automate the evaluation of data and to facilitate de analysis of information. This proposal is a step forward in this direction and the first step toward the development of a mixture of experts tool that integrates different cognitive and statistical approaches to deal with the analysis of exon arrays. The mixture of experts model presented within this work provides great capacities for learning and adaptation to the characteristics of the problem in consideration, using novel algorithms in each of the stages of the analysis process that can be easily configured and combined, and provides results that notably improve those provided by the existing methods for exon arrays analysis. The material used consists of data from exon arrays provided by the Cancer Institute that contain samples from leukemia patients. The methodology used consists of a system based on a mixture of experts. Each one of the experts incorporates novel artificial intelligence techniques that improve the process of carrying out various tasks such as pre-processing, filtering, classification and extraction of knowledge. This article will detail the manner in which individual experts are combined so that together they generate a system capable of extracting knowledge, thus permitting patients to be classified in an automatic and efficient manner that is also comprehensible for medical personnel. The system has been tested in a real setting and has been used for classifying patients who suffer from different forms of leukemia at various stages. Personnel from the Cancer Institute supervised and participated throughout the testing period. Preliminary results are promising, notably improving the results obtained with previously used tools. The medical staff from the Cancer Institute considers the tools that have been developed to be positive and very useful in a supporting capacity for carrying out their daily tasks. Additionally the mixture of experts supplies a tool for the extraction of necessary information in order to explain the associations that have been made in simple terms. That is, it permits the extraction of knowledge for each classification made and generalized in order to be used in subsequent classifications. This allows for a large amount of learning and adaptation within the proposed system.

Learning to Rank Figures within a Biomedical Article

PubMed Central

Liu, Feifan; Yu, Hong

2014-01-01

Hundreds of millions of figures are available in biomedical literature, representing important biomedical experimental evidence. This ever-increasing sheer volume has made it difficult for scientists to effectively and accurately access figures of their interest, the process of which is crucial for validating research facts and for formulating or testing novel research hypotheses. Current figure search applications can't fully meet this challenge as the “bag of figures” assumption doesn't take into account the relationship among figures. In our previous study, hundreds of biomedical researchers have annotated articles in which they serve as corresponding authors. They ranked each figure in their paper based on a figure's importance at their discretion, referred to as “figure ranking”. Using this collection of annotated data, we investigated computational approaches to automatically rank figures. We exploited and extended the state-of-the-art listwise learning-to-rank algorithms and developed a new supervised-learning model BioFigRank. The cross-validation results show that BioFigRank yielded the best performance compared with other state-of-the-art computational models, and the greedy feature selection can further boost the ranking performance significantly. Furthermore, we carry out the evaluation by comparing BioFigRank with three-level competitive domain-specific human experts: (1) First Author, (2) Non-Author-In-Domain-Expert who is not the author nor co-author of an article but who works in the same field of the corresponding author of the article, and (3) Non-Author-Out-Domain-Expert who is not the author nor co-author of an article and who may or may not work in the same field of the corresponding author of an article. Our results show that BioFigRank outperforms Non-Author-Out-Domain-Expert and performs as well as Non-Author-In-Domain-Expert. Although BioFigRank underperforms First Author, since most biomedical researchers are either in- or out-domain-experts for an article, we conclude that BioFigRank represents an artificial intelligence system that offers expert-level intelligence to help biomedical researchers to navigate increasingly proliferated big data efficiently. PMID:24625719

Learning to rank figures within a biomedical article.

PubMed

Liu, Feifan; Yu, Hong

2014-01-01

Hundreds of millions of figures are available in biomedical literature, representing important biomedical experimental evidence. This ever-increasing sheer volume has made it difficult for scientists to effectively and accurately access figures of their interest, the process of which is crucial for validating research facts and for formulating or testing novel research hypotheses. Current figure search applications can't fully meet this challenge as the "bag of figures" assumption doesn't take into account the relationship among figures. In our previous study, hundreds of biomedical researchers have annotated articles in which they serve as corresponding authors. They ranked each figure in their paper based on a figure's importance at their discretion, referred to as "figure ranking". Using this collection of annotated data, we investigated computational approaches to automatically rank figures. We exploited and extended the state-of-the-art listwise learning-to-rank algorithms and developed a new supervised-learning model BioFigRank. The cross-validation results show that BioFigRank yielded the best performance compared with other state-of-the-art computational models, and the greedy feature selection can further boost the ranking performance significantly. Furthermore, we carry out the evaluation by comparing BioFigRank with three-level competitive domain-specific human experts: (1) First Author, (2) Non-Author-In-Domain-Expert who is not the author nor co-author of an article but who works in the same field of the corresponding author of the article, and (3) Non-Author-Out-Domain-Expert who is not the author nor co-author of an article and who may or may not work in the same field of the corresponding author of an article. Our results show that BioFigRank outperforms Non-Author-Out-Domain-Expert and performs as well as Non-Author-In-Domain-Expert. Although BioFigRank underperforms First Author, since most biomedical researchers are either in- or out-domain-experts for an article, we conclude that BioFigRank represents an artificial intelligence system that offers expert-level intelligence to help biomedical researchers to navigate increasingly proliferated big data efficiently.

The limits of electronic fetal heart rate monitoring in the prevention of neonatal metabolic acidemia.

PubMed

Clark, Steven L; Hamilton, Emily F; Garite, Thomas J; Timmins, Audra; Warrick, Philip A; Smith, Samuel

2017-02-01

Despite intensive efforts directed at initial training in fetal heart rate interpretation, continuing medical education, board certification/recertification, team training, and the development of specific protocols for the management of abnormal fetal heart rate patterns, the goals of consistently preventing hypoxia-induced fetal metabolic acidemia and neurologic injury remain elusive. The purpose of this study was to validate a recently published algorithm for the management of category II fetal heart rate tracings, to examine reasons for the birth of infants with significant metabolic acidemia despite the use of electronic fetal heart rate monitoring, and to examine critically the limits of electronic fetal heart rate monitoring in the prevention of neonatal metabolic acidemia. The potential performance of electronic fetal heart rate monitoring under ideal circumstances was evaluated in an outcomes-blinded examination fetal heart rate tracing of infants with metabolic acidemia at birth (base deficit, >12) and matched control infants (base deficit, <8) under the following conditions: (1) expert primary interpretation, (2) use of a published algorithm that was developed and endorsed by a large group of national experts, (3) assumption of a 30-minute period of evaluation for noncritical category II fetal heart rate tracings, followed by delivery within 30 minutes, (4) evaluation without the need to provide patient care simultaneously, and (5) comparison of results under these circumstances with those achieved in actual clinical practice. During the study period, 120 infants were identified with an arterial cord blood base deficit of >12 mM/L. Matched control infants were not demographically different from subjects. In actual practice, operative intervention on the basis of an abnormal fetal heart rate tracings occurred in 36 of 120 fetuses (30.0%) with metabolic acidemia. Based on expert, algorithm-assisted reviews, 55 of 120 patients with acidemia (45.8%) were judged to need operative intervention for abnormal fetal heart rate tracings. This difference was significant (P=.016). In infants who were born with a base deficit of >12 mM/L in which blinded, algorithm-assisted expert review indicated the need for operative delivery, the decision for delivery would have been made an average of 131 minutes before the actual delivery. The rate of expert intervention for fetal heart rate concerns in the nonacidemic control group (22/120; 18.3%) was similar to the actual intervention rate (23/120; 19.2%; P=1.0) Expert review did not mandate earlier delivery in 65 of 120 patients with metabolic acidemia. The primary features of these 65 cases included the occurrence of sentinel events with prolonged deceleration just before delivery, the rapid deterioration of nonemergent category II fetal heart rate tracings before realistic time frames for recognition and intervention, and the failure of recognized fetal heart rate patterns such as variability to identify metabolic acidemia. Expert, algorithm-assisted fetal heart rate interpretation has the potential to improve standard clinical performance by facilitating significantly earlier recognition of some tracings that are associated with metabolic acidemia without increasing the rate of operative intervention. However, this improvement is modest. Of infants who are born with metabolic acidemia, only approximately one-half potentially could be identified and have delivery expedited even under ideal circumstances, which are probably not realistic in current US practice. This represents the limits of electronic fetal heart rate monitoring performance. Additional technologies will be necessary if the goal of the prevention of neonatal metabolic acidemia is to be realized. Copyright © 2016 Elsevier Inc. All rights reserved.

A wavelet based method for automatic detection of slow eye movements: a pilot study.

PubMed

Magosso, Elisa; Provini, Federica; Montagna, Pasquale; Ursino, Mauro

2006-11-01

Electro-oculographic (EOG) activity during the wake-sleep transition is characterized by the appearance of slow eye movements (SEM). The present work describes an algorithm for the automatic localisation of SEM events from EOG recordings. The algorithm is based on a wavelet multiresolution analysis of the difference between right and left EOG tracings, and includes three main steps: (i) wavelet decomposition down to 10 detail levels (i.e., 10 scales), using Daubechies order 4 wavelet; (ii) computation of energy in 0.5s time steps at any level of decomposition; (iii) construction of a non-linear discriminant function expressing the relative energy of high-scale details to both high- and low-scale details. The main assumption is that the value of the discriminant function increases above a given threshold during SEM episodes due to energy redistribution toward higher scales. Ten EOG recordings from ten male patients with obstructive sleep apnea syndrome were used. All tracings included a period from pre-sleep wakefulness to stage 2 sleep. Two experts inspected the tracings separately to score SEMs. A reference set of SEM (gold standard) were obtained by joint examination by both experts. Parameters of the discriminant function were assigned on three tracings (design set) to minimize the disagreement between the system classification and classification by the two experts; the algorithm was then tested on the remaining seven tracings (test set). Results show that the agreement between the algorithm and the gold standard was 80.44+/-4.09%, the sensitivity of the algorithm was 67.2+/-7.37% and the selectivity 83.93+/-8.65%. However, most errors were not caused by an inability of the system to detect intervals with SEM activity against NON-SEM intervals, but were due to a different localisation of the beginning and end of some SEM episodes. The proposed method may be a valuable tool for computerized EOG analysis.

The Swiss Data Science Center on a mission to empower reproducible, traceable and reusable science

NASA Astrophysics Data System (ADS)

Schymanski, Stanislaus; Bouillet, Eric; Verscheure, Olivier

2017-04-01

Our abilities to collect, store and analyse scientific data have sky-rocketed in the past decades, but at the same time, a disconnect between data scientists, domain experts and data providers has begun to emerge. Data scientists are developing more and more powerful algorithms for data mining and analysis, while data providers are making more and more data publicly available, and yet many, if not most, discoveries are based on specific data and/or algorithms that "are available from the authors upon request". In the strong belief that scientific progress would be much faster if reproduction and re-use of such data and algorithms was made easier, the Swiss Data Science Center (SDSC) has committed to provide an open framework for the handling and tracking of scientific data and algorithms, from raw data and first principle equations to final data products and visualisations, modular simulation models and benchmark evaluation algorithms. Led jointly by EPFL and ETH Zurich, the SDSC is composed of a distributed multi-disciplinary team of data scientists and experts in select domains. The center aims to federate data providers, data and computer scientists, and subject-matter experts around a cutting-edge analytics platform offering user-friendly tooling and services to help with the adoption of Open Science, fostering research productivity and excellence. In this presentation, we will discuss our vision of a high-scalable open but secure community-based platform for sharing, accessing, exploring, and analyzing scientific data in easily reproducible workflows, augmented by automated provenance and impact tracking, knowledge graphs, fine-grained access right and digital right management, and a variety of domain-specific software tools. For maximum interoperability, transparency and ease of use, we plan to utilize notebook interfaces wherever possible, such as Apache Zeppelin and Jupyter. Feedback and suggestions from the audience will be gratefully considered.

Machine Learning in Medicine

PubMed Central

Deo, Rahul C.

2015-01-01

Spurred by advances in processing power, memory, storage, and an unprecedented wealth of data, computers are being asked to tackle increasingly complex learning tasks, often with astonishing success. Computers have now mastered a popular variant of poker, learned the laws of physics from experimental data, and become experts in video games – tasks which would have been deemed impossible not too long ago. In parallel, the number of companies centered on applying complex data analysis to varying industries has exploded, and it is thus unsurprising that some analytic companies are turning attention to problems in healthcare. The purpose of this review is to explore what problems in medicine might benefit from such learning approaches and use examples from the literature to introduce basic concepts in machine learning. It is important to note that seemingly large enough medical data sets and adequate learning algorithms have been available for many decades – and yet, although there are thousands of papers applying machine learning algorithms to medical data, very few have contributed meaningfully to clinical care. This lack of impact stands in stark contrast to the enormous relevance of machine learning to many other industries. Thus part of my effort will be to identify what obstacles there may be to changing the practice of medicine through statistical learning approaches, and discuss how these might be overcome. PMID:26572668

Follicular patterned lesions of the thyroid gland: a practical algorithmic approach.

PubMed

Chetty, Runjan

2011-09-01

Follicular patterned lesions of the thyroid are problematic and interpretation is often subjective. While thyroid experts are comfortable with their own criteria and thresholds, those encountering these lesions sporadically have a degree of uncertainty with a proportion of cases. The purpose of this review is to highlight the importance of proper diligent sampling of an encapsulated thyroid lesion (in totality in many cases), examination for capsular and vascular invasion, and finally the assessment of nuclear changes that are pathognomonic of papillary thyroid carcinoma (PTC). Based on these established criteria, an algorithmic approach is suggested using known, accepted terminology. The importance of unequivocal, clear-cut nuclear features of PTC as opposed to inconclusive features is stressed. If the nuclear features in an encapsulated, non-invasive follicular patterned lesion fall short of those encountered in classical PTC, but nonetheless are still worrying or concerning, the term 'uncertain malignant potential or behaviour, most likely benign' is suggested. Indubitable, classical PTC nuclei (whether diffuse or restricted to a single high-power field) are diagnostic of a PTC be it classical, non-invasive or invasive follicular variant PTC. Capsular and vascular invasion remain the only reliable predictors of outcome, as non-invasive, encapsulated follicular variant PTC, even with diffuse PTC nuclear change, behaves in an indolent fashion.

The influence of expert opinions on the selection of wastewater treatment alternatives: a group decision-making approach.

PubMed

Kalbar, Pradip P; Karmakar, Subhankar; Asolekar, Shyam R

2013-10-15

The application of multiple-attribute decision-making (MADM) to real life decision problems suggests that avoiding the loss of information through scenario-based approaches and including expert opinions in the decision-making process are two major challenges that require more research efforts. Recently, a wastewater treatment technology selection effort has been made with a 'scenario-based' method of MADM. This paper focuses on a novel approach to incorporate expert opinions into the scenario-based decision-making process, as expert opinions play a major role in the selection of treatment technologies. The sets of criteria and the indicators that are used consist of both qualitative and quantitative criteria. The group decision-making (GDM) approach that is implemented for aggregating expert opinions is based on an analytical hierarchy process (AHP), which is the most widely used MADM method. The pairwise comparison matrices (PCMs) for qualitative criteria are formed based on expert opinions, whereas, a novel approach is proposed for generating PCMs for quantitative criteria. It has been determined that the experts largely prefer natural treatment systems because they are more sustainable in any scenario. However, PCMs based on expert opinions suggest that advanced technologies such as the sequencing batch reactor (SBR) can also be appropriate for a given decision scenario. The proposed GDM approach is a rationalized process that will be more appropriate in realistic scenarios where multiple stakeholders with local and regional societal priorities are involved in the selection of treatment technology. Copyright © 2013 Elsevier Ltd. All rights reserved.

Breast tumor malignancy modelling using evolutionary neural logic networks.

PubMed

Tsakonas, Athanasios; Dounias, Georgios; Panagi, Georgia; Panourgias, Evangelia

2006-01-01

The present work proposes a computer assisted methodology for the effective modelling of the diagnostic decision for breast tumor malignancy. The suggested approach is based on innovative hybrid computational intelligence algorithms properly applied in related cytological data contained in past medical records. The experimental data used in this study were gathered in the early 1990s in the University of Wisconsin, based in post diagnostic cytological observations performed by expert medical staff. Data were properly encoded in a computer database and accordingly, various alternative modelling techniques were applied on them, in an attempt to form diagnostic models. Previous methods included standard optimisation techniques, as well as artificial intelligence approaches, in a way that a variety of related publications exists in modern literature on the subject. In this report, a hybrid computational intelligence approach is suggested, which effectively combines modern mathematical logic principles, neural computation and genetic programming in an effective manner. The approach proves promising either in terms of diagnostic accuracy and generalization capabilities, or in terms of comprehensibility and practical importance for the related medical staff.

Using simulated fluorescence cell micrographs for the evaluation of cell image segmentation algorithms.

PubMed

Wiesmann, Veit; Bergler, Matthias; Palmisano, Ralf; Prinzen, Martin; Franz, Daniela; Wittenberg, Thomas

2017-03-18

Manual assessment and evaluation of fluorescent micrograph cell experiments is time-consuming and tedious. Automated segmentation pipelines can ensure efficient and reproducible evaluation and analysis with constant high quality for all images of an experiment. Such cell segmentation approaches are usually validated and rated in comparison to manually annotated micrographs. Nevertheless, manual annotations are prone to errors and display inter- and intra-observer variability which influence the validation results of automated cell segmentation pipelines. We present a new approach to simulate fluorescent cell micrographs that provides an objective ground truth for the validation of cell segmentation methods. The cell simulation was evaluated twofold: (1) An expert observer study shows that the proposed approach generates realistic fluorescent cell micrograph simulations. (2) An automated segmentation pipeline on the simulated fluorescent cell micrographs reproduces segmentation performances of that pipeline on real fluorescent cell micrographs. The proposed simulation approach produces realistic fluorescent cell micrographs with corresponding ground truth. The simulated data is suited to evaluate image segmentation pipelines more efficiently and reproducibly than it is possible on manually annotated real micrographs.

Expert systems for automated maintenance of a Mars oxygen production system

NASA Astrophysics Data System (ADS)

Huang, Jen-Kuang; Ho, Ming-Tsang; Ash, Robert L.

1992-08-01

Application of expert system concepts to a breadboard Mars oxygen processor unit have been studied and tested. The research was directed toward developing the methodology required to enable autonomous operation and control of these simple chemical processors at Mars. Failure detection and isolation was the key area of concern, and schemes using forward chaining, backward chaining, knowledge-based expert systems, and rule-based expert systems were examined. Tests and simulations were conducted that investigated self-health checkout, emergency shutdown, and fault detection, in addition to normal control activities. A dynamic system model was developed using the Bond-Graph technique. The dynamic model agreed well with tests involving sudden reductions in throughput. However, nonlinear effects were observed during tests that incorporated step function increases in flow variables. Computer simulations and experiments have demonstrated the feasibility of expert systems utilizing rule-based diagnosis and decision-making algorithms.

Expert systems for space power supply - Design, analysis, and evaluation

NASA Technical Reports Server (NTRS)

Cooper, Ralph S.; Thomson, M. Kemer; Hoshor, Alan

1987-01-01

The feasibility of applying expert systems to the conceptual design, analysis, and evaluation of space power supplies in particular, and complex systems in general is evaluated. To do this, the space power supply design process and its associated knowledge base were analyzed and characterized in a form suitable for computer emulation of a human expert. The existing expert system tools and the results achieved with them were evaluated to assess their applicability to power system design. Some new concepts for combining program architectures (modular expert systems and algorithms) with information about the domain were applied to create a 'deep' system for handling the complex design problem. NOVICE, a code to solve a simplified version of a scoping study of a wide variety of power supply types for a broad range of missions, has been developed, programmed, and tested as a concrete feasibility demonstration.

Real alerts and artifact classification in archived multi-signal vital sign monitoring data: implications for mining big data.

PubMed

Hravnak, Marilyn; Chen, Lujie; Dubrawski, Artur; Bose, Eliezer; Clermont, Gilles; Pinsky, Michael R

2016-12-01

Huge hospital information system databases can be mined for knowledge discovery and decision support, but artifact in stored non-invasive vital sign (VS) high-frequency data streams limits its use. We used machine-learning (ML) algorithms trained on expert-labeled VS data streams to automatically classify VS alerts as real or artifact, thereby "cleaning" such data for future modeling. 634 admissions to a step-down unit had recorded continuous noninvasive VS monitoring data [heart rate (HR), respiratory rate (RR), peripheral arterial oxygen saturation (SpO 2 ) at 1/20 Hz, and noninvasive oscillometric blood pressure (BP)]. Time data were across stability thresholds defined VS event epochs. Data were divided Block 1 as the ML training/cross-validation set and Block 2 the test set. Expert clinicians annotated Block 1 events as perceived real or artifact. After feature extraction, ML algorithms were trained to create and validate models automatically classifying events as real or artifact. The models were then tested on Block 2. Block 1 yielded 812 VS events, with 214 (26 %) judged by experts as artifact (RR 43 %, SpO 2 40 %, BP 15 %, HR 2 %). ML algorithms applied to the Block 1 training/cross-validation set (tenfold cross-validation) gave area under the curve (AUC) scores of 0.97 RR, 0.91 BP and 0.76 SpO 2 . Performance when applied to Block 2 test data was AUC 0.94 RR, 0.84 BP and 0.72 SpO 2 . ML-defined algorithms applied to archived multi-signal continuous VS monitoring data allowed accurate automated classification of VS alerts as real or artifact, and could support data mining for future model building.

Real Alerts and Artifact Classification in Archived Multi-signal Vital Sign Monitoring Data—Implications for Mining Big Data — Implications for Mining Big Data

PubMed Central

Hravnak, Marilyn; Chen, Lujie; Dubrawski, Artur; Bose, Eliezer; Clermont, Gilles; Pinsky, Michael R.

2015-01-01

PURPOSE Huge hospital information system databases can be mined for knowledge discovery and decision support, but artifact in stored non-invasive vital sign (VS) high-frequency data streams limits its use. We used machine-learning (ML) algorithms trained on expert-labeled VS data streams to automatically classify VS alerts as real or artifact, thereby “cleaning” such data for future modeling. METHODS 634 admissions to a step-down unit had recorded continuous noninvasive VS monitoring data (heart rate [HR], respiratory rate [RR], peripheral arterial oxygen saturation [SpO2] at 1/20Hz., and noninvasive oscillometric blood pressure [BP]) Time data were across stability thresholds defined VS event epochs. Data were divided Block 1 as the ML training/cross-validation set and Block 2 the test set. Expert clinicians annotated Block 1 events as perceived real or artifact. After feature extraction, ML algorithms were trained to create and validate models automatically classifying events as real or artifact. The models were then tested on Block 2. RESULTS Block 1 yielded 812 VS events, with 214 (26%) judged by experts as artifact (RR 43%, SpO2 40%, BP 15%, HR 2%). ML algorithms applied to the Block 1 training/cross-validation set (10-fold cross-validation) gave area under the curve (AUC) scores of 0.97 RR, 0.91 BP and 0.76 SpO2. Performance when applied to Block 2 test data was AUC 0.94 RR, 0.84 BP and 0.72 SpO2). CONCLUSIONS ML-defined algorithms applied to archived multi-signal continuous VS monitoring data allowed accurate automated classification of VS alerts as real or artifact, and could support data mining for future model building. PMID:26438655

Automatic peak selection by a Benjamini-Hochberg-based algorithm.

PubMed

Abbas, Ahmed; Kong, Xin-Bing; Liu, Zhi; Jing, Bing-Yi; Gao, Xin

2013-01-01

A common issue in bioinformatics is that computational methods often generate a large number of predictions sorted according to certain confidence scores. A key problem is then determining how many predictions must be selected to include most of the true predictions while maintaining reasonably high precision. In nuclear magnetic resonance (NMR)-based protein structure determination, for instance, computational peak picking methods are becoming more and more common, although expert-knowledge remains the method of choice to determine how many peaks among thousands of candidate peaks should be taken into consideration to capture the true peaks. Here, we propose a Benjamini-Hochberg (B-H)-based approach that automatically selects the number of peaks. We formulate the peak selection problem as a multiple testing problem. Given a candidate peak list sorted by either volumes or intensities, we first convert the peaks into [Formula: see text]-values and then apply the B-H-based algorithm to automatically select the number of peaks. The proposed approach is tested on the state-of-the-art peak picking methods, including WaVPeak [1] and PICKY [2]. Compared with the traditional fixed number-based approach, our approach returns significantly more true peaks. For instance, by combining WaVPeak or PICKY with the proposed method, the missing peak rates are on average reduced by 20% and 26%, respectively, in a benchmark set of 32 spectra extracted from eight proteins. The consensus of the B-H-selected peaks from both WaVPeak and PICKY achieves 88% recall and 83% precision, which significantly outperforms each individual method and the consensus method without using the B-H algorithm. The proposed method can be used as a standard procedure for any peak picking method and straightforwardly applied to some other prediction selection problems in bioinformatics. The source code, documentation and example data of the proposed method is available at http://sfb.kaust.edu.sa/pages/software.aspx.

Automatic Peak Selection by a Benjamini-Hochberg-Based Algorithm

PubMed Central

Abbas, Ahmed; Kong, Xin-Bing; Liu, Zhi; Jing, Bing-Yi; Gao, Xin

2013-01-01

A common issue in bioinformatics is that computational methods often generate a large number of predictions sorted according to certain confidence scores. A key problem is then determining how many predictions must be selected to include most of the true predictions while maintaining reasonably high precision. In nuclear magnetic resonance (NMR)-based protein structure determination, for instance, computational peak picking methods are becoming more and more common, although expert-knowledge remains the method of choice to determine how many peaks among thousands of candidate peaks should be taken into consideration to capture the true peaks. Here, we propose a Benjamini-Hochberg (B-H)-based approach that automatically selects the number of peaks. We formulate the peak selection problem as a multiple testing problem. Given a candidate peak list sorted by either volumes or intensities, we first convert the peaks into -values and then apply the B-H-based algorithm to automatically select the number of peaks. The proposed approach is tested on the state-of-the-art peak picking methods, including WaVPeak [1] and PICKY [2]. Compared with the traditional fixed number-based approach, our approach returns significantly more true peaks. For instance, by combining WaVPeak or PICKY with the proposed method, the missing peak rates are on average reduced by 20% and 26%, respectively, in a benchmark set of 32 spectra extracted from eight proteins. The consensus of the B-H-selected peaks from both WaVPeak and PICKY achieves 88% recall and 83% precision, which significantly outperforms each individual method and the consensus method without using the B-H algorithm. The proposed method can be used as a standard procedure for any peak picking method and straightforwardly applied to some other prediction selection problems in bioinformatics. The source code, documentation and example data of the proposed method is available at http://sfb.kaust.edu.sa/pages/software.aspx. PMID:23308147

DOE Office of Scientific and Technical Information (OSTI.GOV)

Zhang, Jing, E-mail: jing.zhang2@duke.edu; Ghate, Sujata V.; Yoon, Sora C.

Purpose: Mammography is the most widely accepted and utilized screening modality for early breast cancer detection. Providing high quality mammography education to radiology trainees is essential, since excellent interpretation skills are needed to ensure the highest benefit of screening mammography for patients. The authors have previously proposed a computer-aided education system based on trainee models. Those models relate human-assessed image characteristics to trainee error. In this study, the authors propose to build trainee models that utilize features automatically extracted from images using computer vision algorithms to predict likelihood of missing each mass by the trainee. This computer vision-based approach tomore » trainee modeling will allow for automatically searching large databases of mammograms in order to identify challenging cases for each trainee. Methods: The authors’ algorithm for predicting the likelihood of missing a mass consists of three steps. First, a mammogram is segmented into air, pectoral muscle, fatty tissue, dense tissue, and mass using automated segmentation algorithms. Second, 43 features are extracted using computer vision algorithms for each abnormality identified by experts. Third, error-making models (classifiers) are applied to predict the likelihood of trainees missing the abnormality based on the extracted features. The models are developed individually for each trainee using his/her previous reading data. The authors evaluated the predictive performance of the proposed algorithm using data from a reader study in which 10 subjects (7 residents and 3 novices) and 3 experts read 100 mammographic cases. Receiver operating characteristic (ROC) methodology was applied for the evaluation. Results: The average area under the ROC curve (AUC) of the error-making models for the task of predicting which masses will be detected and which will be missed was 0.607 (95% CI,0.564-0.650). This value was statistically significantly different from 0.5 (p < 0.0001). For the 7 residents only, the AUC performance of the models was 0.590 (95% CI,0.537-0.642) and was also significantly higher than 0.5 (p = 0.0009). Therefore, generally the authors’ models were able to predict which masses were detected and which were missed better than chance. Conclusions: The authors proposed an algorithm that was able to predict which masses will be detected and which will be missed by each individual trainee. This confirms existence of error-making patterns in the detection of masses among radiology trainees. Furthermore, the proposed methodology will allow for the optimized selection of difficult cases for the trainees in an automatic and efficient manner.« less

Standardized evaluation framework for evaluating coronary artery stenosis detection, stenosis quantification and lumen segmentation algorithms in computed tomography angiography.

PubMed

Kirişli, H A; Schaap, M; Metz, C T; Dharampal, A S; Meijboom, W B; Papadopoulou, S L; Dedic, A; Nieman, K; de Graaf, M A; Meijs, M F L; Cramer, M J; Broersen, A; Cetin, S; Eslami, A; Flórez-Valencia, L; Lor, K L; Matuszewski, B; Melki, I; Mohr, B; Oksüz, I; Shahzad, R; Wang, C; Kitslaar, P H; Unal, G; Katouzian, A; Örkisz, M; Chen, C M; Precioso, F; Najman, L; Masood, S; Ünay, D; van Vliet, L; Moreno, R; Goldenberg, R; Vuçini, E; Krestin, G P; Niessen, W J; van Walsum, T

2013-12-01

Though conventional coronary angiography (CCA) has been the standard of reference for diagnosing coronary artery disease in the past decades, computed tomography angiography (CTA) has rapidly emerged, and is nowadays widely used in clinical practice. Here, we introduce a standardized evaluation framework to reliably evaluate and compare the performance of the algorithms devised to detect and quantify the coronary artery stenoses, and to segment the coronary artery lumen in CTA data. The objective of this evaluation framework is to demonstrate the feasibility of dedicated algorithms to: (1) (semi-)automatically detect and quantify stenosis on CTA, in comparison with quantitative coronary angiography (QCA) and CTA consensus reading, and (2) (semi-)automatically segment the coronary lumen on CTA, in comparison with expert's manual annotation. A database consisting of 48 multicenter multivendor cardiac CTA datasets with corresponding reference standards are described and made available. The algorithms from 11 research groups were quantitatively evaluated and compared. The results show that (1) some of the current stenosis detection/quantification algorithms may be used for triage or as a second-reader in clinical practice, and that (2) automatic lumen segmentation is possible with a precision similar to that obtained by experts. The framework is open for new submissions through the website, at http://coronary.bigr.nl/stenoses/. Copyright © 2013 Elsevier B.V. All rights reserved.

A parallel expert system for the control of a robotic air vehicle

NASA Technical Reports Server (NTRS)

Shakley, Donald; Lamont, Gary B.

1988-01-01

Expert systems can be used to govern the intelligent control of vehicles, for example the Robotic Air Vehicle (RAV). Due to the nature of the RAV system the associated expert system needs to perform in a demanding real-time environment. The use of a parallel processing capability to support the associated expert system's computational requirement is critical in this application. Thus, algorithms for parallel real-time expert systems must be designed, analyzed, and synthesized. The design process incorporates a consideration of the rule-set/face-set size along with representation issues. These issues are looked at in reference to information movement and various inference mechanisms. Also examined is the process involved with transporting the RAV expert system functions from the TI Explorer, where they are implemented in the Automated Reasoning Tool (ART), to the iPSC Hypercube, where the system is synthesized using Concurrent Common LISP (CCLISP). The transformation process for the ART to CCLISP conversion is described. The performance characteristics of the parallel implementation of these expert systems on the iPSC Hypercube are compared to the TI Explorer implementation.

Best Practices for Reduction of Uncertainty in CFD Results

NASA Technical Reports Server (NTRS)

Mendenhall, Michael R.; Childs, Robert E.; Morrison, Joseph H.

2003-01-01

This paper describes a proposed best-practices system that will present expert knowledge in the use of CFD. The best-practices system will include specific guidelines to assist the user in problem definition, input preparation, grid generation, code selection, parameter specification, and results interpretation. The goal of the system is to assist all CFD users in obtaining high quality CFD solutions with reduced uncertainty and at lower cost for a wide range of flow problems. The best-practices system will be implemented as a software product which includes an expert system made up of knowledge databases of expert information with specific guidelines for individual codes and algorithms. The process of acquiring expert knowledge is discussed, and help from the CFD community is solicited. Benefits and challenges associated with this project are examined.

Active and Passive Haptic Training Approaches in VR Laparoscopic Surgery Training.

PubMed

Marutani, Takafumi; Kato, Toma; Tagawa, Kazuyoshi; Tanaka, Hiromi T; Komori, Masaru; Kurumi, Yoshimasa; Morikawa, Shigehiro

2016-01-01

Laparoscopic surgery has become a widely performed surgery as it is one of the most common minimally invasive surgeries. Doctors perform the surgery by manipulating thin and long surgical instruments precisely with the assistance of laparoscopic video with limited field of view. The power control of the instruments' tip is especially very important, because excessive power may damage internal organs. The training of this surgical technique is mainly supervised by an expert in hands-on coaching program. However, it is difficult for the experts to spend sufficient time for coaching. Therefore, we aim to teach the expert's hand movements in laparoscopic surgery to trainees using VR-based simulator, which is equipped with a guidance force display device. To realize the system, we propose two haptic training approaches for transferring the expert's hand movements to the trainee. One is active training, and the other is passive training. The former approach shows the expert's movements only when the trainee makes large errors while the latter shows the expert's movements continuously. In this study, we validate the applicability of these approaches through tasks in VR laparoscopic surgery training simulator, and identify the differences between these approaches.

Supervised learning of probability distributions by neural networks

NASA Technical Reports Server (NTRS)

Baum, Eric B.; Wilczek, Frank

1988-01-01

Supervised learning algorithms for feedforward neural networks are investigated analytically. The back-propagation algorithm described by Werbos (1974), Parker (1985), and Rumelhart et al. (1986) is generalized by redefining the values of the input and output neurons as probabilities. The synaptic weights are then varied to follow gradients in the logarithm of likelihood rather than in the error. This modification is shown to provide a more rigorous theoretical basis for the algorithm and to permit more accurate predictions. A typical application involving a medical-diagnosis expert system is discussed.

Automatic recognition of fundamental tissues on histology images of the human cardiovascular system.

PubMed

Mazo, Claudia; Trujillo, Maria; Alegre, Enrique; Salazar, Liliana

2016-10-01

Cardiovascular disease is the leading cause of death worldwide. Therefore, techniques for improving diagnosis and treatment in this field have become key areas for research. In particular, approaches for tissue image processing may support education system and medical practice. In this paper, an approach to automatic recognition and classification of fundamental tissues, using morphological information is presented. Taking a 40× or 10× histological image as input, three clusters are created with the k-means algorithm using a structural tensor and the red and the green channels. Loose connective tissue, light regions and cell nuclei are recognised on 40× images. Then, the cell nuclei's features - shape and spatial projection - and light regions are used to recognise and classify epithelial cells and tissue into flat, cubic and cylindrical. In a similar way, light regions, loose connective and muscle tissues are recognised on 10× images. Finally, the tissue's function and composition are used to refine muscle tissue recognition. Experimental validation is then carried out by histologist following expert criteria, along with manually annotated images that are used as a ground-truth. The results revealed that the proposed approach classified the fundamental tissues in a similar way to the conventional method employed by histologists. The proposed automatic recognition approach provides for epithelial tissues a sensitivity of 0.79 for cubic, 0.85 for cylindrical and 0.91 for flat. Furthermore, the experts gave our method an average score of 4.85 out of 5 in the recognition of loose connective tissue and 4.82 out of 5 for muscle tissue recognition. Copyright © 2016 Elsevier Ltd. All rights reserved.

TES: A modular systems approach to expert system development for real-time space applications

NASA Technical Reports Server (NTRS)

Cacace, Ralph; England, Brenda

1988-01-01

A major goal of the Space Station era is to reduce reliance on support from ground based experts. The development of software programs using expert systems technology is one means of reaching this goal without requiring crew members to become intimately familiar with the many complex spacecraft subsystems. Development of an expert systems program requires a validation of the software with actual flight hardware. By combining accurate hardware and software modelling techniques with a modular systems approach to expert systems development, the validation of these software programs can be successfully completed with minimum risk and effort. The TIMES Expert System (TES) is an application that monitors and evaluates real time data to perform fault detection and fault isolation tasks as they would otherwise be carried out by a knowledgeable designer. The development process and primary features of TES, a modular systems approach, and the lessons learned are discussed.

Comparison and assessment of semi-automatic image segmentation in computed tomography scans for image-guided kidney surgery.

PubMed

Glisson, Courtenay L; Altamar, Hernan O; Herrell, S Duke; Clark, Peter; Galloway, Robert L

2011-11-01

Image segmentation is integral to implementing intraoperative guidance for kidney tumor resection. Results seen in computed tomography (CT) data are affected by target organ physiology as well as by the segmentation algorithm used. This work studies variables involved in using level set methods found in the Insight Toolkit to segment kidneys from CT scans and applies the results to an image guidance setting. A composite algorithm drawing on the strengths of multiple level set approaches was built using the Insight Toolkit. This algorithm requires image contrast state and seed points to be identified as input, and functions independently thereafter, selecting and altering method and variable choice as needed. Semi-automatic results were compared to expert hand segmentation results directly and by the use of the resultant surfaces for registration of intraoperative data. Direct comparison using the Dice metric showed average agreement of 0.93 between semi-automatic and hand segmentation results. Use of the segmented surfaces in closest point registration of intraoperative laser range scan data yielded average closest point distances of approximately 1 mm. Application of both inverse registration transforms from the previous step to all hand segmented image space points revealed that the distance variability introduced by registering to the semi-automatically segmented surface versus the hand segmented surface was typically less than 3 mm both near the tumor target and at distal points, including subsurface points. Use of the algorithm shortened user interaction time and provided results which were comparable to the gold standard of hand segmentation. Further, the use of the algorithm's resultant surfaces in image registration provided comparable transformations to surfaces produced by hand segmentation. These data support the applicability and utility of such an algorithm as part of an image guidance workflow.

Neuroprosthetic Decoder Training as Imitation Learning

PubMed Central

Merel, Josh; Paninski, Liam; Cunningham, John P.

2016-01-01

Neuroprosthetic brain-computer interfaces function via an algorithm which decodes neural activity of the user into movements of an end effector, such as a cursor or robotic arm. In practice, the decoder is often learned by updating its parameters while the user performs a task. When the user’s intention is not directly observable, recent methods have demonstrated value in training the decoder against a surrogate for the user’s intended movement. Here we show that training a decoder in this way is a novel variant of an imitation learning problem, where an oracle or expert is employed for supervised training in lieu of direct observations, which are not available. Specifically, we describe how a generic imitation learning meta-algorithm, dataset aggregation (DAgger), can be adapted to train a generic brain-computer interface. By deriving existing learning algorithms for brain-computer interfaces in this framework, we provide a novel analysis of regret (an important metric of learning efficacy) for brain-computer interfaces. This analysis allows us to characterize the space of algorithmic variants and bounds on their regret rates. Existing approaches for decoder learning have been performed in the cursor control setting, but the available design principles for these decoders are such that it has been impossible to scale them to naturalistic settings. Leveraging our findings, we then offer an algorithm that combines imitation learning with optimal control, which should allow for training of arbitrary effectors for which optimal control can generate goal-oriented control. We demonstrate this novel and general BCI algorithm with simulated neuroprosthetic control of a 26 degree-of-freedom model of an arm, a sophisticated and realistic end effector. PMID:27191387

Programming Deep Brain Stimulation for Tremor and Dystonia: The Toronto Western Hospital Algorithms.

PubMed

Picillo, Marina; Lozano, Andres M; Kou, Nancy; Munhoz, Renato Puppi; Fasano, Alfonso

2016-01-01

Deep brain stimulation (DBS) is an effective treatment for essential tremor (ET) and dystonia. After surgery, a number of extensive programming sessions are performed, mainly relying on neurologist's personal experience as no programming guidelines have been provided so far, with the exception of recommendations provided by groups of experts. Finally, fewer information is available for the management of DBS in ET and dystonia compared with Parkinson's disease. Our aim is to review the literature on initial and follow-up DBS programming procedures for ET and dystonia and integrate the results with our current practice at Toronto Western Hospital (TWH) to develop standardized DBS programming protocols. We conducted a literature search of PubMed from inception to July 2014 with the keywords "balance", "bradykinesia", "deep brain stimulation", "dysarthria", "dystonia", "gait disturbances", "initial programming", "loss of benefit", "micrographia", "speech", "speech difficulties" and "tremor". Seventy-six papers were considered for this review. Based on the literature review and our experience at TWH, we refined three algorithms for management of ET, including: (1) initial programming, (2) management of balance and speech issues and (3) loss of stimulation benefit. We also depicted algorithms for the management of dystonia, including: (1) initial programming and (2) management of stimulation-induced hypokinesia (shuffling gait, micrographia and speech impairment). We propose five algorithms tailored to an individualized approach to managing ET and dystonia patients with DBS. We encourage the application of these algorithms to supplement current standards of care in established as well as new DBS centers to test the clinical usefulness of these algorithms in supplementing the current standards of care. Copyright © 2016 Elsevier Inc. All rights reserved.

A novel semi-transductive learning framework for efficient atypicality detection in chest radiographs

NASA Astrophysics Data System (ADS)

Alzubaidi, Mohammad; Balasubramanian, Vineeth; Patel, Ameet; Panchanathan, Sethuraman; Black, John A., Jr.

2012-03-01

Inductive learning refers to machine learning algorithms that learn a model from a set of training data instances. Any test instance is then classified by comparing it to the learned model. When the set of training instances lend themselves well to modeling, the use of a model substantially reduces the computation cost of classification. However, some training data sets are complex, and do not lend themselves well to modeling. Transductive learning refers to machine learning algorithms that classify test instances by comparing them to all of the training instances, without creating an explicit model. This can produce better classification performance, but at a much higher computational cost. Medical images vary greatly across human populations, constituting a data set that does not lend itself well to modeling. Our previous work showed that the wide variations seen across training sets of "normal" chest radiographs make it difficult to successfully classify test radiographs with an inductive (modeling) approach, and that a transductive approach leads to much better performance in detecting atypical regions. The problem with the transductive approach is its high computational cost. This paper develops and demonstrates a novel semi-transductive framework that can address the unique challenges of atypicality detection in chest radiographs. The proposed framework combines the superior performance of transductive methods with the reduced computational cost of inductive methods. Our results show that the proposed semitransductive approach provides both effective and efficient detection of atypical regions within a set of chest radiographs previously labeled by Mayo Clinic expert thoracic radiologists.

Development of a comprehensive software engineering environment

NASA Technical Reports Server (NTRS)

Hartrum, Thomas C.; Lamont, Gary B.

1987-01-01

The generation of a set of tools for software lifecycle is a recurring theme in the software engineering literature. The development of such tools and their integration into a software development environment is a difficult task because of the magnitude (number of variables) and the complexity (combinatorics) of the software lifecycle process. An initial development of a global approach was initiated in 1982 as the Software Development Workbench (SDW). Continuing efforts focus on tool development, tool integration, human interfacing, data dictionaries, and testing algorithms. Current efforts are emphasizing natural language interfaces, expert system software development associates and distributed environments with Ada as the target language. The current implementation of the SDW is on a VAX-11/780. Other software development tools are being networked through engineering workstations.

Automated circumferential construction of first-order aqueous humor outflow pathways using spectral-domain optical coherence tomography

NASA Astrophysics Data System (ADS)

Huang, Alex S.; Belghith, Akram; Dastiridou, Anna; Chopra, Vikas; Zangwill, Linda M.; Weinreb, Robert N.

2017-06-01

The purpose was to create a three-dimensional (3-D) model of circumferential aqueous humor outflow (AHO) in a living human eye with an automated detection algorithm for Schlemm's canal (SC) and first-order collector channels (CC) applied to spectral-domain optical coherence tomography (SD-OCT). Anterior segment SD-OCT scans from a subject were acquired circumferentially around the limbus. A Bayesian Ridge method was used to approximate the location of the SC on infrared confocal laser scanning ophthalmoscopic images with a cross multiplication tool developed to initiate SC/CC detection automated through a fuzzy hidden Markov Chain approach. Automatic segmentation of SC and initial CC's was manually confirmed by two masked graders. Outflow pathways detected by the segmentation algorithm were reconstructed into a 3-D representation of AHO. Overall, only <1% of images (5114 total B-scans) were ungradable. Automatic segmentation algorithm performed well with SC detection 98.3% of the time and <0.1% false positive detection compared to expert grader consensus. CC was detected 84.2% of the time with 1.4% false positive detection. 3-D representation of AHO pathways demonstrated variably thicker and thinner SC with some clear CC roots. Circumferential (360 deg), automated, and validated AHO detection of angle structures in the living human eye with reconstruction was possible.

Automatic Earthquake Detection by Active Learning

NASA Astrophysics Data System (ADS)

Bergen, K.; Beroza, G. C.

2017-12-01

In recent years, advances in machine learning have transformed fields such as image recognition, natural language processing and recommender systems. Many of these performance gains have relied on the availability of large, labeled data sets to train high-accuracy models; labeled data sets are those for which each sample includes a target class label, such as waveforms tagged as either earthquakes or noise. Earthquake seismologists are increasingly leveraging machine learning and data mining techniques to detect and analyze weak earthquake signals in large seismic data sets. One of the challenges in applying machine learning to seismic data sets is the limited labeled data problem; learning algorithms need to be given examples of earthquake waveforms, but the number of known events, taken from earthquake catalogs, may be insufficient to build an accurate detector. Furthermore, earthquake catalogs are known to be incomplete, resulting in training data that may be biased towards larger events and contain inaccurate labels. This challenge is compounded by the class imbalance problem; the events of interest, earthquakes, are infrequent relative to noise in continuous data sets, and many learning algorithms perform poorly on rare classes. In this work, we investigate the use of active learning for automatic earthquake detection. Active learning is a type of semi-supervised machine learning that uses a human-in-the-loop approach to strategically supplement a small initial training set. The learning algorithm incorporates domain expertise through interaction between a human expert and the algorithm, with the algorithm actively posing queries to the user to improve detection performance. We demonstrate the potential of active machine learning to improve earthquake detection performance with limited available training data.

Combining contour detection algorithms for the automatic extraction of the preparation line from a dental 3D measurement

NASA Astrophysics Data System (ADS)

Ahlers, Volker; Weigl, Paul; Schachtzabel, Hartmut

2005-04-01

Due to the increasing demand for high-quality ceramic crowns and bridges, the CAD/CAM-based production of dental restorations has been a subject of intensive research during the last fifteen years. A prerequisite for the efficient processing of the 3D measurement of prepared teeth with a minimal amount of user interaction is the automatic determination of the preparation line, which defines the sealing margin between the restoration and the prepared tooth. Current dental CAD/CAM systems mostly require the interactive definition of the preparation line by the user, at least by means of giving a number of start points. Previous approaches to the automatic extraction of the preparation line rely on single contour detection algorithms. In contrast, we use a combination of different contour detection algorithms to find several independent potential preparation lines from a height profile of the measured data. The different algorithms (gradient-based, contour-based, and region-based) show their strengths and weaknesses in different clinical situations. A classifier consisting of three stages (range check, decision tree, support vector machine), which is trained by human experts with real-world data, finally decides which is the correct preparation line. In a test with 101 clinical preparations, a success rate of 92.0% has been achieved. Thus the combination of different contour detection algorithms yields a reliable method for the automatic extraction of the preparation line, which enables the setup of a turn-key dental CAD/CAM process chain with a minimal amount of interactive screen work.

Modelling effective diagnosis of risk complications in gestational diabetes mellitus: an e-diabetic expert system for pregnant women

NASA Astrophysics Data System (ADS)

Sreedevi, E.; Vijaya Lakshmi, K.; Chaitanya Krishna, E.; Padmavathamma, M.

2012-04-01

Diabetes is a chronic illness that requires continuous medical care and patient self-management education to prevent acute complications and to reduce the risk of long-term complications. This paper deals with study and development of algorithm to develop an initial stage expert system to provide diagnosis to the pregnant women who are suffering from Gestational Diabetes Mellitus (GDM) by means of Oral Glucose Tolerance Test (OGTT).

Neurale Netwerken en Radarsystemen (Neural Networks and Radar Systems)

DTIC Science & Technology

1989-08-01

general issues in cognitive science", Parallel distributed processing, Vol 1: Foundations, Rumelhart et al. 1986 pp 110-146 THO rapport Pagina 151 36 D.E...34Neural networks (part 2)",Expert Focus, IEEE Expert, Spring 1988. 61 J.A. Anderson, " Cognitive and Psychological Computations with Neural Models", IEEE...Pagina 154 69 David H. Ackley, Geoffrey E. Hinton and Terrence J. Sejnowski, "A Learning Algorithm for Boltzmann machines", cognitive science 9, 147-169

An expert system to perform on-line controller restructuring for abrupt model changes

NASA Technical Reports Server (NTRS)

Litt, Jonathan S.

1990-01-01

Work in progress on an expert system used to reconfigure and tune airframe/engine control systems on-line in real time in response to battle damage or structural failures is presented. The closed loop system is monitored constantly for changes in structure and performance, the detection of which prompts the expert system to choose and apply a particular control restructuring algorithm based on the type and severity of the damage. Each algorithm is designed to handle specific types of failures and each is applicable only in certain situations. The expert system uses information about the system model to identify the failure and to select the technique best suited to compensate for it. A depth-first search is used to find a solution. Once a new controller is designed and implemented it must be tuned to recover the original closed-loop handling qualities and responsiveness from the degraded system. Ideally, the pilot should not be able to tell the difference between the original and redesigned systems. The key is that the system must have inherent redundancy so that degraded or missing capabilities can be restored by creative use of alternate functionalities. With enough redundancy in the control system, minor battle damage affecting individual control surfaces or actuators, compressor efficiency, etc., can be compensated for such that the closed-loop performance in not noticeably altered. The work is applied to a Black Hawk/T700 system.

An expert system for diagnosing environmentally induced spacecraft anomalies

NASA Technical Reports Server (NTRS)

Rolincik, Mark; Lauriente, Michael; Koons, Harry C.; Gorney, David

1992-01-01

A new rule-based, machine independent analytical tool was designed for diagnosing spacecraft anomalies using an expert system. Expert systems provide an effective method for saving knowledge, allow computers to sift through large amounts of data pinpointing significant parts, and most importantly, use heuristics in addition to algorithms, which allow approximate reasoning and inference and the ability to attack problems not rigidly defined. The knowledge base consists of over two-hundred (200) rules and provides links to historical and environmental databases. The environmental causes considered are bulk charging, single event upsets (SEU), surface charging, and total radiation dose. The system's driver translates forward chaining rules into a backward chaining sequence, prompting the user for information pertinent to the causes considered. The use of heuristics frees the user from searching through large amounts of irrelevant information and allows the user to input partial information (varying degrees of confidence in an answer) or 'unknown' to any question. The modularity of the expert system allows for easy updates and modifications. It not only provides scientists with needed risk analysis and confidence not found in algorithmic programs, but is also an effective learning tool, and the window implementation makes it very easy to use. The system currently runs on a Micro VAX II at Goddard Space Flight Center (GSFC). The inference engine used is NASA's C Language Integrated Production System (CLIPS).

Efficient discovery of risk patterns in medical data.

PubMed

Li, Jiuyong; Fu, Ada Wai-chee; Fahey, Paul

2009-01-01

This paper studies a problem of efficiently discovering risk patterns in medical data. Risk patterns are defined by a statistical metric, relative risk, which has been widely used in epidemiological research. To avoid fruitless search in the complete exploration of risk patterns, we define optimal risk pattern set to exclude superfluous patterns, i.e. complicated patterns with lower relative risk than their corresponding simpler form patterns. We prove that mining optimal risk pattern sets conforms an anti-monotone property that supports an efficient mining algorithm. We propose an efficient algorithm for mining optimal risk pattern sets based on this property. We also propose a hierarchical structure to present discovered patterns for the easy perusal by domain experts. The proposed approach is compared with two well-known rule discovery methods, decision tree and association rule mining approaches on benchmark data sets and applied to a real world application. The proposed method discovers more and better quality risk patterns than a decision tree approach. The decision tree method is not designed for such applications and is inadequate for pattern exploring. The proposed method does not discover a large number of uninteresting superfluous patterns as an association mining approach does. The proposed method is more efficient than an association rule mining method. A real world case study shows that the method reveals some interesting risk patterns to medical practitioners. The proposed method is an efficient approach to explore risk patterns. It quickly identifies cohorts of patients that are vulnerable to a risk outcome from a large data set. The proposed method is useful for exploratory study on large medical data to generate and refine hypotheses. The method is also useful for designing medical surveillance systems.

Comparing the performance of expert user heuristics and an integer linear program in aircraft carrier deck operations.

PubMed

Ryan, Jason C; Banerjee, Ashis Gopal; Cummings, Mary L; Roy, Nicholas

2014-06-01

Planning operations across a number of domains can be considered as resource allocation problems with timing constraints. An unexplored instance of such a problem domain is the aircraft carrier flight deck, where, in current operations, replanning is done without the aid of any computerized decision support. Rather, veteran operators employ a set of experience-based heuristics to quickly generate new operating schedules. These expert user heuristics are neither codified nor evaluated by the United States Navy; they have grown solely from the convergent experiences of supervisory staff. As unmanned aerial vehicles (UAVs) are introduced in the aircraft carrier domain, these heuristics may require alterations due to differing capabilities. The inclusion of UAVs also allows for new opportunities for on-line planning and control, providing an alternative to the current heuristic-based replanning methodology. To investigate these issues formally, we have developed a decision support system for flight deck operations that utilizes a conventional integer linear program-based planning algorithm. In this system, a human operator sets both the goals and constraints for the algorithm, which then returns a proposed schedule for operator approval. As a part of validating this system, the performance of this collaborative human-automation planner was compared with that of the expert user heuristics over a set of test scenarios. The resulting analysis shows that human heuristics often outperform the plans produced by an optimization algorithm, but are also often more conservative.

Web-based health services and clinical decision support.

PubMed

Jegelevicius, Darius; Marozas, Vaidotas; Lukosevicius, Arunas; Patasius, Martynas

2004-01-01

The purpose of this study was the development of a Web-based e-health service for comprehensive assistance and clinical decision support. The service structure consists of a Web server, a PHP-based Web interface linked to a clinical SQL database, Java applets for interactive manipulation and visualization of signals and a Matlab server linked with signal and data processing algorithms implemented by Matlab programs. The service ensures diagnostic signal- and image analysis-sbased clinical decision support. By using the discussed methodology, a pilot service for pathology specialists for automatic calculation of the proliferation index has been developed. Physicians use a simple Web interface for uploading the pictures under investigation to the server; subsequently a Java applet interface is used for outlining the region of interest and, after processing on the server, the requested proliferation index value is calculated. There is also an "expert corner", where experts can submit their index estimates and comments on particular images, which is especially important for system developers. These expert evaluations are used for optimization and verification of automatic analysis algorithms. Decision support trials have been conducted for ECG and ophthalmology ultrasonic investigations of intraocular tumor differentiation. Data mining algorithms have been applied and decision support trees constructed. These services are under implementation by a Web-based system too. The study has shown that the Web-based structure ensures more effective, flexible and accessible services compared with standalone programs and is very convenient for biomedical engineers and physicians, especially in the development phase.

Automated planning of computer assisted mosaic arthroplasty.

PubMed

Inoue, Jiro; Kunz, Manuela; Hurtig, Mark B; Waldman, Stephen D; Stewart, A James

2011-01-01

We describe and evaluate a computer algorithm that automatically develops a surgical plan for computer assisted mosaic arthroplasty, a technically demanding procedure in which a set of osteochondral plugs are transplanted from a non-load-bearing area of the joint to the site of a cartilage defect. We found that the algorithm produced plans that were at least as good as a human expert, had less variability, and took less time.

What kind of computation is intelligence. A framework for integrating different kinds of expertise

NASA Technical Reports Server (NTRS)

Chandrasekaran, B.

1989-01-01

The view that the deliberative aspect of intelligent behavior is a distinct type of algorithm; in particular, a goal-seeking exploratory process using qualitative representations of knowledge and inference is elaborated. There are other kinds of algorithms that also embody expertise in domains. The different types of expertise and how they can and should be integrated to give full account of expert behavior are discussed.

CLEAR: Communications Link Expert Assistance Resource

NASA Technical Reports Server (NTRS)

Hull, Larry G.; Hughes, Peter M.

1987-01-01

Communications Link Expert Assistance Resource (CLEAR) is a real time, fault diagnosis expert system for the Cosmic Background Explorer (COBE) Mission Operations Room (MOR). The CLEAR expert system is an operational prototype which assists the MOR operator/analyst by isolating and diagnosing faults in the spacecraft communication link with the Tracking and Data Relay Satellite (TDRS) during periods of realtime data acquisition. The mission domain, user requirements, hardware configuration, expert system concept, tool selection, development approach, and system design were discussed. Development approach and system implementation are emphasized. Also discussed are system architecture, tool selection, operation, and future plans.

Automatic elastic image registration by interpolation of 3D rotations and translations from discrete rigid-body transformations.

PubMed

Walimbe, Vivek; Shekhar, Raj

2006-12-01

We present an algorithm for automatic elastic registration of three-dimensional (3D) medical images. Our algorithm initially recovers the global spatial mismatch between the reference and floating images, followed by hierarchical octree-based subdivision of the reference image and independent registration of the floating image with the individual subvolumes of the reference image at each hierarchical level. Global as well as local registrations use the six-parameter full rigid-body transformation model and are based on maximization of normalized mutual information (NMI). To ensure robustness of the subvolume registration with low voxel counts, we calculate NMI using a combination of current and prior mutual histograms. To generate a smooth deformation field, we perform direct interpolation of six-parameter rigid-body subvolume transformations obtained at the last subdivision level. Our interpolation scheme involves scalar interpolation of the 3D translations and quaternion interpolation of the 3D rotational pose. We analyzed the performance of our algorithm through experiments involving registration of synthetically deformed computed tomography (CT) images. Our algorithm is general and can be applied to image pairs of any two modalities of most organs. We have demonstrated successful registration of clinical whole-body CT and positron emission tomography (PET) images using this algorithm. The registration accuracy for this application was evaluated, based on validation using expert-identified anatomical landmarks in 15 CT-PET image pairs. The algorithm's performance was comparable to the average accuracy observed for three expert-determined registrations in the same 15 image pairs.

Automatic detection of end-diastolic and end-systolic frames in 2D echocardiography.

PubMed

Zolgharni, Massoud; Negoita, Madalina; Dhutia, Niti M; Mielewczik, Michael; Manoharan, Karikaran; Sohaib, S M Afzal; Finegold, Judith A; Sacchi, Stefania; Cole, Graham D; Francis, Darrel P

2017-07-01

Correctly selecting the end-diastolic and end-systolic frames on a 2D echocardiogram is important and challenging, for both human experts and automated algorithms. Manual selection is time-consuming and subject to uncertainty, and may affect the results obtained, especially for advanced measurements such as myocardial strain. We developed and evaluated algorithms which can automatically extract global and regional cardiac velocity, and identify end-diastolic and end-systolic frames. We acquired apical four-chamber 2D echocardiographic video recordings, each at least 10 heartbeats long, acquired twice at frame rates of 52 and 79 frames/s from 19 patients, yielding 38 recordings. Five experienced echocardiographers independently marked end-systolic and end-diastolic frames for the first 10 heartbeats of each recording. The automated algorithm also did this. Using the average of time points identified by five human operators as the reference gold standard, the individual operators had a root mean square difference from that gold standard of 46.5 ms. The algorithm had a root mean square difference from the human gold standard of 40.5 ms (P<.0001). Put another way, the algorithm-identified time point was an outlier in 122/564 heartbeats (21.6%), whereas the average human operator was an outlier in 254/564 heartbeats (45%). An automated algorithm can identify the end-systolic and end-diastolic frames with performance indistinguishable from that of human experts. This saves staff time, which could therefore be invested in assessing more beats, and reduces uncertainty about the reliability of the choice of frame. © 2017, Wiley Periodicals, Inc.

Problem-solving strategies in psychiatry: differences between experts and novices in diagnostic accuracy and reasoning.

PubMed

Gabriel, Adel; Violato, Claudio

2013-01-01

The purpose of this study was to examine and compare diagnostic success and its relationship with the diagnostic reasoning process between novices and experts in psychiatry. Nine volunteers, comprising five expert psychiatrists and four clinical clerks, completed a think-aloud protocol while attempting to make a DSM-IV (Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition) diagnosis of a selected case with both Axis I and Axis III diagnoses. Expert psychiatrists made significantly more successful diagnoses for both the primary psychiatric and medical diagnoses than clinical clerks. Expert psychiatrists also gave fewer differential options. Analyzing the think-aloud protocols, expert psychiatrists were much more organized, made fewer mistakes, and utilized significantly less time to access their knowledge than clinical clerks. Both novices and experts seemed to use the hypothetic-deductive and scheme-inductive approaches to diagnosis. However, experts utilized hypothetic-deductive approaches significantly more often than novices. The hypothetic-deductive diagnostic strategy was utilized more than the scheme-inductive approach by both expert psychiatrists and clinical clerks. However, a specific relationship between diagnostic reasoning and diagnostic success could not be identified in this small pilot study. The author recommends a larger study that would include a detailed analysis of the think-aloud protocols.

Calibrating Building Energy Models Using Supercomputer Trained Machine Learning Agents

DOE Office of Scientific and Technical Information (OSTI.GOV)

Sanyal, Jibonananda; New, Joshua Ryan; Edwards, Richard

2014-01-01

Building Energy Modeling (BEM) is an approach to model the energy usage in buildings for design and retrofit purposes. EnergyPlus is the flagship Department of Energy software that performs BEM for different types of buildings. The input to EnergyPlus can often extend in the order of a few thousand parameters which have to be calibrated manually by an expert for realistic energy modeling. This makes it challenging and expensive thereby making building energy modeling unfeasible for smaller projects. In this paper, we describe the Autotune research which employs machine learning algorithms to generate agents for the different kinds of standardmore » reference buildings in the U.S. building stock. The parametric space and the variety of building locations and types make this a challenging computational problem necessitating the use of supercomputers. Millions of EnergyPlus simulations are run on supercomputers which are subsequently used to train machine learning algorithms to generate agents. These agents, once created, can then run in a fraction of the time thereby allowing cost-effective calibration of building models.« less

A computationally efficient method for incorporating spike waveform information into decoding algorithms.

PubMed

Ventura, Valérie; Todorova, Sonia

2015-05-01

Spike-based brain-computer interfaces (BCIs) have the potential to restore motor ability to people with paralysis and amputation, and have shown impressive performance in the lab. To transition BCI devices from the lab to the clinic, decoding must proceed automatically and in real time, which prohibits the use of algorithms that are computationally intensive or require manual tweaking. A common choice is to avoid spike sorting and treat the signal on each electrode as if it came from a single neuron, which is fast, easy, and therefore desirable for clinical use. But this approach ignores the kinematic information provided by individual neurons recorded on the same electrode. The contribution of this letter is a linear decoding model that extracts kinematic information from individual neurons without spike-sorting the electrode signals. The method relies on modeling sample averages of waveform features as functions of kinematics, which is automatic and requires minimal data storage and computation. In offline reconstruction of arm trajectories of a nonhuman primate performing reaching tasks, the proposed method performs as well as decoders based on expertly manually and automatically sorted spikes.

Knowledge-based modularization and global optimization of artificial neural network models in hydrological forecasting.

PubMed

Corzo, Gerald; Solomatine, Dimitri

2007-05-01

Natural phenomena are multistationary and are composed of a number of interacting processes, so one single model handling all processes often suffers from inaccuracies. A solution is to partition data in relation to such processes using the available domain knowledge or expert judgment, to train separate models for each of the processes, and to merge them in a modular model (committee). In this paper a problem of water flow forecast in watershed hydrology is considered where the flow process can be presented as consisting of two subprocesses -- base flow and excess flow, so that these two processes can be separated. Several approaches to data separation techniques are studied. Two case studies with different forecast horizons are considered. Parameters of the algorithms responsible for data partitioning are optimized using genetic algorithms and global pattern search. It was found that modularization of ANN models using domain knowledge makes models more accurate, if compared with a global model trained on the whole data set, especially when forecast horizon (and hence the complexity of the modelled processes) is increased.

Automated EEG artifact elimination by applying machine learning algorithms to ICA-based features.

PubMed

Radüntz, Thea; Scouten, Jon; Hochmuth, Olaf; Meffert, Beate

2017-08-01

Biological and non-biological artifacts cause severe problems when dealing with electroencephalogram (EEG) recordings. Independent component analysis (ICA) is a widely used method for eliminating various artifacts from recordings. However, evaluating and classifying the calculated independent components (IC) as artifact or EEG is not fully automated at present. In this study, we propose a new approach for automated artifact elimination, which applies machine learning algorithms to ICA-based features. We compared the performance of our classifiers with the visual classification results given by experts. The best result with an accuracy rate of 95% was achieved using features obtained by range filtering of the topoplots and IC power spectra combined with an artificial neural network. Compared with the existing automated solutions, our proposed method is not limited to specific types of artifacts, electrode configurations, or number of EEG channels. The main advantages of the proposed method is that it provides an automatic, reliable, real-time capable, and practical tool, which avoids the need for the time-consuming manual selection of ICs during artifact removal.

Sensor management in RADAR/IRST track fusion

NASA Astrophysics Data System (ADS)

Hu, Shi-qiang; Jing, Zhong-liang

2004-07-01

In this paper, a novel radar management strategy technique suitable for RADAR/IRST track fusion, which is based on Fisher Information Matrix (FIM) and fuzzy stochastic decision approach, is put forward. Firstly, optimal radar measurements' scheduling is obtained by the method of maximizing determinant of the Fisher information matrix of radar and IRST measurements, which is managed by the expert system. Then, suggested a "pseudo sensor" to predict the possible target position using the polynomial method based on the radar and IRST measurements, using "pseudo sensor" model to estimate the target position even if the radar is turned off. At last, based on the tracking performance and the state of target maneuver, fuzzy stochastic decision is used to adjust the optimal radar scheduling and retrieve the module parameter of "pseudo sensor". The experiment result indicates that the algorithm can not only limit Radar activity effectively but also keep the tracking accuracy of active/passive system well. And this algorithm eliminates the drawback of traditional Radar management methods that the Radar activity is fixed and not easy to control and protect.

SUVI Thematic Maps: A new tool for space weather forecasting

NASA Astrophysics Data System (ADS)

Hughes, J. M.; Seaton, D. B.; Darnel, J.

2017-12-01

The new Solar Ultraviolet Imager (SUVI) instruments aboard NOAA's GOES-R series satellites collect continuous, high-quality imagery of the Sun in six wavelengths. SUVI imagers produce at least one image every 10 seconds, or 8,640 images per day, considerably more data than observers can digest in real time. Over the projected 20-year lifetime of the four GOES-R series spacecraft, SUVI will provide critical imagery for space weather forecasters and produce an extensive but unwieldy archive. In order to condense the database into a dynamic and searchable form we have developed solar thematic maps, maps of the Sun with key features, such as coronal holes, flares, bright regions, quiet corona, and filaments, identified. Thematic maps will be used in NOAA's Space Weather Prediction Center to improve forecaster response time to solar events and generate several derivative products. Likewise, scientists use thematic maps to find observations of interest more easily. Using an expert-trained, naive Bayesian classifier to label each pixel, we create thematic maps in real-time. We created software to collect expert classifications of solar features based on SUVI images. Using this software, we compiled a database of expert classifications, from which we could characterize the distribution of pixels associated with each theme. Given new images, the classifier assigns each pixel the most appropriate label according to the trained distribution. Here we describe the software to collect expert training and the successes and limitations of the classifier. The algorithm excellently identifies coronal holes but fails to consistently detect filaments and prominences. We compare the Bayesian classifier to an artificial neural network, one of our attempts to overcome the aforementioned limitations. These results are very promising and encourage future research into an ensemble classification approach.

A Qualitative Approach to the Evaluation of Expert Systems Shells.

ERIC Educational Resources Information Center

Slawson, Dean A.; And Others

This study explores an approach to the evaluation of expert system shells using case studies. The methodology and some of the results of an evaluation of the prototype development of an expert system using the shell "M1" are detailed, including a description of the participants and the project, the data collection process and materials,…

The Coach2Move Approach: Development and Acceptability of an Individually Tailored Physical Therapy Strategy to Increase Activity Levels in Older Adults With Mobility Problems.

PubMed

de Vries, Nienke M; van Ravensberg, C Dorine; Hobbelen, Johannes S M; van der Wees, Philip J; Olde Rikkert, Marcel G M; Staal, J Bart; Nijhuis-van der Sanden, Maria W G

2015-01-01

Despite the positive effects of physical activity on numerous aspects of health, many older adults remain sedentary even after participating in physical activity interventions. Standardized exercise programs do not necessarily bring about the behavioral change that is necessary. Therefore, a patient-centered approach is needed. The purpose of this study was to develop and assess the acceptability and potential effectiveness of the Coach2Move strategy; a physical therapy (PT) approach aimed at improving the long-term level of physical activity in mobility-limited older adults. The Coach2Move strategy was developed on the basis of 2 systematic literature studies and expert consultations. Multiple focus group meetings and a Delphi procedure were organized to gain consensus on the Coach2Move strategy. Acceptability and potential effectiveness were studied in a pilot study with a pre-/postdesign in which 2 physical therapists and 12 patients participated. To assess acceptability, patients were interviewed, discussion were held with the involved physical therapists was held, and health records were studied. Potential effectiveness was tested measuring the level of physical activity, frailty, quality of life, and mobility before and after treatment. On the basis of the literature study and expert consultations, an algorithm based on the Hypothesis Oriented Algorithm for Clinicians Part II was developed: the Coach2Move approach. Key elements of the Coach2Move approach include an extensive intake using motivational interviewing, clinical reasoning, coaching to increase physical activity and self-management, focusing on meaningful activities, and working according to 3 patient-tailored intervention profiles with a predefined number of sessions. The pilot study showed high appraisal of the strategy by both physical therapists and patients. Moreover, a potential effect on the level of physical activity, frailty, quality of life, and mobility was observed. Because the pilot study was not randomized or controlled and included a small sample, no conclusions can be drawn about the effectiveness of the Coach2Move strategy. However, all suggestions made in this study were implemented in an ongoing, randomized controlled trial in which the Coach2Move strategy will be compared to usual care PT. In conclusion, the Coach2Move strategy can be considered acceptable in PT practice and showed potential benefits. The results on the (cost-)effectiveness of this strategy based on a large, randomized, controlled trial are expected in 2014.

Comparing experts and novices in Martian surface feature change detection and identification

NASA Astrophysics Data System (ADS)

Wardlaw, Jessica; Sprinks, James; Houghton, Robert; Muller, Jan-Peter; Sidiropoulos, Panagiotis; Bamford, Steven; Marsh, Stuart

2018-02-01

Change detection in satellite images is a key concern of the Earth Observation field for environmental and climate change monitoring. Satellite images also provide important clues to both the past and present surface conditions of other planets, which cannot be validated on the ground. With the volume of satellite imagery continuing to grow, the inadequacy of computerised solutions to manage and process imagery to the required professional standard is of critical concern. Whilst studies find the crowd sourcing approach suitable for the counting of impact craters in single images, images of higher resolution contain a much wider range of features, and the performance of novices in identifying more complex features and detecting change, remains unknown. This paper presents a first step towards understanding whether novices can identify and annotate changes in different geomorphological features. A website was developed to enable visitors to flick between two images of the same location on Mars taken at different times and classify 1) if a surface feature changed and if so, 2) what feature had changed from a pre-defined list of six. Planetary scientists provided ;expert; data against which classifications made by novices could be compared when the project subsequently went public. Whilst no significant difference was found in images identified with surface changes by expert and novices, results exhibited differences in consensus within and between experts and novices when asked to classify the type of change. Experts demonstrated higher levels of agreement in classification of changes as dust devil tracks, slope streaks and impact craters than other features, whilst the consensus of novices was consistent across feature types; furthermore, the level of consensus amongst regardless of feature type. These trends are secondary to the low levels of consensus found, regardless of feature type or classifier expertise. These findings demand the attention of researchers who want to use crowd-sourcing for similar scientific purposes, particularly for the supervised training of computer algorithms, and inform the scope and design of future projects.

Fusing Continuous-Valued Medical Labels Using a Bayesian Model.

PubMed

Zhu, Tingting; Dunkley, Nic; Behar, Joachim; Clifton, David A; Clifford, Gari D

2015-12-01

With the rapid increase in volume of time series medical data available through wearable devices, there is a need to employ automated algorithms to label data. Examples of labels include interventions, changes in activity (e.g. sleep) and changes in physiology (e.g. arrhythmias). However, automated algorithms tend to be unreliable resulting in lower quality care. Expert annotations are scarce, expensive, and prone to significant inter- and intra-observer variance. To address these problems, a Bayesian Continuous-valued Label Aggregator (BCLA) is proposed to provide a reliable estimation of label aggregation while accurately infer the precision and bias of each algorithm. The BCLA was applied to QT interval (pro-arrhythmic indicator) estimation from the electrocardiogram using labels from the 2006 PhysioNet/Computing in Cardiology Challenge database. It was compared to the mean, median, and a previously proposed Expectation Maximization (EM) label aggregation approaches. While accurately predicting each labelling algorithm's bias and precision, the root-mean-square error of the BCLA was 11.78 ± 0.63 ms, significantly outperforming the best Challenge entry (15.37 ± 2.13 ms) as well as the EM, mean, and median voting strategies (14.76 ± 0.52, 17.61 ± 0.55, and 14.43 ± 0.57 ms respectively with p < 0.0001). The BCLA could therefore provide accurate estimation for medical continuous-valued label tasks in an unsupervised manner even when the ground truth is not available.

Towards human-computer synergetic analysis of large-scale biological data.

PubMed

Singh, Rahul; Yang, Hui; Dalziel, Ben; Asarnow, Daniel; Murad, William; Foote, David; Gormley, Matthew; Stillman, Jonathan; Fisher, Susan

2013-01-01

Advances in technology have led to the generation of massive amounts of complex and multifarious biological data in areas ranging from genomics to structural biology. The volume and complexity of such data leads to significant challenges in terms of its analysis, especially when one seeks to generate hypotheses or explore the underlying biological processes. At the state-of-the-art, the application of automated algorithms followed by perusal and analysis of the results by an expert continues to be the predominant paradigm for analyzing biological data. This paradigm works well in many problem domains. However, it also is limiting, since domain experts are forced to apply their instincts and expertise such as contextual reasoning, hypothesis formulation, and exploratory analysis after the algorithm has produced its results. In many areas where the organization and interaction of the biological processes is poorly understood and exploratory analysis is crucial, what is needed is to integrate domain expertise during the data analysis process and use it to drive the analysis itself. In context of the aforementioned background, the results presented in this paper describe advancements along two methodological directions. First, given the context of biological data, we utilize and extend a design approach called experiential computing from multimedia information system design. This paradigm combines information visualization and human-computer interaction with algorithms for exploratory analysis of large-scale and complex data. In the proposed approach, emphasis is laid on: (1) allowing users to directly visualize, interact, experience, and explore the data through interoperable visualization-based and algorithmic components, (2) supporting unified query and presentation spaces to facilitate experimentation and exploration, (3) providing external contextual information by assimilating relevant supplementary data, and (4) encouraging user-directed information visualization, data exploration, and hypotheses formulation. Second, to illustrate the proposed design paradigm and measure its efficacy, we describe two prototype web applications. The first, called XMAS (Experiential Microarray Analysis System) is designed for analysis of time-series transcriptional data. The second system, called PSPACE (Protein Space Explorer) is designed for holistic analysis of structural and structure-function relationships using interactive low-dimensional maps of the protein structure space. Both these systems promote and facilitate human-computer synergy, where cognitive elements such as domain knowledge, contextual reasoning, and purpose-driven exploration, are integrated with a host of powerful algorithmic operations that support large-scale data analysis, multifaceted data visualization, and multi-source information integration. The proposed design philosophy, combines visualization, algorithmic components and cognitive expertise into a seamless processing-analysis-exploration framework that facilitates sense-making, exploration, and discovery. Using XMAS, we present case studies that analyze transcriptional data from two highly complex domains: gene expression in the placenta during human pregnancy and reaction of marine organisms to heat stress. With PSPACE, we demonstrate how complex structure-function relationships can be explored. These results demonstrate the novelty, advantages, and distinctions of the proposed paradigm. Furthermore, the results also highlight how domain insights can be combined with algorithms to discover meaningful knowledge and formulate evidence-based hypotheses during the data analysis process. Finally, user studies against comparable systems indicate that both XMAS and PSPACE deliver results with better interpretability while placing lower cognitive loads on the users. XMAS is available at: http://tintin.sfsu.edu:8080/xmas. PSPACE is available at: http://pspace.info/.

Towards human-computer synergetic analysis of large-scale biological data

PubMed Central

2013-01-01

Background Advances in technology have led to the generation of massive amounts of complex and multifarious biological data in areas ranging from genomics to structural biology. The volume and complexity of such data leads to significant challenges in terms of its analysis, especially when one seeks to generate hypotheses or explore the underlying biological processes. At the state-of-the-art, the application of automated algorithms followed by perusal and analysis of the results by an expert continues to be the predominant paradigm for analyzing biological data. This paradigm works well in many problem domains. However, it also is limiting, since domain experts are forced to apply their instincts and expertise such as contextual reasoning, hypothesis formulation, and exploratory analysis after the algorithm has produced its results. In many areas where the organization and interaction of the biological processes is poorly understood and exploratory analysis is crucial, what is needed is to integrate domain expertise during the data analysis process and use it to drive the analysis itself. Results In context of the aforementioned background, the results presented in this paper describe advancements along two methodological directions. First, given the context of biological data, we utilize and extend a design approach called experiential computing from multimedia information system design. This paradigm combines information visualization and human-computer interaction with algorithms for exploratory analysis of large-scale and complex data. In the proposed approach, emphasis is laid on: (1) allowing users to directly visualize, interact, experience, and explore the data through interoperable visualization-based and algorithmic components, (2) supporting unified query and presentation spaces to facilitate experimentation and exploration, (3) providing external contextual information by assimilating relevant supplementary data, and (4) encouraging user-directed information visualization, data exploration, and hypotheses formulation. Second, to illustrate the proposed design paradigm and measure its efficacy, we describe two prototype web applications. The first, called XMAS (Experiential Microarray Analysis System) is designed for analysis of time-series transcriptional data. The second system, called PSPACE (Protein Space Explorer) is designed for holistic analysis of structural and structure-function relationships using interactive low-dimensional maps of the protein structure space. Both these systems promote and facilitate human-computer synergy, where cognitive elements such as domain knowledge, contextual reasoning, and purpose-driven exploration, are integrated with a host of powerful algorithmic operations that support large-scale data analysis, multifaceted data visualization, and multi-source information integration. Conclusions The proposed design philosophy, combines visualization, algorithmic components and cognitive expertise into a seamless processing-analysis-exploration framework that facilitates sense-making, exploration, and discovery. Using XMAS, we present case studies that analyze transcriptional data from two highly complex domains: gene expression in the placenta during human pregnancy and reaction of marine organisms to heat stress. With PSPACE, we demonstrate how complex structure-function relationships can be explored. These results demonstrate the novelty, advantages, and distinctions of the proposed paradigm. Furthermore, the results also highlight how domain insights can be combined with algorithms to discover meaningful knowledge and formulate evidence-based hypotheses during the data analysis process. Finally, user studies against comparable systems indicate that both XMAS and PSPACE deliver results with better interpretability while placing lower cognitive loads on the users. XMAS is available at: http://tintin.sfsu.edu:8080/xmas. PSPACE is available at: http://pspace.info/. PMID:24267485

Fully automatic, multiorgan segmentation in normal whole body magnetic resonance imaging (MRI), using classification forests (CFs), convolutional neural networks (CNNs), and a multi-atlas (MA) approach.

PubMed

Lavdas, Ioannis; Glocker, Ben; Kamnitsas, Konstantinos; Rueckert, Daniel; Mair, Henrietta; Sandhu, Amandeep; Taylor, Stuart A; Aboagye, Eric O; Rockall, Andrea G

2017-10-01

As part of a program to implement automatic lesion detection methods for whole body magnetic resonance imaging (MRI) in oncology, we have developed, evaluated, and compared three algorithms for fully automatic, multiorgan segmentation in healthy volunteers. The first algorithm is based on classification forests (CFs), the second is based on 3D convolutional neural networks (CNNs) and the third algorithm is based on a multi-atlas (MA) approach. We examined data from 51 healthy volunteers, scanned prospectively with a standardized, multiparametric whole body MRI protocol at 1.5 T. The study was approved by the local ethics committee and written consent was obtained from the participants. MRI data were used as input data to the algorithms, while training was based on manual annotation of the anatomies of interest by clinical MRI experts. Fivefold cross-validation experiments were run on 34 artifact-free subjects. We report three overlap and three surface distance metrics to evaluate the agreement between the automatic and manual segmentations, namely the dice similarity coefficient (DSC), recall (RE), precision (PR), average surface distance (ASD), root-mean-square surface distance (RMSSD), and Hausdorff distance (HD). Analysis of variances was used to compare pooled label metrics between the three algorithms and the DSC on a 'per-organ' basis. A Mann-Whitney U test was used to compare the pooled metrics between CFs and CNNs and the DSC on a 'per-organ' basis, when using different imaging combinations as input for training. All three algorithms resulted in robust segmenters that were effectively trained using a relatively small number of datasets, an important consideration in the clinical setting. Mean overlap metrics for all the segmented structures were: CFs: DSC = 0.70 ± 0.18, RE = 0.73 ± 0.18, PR = 0.71 ± 0.14, CNNs: DSC = 0.81 ± 0.13, RE = 0.83 ± 0.14, PR = 0.82 ± 0.10, MA: DSC = 0.71 ± 0.22, RE = 0.70 ± 0.34, PR = 0.77 ± 0.15. Mean surface distance metrics for all the segmented structures were: CFs: ASD = 13.5 ± 11.3 mm, RMSSD = 34.6 ± 37.6 mm and HD = 185.7 ± 194.0 mm, CNNs; ASD = 5.48 ± 4.84 mm, RMSSD = 17.0 ± 13.3 mm and HD = 199.0 ± 101.2 mm, MA: ASD = 4.22 ± 2.42 mm, RMSSD = 6.13 ± 2.55 mm, and HD = 38.9 ± 28.9 mm. The pooled performance of CFs improved when all imaging combinations (T2w + T1w + DWI) were used as input, while the performance of CNNs deteriorated, but in neither case, significantly. CNNs with T2w images as input, performed significantly better than CFs with all imaging combinations as input for all anatomical labels, except for the bladder. Three state-of-the-art algorithms were developed and used to automatically segment major organs and bones in whole body MRI; good agreement to manual segmentations performed by clinical MRI experts was observed. CNNs perform favorably, when using T2w volumes as input. Using multimodal MRI data as input to CNNs did not improve the segmentation performance. © 2017 American Association of Physicists in Medicine.

How to control if even experts are not sure: Robust fuzzy control

NASA Technical Reports Server (NTRS)

Nguyen, Hung T.; Kreinovich, Vladik YA.; Lea, Robert; Tolbert, Dana

1992-01-01

In real life, the degrees of certainty that correspond to one of the same expert can differ drastically, and fuzzy control algorithms translate these different degrees of uncertainty into different control strategies. In such situations, it is reasonable to choose a fuzzy control methodology that is the least vulnerable to this kind of uncertainty. It is shown that this 'robustness' demand leads to min and max for &- and V-operations, to 1-x for negation, and to centroid as a defuzzification procedure.

Biased normalized cuts for target detection in hyperspectral imagery

NASA Astrophysics Data System (ADS)

Zhang, Xuewen; Dorado-Munoz, Leidy P.; Messinger, David W.; Cahill, Nathan D.

2016-05-01

The Biased Normalized Cuts (BNC) algorithm is a useful technique for detecting targets or objects in RGB imagery. In this paper, we propose modifying BNC for the purpose of target detection in hyperspectral imagery. As opposed to other target detection algorithms that typically encode target information prior to dimensionality reduction, our proposed algorithm encodes target information after dimensionality reduction, enabling a user to detect different targets in interactive mode. To assess the proposed BNC algorithm, we utilize hyperspectral imagery (HSI) from the SHARE 2012 data campaign, and we explore the relationship between the number and the position of expert-provided target labels and the precision/recall of the remaining targets in the scene.

Development of Increasingly Autonomous Traffic Data Manager Using Pilot Relevancy and Ranking Data

NASA Technical Reports Server (NTRS)

Le Vie, Lisa R.; Houston, Vincent E.

2017-01-01

NASA's Safe Autonomous Systems Operations (SASO) project goal is to define and safely enable all future airspace operations by justifiable and optimal autonomy for advanced air, ground, and connected capabilities. This work showcases how Increasingly Autonomous Systems (IAS) could create operational transformations beneficial to the enhancement of civil aviation safety and efficiency. One such IAS under development is the Traffic Data Manager (TDM). This concept is a prototype 'intelligent party-line' system that would declutter and parse out non-relevant air traffic, displaying only relevant air traffic to the aircrew in a digital data communications (Data Comm) environment. As an initial step, over 22,000 data points were gathered from 31 Airline Transport Pilots to train the machine learning algorithms designed to mimic human experts and expertise. The test collection used an analog of the Navigation Display. Pilots were asked to rate the relevancy of the displayed traffic using an interactive tablet application. Pilots were also asked to rank the order of importance of the information given, to better weight the variables within the algorithm. They were also asked if the information given was enough data, and more importantly the "right" data to best inform the algorithm. The paper will describe the findings and their impact to the further development of the algorithm for TDM and, in general, address the issue of how can we train supervised machine learning algorithms, critical to increasingly autonomous systems, with the knowledge and expertise of expert human pilots.

Automated image alignment and segmentation to follow progression of geographic atrophy in age-related macular degeneration.

PubMed

Ramsey, David J; Sunness, Janet S; Malviya, Poorva; Applegate, Carol; Hager, Gregory D; Handa, James T

2014-07-01

To develop a computer-based image segmentation method for standardizing the quantification of geographic atrophy (GA). The authors present an automated image segmentation method based on the fuzzy c-means clustering algorithm for the detection of GA lesions. The method is evaluated by comparing computerized segmentation against outlines of GA drawn by an expert grader for a longitudinal series of fundus autofluorescence images with paired 30° color fundus photographs for 10 patients. The automated segmentation method showed excellent agreement with an expert grader for fundus autofluorescence images, achieving a performance level of 94 ± 5% sensitivity and 98 ± 2% specificity on a per-pixel basis for the detection of GA area, but performed less well on color fundus photographs with a sensitivity of 47 ± 26% and specificity of 98 ± 2%. The segmentation algorithm identified 75 ± 16% of the GA border correctly in fundus autofluorescence images compared with just 42 ± 25% for color fundus photographs. The results of this study demonstrate a promising computerized segmentation method that may enhance the reproducibility of GA measurement and provide an objective strategy to assist an expert in the grading of images.

Automating expert role to determine design concept in Kansei Engineering

NASA Astrophysics Data System (ADS)

Lokman, Anitawati Mohd; Haron, Mohammad Bakri Che; Abidin, Siti Zaleha Zainal; Khalid, Noor Elaiza Abd

2016-02-01

Affect has become imperative in product quality. In affective design field, Kansei Engineering (KE) has been recognized as a technology that enables discovery of consumer's emotion and formulation of guide to design products that win consumers in the competitive market. Albeit powerful technology, there is no rule of thumb in its analysis and interpretation process. KE expertise is required to determine sets of related Kansei and the significant concept of emotion. Many research endeavors become handicapped with the limited number of available and accessible KE experts. This work is performed to simulate the role of experts with the use of Natphoric algorithm thus providing sound solution to the complexity and flexibility in KE. The algorithm is designed to learn the process by implementing training datasets taken from previous KE research works. A framework for automated KE is then designed to realize the development of automated KE system. A comparative analysis is performed to determine feasibility of the developed prototype to automate the process. The result shows that the significant Kansei is determined by manual KE implementation and the automated process is highly similar. KE research advocates will benefit this system to automatically determine significant design concepts.

Fully automated, deep learning segmentation of oxygen-induced retinopathy images

PubMed Central

Xiao, Sa; Bucher, Felicitas; Wu, Yue; Rokem, Ariel; Lee, Cecilia S.; Marra, Kyle V.; Fallon, Regis; Diaz-Aguilar, Sophia; Aguilar, Edith; Friedlander, Martin; Lee, Aaron Y.

2017-01-01

Oxygen-induced retinopathy (OIR) is a widely used model to study ischemia-driven neovascularization (NV) in the retina and to serve in proof-of-concept studies in evaluating antiangiogenic drugs for ocular, as well as nonocular, diseases. The primary parameters that are analyzed in this mouse model include the percentage of retina with vaso-obliteration (VO) and NV areas. However, quantification of these two key variables comes with a great challenge due to the requirement of human experts to read the images. Human readers are costly, time-consuming, and subject to bias. Using recent advances in machine learning and computer vision, we trained deep learning neural networks using over a thousand segmentations to fully automate segmentation in OIR images. While determining the percentage area of VO, our algorithm achieved a similar range of correlation coefficients to that of expert inter-human correlation coefficients. In addition, our algorithm achieved a higher range of correlation coefficients compared with inter-expert correlation coefficients for quantification of the percentage area of neovascular tufts. In summary, we have created an open-source, fully automated pipeline for the quantification of key values of OIR images using deep learning neural networks. PMID:29263301

The Burn Medical Assistant: Developing Machine Learning Algorithms to Aid in the Estimation of Burn Wound Size

DTIC Science & Technology

2017-10-01

hypothesis that a computer machine learning algorithm can analyze and classify burn injures using multispectral imaging within 5% of an expert clinician...morbidity. In response to these challenges, the USAISR developed and obtained FDA 510(k) clearance of the Burn Navigator™, a computer decision support... computer decision support software (CDSS), can significantly change the CDSS algorithm’s recommendations and thus the total fluid administered to a

Interactive Inverse Groundwater Modeling - Addressing User Fatigue

NASA Astrophysics Data System (ADS)

Singh, A.; Minsker, B. S.

2006-12-01

This paper builds on ongoing research on developing an interactive and multi-objective framework to solve the groundwater inverse problem. In this work we solve the classic groundwater inverse problem of estimating a spatially continuous conductivity field, given field measurements of hydraulic heads. The proposed framework is based on an interactive multi-objective genetic algorithm (IMOGA) that not only considers quantitative measures such as calibration error and degree of regularization, but also takes into account expert knowledge about the structure of the underlying conductivity field expressed as subjective rankings of potential conductivity fields by the expert. The IMOGA converges to the optimal Pareto front representing the best trade- off among the qualitative as well as quantitative objectives. However, since the IMOGA is a population-based iterative search it requires the user to evaluate hundreds of solutions. This leads to the problem of 'user fatigue'. We propose a two step methodology to combat user fatigue in such interactive systems. The first step is choosing only a few highly representative solutions to be shown to the expert for ranking. Spatial clustering is used to group the search space based on the similarity of the conductivity fields. Sampling is then carried out from different clusters to improve the diversity of solutions shown to the user. Once the expert has ranked representative solutions from each cluster a machine learning model is used to 'learn user preference' and extrapolate these for the solutions not ranked by the expert. We investigate different machine learning models such as Decision Trees, Bayesian learning model, and instance based weighting to model user preference. In addition, we also investigate ways to improve the performance of these models by providing information about the spatial structure of the conductivity fields (which is what the expert bases his or her rank on). Results are shown for each of these machine learning models and the advantages and disadvantages for each approach are discussed. These results indicate that using the proposed two-step methodology leads to significant reduction in user-fatigue without deteriorating the solution quality of the IMOGA.

Knowledge of Fecal Calprotectin and Infliximab Trough Levels Alters Clinical Decision-making for IBD Outpatients on Maintenance Infliximab Therapy.

PubMed

Huang, Vivian W; Prosser, Connie; Kroeker, Karen I; Wang, Haili; Shalapay, Carol; Dhami, Neil; Fedorak, Darryl K; Halloran, Brendan; Dieleman, Levinus A; Goodman, Karen J; Fedorak, Richard N

2015-06-01

Infliximab is an effective therapy for inflammatory bowel disease (IBD). However, more than 50% of patients lose response. Empiric dose intensification is not effective for all patients because not all patients have objective disease activity or subtherapeutic drug level. The aim was to determine how an objective marker of disease activity or therapeutic drug monitoring affects clinical decisions regarding maintenance infliximab therapy in outpatients with IBD. Consecutive patients with IBD on maintenance infliximab therapy were invited to participate by providing preinfusion stool and blood samples. Fecal calprotectin (FCP) and infliximab trough levels (ITLs) were measured by enzyme linked immunosorbent assay. Three decisions were compared: (1) actual clinical decision, (2) algorithmic FCP or ITL decisions, and (3) expert panel decision based on (a) clinical data, (b) clinical data plus FCP, and (c) clinical data plus FCP plus ITL. In secondary analysis, Receiver-operating curves were used to assess the ability of FCP and ITL in predicting clinical disease activity or remission. A total of 36 sets of blood and stool were available for analysis; median FCP 191.5 μg/g, median ITLs 7.3 μg/mL. The actual clinical decision differed from the hypothetical decision in 47.2% (FCP algorithm); 69.4% (ITL algorithm); 25.0% (expert panel clinical decision); 44.4% (expert panel clinical plus FCP); 58.3% (expert panel clinical plus FCP plus ITL) cases. FCP predicted clinical relapse (area under the curve [AUC] = 0.417; 95% confidence interval [CI], 0.197-0.641) and subtherapeutic ITL (AUC = 0.774; 95% CI, 0.536-1.000). ITL predicted clinical remission (AUC = 0.498; 95% CI, 0.254-0.742) and objective remission (AUC = 0.773; 95% CI, 0.622-0.924). Using FCP and ITLs in addition to clinical data results in an increased number of decisions to optimize management in outpatients with IBD on stable maintenance infliximab therapy.

Cardiac risk stratification in renal transplantation using a form of artificial intelligence.

PubMed

Heston, T F; Norman, D J; Barry, J M; Bennett, W M; Wilson, R A

1997-02-15

The purpose of this study was to determine if an expert network, a form of artificial intelligence, could effectively stratify cardiac risk in candidates for renal transplant. Input into the expert network consisted of clinical risk factors and thallium-201 stress test data. Clinical risk factor screening alone identified 95 of 189 patients as high risk. These 95 patients underwent thallium-201 stress testing, and 53 had either reversible or fixed defects. The other 42 patients were classified as low risk. This algorithm made up the "expert system," and during the 4-year follow-up period had a sensitivity of 82%, specificity of 77%, and accuracy of 78%. An artificial neural network was added to the expert system, creating an expert network. Input into the neural network consisted of both clinical variables and thallium-201 stress test data. There were 5 hidden nodes and the output (end point) was cardiac death. The expert network increased the specificity of the expert system alone from 77% to 90% (p < 0.001), the accuracy from 78% to 89% (p < 0.005), and maintained the overall sensitivity at 88%. An expert network based on clinical risk factor screening and thallium-201 stress testing had an accuracy of 89% in predicting the 4-year cardiac mortality among 189 renal transplant candidates.

A Bayesian approach to the characterization of electroencephalographic recordings in premature infants

NASA Astrophysics Data System (ADS)

Mitchell, Timothy J.

Preterm infants are particularly susceptible to cerebral injury, and electroencephalographic (EEG) recordings provide an important diagnostic tool for determining cerebral health. However, interpreting these EEG recordings is challenging and requires the skills of a trained electroencephalographer. Because these EEG specialists are rare, an automated interpretation of newborn EEG recordings would increase access to an important diagnostic tool for physicians. To automate this procedure, we employ a novel Bayesian approach to compute the probability of EEG features (waveforms) including suppression, delta brushes, and delta waves. The power of this approach lies not only in its ability to closely mimic the techniques used by EEG specialists, but also its ability to be generalized to identify other waveforms that may be of interest for future work. The results of these calculations are used in a program designed to output simple statistics related to the presence or absence of such features. Direct comparison of the software with expert human readers has indicated satisfactory performance, and the algorithm has shown promise in its ability to distinguish between infants with normal neurodevelopmental outcome and those with poor neurodevelopmental outcome.

Support vector machines

NASA Technical Reports Server (NTRS)

Garay, Michael J.; Mazzoni, Dominic; Davies, Roger; Wagstaff, Kiri

2004-01-01

Support Vector Machines (SVMs) are a type of supervised learning algorith,, other examples of which are Artificial Neural Networks (ANNs), Decision Trees, and Naive Bayesian Classifiers. Supervised learning algorithms are used to classify objects labled by a 'supervisor' - typically a human 'expert.'.

Uncertain programming models for portfolio selection with uncertain returns

NASA Astrophysics Data System (ADS)

Zhang, Bo; Peng, Jin; Li, Shengguo

2015-10-01

In an indeterminacy economic environment, experts' knowledge about the returns of securities consists of much uncertainty instead of randomness. This paper discusses portfolio selection problem in uncertain environment in which security returns cannot be well reflected by historical data, but can be evaluated by the experts. In the paper, returns of securities are assumed to be given by uncertain variables. According to various decision criteria, the portfolio selection problem in uncertain environment is formulated as expected-variance-chance model and chance-expected-variance model by using the uncertainty programming. Within the framework of uncertainty theory, for the convenience of solving the models, some crisp equivalents are discussed under different conditions. In addition, a hybrid intelligent algorithm is designed in the paper to provide a general method for solving the new models in general cases. At last, two numerical examples are provided to show the performance and applications of the models and algorithm.

An expert support system for breast cancer diagnosis using color wavelet features.

PubMed

Issac Niwas, S; Palanisamy, P; Chibbar, Rajni; Zhang, W J

2012-10-01

Breast cancer diagnosis can be done through the pathologic assessments of breast tissue samples such as core needle biopsy technique. The result of analysis on this sample by pathologist is crucial for breast cancer patient. In this paper, nucleus of tissue samples are investigated after decomposition by means of the Log-Gabor wavelet on HSV color domain and an algorithm is developed to compute the color wavelet features. These features are used for breast cancer diagnosis using Support Vector Machine (SVM) classifier algorithm. The ability of properly trained SVM is to correctly classify patterns and make them particularly suitable for use in an expert system that aids in the diagnosis of cancer tissue samples. The results are compared with other multivariate classifiers such as Naïves Bayes classifier and Artificial Neural Network. The overall accuracy of the proposed method using SVM classifier will be further useful for automation in cancer diagnosis.

Geriatric Assessment-Guided Care Processes for Older Adults: A Delphi Consensus of Geriatric Oncology Experts.

PubMed

Mohile, Supriya Gupta; Velarde, Carla; Hurria, Arti; Magnuson, Allison; Lowenstein, Lisa; Pandya, Chintan; O'Donovan, Anita; Gorawara-Bhat, Rita; Dale, William

2015-09-01

Structured care processes that provide a framework for how oncologists can incorporate geriatric assessment (GA) into clinical practice could improve outcomes for vulnerable older adults with cancer, a growing population at high risk of toxicity from cancer treatment. We sought to obtain consensus from an expert panel on the use of GA in clinical practice and to develop algorithms of GA-guided care processes. The Delphi technique, a well-recognized structured and reiterative process to reach consensus, was used. Participants were geriatric oncology experts who attended NIH-funded U13 or Cancer and Aging Research Group conferences. Consensus was defined as an interquartile range of 2 or more units, or 66.7% or greater, selecting a utility/helpfulness rating of 7 or greater on a 10-point Likert scale. For nominal data, consensus was defined as agreement among 66.7% or more of the group. From 33 invited, 30 participants completed all 3 rounds. Most experts (75%) used GA in clinical care, and the remainder were involved in geriatric oncology research. The panel met consensus that "all patients aged 75 years or older and those who are younger with age-related health concerns" should undergo GA and that all domains (function, physical performance, comorbidity/polypharmacy, cognition, nutrition, psychological status, and social support) should be included. Consensus was met for how GA could guide nononcologic interventions and cancer treatment decisions. Algorithms for GA-guided care processes were developed. This Delphi investigation of geriatric oncology experts demonstrated that GA should be performed for older patients with cancer to guide care processes. Copyright © 2015 by the National Comprehensive Cancer Network.

Korean Medication Algorithm Project for Bipolar Disorder: third revision

PubMed Central

Woo, Young Sup; Lee, Jung Goo; Jeong, Jong-Hyun; Kim, Moon-Doo; Sohn, Inki; Shim, Se-Hoon; Jon, Duk-In; Seo, Jeong Seok; Shin, Young-Chul; Min, Kyung Joon; Yoon, Bo-Hyun; Bahk, Won-Myong

2015-01-01

Objective To constitute the third revision of the guidelines for the treatment of bipolar disorder issued by the Korean Medication Algorithm Project for Bipolar Disorder (KMAP-BP 2014). Methods A 56-item questionnaire was used to obtain the consensus of experts regarding pharmacological treatment strategies for the various phases of bipolar disorder and for special populations. The review committee included 110 Korean psychiatrists and 38 experts for child and adolescent psychiatry. Of the committee members, 64 general psychiatrists and 23 child and adolescent psychiatrists responded to the survey. Results The treatment of choice (TOC) for euphoric, mixed, and psychotic mania was the combination of a mood stabilizer (MS) and an atypical antipsychotic (AAP); the TOC for acute mild depression was monotherapy with MS or AAP; and the TOC for moderate or severe depression was MS plus AAP/antidepressant. The first-line maintenance treatment following mania or depression was MS monotherapy or MS plus AAP; the first-line treatment after mania was AAP monotherapy; and the first-line treatment after depression was lamotrigine (LTG) monotherapy, LTG plus MS/AAP, or MS plus AAP plus LTG. The first-line treatment strategy for mania in children and adolescents was MS plus AAP or AAP monotherapy. For geriatric bipolar patients, the TOC for mania was AAP/MS monotherapy, and the TOC for depression was AAP plus MS or AAP monotherapy. Conclusion The expert consensus in the KMAP-BP 2014 differed from that in previous publications; most notably, the preference for AAP was increased in the treatment of acute mania, depression, and maintenance treatment. There was increased expert preference for the use of AAP and LTG. The major limitation of the present study is that it was based on the consensus of Korean experts rather than on experimental evidence. PMID:25750530

Korean Medication Algorithm Project for Bipolar Disorder: third revision.

PubMed

Woo, Young Sup; Lee, Jung Goo; Jeong, Jong-Hyun; Kim, Moon-Doo; Sohn, Inki; Shim, Se-Hoon; Jon, Duk-In; Seo, Jeong Seok; Shin, Young-Chul; Min, Kyung Joon; Yoon, Bo-Hyun; Bahk, Won-Myong

2015-01-01

To constitute the third revision of the guidelines for the treatment of bipolar disorder issued by the Korean Medication Algorithm Project for Bipolar Disorder (KMAP-BP 2014). A 56-item questionnaire was used to obtain the consensus of experts regarding pharmacological treatment strategies for the various phases of bipolar disorder and for special populations. The review committee included 110 Korean psychiatrists and 38 experts for child and adolescent psychiatry. Of the committee members, 64 general psychiatrists and 23 child and adolescent psychiatrists responded to the survey. The treatment of choice (TOC) for euphoric, mixed, and psychotic mania was the combination of a mood stabilizer (MS) and an atypical antipsychotic (AAP); the TOC for acute mild depression was monotherapy with MS or AAP; and the TOC for moderate or severe depression was MS plus AAP/antidepressant. The first-line maintenance treatment following mania or depression was MS monotherapy or MS plus AAP; the first-line treatment after mania was AAP monotherapy; and the first-line treatment after depression was lamotrigine (LTG) monotherapy, LTG plus MS/AAP, or MS plus AAP plus LTG. The first-line treatment strategy for mania in children and adolescents was MS plus AAP or AAP monotherapy. For geriatric bipolar patients, the TOC for mania was AAP/MS monotherapy, and the TOC for depression was AAP plus MS or AAP monotherapy. The expert consensus in the KMAP-BP 2014 differed from that in previous publications; most notably, the preference for AAP was increased in the treatment of acute mania, depression, and maintenance treatment. There was increased expert preference for the use of AAP and LTG. The major limitation of the present study is that it was based on the consensus of Korean experts rather than on experimental evidence.

Towards a 3d Based Platform for Cultural Heritage Site Survey and Virtual Exploration

NASA Astrophysics Data System (ADS)

Seinturier, J.; Riedinger, C.; Mahiddine, A.; Peloso, D.; Boï, J.-M.; Merad, D.; Drap, P.

2013-07-01

This paper present a 3D platform that enables to make both cultural heritage site survey and its virtual exploration. It provides a single and easy way to use framework for merging multi scaled 3D measurements based on photogrammetry, documentation produced by experts and the knowledge of involved domains leaving the experts able to extract and choose the relevant information to produce the final survey. Taking into account the interpretation of the real world during the process of archaeological surveys is in fact the main goal of a survey. New advances in photogrammetry and the capability to produce dense 3D point clouds do not solve the problem of surveys. New opportunities for 3D representation are now available and we must to use them and find new ways to link geometry and knowledge. The new platform is able to efficiently manage and process large 3D data (points set, meshes) thanks to the implementation of space partition methods coming from the state of the art such as octrees and kd-trees and thus can interact with dense point clouds (thousands to millions of points) in real time. The semantisation of raw 3D data relies on geometric algorithms such as geodetic path computation, surface extraction from dense points cloud and geometrical primitive optimization. The platform provide an interface that enables expert to describe geometric representations of interesting objects like ashlar blocs, stratigraphic units or generic items (contour, lines, … ) directly onto the 3D representation of the site and without explicit links to underlying algorithms. The platform provide two ways for describing geometric representation. If oriented photographs are available, the expert can draw geometry on a photograph and the system computes its 3D representation by projection on the underlying mesh or the points cloud. If photographs are not available or if the expert wants to only use the 3D representation then he can simply draw objects shape on it. When 3D representations of objects of a surveyed site are extracted from the mesh, the link with domain related documentation is done by means of a set of forms designed by experts. Information from these forms are linked with geometry such that documentation can be attached to the viewed objects. Additional semantisation methods related to specific domains have been added to the platform. Beyond realistic rendering of surveyed site, the platform embeds non photorealistic rendering (NPR) algorithms. These algorithms enable to dynamically illustrate objects of interest that are related to knowledge with specific styles. The whole platform is implemented with a Java framework and relies on an actual and effective 3D engine that make available latest rendering methods. We illustrate this work on various photogrammetric survey, in medieval archaeology with the Shawbak castle in Jordan and in underwater archaeology on different marine sites.

Unsupervised detection and removal of muscle artifacts from scalp EEG recordings using canonical correlation analysis, wavelets and random forests.

PubMed

Anastasiadou, Maria N; Christodoulakis, Manolis; Papathanasiou, Eleftherios S; Papacostas, Savvas S; Mitsis, Georgios D

2017-09-01

This paper proposes supervised and unsupervised algorithms for automatic muscle artifact detection and removal from long-term EEG recordings, which combine canonical correlation analysis (CCA) and wavelets with random forests (RF). The proposed algorithms first perform CCA and continuous wavelet transform of the canonical components to generate a number of features which include component autocorrelation values and wavelet coefficient magnitude values. A subset of the most important features is subsequently selected using RF and labelled observations (supervised case) or synthetic data constructed from the original observations (unsupervised case). The proposed algorithms are evaluated using realistic simulation data as well as 30min epochs of non-invasive EEG recordings obtained from ten patients with epilepsy. We assessed the performance of the proposed algorithms using classification performance and goodness-of-fit values for noisy and noise-free signal windows. In the simulation study, where the ground truth was known, the proposed algorithms yielded almost perfect performance. In the case of experimental data, where expert marking was performed, the results suggest that both the supervised and unsupervised algorithm versions were able to remove artifacts without affecting noise-free channels considerably, outperforming standard CCA, independent component analysis (ICA) and Lagged Auto-Mutual Information Clustering (LAMIC). The proposed algorithms achieved excellent performance for both simulation and experimental data. Importantly, for the first time to our knowledge, we were able to perform entirely unsupervised artifact removal, i.e. without using already marked noisy data segments, achieving performance that is comparable to the supervised case. Overall, the results suggest that the proposed algorithms yield significant future potential for improving EEG signal quality in research or clinical settings without the need for marking by expert neurophysiologists, EMG signal recording and user visual inspection. Copyright © 2017 International Federation of Clinical Neurophysiology. Published by Elsevier B.V. All rights reserved.

Multispectral imaging burn wound tissue classification system: a comparison of test accuracies between several common machine learning algorithms

NASA Astrophysics Data System (ADS)

Squiers, John J.; Li, Weizhi; King, Darlene R.; Mo, Weirong; Zhang, Xu; Lu, Yang; Sellke, Eric W.; Fan, Wensheng; DiMaio, J. Michael; Thatcher, Jeffrey E.

2016-03-01

The clinical judgment of expert burn surgeons is currently the standard on which diagnostic and therapeutic decisionmaking regarding burn injuries is based. Multispectral imaging (MSI) has the potential to increase the accuracy of burn depth assessment and the intraoperative identification of viable wound bed during surgical debridement of burn injuries. A highly accurate classification model must be developed using machine-learning techniques in order to translate MSI data into clinically-relevant information. An animal burn model was developed to build an MSI training database and to study the burn tissue classification ability of several models trained via common machine-learning algorithms. The algorithms tested, from least to most complex, were: K-nearest neighbors (KNN), decision tree (DT), linear discriminant analysis (LDA), weighted linear discriminant analysis (W-LDA), quadratic discriminant analysis (QDA), ensemble linear discriminant analysis (EN-LDA), ensemble K-nearest neighbors (EN-KNN), and ensemble decision tree (EN-DT). After the ground-truth database of six tissue types (healthy skin, wound bed, blood, hyperemia, partial injury, full injury) was generated by histopathological analysis, we used 10-fold cross validation to compare the algorithms' performances based on their accuracies in classifying data against the ground truth, and each algorithm was tested 100 times. The mean test accuracy of the algorithms were KNN 68.3%, DT 61.5%, LDA 70.5%, W-LDA 68.1%, QDA 68.9%, EN-LDA 56.8%, EN-KNN 49.7%, and EN-DT 36.5%. LDA had the highest test accuracy, reflecting the bias-variance tradeoff over the range of complexities inherent to the algorithms tested. Several algorithms were able to match the current standard in burn tissue classification, the clinical judgment of expert burn surgeons. These results will guide further development of an MSI burn tissue classification system. Given that there are few surgeons and facilities specializing in burn care, this technology may improve the standard of burn care for patients without access to specialized facilities.

An Interpretable Machine Learning Model for Accurate Prediction of Sepsis in the ICU.

PubMed

Nemati, Shamim; Holder, Andre; Razmi, Fereshteh; Stanley, Matthew D; Clifford, Gari D; Buchman, Timothy G

2018-04-01

Sepsis is among the leading causes of morbidity, mortality, and cost overruns in critically ill patients. Early intervention with antibiotics improves survival in septic patients. However, no clinically validated system exists for real-time prediction of sepsis onset. We aimed to develop and validate an Artificial Intelligence Sepsis Expert algorithm for early prediction of sepsis. Observational cohort study. Academic medical center from January 2013 to December 2015. Over 31,000 admissions to the ICUs at two Emory University hospitals (development cohort), in addition to over 52,000 ICU patients from the publicly available Medical Information Mart for Intensive Care-III ICU database (validation cohort). Patients who met the Third International Consensus Definitions for Sepsis (Sepsis-3) prior to or within 4 hours of their ICU admission were excluded, resulting in roughly 27,000 and 42,000 patients within our development and validation cohorts, respectively. None. High-resolution vital signs time series and electronic medical record data were extracted. A set of 65 features (variables) were calculated on hourly basis and passed to the Artificial Intelligence Sepsis Expert algorithm to predict onset of sepsis in the proceeding T hours (where T = 12, 8, 6, or 4). Artificial Intelligence Sepsis Expert was used to predict onset of sepsis in the proceeding T hours and to produce a list of the most significant contributing factors. For the 12-, 8-, 6-, and 4-hour ahead prediction of sepsis, Artificial Intelligence Sepsis Expert achieved area under the receiver operating characteristic in the range of 0.83-0.85. Performance of the Artificial Intelligence Sepsis Expert on the development and validation cohorts was indistinguishable. Using data available in the ICU in real-time, Artificial Intelligence Sepsis Expert can accurately predict the onset of sepsis in an ICU patient 4-12 hours prior to clinical recognition. A prospective study is necessary to determine the clinical utility of the proposed sepsis prediction model.

TrajGraph: A Graph-Based Visual Analytics Approach to Studying Urban Network Centralities Using Taxi Trajectory Data.

PubMed

Huang, Xiaoke; Zhao, Ye; Yang, Jing; Zhang, Chong; Ma, Chao; Ye, Xinyue

2016-01-01

We propose TrajGraph, a new visual analytics method, for studying urban mobility patterns by integrating graph modeling and visual analysis with taxi trajectory data. A special graph is created to store and manifest real traffic information recorded by taxi trajectories over city streets. It conveys urban transportation dynamics which can be discovered by applying graph analysis algorithms. To support interactive, multiscale visual analytics, a graph partitioning algorithm is applied to create region-level graphs which have smaller size than the original street-level graph. Graph centralities, including Pagerank and betweenness, are computed to characterize the time-varying importance of different urban regions. The centralities are visualized by three coordinated views including a node-link graph view, a map view and a temporal information view. Users can interactively examine the importance of streets to discover and assess city traffic patterns. We have implemented a fully working prototype of this approach and evaluated it using massive taxi trajectories of Shenzhen, China. TrajGraph's capability in revealing the importance of city streets was evaluated by comparing the calculated centralities with the subjective evaluations from a group of drivers in Shenzhen. Feedback from a domain expert was collected. The effectiveness of the visual interface was evaluated through a formal user study. We also present several examples and a case study to demonstrate the usefulness of TrajGraph in urban transportation analysis.

A scalable method to improve gray matter segmentation at ultra high field MRI.

PubMed

Gulban, Omer Faruk; Schneider, Marian; Marquardt, Ingo; Haast, Roy A M; De Martino, Federico

2018-01-01

High-resolution (functional) magnetic resonance imaging (MRI) at ultra high magnetic fields (7 Tesla and above) enables researchers to study how anatomical and functional properties change within the cortical ribbon, along surfaces and across cortical depths. These studies require an accurate delineation of the gray matter ribbon, which often suffers from inclusion of blood vessels, dura mater and other non-brain tissue. Residual segmentation errors are commonly corrected by browsing the data slice-by-slice and manually changing labels. This task becomes increasingly laborious and prone to error at higher resolutions since both work and error scale with the number of voxels. Here we show that many mislabeled, non-brain voxels can be corrected more efficiently and semi-automatically by representing three-dimensional anatomical images using two-dimensional histograms. We propose both a uni-modal (based on first spatial derivative) and multi-modal (based on compositional data analysis) approach to this representation and quantify the benefits in 7 Tesla MRI data of nine volunteers. We present an openly accessible Python implementation of these approaches and demonstrate that editing cortical segmentations using two-dimensional histogram representations as an additional post-processing step aids existing algorithms and yields improved gray matter borders. By making our data and corresponding expert (ground truth) segmentations openly available, we facilitate future efforts to develop and test segmentation algorithms on this challenging type of data.

A scalable method to improve gray matter segmentation at ultra high field MRI

PubMed Central

De Martino, Federico

2018-01-01

High-resolution (functional) magnetic resonance imaging (MRI) at ultra high magnetic fields (7 Tesla and above) enables researchers to study how anatomical and functional properties change within the cortical ribbon, along surfaces and across cortical depths. These studies require an accurate delineation of the gray matter ribbon, which often suffers from inclusion of blood vessels, dura mater and other non-brain tissue. Residual segmentation errors are commonly corrected by browsing the data slice-by-slice and manually changing labels. This task becomes increasingly laborious and prone to error at higher resolutions since both work and error scale with the number of voxels. Here we show that many mislabeled, non-brain voxels can be corrected more efficiently and semi-automatically by representing three-dimensional anatomical images using two-dimensional histograms. We propose both a uni-modal (based on first spatial derivative) and multi-modal (based on compositional data analysis) approach to this representation and quantify the benefits in 7 Tesla MRI data of nine volunteers. We present an openly accessible Python implementation of these approaches and demonstrate that editing cortical segmentations using two-dimensional histogram representations as an additional post-processing step aids existing algorithms and yields improved gray matter borders. By making our data and corresponding expert (ground truth) segmentations openly available, we facilitate future efforts to develop and test segmentation algorithms on this challenging type of data. PMID:29874295

Machine Learning in Medicine.

PubMed

Deo, Rahul C

2015-11-17

Spurred by advances in processing power, memory, storage, and an unprecedented wealth of data, computers are being asked to tackle increasingly complex learning tasks, often with astonishing success. Computers have now mastered a popular variant of poker, learned the laws of physics from experimental data, and become experts in video games - tasks that would have been deemed impossible not too long ago. In parallel, the number of companies centered on applying complex data analysis to varying industries has exploded, and it is thus unsurprising that some analytic companies are turning attention to problems in health care. The purpose of this review is to explore what problems in medicine might benefit from such learning approaches and use examples from the literature to introduce basic concepts in machine learning. It is important to note that seemingly large enough medical data sets and adequate learning algorithms have been available for many decades, and yet, although there are thousands of papers applying machine learning algorithms to medical data, very few have contributed meaningfully to clinical care. This lack of impact stands in stark contrast to the enormous relevance of machine learning to many other industries. Thus, part of my effort will be to identify what obstacles there may be to changing the practice of medicine through statistical learning approaches, and discuss how these might be overcome. © 2015 American Heart Association, Inc.

Columbia Classification Algorithm of Suicide Assessment (C-CASA): classification of suicidal events in the FDA's pediatric suicidal risk analysis of antidepressants.

PubMed

Posner, Kelly; Oquendo, Maria A; Gould, Madelyn; Stanley, Barbara; Davies, Mark

2007-07-01

To evaluate the link between antidepressants and suicidal behavior and ideation (suicidality) in youth, adverse events from pediatric clinical trials were classified in order to identify suicidal events. The authors describe the Columbia Classification Algorithm for Suicide Assessment (C-CASA), a standardized suicidal rating system that provided data for the pediatric suicidal risk analysis of antidepressants conducted by the Food and Drug Administration (FDA). Adverse events (N=427) from 25 pediatric antidepressant clinical trials were systematically identified by pharmaceutical companies. Randomly assigned adverse events were evaluated by three of nine independent expert suicidologists using the Columbia classification algorithm. Reliability of the C-CASA ratings and agreement with pharmaceutical company classification were estimated. Twenty-six new, possibly suicidal events (behavior and ideation) that were not originally identified by pharmaceutical companies were identified in the C-CASA, and 12 events originally labeled as suicidal by pharmaceutical companies were eliminated, which resulted in a total of 38 discrepant ratings. For the specific label of "suicide attempt," a relatively low level of agreement was observed between the C-CASA and pharmaceutical company ratings, with the C-CASA reporting a 50% reduction in ratings. Thus, although the C-CASA resulted in the identification of more suicidal events overall, fewer events were classified as suicide attempts. Additionally, the C-CASA ratings were highly reliable (intraclass correlation coefficient [ICC]=0.89). Utilizing a methodical, anchored approach to categorizing suicidality provides an accurate and comprehensive identification of suicidal events. The FDA's audit of the C-CASA demonstrated excellent transportability of this approach. The Columbia algorithm was used to classify suicidal adverse events in the recent FDA adult antidepressant safety analyses and has also been mandated to be applied to all anticonvulsant trials and other centrally acting agents and nonpsychotropic drugs.

Automated Quantification of Volumetric Optic Disc Swelling in Papilledema Using Spectral-Domain Optical Coherence Tomography

PubMed Central

Wang, Jui-Kai; Kardon, Randy H.; Kupersmith, Mark J.; Garvin, Mona K.

2012-01-01

Purpose. To develop an automated method for the quantification of volumetric optic disc swelling in papilledema subjects using spectral-domain optical coherence tomography (SD-OCT) and to determine the extent that such volumetric measurements correlate with Frisén scale grades (from fundus photographs) and two-dimensional (2-D) peripapillary retinal nerve fiber layer (RNFL) and total retinal (TR) thickness measurements from SD-OCT. Methods. A custom image-analysis algorithm was developed to obtain peripapillary circular RNFL thickness, TR thickness, and TR volume measurements from SD-OCT volumes of subjects with papilledema. In addition, peripapillary RNFL thickness measures from the commercially available Zeiss SD-OCT machine were obtained. Expert Frisén scale grades were independently obtained from corresponding fundus photographs. Results. In 71 SD-OCT scans, the mean (± standard deviation) resulting TR volumes for Frisén scale 0 to scale 4 were 11.36 ± 0.56, 12.53 ± 1.21, 14.42 ± 2.11, 17.48 ± 2.63, and 21.81 ± 3.16 mm3, respectively. The Spearman's rank correlation coefficient was 0.737. Using 55 eyes with valid Zeiss RNFL measurements, Pearson's correlation coefficient (r) between the TR volume and the custom algorithm's TR thickness, the custom algorithm's RNFL thickness, and Zeiss' RNFL thickness was 0.980, 0.929, and 0.946, respectively. Between Zeiss' RNFL and the custom algorithm's RNFL, and the study's TR thickness, r was 0.901 and 0.961, respectively. Conclusions. Volumetric measurements of the degree of disc swelling in subjects with papilledema can be obtained from SD-OCT volumes, with the mean volume appearing to be roughly linearly related to the Frisén scale grade. Using such an approach can provide a more continuous, objective, and robust means for assessing the degree of disc swelling compared with presently available approaches. PMID:22599584

Columbia Classification Algorithm of Suicide Assessment (C-CASA): Classification of Suicidal Events in the FDA’s Pediatric Suicidal Risk Analysis of Antidepressants

PubMed Central

Posner, Kelly; Oquendo, Maria A.; Gould, Madelyn; Stanley, Barbara; Davies, Mark

2013-01-01

Objective To evaluate the link between antidepressants and suicidal behavior and ideation (suicidality) in youth, adverse events from pediatric clinical trials were classified in order to identify suicidal events. The authors describe the Columbia Classification Algorithm for Suicide Assessment (C-CASA), a standardized suicidal rating system that provided data for the pediatric suicidal risk analysis of antide-pressants conducted by the Food and Drug Administration (FDA). Method Adverse events (N=427) from 25 pediatric antidepressant clinical trials were systematically identified by pharmaceutical companies. Randomly assigned adverse events were evaluated by three of nine independent expert suicidologists using the Columbia classification algorithm. Reliability of the C-CASA ratings and agreement with pharmaceutical company classification were estimated. Results Twenty-six new, possibly suicidal events (behavior and ideation) that were not originally identified by pharmaceutical companies were identified in the C-CASA, and 12 events originally labeled as suicidal by pharmaceutical companies were eliminated, which resulted in a total of 38 discrepant ratings. For the specific label of “suicide attempt,” a relatively low level of agreement was observed between the C-CASA and pharmaceutical company ratings, with the C-CASA reporting a 50% reduction in ratings. Thus, although the C-CASA resulted in the identification of more suicidal events overall, fewer events were classified as suicide attempts. Additionally, the C-CASA ratings were highly reliable (intraclass correlation coefficient [ICC]=0.89). Conclusions Utilizing a methodical, anchored approach to categorizing suicidality provides an accurate and comprehensive identification of suicidal events. The FDA’s audit of the C-CASA demonstrated excellent transportability of this approach. The Columbia algorithm was used to classify suicidal adverse events in the recent FDA adult antidepressant safety analyses and has also been mandated to be applied to all anticonvulsant trials and other centrally acting agents and nonpsychotropic drugs. PMID:17606655

Korean Medication Algorithm for Depressive Disorders 2017: Third Revision

PubMed Central

Seo, Jeong Seok; Wang, Hee Ryung; Woo, Young Sup; Park, Young-Min; Jeong, Jong-Hyun; Kim, Won; Shim, Se-Hoon; Lee, Jung Goo; Jon, Duk-In

2018-01-01

Objective In 2002, the Korean Society for Affective Disorders developed the guidelines for the treatment of major depressive disorder (MDD), and revised it in 2006 and 2012. The third revision of these guidelines was undertaken to reflect advances in the field. Methods Using a 44-item questionnaire, an expert consensus was obtained on pharmacological treatment strategies for MDD 1) without or 2) with psychotic features, 3) depression subtypes, 4) maintenance, 5) special populations, 6) the choice of an antidepressant (AD) regarding safety and adverse effects, and 7) non-pharmacological biological therapies. Recommended first, second, and third-line strategies were derived statistically. Results AD monotherapy is recommended as the first-line strategy for non-psychotic depression in adults, children/adolescents, elderly adults, patient with persistent depressive disorder, and pregnant women or patients with postpartum depression or premenstrual dysphoric disorder. The combination of AD and atypical antipsychotics (AAP) was recommended for psychotic depression in adult, child/adolescent, postpartum depression, and mixed features or anxious distress. Most experts recommended stopping the ongoing initial AD and AAP after a certain period in patients with one or two depressive episodes. As an MDD treatment modality, 92% of experts are considering electroconvulsive therapy and 46.8% are applying it clinically, while 86% of experts are considering repetitive transcranial magnetic stimulation but only 31.6% are applying it clinically. Conclusion The pharmacological treatment strategy in 2017 is similar to that of Korean Medication Algorithm for Depressive Disorder 2012. The preference of AAPs was more increased. PMID:29397669

Automated algorithms for detecting sleep period time using a multi-sensor pattern-recognition activity monitor from 24 h free-living data in older adults.

PubMed

Cabanas-Sánchez, Verónica; Higueras-Fresnillo, Sara; De la Cámara, Miguel Ángel; Veiga, Oscar L; Martinez-Gomez, David

2018-05-16

The aims of the present study were (i) to develop automated algorithms to identify the sleep period time in 24 h data from the Intelligent Device for Energy Expenditure and Activity (IDEEA) in older adults, and (ii) to analyze the agreement between these algorithms to identify the sleep period time as compared to self-reported data and expert visual analysis of accelerometer raw data. This study comprised 50 participants, aged 65-85 years. Fourteen automated algorithms were developed. Participants reported their bedtime and waking time on the days on which they wore the device. A well-trained expert reviewed each IDEEA file in order to visually identify bedtime and waking time on each day. To explore the agreement between methods, Pearson correlations, mean differences, mean percentage errors, accuracy, sensitivity and specificity, and the Bland-Altman method were calculated. With 87 d of valid data, algorithms 6, 7, 11 and 12 achieved higher levels of agreement in determining sleep period time when compared to self-reported data (mean difference  =  -0.34 to 0.01 h d -1 ; mean absolute error  =  10.66%-11.44%; r  =  0.515-0.686; accuracy  =  95.0%-95.6%; sensitivity  =  93.0%-95.8%; specificity  =  95.7%-96.4%) and expert visual analysis (mean difference  =  -0.04 to 0.31 h d -1 ; mean absolute error  =  5.0%-6.97%; r  =  0.620-0.766; accuracy  =  97.2%-98.0%; sensitivity  =  94.5%-97.6%; specificity  =  98.4%-98.8%). Bland-Altman plots showed no systematic biases in these comparisons (all p  >  0.05). Differences between methods did not vary significantly by gender, age, obesity, self-rated health, or the presence of chronic conditions. These four algorithms can be used to identify easily and with adequate accuracy the sleep period time using the IDEEA activity monitor from 24 h free-living data in older adults.

Toward a theory of distributed word expert natural language parsing

NASA Technical Reports Server (NTRS)

Rieger, C.; Small, S.

1981-01-01

An approach to natural language meaning-based parsing in which the unit of linguistic knowledge is the word rather than the rewrite rule is described. In the word expert parser, knowledge about language is distributed across a population of procedural experts, each representing a word of the language, and each an expert at diagnosing that word's intended usage in context. The parser is structured around a coroutine control environment in which the generator-like word experts ask questions and exchange information in coming to collective agreement on sentence meaning. The word expert theory is advanced as a better cognitive model of human language expertise than the traditional rule-based approach. The technical discussion is organized around examples taken from the prototype LISP system which implements parts of the theory.

Development of an evolutionary fuzzy expert system for estimating future behavior of stock price

NASA Astrophysics Data System (ADS)

Mehmanpazir, Farhad; Asadi, Shahrokh

2017-03-01

The stock market has always been an attractive area for researchers since no method has been found yet to predict the stock price behavior precisely. Due to its high rate of uncertainty and volatility, it carries a higher risk than any other investment area, thus the stock price behavior is difficult to simulation. This paper presents a "data mining-based evolutionary fuzzy expert system" (DEFES) approach to estimate the behavior of stock price. This tool is developed in seven-stage architecture. Data mining is used in three stages to reduce the complexity of the whole data space. The first stage, noise filtering, is used to make our raw data clean and smooth. Variable selection is second stage; we use stepwise regression analysis to choose the key variables been considered in the model. In the third stage, K-means is used to divide the data into sub-populations to decrease the effects of noise and rebate complexity of the patterns. At next stage, extraction of Mamdani type fuzzy rule-based system will be carried out for each cluster by means of genetic algorithm and evolutionary strategy. In the fifth stage, we use binary genetic algorithm to rule filtering to remove the redundant rules in order to solve over learning phenomenon. In the sixth stage, we utilize the genetic tuning process to slightly adjust the shape of the membership functions. Last stage is the testing performance of tool and adjusts parameters. This is the first study on using an approximate fuzzy rule base system and evolutionary strategy with the ability of extracting the whole knowledge base of fuzzy expert system for stock price forecasting problems. The superiority and applicability of DEFES are shown for International Business Machines Corporation and compared the outcome with the results of the other methods. Results with MAPE metric and Wilcoxon signed ranks test indicate that DEFES provides more accuracy and outperforms all previous methods, so it can be considered as a superior tool for stock price forecasting problems.

Creating a test blueprint for a progress testing program: A paired-comparisons approach.

PubMed

von Bergmann, HsingChi; Childs, Ruth A

2018-03-01

Creating a new testing program requires the development of a test blueprint that will determine how the items on each test form are distributed across possible content areas and practice domains. To achieve validity, categories of a blueprint are typically based on the judgments of content experts. How experts judgments are elicited and combined is important to the quality of resulting test blueprints. Content experts in dentistry participated in a day-long faculty-wide workshop to discuss, refine, and confirm the categories and their relative weights. After reaching agreement on categories and their definitions, experts judged the relative importance between category pairs, registering their judgments anonymously using iClicker, an audience response system. Judgments were combined in two ways: a simple calculation that could be performed during the workshop and a multidimensional scaling of the judgments performed later. Content experts were able to produce a set of relative weights using this approach. The multidimensional scaling yielded a three-dimensional model with the potential to provide deeper insights into the basis of the experts' judgments. The approach developed and demonstrated in this study can be applied across academic disciplines to elicit and combine content experts judgments for the development of test blueprints.

An ensemble boosting model for predicting transfer to the pediatric intensive care unit.

PubMed

Rubin, Jonathan; Potes, Cristhian; Xu-Wilson, Minnan; Dong, Junzi; Rahman, Asif; Nguyen, Hiep; Moromisato, David

2018-04-01

Early deterioration indicators have the potential to alert hospital care staff in advance of adverse events, such as patients requiring an increased level of care, or the need for rapid response teams to be called. Our work focuses on the problem of predicting the transfer of pediatric patients from the general ward of a hospital to the pediatric intensive care unit. The development of a data-driven pediatric early deterioration indicator for use by clinicians with the purpose of predicting encounters where transfer from the general ward to the PICU is likely. Using data collected over 5.5 years from the electronic health records of two medical facilities, we develop machine learning classifiers based on adaptive boosting and gradient tree boosting. We further combine these learned classifiers into an ensemble model and compare its performance to a modified pediatric early warning score (PEWS) baseline that relies on expert defined guidelines. To gauge model generalizability, we perform an inter-facility evaluation where we train our algorithm on data from one facility and perform evaluation on a hidden test dataset from a separate facility. We show that improvements are witnessed over the modified PEWS baseline in accuracy (0.77 vs. 0.69), sensitivity (0.80 vs. 0.68), specificity (0.74 vs. 0.70) and AUROC (0.85 vs. 0.73). Data-driven, machine learning algorithms can improve PICU transfer prediction accuracy compared to expertly defined systems, such as a modified PEWS, but care must be taken in the training of such approaches to avoid inadvertently introducing bias into the outcomes of these systems. Copyright © 2018 Elsevier B.V. All rights reserved.

Classification of fMRI independent components using IC-fingerprints and support vector machine classifiers.

PubMed

De Martino, Federico; Gentile, Francesco; Esposito, Fabrizio; Balsi, Marco; Di Salle, Francesco; Goebel, Rainer; Formisano, Elia

2007-01-01

We present a general method for the classification of independent components (ICs) extracted from functional MRI (fMRI) data sets. The method consists of two steps. In the first step, each fMRI-IC is associated with an IC-fingerprint, i.e., a representation of the component in a multidimensional space of parameters. These parameters are post hoc estimates of global properties of the ICs and are largely independent of a specific experimental design and stimulus timing. In the second step a machine learning algorithm automatically separates the IC-fingerprints into six general classes after preliminary training performed on a small subset of expert-labeled components. We illustrate this approach in a multisubject fMRI study employing visual structure-from-motion stimuli encoding faces and control random shapes. We show that: (1) IC-fingerprints are a valuable tool for the inspection, characterization and selection of fMRI-ICs and (2) automatic classifications of fMRI-ICs in new subjects present a high correspondence with those obtained by expert visual inspection of the components. Importantly, our classification procedure highlights several neurophysiologically interesting processes. The most intriguing of which is reflected, with high intra- and inter-subject reproducibility, in one IC exhibiting a transiently task-related activation in the 'face' region of the primary sensorimotor cortex. This suggests that in addition to or as part of the mirror system, somatotopic regions of the sensorimotor cortex are involved in disambiguating the perception of a moving body part. Finally, we show that the same classification algorithm can be successfully applied, without re-training, to fMRI collected using acquisition parameters, stimulation modality and timing considerably different from those used for training.

Planet Hunters: Kepler by Eye

NASA Astrophysics Data System (ADS)

Schwamb, Megan E.; Lintott, C.; Fischer, D.; Smith, A. M.; Boyajian, T. S.; Brewer, J. M.; Giguere, M. J.; Lynn, S.; Parrish, M.; Schawinski, K.; Schmitt, J.; Simpson, R.; Wang, J.

2014-01-01

Planet Hunters (http://www.planethunters.org), part of the Zooniverse's (http://www.zooniverse.org) collection of online citizen science projects, uses the World Wide Web to enlist the general public to identify transits in the pubic Kepler light curves. Planet Hunters utilizes human pattern recognition to identify planet transits that may be missed by automated detection algorithms looking for periodic events. Referred to as ‘crowdsourcing’ or ‘citizen science’, the combined assessment of many non-expert human classifiers with minimal training can often equal or best that of a trained expert and in many cases outperform the best machine-learning algorithm. Visitors to the Planet Hunters' website are presented with a randomly selected ~30-day light curve segment from one of Kepler’s ~160,000 target stars and are asked to draw boxes to mark the locations of visible transits in the web interface. 5-10 classifiers review each 30-day light curve segment. Since December 2010, more than 260,000 volunteers world wide have participated, contributing over 20 million classifications. We have demonstrated the success of a citizen science approach with the project’s more than 20 planet candidates, the discovery of PH1b, a transiting circumbinary planet in a quadruple star system, and the discovery of PH2-b, a confirmed Jupiter-sized planet in the habitable zone of a Sun-like star. I will provide an overview of Planet Hunters, highlighting several of project's most recent exoplanet and astrophysical discoveries. Acknowledgements: MES was supported in part by a NSF AAPF under award AST-1003258 and a American Philosophical Society Franklin Grant. We acknowledge support from NASA ADAP12-0172 grant to PI Fischer.

Recommending Education Materials for Diabetic Questions Using Information Retrieval Approaches

PubMed Central

Wang, Yanshan; Shen, Feichen; Liu, Sijia; Rastegar-Mojarad, Majid; Wang, Liwei

2017-01-01

Background Self-management is crucial to diabetes care and providing expert-vetted content for answering patients’ questions is crucial in facilitating patient self-management. Objective The aim is to investigate the use of information retrieval techniques in recommending patient education materials for diabetic questions of patients. Methods We compared two retrieval algorithms, one based on Latent Dirichlet Allocation topic modeling (topic modeling-based model) and one based on semantic group (semantic group-based model), with the baseline retrieval models, vector space model (VSM), in recommending diabetic patient education materials to diabetic questions posted on the TuDiabetes forum. The evaluation was based on a gold standard dataset consisting of 50 randomly selected diabetic questions where the relevancy of diabetic education materials to the questions was manually assigned by two experts. The performance was assessed using precision of top-ranked documents. Results We retrieved 7510 diabetic questions on the forum and 144 diabetic patient educational materials from the patient education database at Mayo Clinic. The mapping rate of words in each corpus mapped to the Unified Medical Language System (UMLS) was significantly different (P<.001). The topic modeling-based model outperformed the other retrieval algorithms. For example, for the top-retrieved document, the precision of the topic modeling-based, semantic group-based, and VSM models was 67.0%, 62.8%, and 54.3%, respectively. Conclusions This study demonstrated that topic modeling can mitigate the vocabulary difference and it achieved the best performance in recommending education materials for answering patients’ questions. One direction for future work is to assess the generalizability of our findings and to extend our study to other disease areas, other patient education material resources, and online forums. PMID:29038097

Esophageal achalasia: current diagnosis and treatment.

PubMed

Schlottmann, Francisco; Patti, Marco G

2018-05-27

Esophageal achalasia is a primary esophageal motility disorder of unknown origin, characterized by lack of peristalsis and by incomplete or absent relaxation of the lower esophageal sphincter in response to swallowing. The goal of treatment is to eliminate the functional obstruction at the level of the gastroesophageal junction Areas covered: This comprehensive review will evaluate the current literature, illustrating the diagnostic evaluation and providing an evidence-based treatment algorithm for this disease Expert commentary: Today we have three very effective therapeutic modalities to treat patients with achalasia - pneumatic dilatation, per-oral endoscopic myotomy and laparoscopic Heller myotomy with fundoplication. Treatment should be tailored to the individual patient, in centers where a multidisciplinary approach is available. Esophageal resection should be considered as a last resort for patients who have failed prior therapeutic attempts.

Segmentation of mosaicism in cervicographic images using support vector machines

NASA Astrophysics Data System (ADS)

Xue, Zhiyun; Long, L. Rodney; Antani, Sameer; Jeronimo, Jose; Thoma, George R.

2009-02-01

The National Library of Medicine (NLM), in collaboration with the National Cancer Institute (NCI), is creating a large digital repository of cervicographic images for the study of uterine cervix cancer prevention. One of the research goals is to automatically detect diagnostic bio-markers in these images. Reliable bio-marker segmentation in large biomedical image collections is a challenging task due to the large variation in image appearance. Methods described in this paper focus on segmenting mosaicism, which is an important vascular feature used to visually assess the degree of cervical intraepithelial neoplasia. The proposed approach uses support vector machines (SVM) trained on a ground truth dataset annotated by medical experts (which circumvents the need for vascular structure extraction). We have evaluated the performance of the proposed algorithm and experimentally demonstrated its feasibility.

An expert system for the evaluation of reinforced concrete structure durability

DOE Office of Scientific and Technical Information (OSTI.GOV)

Berra, M.; Bertolini, L.; Briglia, M.C.

1999-11-01

A user-friendly expert system has been developed to evaluate primarily the durability of reinforced concrete structures, either in the design phase or during service life related to reinforcement corrosion. Besides the durability module, the ES has been provided with three other expert modules in order to support the user during the following activities: inspections, corrosion diagnosis and repair strategy (of concrete and reinforcement). Corrosion induced by carbonation and chlorides penetration and caused by concrete degradation such as sulfate attack, freeze/thaw cycles, alkali silica reaction are considered. The knowledge used for the expert system is based both on open literature andmore » international standards as well as on specific experiences and proprietary databases. The paper describes main features of the system, including the modeling of the knowledge, input data, the algorithms, the rules and the outputs for each module.« less

Automated wholeslide analysis of multiplex-brightfield IHC images for cancer cells and carcinoma-associated fibroblasts

NASA Astrophysics Data System (ADS)

Lorsakul, Auranuch; Andersson, Emilia; Vega Harring, Suzana; Sade, Hadassah; Grimm, Oliver; Bredno, Joerg

2017-03-01

Multiplex-brightfield immunohistochemistry (IHC) staining and quantitative measurement of multiple biomarkers can support therapeutic targeting of carcinoma-associated fibroblasts (CAF). This paper presents an automated digitalpathology solution to simultaneously analyze multiple biomarker expressions within a single tissue section stained with an IHC duplex assay. Our method was verified against ground truth provided by expert pathologists. In the first stage, the automated method quantified epithelial-carcinoma cells expressing cytokeratin (CK) using robust nucleus detection and supervised cell-by-cell classification algorithms with a combination of nucleus and contextual features. Using fibroblast activation protein (FAP) as biomarker for CAFs, the algorithm was trained, based on ground truth obtained from pathologists, to automatically identify tumor-associated stroma using a supervised-generation rule. The algorithm reported distance to nearest neighbor in the populations of tumor cells and activated-stromal fibroblasts as a wholeslide measure of spatial relationships. A total of 45 slides from six indications (breast, pancreatic, colorectal, lung, ovarian, and head-and-neck cancers) were included for training and verification. CK-positive cells detected by the algorithm were verified by a pathologist with good agreement (R2=0.98) to ground-truth count. For the area occupied by FAP-positive cells, the inter-observer agreement between two sets of ground-truth measurements was R2=0.93 whereas the algorithm reproduced the pathologists' areas with R2=0.96. The proposed methodology enables automated image analysis to measure spatial relationships of cells stained in an IHC-multiplex assay. Our proof-of-concept results show an automated algorithm can be trained to reproduce the expert assessment and provide quantitative readouts that potentially support a cutoff determination in hypothesis testing related to CAF-targeting-therapy decisions.

Integration of QSAR and SAR methods for the mechanistic interpretation of predictive models for carcinogenicity

PubMed Central

Fjodorova, Natalja; Novič, Marjana

2012-01-01

The knowledge-based Toxtree expert system (SAR approach) was integrated with the statistically based counter propagation artificial neural network (CP ANN) model (QSAR approach) to contribute to a better mechanistic understanding of a carcinogenicity model for non-congeneric chemicals using Dragon descriptors and carcinogenic potency for rats as a response. The transparency of the CP ANN algorithm was demonstrated using intrinsic mapping technique specifically Kohonen maps. Chemical structures were represented by Dragon descriptors that express the structural and electronic features of molecules such as their shape and electronic surrounding related to reactivity of molecules. It was illustrated how the descriptors are correlated with particular structural alerts (SAs) for carcinogenicity with recognized mechanistic link to carcinogenic activity. Moreover, the Kohonen mapping technique enables one to examine the separation of carcinogens and non-carcinogens (for rats) within a family of chemicals with a particular SA for carcinogenicity. The mechanistic interpretation of models is important for the evaluation of safety of chemicals. PMID:24688639

Automation of data processing and calculation of retention parameters and thermodynamic data for gas chromatography

NASA Astrophysics Data System (ADS)

Makarycheva, A. I.; Faerman, V. A.

2017-02-01

The analyses of automation patterns is performed and the programming solution for the automation of data processing of the chromatographic data and their further information storage with a help of a software package, Mathcad and MS Excel spreadsheets, is developed. The offered approach concedes the ability of data processing algorithm modification and does not require any programming experts participation. The approach provides making a measurement of the given time and retention volumes, specific retention volumes, a measurement of differential molar free adsorption energy, and a measurement of partial molar solution enthalpies and isosteric heats of adsorption. The developed solution is focused on the appliance in a small research group and is tested on the series of some new gas chromatography sorbents. More than 20 analytes were submitted to calculation of retention parameters and thermodynamic sorption quantities. The received data are provided in the form accessible to comparative analysis, and they are able to find sorbing agents with the most profitable properties to solve some concrete analytic issues.

Automatic Identification of Critical Follow-Up Recommendation Sentences in Radiology Reports

PubMed Central

Yetisgen-Yildiz, Meliha; Gunn, Martin L.; Xia, Fei; Payne, Thomas H.

2011-01-01

Communication of follow-up recommendations when abnormalities are identified on imaging studies is prone to error. When recommendations are not systematically identified and promptly communicated to referrers, poor patient outcomes can result. Using information technology can improve communication and improve patient safety. In this paper, we describe a text processing approach that uses natural language processing (NLP) and supervised text classification methods to automatically identify critical recommendation sentences in radiology reports. To increase the classification performance we enhanced the simple unigram token representation approach with lexical, semantic, knowledge-base, and structural features. We tested different combinations of those features with the Maximum Entropy (MaxEnt) classification algorithm. Classifiers were trained and tested with a gold standard corpus annotated by a domain expert. We applied 5-fold cross validation and our best performing classifier achieved 95.60% precision, 79.82% recall, 87.0% F-score, and 99.59% classification accuracy in identifying the critical recommendation sentences in radiology reports. PMID:22195225

Automatic identification of critical follow-up recommendation sentences in radiology reports.

PubMed

Yetisgen-Yildiz, Meliha; Gunn, Martin L; Xia, Fei; Payne, Thomas H

2011-01-01

Communication of follow-up recommendations when abnormalities are identified on imaging studies is prone to error. When recommendations are not systematically identified and promptly communicated to referrers, poor patient outcomes can result. Using information technology can improve communication and improve patient safety. In this paper, we describe a text processing approach that uses natural language processing (NLP) and supervised text classification methods to automatically identify critical recommendation sentences in radiology reports. To increase the classification performance we enhanced the simple unigram token representation approach with lexical, semantic, knowledge-base, and structural features. We tested different combinations of those features with the Maximum Entropy (MaxEnt) classification algorithm. Classifiers were trained and tested with a gold standard corpus annotated by a domain expert. We applied 5-fold cross validation and our best performing classifier achieved 95.60% precision, 79.82% recall, 87.0% F-score, and 99.59% classification accuracy in identifying the critical recommendation sentences in radiology reports.

A New Data Mining Scheme Using Artificial Neural Networks

PubMed Central

Kamruzzaman, S. M.; Jehad Sarkar, A. M.

2011-01-01

Classification is one of the data mining problems receiving enormous attention in the database community. Although artificial neural networks (ANNs) have been successfully applied in a wide range of machine learning applications, they are however often regarded as black boxes, i.e., their predictions cannot be explained. To enhance the explanation of ANNs, a novel algorithm to extract symbolic rules from ANNs has been proposed in this paper. ANN methods have not been effectively utilized for data mining tasks because how the classifications were made is not explicitly stated as symbolic rules that are suitable for verification or interpretation by human experts. With the proposed approach, concise symbolic rules with high accuracy, that are easily explainable, can be extracted from the trained ANNs. Extracted rules are comparable with other methods in terms of number of rules, average number of conditions for a rule, and the accuracy. The effectiveness of the proposed approach is clearly demonstrated by the experimental results on a set of benchmark data mining classification problems. PMID:22163866

PEGASO: A Personalised and Motivational ICT System to Empower Adolescents Towards Healthy Lifestyles.

PubMed

Carrino, Stefano; Caon, Maurizio; Angelini, Leonardo; Mugellini, Elena; Abou Khaled, Omar; Orte, Silvia; Vargiu, Eloisa; Coulson, Neil; Serrano, José C E; Tabozzi, Sarah; Lafortuna, Claudio; Rizzo, Giovanna

2014-01-01

Unhealthy alimentary behaviours and physical inactivity habits are key risk factors for major non communicable diseases. Several researches demonstrate that juvenile obesity can lead to serious medical conditions, pathologies and have important psycho-social consequences. PEGASO is a multidisciplinary project aimed at promoting healthy lifestyles among teenagers through assistive technology. The core of this project is represented by the ICT system, which allows providing tailored interventions to the users through their smartphones in order to motivate them. The novelty of this approach consists of developing a Virtual Individual Model (VIM) for user characterization, which is based on physical, functional and behavioural parameters opportunely selected by experts. These parameters are digitised and updated thanks to the user monitoring through smartphone; data mining algorithms are applied for the detection of activity and nutrition habits and this information is used to provide personalised feedback. The user interface will be developed using gamified approaches and integrating serious games to effectively promote health literacy and facilitate behaviour change.

An artificial intelligence approach to lithostratigraphic correlation using geophysical well logs

DOE Office of Scientific and Technical Information (OSTI.GOV)

Olea, R.A.; Davis, J.C.

1986-01-01

Computer programs for lithostratigraphic correlation of well logs have achieved limited success. Their algorithms are based on an oversimplified view of the manual process used by analysts to establish geologically correct correlations. The programs experience difficulties if the correlated rocks deviate from an ideal geometry of perfectly homogeneous, parallel layers of infinite extent. Artificial intelligence provides a conceptual basis for formulating the task of lithostratigraphic correlation, leading to more realistic procedures. A prototype system using the ''production rule'' approach of expert systems successfully correlates well logs in areas of stratigraphic complexity. Two digitized logs are used per well, one formore » curve matching and the other for lithologic comparison. The software has been successfully used to correlate more than 100,000 ft (30 480 m) of section, through clastic sequences in Louisiana and through carbonate sequences in Kansas. Correlations have been achieved even in the presence of faults, unconformities, facies changes, and lateral variations in bed thickness.« less

Identifying Medical Diagnoses and Treatable Diseases by Image-Based Deep Learning.

PubMed

Kermany, Daniel S; Goldbaum, Michael; Cai, Wenjia; Valentim, Carolina C S; Liang, Huiying; Baxter, Sally L; McKeown, Alex; Yang, Ge; Wu, Xiaokang; Yan, Fangbing; Dong, Justin; Prasadha, Made K; Pei, Jacqueline; Ting, Magdalene Y L; Zhu, Jie; Li, Christina; Hewett, Sierra; Dong, Jason; Ziyar, Ian; Shi, Alexander; Zhang, Runze; Zheng, Lianghong; Hou, Rui; Shi, William; Fu, Xin; Duan, Yaou; Huu, Viet A N; Wen, Cindy; Zhang, Edward D; Zhang, Charlotte L; Li, Oulan; Wang, Xiaobo; Singer, Michael A; Sun, Xiaodong; Xu, Jie; Tafreshi, Ali; Lewis, M Anthony; Xia, Huimin; Zhang, Kang

2018-02-22

The implementation of clinical-decision support algorithms for medical imaging faces challenges with reliability and interpretability. Here, we establish a diagnostic tool based on a deep-learning framework for the screening of patients with common treatable blinding retinal diseases. Our framework utilizes transfer learning, which trains a neural network with a fraction of the data of conventional approaches. Applying this approach to a dataset of optical coherence tomography images, we demonstrate performance comparable to that of human experts in classifying age-related macular degeneration and diabetic macular edema. We also provide a more transparent and interpretable diagnosis by highlighting the regions recognized by the neural network. We further demonstrate the general applicability of our AI system for diagnosis of pediatric pneumonia using chest X-ray images. This tool may ultimately aid in expediting the diagnosis and referral of these treatable conditions, thereby facilitating earlier treatment, resulting in improved clinical outcomes. VIDEO ABSTRACT. Copyright © 2018 Elsevier Inc. All rights reserved.

Classifying smoking urges via machine learning.

PubMed

Dumortier, Antoine; Beckjord, Ellen; Shiffman, Saul; Sejdić, Ervin

2016-12-01

Smoking is the largest preventable cause of death and diseases in the developed world, and advances in modern electronics and machine learning can help us deliver real-time intervention to smokers in novel ways. In this paper, we examine different machine learning approaches to use situational features associated with having or not having urges to smoke during a quit attempt in order to accurately classify high-urge states. To test our machine learning approaches, specifically, Bayes, discriminant analysis and decision tree learning methods, we used a dataset collected from over 300 participants who had initiated a quit attempt. The three classification approaches are evaluated observing sensitivity, specificity, accuracy and precision. The outcome of the analysis showed that algorithms based on feature selection make it possible to obtain high classification rates with only a few features selected from the entire dataset. The classification tree method outperformed the naive Bayes and discriminant analysis methods, with an accuracy of the classifications up to 86%. These numbers suggest that machine learning may be a suitable approach to deal with smoking cessation matters, and to predict smoking urges, outlining a potential use for mobile health applications. In conclusion, machine learning classifiers can help identify smoking situations, and the search for the best features and classifier parameters significantly improves the algorithms' performance. In addition, this study also supports the usefulness of new technologies in improving the effect of smoking cessation interventions, the management of time and patients by therapists, and thus the optimization of available health care resources. Future studies should focus on providing more adaptive and personalized support to people who really need it, in a minimum amount of time by developing novel expert systems capable of delivering real-time interventions. Copyright © 2016 Elsevier Ireland Ltd. All rights reserved.

A semiautomatic segmentation method for prostate in CT images using local texture classification and statistical shape modeling.

PubMed

Shahedi, Maysam; Halicek, Martin; Guo, Rongrong; Zhang, Guoyi; Schuster, David M; Fei, Baowei

2018-06-01

Prostate segmentation in computed tomography (CT) images is useful for treatment planning and procedure guidance such as external beam radiotherapy and brachytherapy. However, because of the low, soft tissue contrast of CT images, manual segmentation of the prostate is a time-consuming task with high interobserver variation. In this study, we proposed a semiautomated, three-dimensional (3D) segmentation for prostate CT images using shape and texture analysis and we evaluated the method against manual reference segmentations. The prostate gland usually has a globular shape with a smoothly curved surface, and its shape could be accurately modeled or reconstructed having a limited number of well-distributed surface points. In a training dataset, using the prostate gland centroid point as the origin of a coordination system, we defined an intersubject correspondence between the prostate surface points based on the spherical coordinates. We applied this correspondence to generate a point distribution model for prostate shape using principal component analysis and to study the local texture difference between prostate and nonprostate tissue close to the different prostate surface subregions. We used the learned shape and texture characteristics of the prostate in CT images and then combined them with user inputs to segment a new image. We trained our segmentation algorithm using 23 CT images and tested the algorithm on two sets of 10 nonbrachytherapy and 37 postlow dose rate brachytherapy CT images. We used a set of error metrics to evaluate the segmentation results using two experts' manual reference segmentations. For both nonbrachytherapy and post-brachytherapy image sets, the average measured Dice similarity coefficient (DSC) was 88% and the average mean absolute distance (MAD) was 1.9 mm. The average measured differences between the two experts on both datasets were 92% (DSC) and 1.1 mm (MAD). The proposed, semiautomatic segmentation algorithm showed a fast, robust, and accurate performance for 3D prostate segmentation of CT images, specifically when no previous, intrapatient information, that is, previously segmented images, was available. The accuracy of the algorithm is comparable to the best performance results reported in the literature and approaches the interexpert variability observed in manual segmentation. © 2018 American Association of Physicists in Medicine.

DEVELOPMENT AND TESTING OF FAULT-DIAGNOSIS ALGORITHMS FOR REACTOR PLANT SYSTEMS

DOE Office of Scientific and Technical Information (OSTI.GOV)

Grelle, Austin L.; Park, Young S.; Vilim, Richard B.

Argonne National Laboratory is further developing fault diagnosis algorithms for use by the operator of a nuclear plant to aid in improved monitoring of overall plant condition and performance. The objective is better management of plant upsets through more timely, informed decisions on control actions with the ultimate goal of improved plant safety, production, and cost management. Integration of these algorithms with visual aids for operators is taking place through a collaboration under the concept of an operator advisory system. This is a software entity whose purpose is to manage and distill the enormous amount of information an operator mustmore » process to understand the plant state, particularly in off-normal situations, and how the state trajectory will unfold in time. The fault diagnosis algorithms were exhaustively tested using computer simulations of twenty different faults introduced into the chemical and volume control system (CVCS) of a pressurized water reactor (PWR). The algorithms are unique in that each new application to a facility requires providing only the piping and instrumentation diagram (PID) and no other plant-specific information; a subject-matter expert is not needed to install and maintain each instance of an application. The testing approach followed accepted procedures for verifying and validating software. It was shown that the code satisfies its functional requirement which is to accept sensor information, identify process variable trends based on this sensor information, and then to return an accurate diagnosis based on chains of rules related to these trends. The validation and verification exercise made use of GPASS, a one-dimensional systems code, for simulating CVCS operation. Plant components were failed and the code generated the resulting plant response. Parametric studies with respect to the severity of the fault, the richness of the plant sensor set, and the accuracy of sensors were performed as part of the validation exercise. The background and overview of the software will be presented to give an overview of the approach. Following, the verification and validation effort using the GPASS code for simulation of plant transients including a sensitivity study on important parameters will be presented« less

Strategies for adding adaptive learning mechanisms to rule-based diagnostic expert systems

NASA Technical Reports Server (NTRS)

Stclair, D. C.; Sabharwal, C. L.; Bond, W. E.; Hacke, Keith

1988-01-01

Rule-based diagnostic expert systems can be used to perform many of the diagnostic chores necessary in today's complex space systems. These expert systems typically take a set of symptoms as input and produce diagnostic advice as output. The primary objective of such expert systems is to provide accurate and comprehensive advice which can be used to help return the space system in question to nominal operation. The development and maintenance of diagnostic expert systems is time and labor intensive since the services of both knowledge engineer(s) and domain expert(s) are required. The use of adaptive learning mechanisms to increment evaluate and refine rules promises to reduce both time and labor costs associated with such systems. This paper describes the basic adaptive learning mechanisms of strengthening, weakening, generalization, discrimination, and discovery. Next basic strategies are discussed for adding these learning mechanisms to rule-based diagnostic expert systems. These strategies support the incremental evaluation and refinement of rules in the knowledge base by comparing the set of advice given by the expert system (A) with the correct diagnosis (C). Techniques are described for selecting those rules in the in the knowledge base which should participate in adaptive learning. The strategies presented may be used with a wide variety of learning algorithms. Further, these strategies are applicable to a large number of rule-based diagnostic expert systems. They may be used to provide either immediate or deferred updating of the knowledge base.

Polymeric endovascular strut and lumen detection algorithm for intracoronary optical coherence tomography images

NASA Astrophysics Data System (ADS)

Amrute, Junedh M.; Athanasiou, Lambros S.; Rikhtegar, Farhad; de la Torre Hernández, José M.; Camarero, Tamara García; Edelman, Elazer R.

2018-03-01

Polymeric endovascular implants are the next step in minimally invasive vascular interventions. As an alternative to traditional metallic drug-eluting stents, these often-erodible scaffolds present opportunities and challenges for patients and clinicians. Theoretically, as they resorb and are absorbed over time, they obviate the long-term complications of permanent implants, but in the short-term visualization and therefore positioning is problematic. Polymeric scaffolds can only be fully imaged using optical coherence tomography (OCT) imaging-they are relatively invisible via angiography-and segmentation of polymeric struts in OCT images is performed manually, a laborious and intractable procedure for large datasets. Traditional lumen detection methods using implant struts as boundary limits fail in images with polymeric implants. Therefore, it is necessary to develop an automated method to detect polymeric struts and luminal borders in OCT images; we present such a fully automated algorithm. Accuracy was validated using expert annotations on 1140 OCT images with a positive predictive value of 0.93 for strut detection and an R2 correlation coefficient of 0.94 between detected and expert-annotated lumen areas. The proposed algorithm allows for rapid, accurate, and automated detection of polymeric struts and the luminal border in OCT images.

A novel time-domain signal processing algorithm for real time ventricular fibrillation detection

NASA Astrophysics Data System (ADS)

Monte, G. E.; Scarone, N. C.; Liscovsky, P. O.; Rotter S/N, P.

2011-12-01

This paper presents an application of a novel algorithm for real time detection of ECG pathologies, especially ventricular fibrillation. It is based on segmentation and labeling process of an oversampled signal. After this treatment, analyzing sequence of segments, global signal behaviours are obtained in the same way like a human being does. The entire process can be seen as a morphological filtering after a smart data sampling. The algorithm does not require any ECG digital signal pre-processing, and the computational cost is low, so it can be embedded into the sensors for wearable and permanent applications. The proposed algorithms could be the input signal description to expert systems or to artificial intelligence software in order to detect other pathologies.

Towards a Fuzzy Expert System on Toxicological Data Quality Assessment.

PubMed

Yang, Longzhi; Neagu, Daniel; Cronin, Mark T D; Hewitt, Mark; Enoch, Steven J; Madden, Judith C; Przybylak, Katarzyna

2013-01-01

Quality assessment (QA) requires high levels of domain-specific experience and knowledge. QA tasks for toxicological data are usually performed by human experts manually, although a number of quality evaluation schemes have been proposed in the literature. For instance, the most widely utilised Klimisch scheme1 defines four data quality categories in order to tag data instances with respect to their qualities; ToxRTool2 is an extension of the Klimisch approach aiming to increase the transparency and harmonisation of the approach. Note that the processes of QA in many other areas have been automatised by employing expert systems. Briefly, an expert system is a computer program that uses a knowledge base built upon human expertise, and an inference engine that mimics the reasoning processes of human experts to infer new statements from incoming data. In particular, expert systems have been extended to deal with the uncertainty of information by representing uncertain information (such as linguistic terms) as fuzzy sets under the framework of fuzzy set theory and performing inferences upon fuzzy sets according to fuzzy arithmetic. This paper presents an experimental fuzzy expert system for toxicological data QA which is developed on the basis of the Klimisch approach and the ToxRTool in an effort to illustrate the power of expert systems to toxicologists, and to examine if fuzzy expert systems are a viable solution for QA of toxicological data. Such direction still faces great difficulties due to the well-known common challenge of toxicological data QA that "five toxicologists may have six opinions". In the meantime, this challenge may offer an opportunity for expert systems because the construction and refinement of the knowledge base could be a converging process of different opinions which is of significant importance for regulatory policy making under the regulation of REACH, though a consensus may never be reached. Also, in order to facilitate the implementation of Weight of Evidence approaches and in silico modelling proposed by REACH, there is a higher appeal of numerical quality values than nominal (categorical) ones, where the proposed fuzzy expert system could help. Most importantly, the deriving processes of quality values generated in this way are fully transparent, and thus comprehensible, for final users, which is another vital point for policy making specified in REACH. Case studies have been conducted and this report not only shows the promise of the approach, but also demonstrates the difficulties of the approach and thus indicates areas for future development. Copyright © 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

A domain specific language for performance portable molecular dynamics algorithms

NASA Astrophysics Data System (ADS)

Saunders, William Robert; Grant, James; Müller, Eike Hermann

2018-03-01

Developers of Molecular Dynamics (MD) codes face significant challenges when adapting existing simulation packages to new hardware. In a continuously diversifying hardware landscape it becomes increasingly difficult for scientists to be experts both in their own domain (physics/chemistry/biology) and specialists in the low level parallelisation and optimisation of their codes. To address this challenge, we describe a "Separation of Concerns" approach for the development of parallel and optimised MD codes: the science specialist writes code at a high abstraction level in a domain specific language (DSL), which is then translated into efficient computer code by a scientific programmer. In a related context, an abstraction for the solution of partial differential equations with grid based methods has recently been implemented in the (Py)OP2 library. Inspired by this approach, we develop a Python code generation system for molecular dynamics simulations on different parallel architectures, including massively parallel distributed memory systems and GPUs. We demonstrate the efficiency of the auto-generated code by studying its performance and scalability on different hardware and compare it to other state-of-the-art simulation packages. With growing data volumes the extraction of physically meaningful information from the simulation becomes increasingly challenging and requires equally efficient implementations. A particular advantage of our approach is the easy expression of such analysis algorithms. We consider two popular methods for deducing the crystalline structure of a material from the local environment of each atom, show how they can be expressed in our abstraction and implement them in the code generation framework.

Automatic segmentation for detecting uterine fibroid regions treated with MR-guided high intensity focused ultrasound (MR-HIFU).

PubMed

Antila, Kari; Nieminen, Heikki J; Sequeiros, Roberto Blanco; Ehnholm, Gösta

2014-07-01

Up to 25% of women suffer from uterine fibroids (UF) that cause infertility, pain, and discomfort. MR-guided high intensity focused ultrasound (MR-HIFU) is an emerging technique for noninvasive, computer-guided thermal ablation of UFs. The volume of induced necrosis is a predictor of the success of the treatment. However, accurate volume assessment by hand can be time consuming, and quick tools produce biased results. Therefore, fast and reliable tools are required in order to estimate the technical treatment outcome during the therapy event so as to predict symptom relief. A novel technique has been developed for the segmentation and volume assessment of the treated region. Conventional algorithms typically require user interaction ora priori knowledge of the target. The developed algorithm exploits the treatment plan, the coordinates of the intended ablation, for fully automatic segmentation with no user input. A good similarity to an expert-segmented manual reference was achieved (Dice similarity coefficient = 0.880 ± 0.074). The average automatic segmentation time was 1.6 ± 0.7 min per patient against an order of tens of minutes when done manually. The results suggest that the segmentation algorithm developed, requiring no user-input, provides a feasible and practical approach for the automatic evaluation of the boundary and volume of the HIFU-treated region.

Automated circumferential construction of first-order aqueous humor outflow pathways using spectral-domain optical coherence tomography.

PubMed

Huang, Alex S; Belghith, Akram; Dastiridou, Anna; Chopra, Vikas; Zangwill, Linda M; Weinreb, Robert N

2017-06-01

The purpose was to create a three-dimensional (3-D) model of circumferential aqueous humor outflow (AHO) in a living human eye with an automated detection algorithm for Schlemm’s canal (SC) and first-order collector channels (CC) applied to spectral-domain optical coherence tomography (SD-OCT). Anterior segment SD-OCT scans from a subject were acquired circumferentially around the limbus. A Bayesian Ridge method was used to approximate the location of the SC on infrared confocal laser scanning ophthalmoscopic images with a cross multiplication tool developed to initiate SC/CC detection automated through a fuzzy hidden Markov Chain approach. Automatic segmentation of SC and initial CC’s was manually confirmed by two masked graders. Outflow pathways detected by the segmentation algorithm were reconstructed into a 3-D representation of AHO. Overall, only <1% of images (5114 total B-scans) were ungradable. Automatic segmentation algorithm performed well with SC detection 98.3% of the time and <0.1% false positive detection compared to expert grader consensus. CC was detected 84.2% of the time with 1.4% false positive detection. 3-D representation of AHO pathways demonstrated variably thicker and thinner SC with some clear CC roots. Circumferential (360 deg), automated, and validated AHO detection of angle structures in the living human eye with reconstruction was possible.

Collaborative autonomous sensing with Bayesians in the loop

NASA Astrophysics Data System (ADS)

Ahmed, Nisar

2016-10-01

There is a strong push to develop intelligent unmanned autonomy that complements human reasoning for applications as diverse as wilderness search and rescue, military surveillance, and robotic space exploration. More than just replacing humans for `dull, dirty and dangerous' work, autonomous agents are expected to cope with a whole host of uncertainties while working closely together with humans in new situations. The robotics revolution firmly established the primacy of Bayesian algorithms for tackling challenging perception, learning and decision-making problems. Since the next frontier of autonomy demands the ability to gather information across stretches of time and space that are beyond the reach of a single autonomous agent, the next generation of Bayesian algorithms must capitalize on opportunities to draw upon the sensing and perception abilities of humans-in/on-the-loop. This work summarizes our recent research toward harnessing `human sensors' for information gathering tasks. The basic idea behind is to allow human end users (i.e. non-experts in robotics, statistics, machine learning, etc.) to directly `talk to' the information fusion engine and perceptual processes aboard any autonomous agent. Our approach is grounded in rigorous Bayesian modeling and fusion of flexible semantic information derived from user-friendly interfaces, such as natural language chat and locative hand-drawn sketches. This naturally enables `plug and play' human sensing with existing probabilistic algorithms for planning and perception, and has been successfully demonstrated with human-robot teams in target localization applications.

PhyloPythiaS+: a self-training method for the rapid reconstruction of low-ranking taxonomic bins from metagenomes.

PubMed

Gregor, Ivan; Dröge, Johannes; Schirmer, Melanie; Quince, Christopher; McHardy, Alice C

2016-01-01

Background. Metagenomics is an approach for characterizing environmental microbial communities in situ, it allows their functional and taxonomic characterization and to recover sequences from uncultured taxa. This is often achieved by a combination of sequence assembly and binning, where sequences are grouped into 'bins' representing taxa of the underlying microbial community. Assignment to low-ranking taxonomic bins is an important challenge for binning methods as is scalability to Gb-sized datasets generated with deep sequencing techniques. One of the best available methods for species bins recovery from deep-branching phyla is the expert-trained PhyloPythiaS package, where a human expert decides on the taxa to incorporate in the model and identifies 'training' sequences based on marker genes directly from the sample. Due to the manual effort involved, this approach does not scale to multiple metagenome samples and requires substantial expertise, which researchers who are new to the area do not have. Results. We have developed PhyloPythiaS+, a successor to our PhyloPythia(S) software. The new (+) component performs the work previously done by the human expert. PhyloPythiaS+ also includes a new k-mer counting algorithm, which accelerated the simultaneous counting of 4-6-mers used for taxonomic binning 100-fold and reduced the overall execution time of the software by a factor of three. Our software allows to analyze Gb-sized metagenomes with inexpensive hardware, and to recover species or genera-level bins with low error rates in a fully automated fashion. PhyloPythiaS+ was compared to MEGAN, taxator-tk, Kraken and the generic PhyloPythiaS model. The results showed that PhyloPythiaS+ performs especially well for samples originating from novel environments in comparison to the other methods. Availability. PhyloPythiaS+ in a virtual machine is available for installation under Windows, Unix systems or OS X on: https://github.com/algbioi/ppsp/wiki.

The Development of Expert Learners in the Classroom

ERIC Educational Resources Information Center

Rahman, Saemah; Mahmud, Zuria; Yassin, Siti Fatimah Mohd; Amir, Ruslin; Ilias, Khadijah Wan

2010-01-01

The term "expert learner" refers to students who are actively engaged with the materials learned and take responsibility for their own learning. Literature reviews suggested the use of metacognitive approach to help develop students to become expert learners. Research on development of expert learners can be traced from movements that…

Software for Data Analysis with Graphical Models

NASA Technical Reports Server (NTRS)

Buntine, Wray L.; Roy, H. Scott

1994-01-01

Probabilistic graphical models are being used widely in artificial intelligence and statistics, for instance, in diagnosis and expert systems, as a framework for representing and reasoning with probabilities and independencies. They come with corresponding algorithms for performing statistical inference. This offers a unifying framework for prototyping and/or generating data analysis algorithms from graphical specifications. This paper illustrates the framework with an example and then presents some basic techniques for the task: problem decomposition and the calculation of exact Bayes factors. Other tools already developed, such as automatic differentiation, Gibbs sampling, and use of the EM algorithm, make this a broad basis for the generation of data analysis software.

Comparison of three methods for materials identification and mapping with imaging spectroscopy

NASA Technical Reports Server (NTRS)

Clark, Roger N.; Swayze, Gregg; Boardman, Joe; Kruse, Fred

1993-01-01

We are comparing three methods of mapping analysis tools for imaging spectroscopy data. The purpose of this comparison is to understand the advantages and disadvantages of each algorithm so others would be better able to choose the best algorithm or combinations of algorithms for a particular problem. The three algorithms are: (1) the spectralfeature modified least squares mapping algorithm of Clark et al (1990, 1991): programs mbandmap and tricorder; (2) the Spectral Angle Mapper Algorithm(Boardman, 1993) found in the CU CSES SIPS package; and (3) the Expert System of Kruse et al. (1993). The comparison uses a ground-calibrated 1990 AVIRIS scene of 400 by 410 pixels over Cuprite, Nevada. Along with the test data set is a spectral library of 38 minerals. Each algorithm is tested with the same AVIRIS data set and spectral library. Field work has confirmed the presence of many of these minerals in the AVIRIS scene (Swayze et al. 1992).

Collaborative development of predictive toxicology applications

PubMed Central

2010-01-01

OpenTox provides an interoperable, standards-based Framework for the support of predictive toxicology data management, algorithms, modelling, validation and reporting. It is relevant to satisfying the chemical safety assessment requirements of the REACH legislation as it supports access to experimental data, (Quantitative) Structure-Activity Relationship models, and toxicological information through an integrating platform that adheres to regulatory requirements and OECD validation principles. Initial research defined the essential components of the Framework including the approach to data access, schema and management, use of controlled vocabularies and ontologies, architecture, web service and communications protocols, and selection and integration of algorithms for predictive modelling. OpenTox provides end-user oriented tools to non-computational specialists, risk assessors, and toxicological experts in addition to Application Programming Interfaces (APIs) for developers of new applications. OpenTox actively supports public standards for data representation, interfaces, vocabularies and ontologies, Open Source approaches to core platform components, and community-based collaboration approaches, so as to progress system interoperability goals. The OpenTox Framework includes APIs and services for compounds, datasets, features, algorithms, models, ontologies, tasks, validation, and reporting which may be combined into multiple applications satisfying a variety of different user needs. OpenTox applications are based on a set of distributed, interoperable OpenTox API-compliant REST web services. The OpenTox approach to ontology allows for efficient mapping of complementary data coming from different datasets into a unifying structure having a shared terminology and representation. Two initial OpenTox applications are presented as an illustration of the potential impact of OpenTox for high-quality and consistent structure-activity relationship modelling of REACH-relevant endpoints: ToxPredict which predicts and reports on toxicities for endpoints for an input chemical structure, and ToxCreate which builds and validates a predictive toxicity model based on an input toxicology dataset. Because of the extensible nature of the standardised Framework design, barriers of interoperability between applications and content are removed, as the user may combine data, models and validation from multiple sources in a dependable and time-effective way. PMID:20807436

Collaborative development of predictive toxicology applications.

PubMed

Hardy, Barry; Douglas, Nicki; Helma, Christoph; Rautenberg, Micha; Jeliazkova, Nina; Jeliazkov, Vedrin; Nikolova, Ivelina; Benigni, Romualdo; Tcheremenskaia, Olga; Kramer, Stefan; Girschick, Tobias; Buchwald, Fabian; Wicker, Joerg; Karwath, Andreas; Gütlein, Martin; Maunz, Andreas; Sarimveis, Haralambos; Melagraki, Georgia; Afantitis, Antreas; Sopasakis, Pantelis; Gallagher, David; Poroikov, Vladimir; Filimonov, Dmitry; Zakharov, Alexey; Lagunin, Alexey; Gloriozova, Tatyana; Novikov, Sergey; Skvortsova, Natalia; Druzhilovsky, Dmitry; Chawla, Sunil; Ghosh, Indira; Ray, Surajit; Patel, Hitesh; Escher, Sylvia

2010-08-31

OpenTox provides an interoperable, standards-based Framework for the support of predictive toxicology data management, algorithms, modelling, validation and reporting. It is relevant to satisfying the chemical safety assessment requirements of the REACH legislation as it supports access to experimental data, (Quantitative) Structure-Activity Relationship models, and toxicological information through an integrating platform that adheres to regulatory requirements and OECD validation principles. Initial research defined the essential components of the Framework including the approach to data access, schema and management, use of controlled vocabularies and ontologies, architecture, web service and communications protocols, and selection and integration of algorithms for predictive modelling. OpenTox provides end-user oriented tools to non-computational specialists, risk assessors, and toxicological experts in addition to Application Programming Interfaces (APIs) for developers of new applications. OpenTox actively supports public standards for data representation, interfaces, vocabularies and ontologies, Open Source approaches to core platform components, and community-based collaboration approaches, so as to progress system interoperability goals.The OpenTox Framework includes APIs and services for compounds, datasets, features, algorithms, models, ontologies, tasks, validation, and reporting which may be combined into multiple applications satisfying a variety of different user needs. OpenTox applications are based on a set of distributed, interoperable OpenTox API-compliant REST web services. The OpenTox approach to ontology allows for efficient mapping of complementary data coming from different datasets into a unifying structure having a shared terminology and representation.Two initial OpenTox applications are presented as an illustration of the potential impact of OpenTox for high-quality and consistent structure-activity relationship modelling of REACH-relevant endpoints: ToxPredict which predicts and reports on toxicities for endpoints for an input chemical structure, and ToxCreate which builds and validates a predictive toxicity model based on an input toxicology dataset. Because of the extensible nature of the standardised Framework design, barriers of interoperability between applications and content are removed, as the user may combine data, models and validation from multiple sources in a dependable and time-effective way.

Perceptual expertise in forensic facial image comparison

PubMed Central

White, David; Phillips, P. Jonathon; Hahn, Carina A.; Hill, Matthew; O'Toole, Alice J.

2015-01-01

Forensic facial identification examiners are required to match the identity of faces in images that vary substantially, owing to changes in viewing conditions and in a person's appearance. These identifications affect the course and outcome of criminal investigations and convictions. Despite calls for research on sources of human error in forensic examination, existing scientific knowledge of face matching accuracy is based, almost exclusively, on people without formal training. Here, we administered three challenging face matching tests to a group of forensic examiners with many years' experience of comparing face images for law enforcement and government agencies. Examiners outperformed untrained participants and computer algorithms, thereby providing the first evidence that these examiners are experts at this task. Notably, computationally fusing responses of multiple experts produced near-perfect performance. Results also revealed qualitative differences between expert and non-expert performance. First, examiners' superiority was greatest at longer exposure durations, suggestive of more entailed comparison in forensic examiners. Second, experts were less impaired by image inversion than non-expert students, contrasting with face memory studies that show larger face inversion effects in high performers. We conclude that expertise in matching identity across unfamiliar face images is supported by processes that differ qualitatively from those supporting memory for individual faces. PMID:26336174

Informing Public Perceptions About Climate Change: A 'Mental Models' Approach.

PubMed

Wong-Parodi, Gabrielle; Bruine de Bruin, Wändi

2017-10-01

As the specter of climate change looms on the horizon, people will face complex decisions about whether to support climate change policies and how to cope with climate change impacts on their lives. Without some grasp of the relevant science, they may find it hard to make informed decisions. Climate experts therefore face the ethical need to effectively communicate to non-expert audiences. Unfortunately, climate experts may inadvertently violate the maxims of effective communication, which require sharing communications that are truthful, brief, relevant, clear, and tested for effectiveness. Here, we discuss the 'mental models' approach towards developing communications, which aims to help experts to meet the maxims of effective communications, and to better inform the judgments and decisions of non-expert audiences.

Expert judgement and uncertainty quantification for climate change

NASA Astrophysics Data System (ADS)

Oppenheimer, Michael; Little, Christopher M.; Cooke, Roger M.

2016-05-01

Expert judgement is an unavoidable element of the process-based numerical models used for climate change projections, and the statistical approaches used to characterize uncertainty across model ensembles. Here, we highlight the need for formalized approaches to unifying numerical modelling with expert judgement in order to facilitate characterization of uncertainty in a reproducible, consistent and transparent fashion. As an example, we use probabilistic inversion, a well-established technique used in many other applications outside of climate change, to fuse two recent analyses of twenty-first century Antarctic ice loss. Probabilistic inversion is but one of many possible approaches to formalizing the role of expert judgement, and the Antarctic ice sheet is only one possible climate-related application. We recommend indicators or signposts that characterize successful science-based uncertainty quantification.

Systematically evaluating read-across prediction and ...

EPA Pesticide Factsheets

Read-across is a popular data gap filling technique within category and analogue approaches for regulatory purposes. Acceptance of read-across remains an ongoing challenge with several efforts underway for identifying and addressing uncertainties. Here we demonstrate an algorithmic, automated approach to evaluate the utility of using in vitro bioactivity data (“bioactivity descriptors”, from EPA’s ToxCast program) in conjunction with chemical descriptor information to derive local validity domains (specific sets of nearest neighbors) to facilitate read-across for a number of in vivo repeated dose toxicity study types. Over 3400 different chemical structure descriptors were generated for a set of 976 chemicals and supplemented with the outcomes from 821 in vitro assays. The read-across prediction for a given chemical was based on the similarity weighted endpoint outcomes of its nearest neighbors. The approach enabled a performance baseline for read-across predictions of specific study outcomes to be established. Bioactivity descriptors were often found to be more predictive of in vivo toxicity outcomes than chemical descriptors or a combination of both. The approach shows promise as part of a screening assessment in the absence of prior knowledge. Future work will investigate to what extent encoding expert knowledge leads to an improvement in read-across prediction. Read-across is a popular data gap filling technique within category and analogue approaches

A Deep Learning Approach for Fault Diagnosis of Induction Motors in Manufacturing

NASA Astrophysics Data System (ADS)

Shao, Si-Yu; Sun, Wen-Jun; Yan, Ru-Qiang; Wang, Peng; Gao, Robert X.

2017-11-01

Extracting features from original signals is a key procedure for traditional fault diagnosis of induction motors, as it directly influences the performance of fault recognition. However, high quality features need expert knowledge and human intervention. In this paper, a deep learning approach based on deep belief networks (DBN) is developed to learn features from frequency distribution of vibration signals with the purpose of characterizing working status of induction motors. It combines feature extraction procedure with classification task together to achieve automated and intelligent fault diagnosis. The DBN model is built by stacking multiple-units of restricted Boltzmann machine (RBM), and is trained using layer-by-layer pre-training algorithm. Compared with traditional diagnostic approaches where feature extraction is needed, the presented approach has the ability of learning hierarchical representations, which are suitable for fault classification, directly from frequency distribution of the measurement data. The structure of the DBN model is investigated as the scale and depth of the DBN architecture directly affect its classification performance. Experimental study conducted on a machine fault simulator verifies the effectiveness of the deep learning approach for fault diagnosis of induction motors. This research proposes an intelligent diagnosis method for induction motor which utilizes deep learning model to automatically learn features from sensor data and realize working status recognition.

An adaptive algorithm for the detection of microcalcifications in simulated low-dose mammography.

PubMed

Treiber, O; Wanninger, F; Führ, H; Panzer, W; Regulla, D; Winkler, G

2003-02-21

This paper uses the task of microcalcification detection as a benchmark problem to assess the potential for dose reduction in x-ray mammography. We present the results of a newly developed algorithm for detection of microcalcifications as a case study for a typical commercial film-screen system (Kodak Min-R 2000/2190). The first part of the paper deals with the simulation of dose reduction for film-screen mammography based on a physical model of the imaging process. Use of a more sensitive film-screen system is expected to result in additional smoothing of the image. We introduce two different models of that behaviour, called moderate and strong smoothing. We then present an adaptive, model-based microcalcification detection algorithm. Comparing detection results with ground-truth images obtained under the supervision of an expert radiologist allows us to establish the soundness of the detection algorithm. We measure the performance on the dose-reduced images in order to assess the loss of information due to dose reduction. It turns out that the smoothing behaviour has a strong influence on detection rates. For moderate smoothing. a dose reduction by 25% has no serious influence on the detection results. whereas a dose reduction by 50% already entails a marked deterioration of the performance. Strong smoothing generally leads to an unacceptable loss of image quality. The test results emphasize the impact of the more sensitive film-screen system and its characteristics on the problem of assessing the potential for dose reduction in film-screen mammography. The general approach presented in the paper can be adapted to fully digital mammography.

An adaptive algorithm for the detection of microcalcifications in simulated low-dose mammography

NASA Astrophysics Data System (ADS)

Treiber, O.; Wanninger, F.; Führ, H.; Panzer, W.; Regulla, D.; Winkler, G.

2003-02-01

This paper uses the task of microcalcification detection as a benchmark problem to assess the potential for dose reduction in x-ray mammography. We present the results of a newly developed algorithm for detection of microcalcifications as a case study for a typical commercial film-screen system (Kodak Min-R 2000/2190). The first part of the paper deals with the simulation of dose reduction for film-screen mammography based on a physical model of the imaging process. Use of a more sensitive film-screen system is expected to result in additional smoothing of the image. We introduce two different models of that behaviour, called moderate and strong smoothing. We then present an adaptive, model-based microcalcification detection algorithm. Comparing detection results with ground-truth images obtained under the supervision of an expert radiologist allows us to establish the soundness of the detection algorithm. We measure the performance on the dose-reduced images in order to assess the loss of information due to dose reduction. It turns out that the smoothing behaviour has a strong influence on detection rates. For moderate smoothing, a dose reduction by 25% has no serious influence on the detection results, whereas a dose reduction by 50% already entails a marked deterioration of the performance. Strong smoothing generally leads to an unacceptable loss of image quality. The test results emphasize the impact of the more sensitive film-screen system and its characteristics on the problem of assessing the potential for dose reduction in film-screen mammography. The general approach presented in the paper can be adapted to fully digital mammography.

Automated Quantification of Pneumothorax in CT

PubMed Central

Do, Synho; Salvaggio, Kristen; Gupta, Supriya; Kalra, Mannudeep; Ali, Nabeel U.; Pien, Homer

2012-01-01

An automated, computer-aided diagnosis (CAD) algorithm for the quantification of pneumothoraces from Multidetector Computed Tomography (MDCT) images has been developed. Algorithm performance was evaluated through comparison to manual segmentation by expert radiologists. A combination of two-dimensional and three-dimensional processing techniques was incorporated to reduce required processing time by two-thirds (as compared to similar techniques). Volumetric measurements on relative pneumothorax size were obtained and the overall performance of the automated method shows an average error of just below 1%. PMID:23082091

A Step-Wise Approach to Elicit Triangular Distributions

NASA Technical Reports Server (NTRS)

Greenberg, Marc W.

2013-01-01

Adapt/combine known methods to demonstrate an expert judgment elicitation process that: 1.Models expert's inputs as a triangular distribution, 2.Incorporates techniques to account for expert bias and 3.Is structured in a way to help justify expert's inputs. This paper will show one way of "extracting" expert opinion for estimating purposes. Nevertheless, as with most subjective methods, there are many ways to do this.

Developing expertise in surgery.

PubMed

Alderson, David

2010-01-01

The concept of expertise is widely embraced but poorly defined in surgery. Dictionary definitions differentiate between authority and experience, while a third view sees expertise as a mind-set rather than a status. Both absolute and relative models of expertise have been developed, and each allows a richer understanding of the application of these concepts to emerge. Trainees must develop both independent and interdependent expertise, and an appreciation of the essentially constructivist and uncertain nature of medical knowledge. Approach may be more important than innate talent; the concepts of 'flow', sustained 'deliberate practice' and 'adaptive expertise' are examples of expert approaches to learning. Non-analytical reasoning plays a key role in decision making at expert levels of practice. A technically gifted surgeon may be seen as a safety hazard rather than an expert if inter-dependent expertise has not been developed. Key roles of a surgical educator are to facilitate the development of an expert approach to education and to enable entry into and movement towards the centre of an expert community of practice.

A study on the value of computer-assisted assessment for SPECT/CT-scans in sentinel lymph node diagnostics of penile cancer as well as clinical reliability and morbidity of this procedure.

PubMed

Lützen, Ulf; Naumann, Carsten Maik; Marx, Marlies; Zhao, Yi; Jüptner, Michael; Baumann, René; Papp, László; Zsótér, Norbert; Aksenov, Alexey; Jünemann, Klaus-Peter; Zuhayra, Maaz

2016-09-07

Because of the increasing importance of computer-assisted post processing of image data in modern medical diagnostic we studied the value of an algorithm for assessment of single photon emission computed tomography/computed tomography (SPECT/CT)-data, which has been used for the first time for lymph node staging in penile cancer with non-palpable inguinal lymph nodes. In the guidelines of the relevant international expert societies, sentinel lymph node-biopsy (SLNB) is recommended as a diagnostic method of choice. The aim of this study is to evaluate the value of the afore-mentioned algorithm and in the clinical context the reliability and the associated morbidity of this procedure. Between 2008 and 2015, 25 patients with invasive penile cancer and inconspicuous inguinal lymph node status underwent SLNB after application of the radiotracer Tc-99m labelled nanocolloid. We recorded in a prospective approach the reliability and the complication rate of the procedure. In addition, we evaluated the results of an algorithm for SPECT/CT-data assessment of these patients. SLNB was carried out in 44 groins of 25 patients. In three patients, inguinal lymph node metastases were detected via SLNB. In one patient, bilateral lymph node recurrence of the groins occurred after negative SLNB. There was a false-negative rate of 4 % in relation to the number of patients (1/25), resp. 4.5 % in relation to the number of groins (2/44). Morbidity was 4 % in relation to the number of patients (1/25), resp. 2.3 % in relation to the number of groins (1/44). The results of computer-assisted assessment of SPECT/CT data for sentinel lymph node (SLN)-diagnostics demonstrated high sensitivity of 88.8 % and specificity of 86.7 %. SLNB is a very reliable method, associated with low morbidity. Computer-assisted assessment of SPECT/CT data of the SLN-diagnostics shows high sensitivity and specificity. While it cannot replace the assessment by medical experts, it can still provide substantial supplement and assistance.

The need for a comprehensive expert system development methodology

NASA Technical Reports Server (NTRS)

Baumert, John; Critchfield, Anna; Leavitt, Karen

1988-01-01

In a traditional software development environment, the introduction of standardized approaches has led to higher quality, maintainable products on the technical side and greater visibility into the status of the effort on the management side. This study examined expert system development to determine whether it differed enough from traditional systems to warrant a reevaluation of current software development methodologies. Its purpose was to identify areas of similarity with traditional software development and areas requiring tailoring to the unique needs of expert systems. A second purpose was to determine whether existing expert system development methodologies meet the needs of expert system development, management, and maintenance personnel. The study consisted of a literature search and personal interviews. It was determined that existing methodologies and approaches to developing expert systems are not comprehensive nor are they easily applied, especially to cradle to grave system development. As a result, requirements were derived for an expert system development methodology and an initial annotated outline derived for such a methodology.

Probabilistic inversion of expert assessments to inform projections about Antarctic ice sheet responses.

PubMed

Fuller, Robert William; Wong, Tony E; Keller, Klaus

2017-01-01

The response of the Antarctic ice sheet (AIS) to changing global temperatures is a key component of sea-level projections. Current projections of the AIS contribution to sea-level changes are deeply uncertain. This deep uncertainty stems, in part, from (i) the inability of current models to fully resolve key processes and scales, (ii) the relatively sparse available data, and (iii) divergent expert assessments. One promising approach to characterizing the deep uncertainty stemming from divergent expert assessments is to combine expert assessments, observations, and simple models by coupling probabilistic inversion and Bayesian inversion. Here, we present a proof-of-concept study that uses probabilistic inversion to fuse a simple AIS model and diverse expert assessments. We demonstrate the ability of probabilistic inversion to infer joint prior probability distributions of model parameters that are consistent with expert assessments. We then confront these inferred expert priors with instrumental and paleoclimatic observational data in a Bayesian inversion. These additional constraints yield tighter hindcasts and projections. We use this approach to quantify how the deep uncertainty surrounding expert assessments affects the joint probability distributions of model parameters and future projections.

a Study on Satellite Diagnostic Expert Systems Using Case-Based Approach

NASA Astrophysics Data System (ADS)

Park, Young-Tack; Kim, Jae-Hoon; Park, Hyun-Soo

1997-06-01

Many research works are on going to monitor and diagnose diverse malfunctions of satellite systems as the complexity and number of satellites increase. Currently, many works on monitoring and diagnosis are carried out by human experts but there are needs to automate much of the routine works of them. Hence, it is necessary to study on using expert systems which can assist human experts routine work by doing automatically, thereby allow human experts devote their expertise more critical and important areas of monitoring and diagnosis. In this paper, we are employing artificial intelligence techniques to model human experts' knowledge and inference the constructed knowledge. Especially, case-based approaches are used to construct a knowledge base to model human expert capabilities which use previous typical exemplars. We have designed and implemented a prototype case-based system for diagnosing satellite malfunctions using cases. Our system remembers typical failure cases and diagnoses a current malfunction by indexing the case base. Diverse methods are used to build a more user friendly interface which allows human experts can build a knowledge base in as easy way.

Ask-the-expert: Active Learning Based Knowledge Discovery Using the Expert

NASA Technical Reports Server (NTRS)

Das, Kamalika; Avrekh, Ilya; Matthews, Bryan; Sharma, Manali; Oza, Nikunj

2017-01-01

Often the manual review of large data sets, either for purposes of labeling unlabeled instances or for classifying meaningful results from uninteresting (but statistically significant) ones is extremely resource intensive, especially in terms of subject matter expert (SME) time. Use of active learning has been shown to diminish this review time significantly. However, since active learning is an iterative process of learning a classifier based on a small number of SME-provided labels at each iteration, the lack of an enabling tool can hinder the process of adoption of these technologies in real-life, in spite of their labor-saving potential. In this demo we present ASK-the-Expert, an interactive tool that allows SMEs to review instances from a data set and provide labels within a single framework. ASK-the-Expert is powered by an active learning algorithm for training a classifier in the backend. We demonstrate this system in the context of an aviation safety application, but the tool can be adopted to work as a simple review and labeling tool as well, without the use of active learning.

Ask-the-Expert: Active Learning Based Knowledge Discovery Using the Expert

NASA Technical Reports Server (NTRS)

Das, Kamalika

2017-01-01

Often the manual review of large data sets, either for purposes of labeling unlabeled instances or for classifying meaningful results from uninteresting (but statistically significant) ones is extremely resource intensive, especially in terms of subject matter expert (SME) time. Use of active learning has been shown to diminish this review time significantly. However, since active learning is an iterative process of learning a classifier based on a small number of SME-provided labels at each iteration, the lack of an enabling tool can hinder the process of adoption of these technologies in real-life, in spite of their labor-saving potential. In this demo we present ASK-the-Expert, an interactive tool that allows SMEs to review instances from a data set and provide labels within a single framework. ASK-the-Expert is powered by an active learning algorithm for training a classifier in the back end. We demonstrate this system in the context of an aviation safety application, but the tool can be adopted to work as a simple review and labeling tool as well, without the use of active learning.

An SSME High Pressure Oxidizer Turbopump diagnostic system using G2 real-time expert system

NASA Technical Reports Server (NTRS)

Guo, Ten-Huei

1991-01-01

An expert system which diagnoses various seal leakage faults in the High Pressure Oxidizer Turbopump of the SSME was developed using G2 real-time expert system. Three major functions of the software were implemented: model-based data generation, real-time expert system reasoning, and real-time input/output communication. This system is proposed as one module of a complete diagnostic system for the SSME. Diagnosis of a fault is defined as the determination of its type, severity, and likelihood. Since fault diagnosis is often accomplished through the use of heuristic human knowledge, an expert system based approach has been adopted as a paradigm to develop this diagnostic system. To implement this approach, a software shell which can be easily programmed to emulate the human decision process, the G2 Real-Time Expert System, was selected. Lessons learned from this implementation are discussed.

An SSME high pressure oxidizer turbopump diagnostic system using G2(TM) real-time expert system

NASA Technical Reports Server (NTRS)

Guo, Ten-Huei

1991-01-01

An expert system which diagnoses various seal leakage faults in the High Pressure Oxidizer Turbopump of the SSME was developed using G2(TM) real-time expert system. Three major functions of the software were implemented: model-based data generation, real-time expert system reasoning, and real-time input/output communication. This system is proposed as one module of a complete diagnostic system for Space Shuttle Main Engine. Diagnosis of a fault is defined as the determination of its type, severity, and likelihood. Since fault diagnosis is often accomplished through the use of heuristic human knowledge, an expert system based approach was adopted as a paradigm to develop this diagnostic system. To implement this approach, a software shell which can be easily programmed to emulate the human decision process, the G2 Real-Time Expert System, was selected. Lessons learned from this implementation are discussed.

Ocular biometrics by score-level fusion of disparate experts.

PubMed

Proença, Hugo

2014-12-01

The concept of periocular biometrics emerged to improve the robustness of iris recognition to degraded data. Being a relatively recent topic, most of the periocular recognition algorithms work in a holistic way and apply a feature encoding/matching strategy without considering each biological component in the periocular area. This not only augments the correlation between the components in the resulting biometric signature, but also increases the sensitivity to particular data covariates. The main novelty in this paper is to propose a periocular recognition ensemble made of two disparate components: 1) one expert analyses the iris texture and exhaustively exploits the multispectral information in visible-light data and 2) another expert parameterizes the shape of eyelids and defines a surrounding dimensionless region-of-interest, from where statistics of the eyelids, eyelashes, and skin wrinkles/furrows are encoded. Both experts work on disjoint regions of the periocular area and meet three important properties. First, they produce practically independent responses, which is behind the better performance of the ensemble when compared to the best individual recognizer. Second, they do not share particularly sensitivity to any image covariate, which accounts for augmenting the robustness against degraded data. Finally, it should be stressed that we disregard information in the periocular region that can be easily forged (e.g., shape of eyebrows), which constitutes an active anticounterfeit measure. An empirical evaluation was conducted on two public data sets (FRGC and UBIRIS.v2), and points for consistent improvements in performance of the proposed ensemble over the state-of-the-art periocular recognition algorithms.

Adolescence transitional care in neurogenic detrusor overactivity and the use of OnabotulinumtoxinA: A clinical algorithm from an Italian consensus statement.

PubMed

Palleschi, Giovanni; Mosiello, Giovanni; Iacovelli, Valerio; Musco, Stefania; Del Popolo, Giulio; Giannantoni, Antonella; Carbone, Antonio; Carone, Roberto; Tubaro, Andrea; De Gennaro, Mario; Marte, Antonio; Finazzi Agrò, Enrico

2018-03-01

OnabotulinumtoxinA (onaBNTa) for treating neurogenic detrusor overactivity (NDO) is widely used after its regulatory approval in adults. Although the administration of onaBNTa is still considered off-label in children, data have already been reported on its efficacy and safety. Nowadays, there is a lack of standardized protocols for treatment of NDO with onaBNTa in adolescent patients in their transition from the childhood to the adult age. With the aim to address this issue a consensus panel was obtained. A panel of leading urologists and urogynaecologists skilled in functional urology, neuro-urology, urogynaecology, and pediatric urology participated in a consensus-forming project using a Delphi method to reach national consensus on NDO-onaBNTa treatment in adolescence transitional care. In total, 11 experts participated. All panelists participated in the four phases of the consensus process. Consensus was reached if ≥70% of the experts agreed on recommendations. To facilitate a common understanding among all experts, a face-to-face consensus meeting was held in Rome in march 2015 and then with a follow-up teleconference in march 2017. By the end of the Delphi process, formal consensus was achieved for 100% of the items and an algorithm was then developed. This manuscript represents the first report on the onaBNTa in adolescents. Young adults should be treated as a distinct sub-population in policy, planning, programming, and research, as strongly sustained by national public health care. This consensus and the algorithm could support multidisciplinary communication, reduce the extent of variations in clinical practice and optimize clinical decision making. © 2017 Wiley Periodicals, Inc.

Methods to elicit probability distributions from experts: a systematic review of reported practice in health technology assessment.

PubMed

Grigore, Bogdan; Peters, Jaime; Hyde, Christopher; Stein, Ken

2013-11-01

Elicitation is a technique that can be used to obtain probability distribution from experts about unknown quantities. We conducted a methodology review of reports where probability distributions had been elicited from experts to be used in model-based health technology assessments. Databases including MEDLINE, EMBASE and the CRD database were searched from inception to April 2013. Reference lists were checked and citation mapping was also used. Studies describing their approach to the elicitation of probability distributions were included. Data was abstracted on pre-defined aspects of the elicitation technique. Reports were critically appraised on their consideration of the validity, reliability and feasibility of the elicitation exercise. Fourteen articles were included. Across these studies, the most marked features were heterogeneity in elicitation approach and failure to report key aspects of the elicitation method. The most frequently used approaches to elicitation were the histogram technique and the bisection method. Only three papers explicitly considered the validity, reliability and feasibility of the elicitation exercises. Judged by the studies identified in the review, reports of expert elicitation are insufficient in detail and this impacts on the perceived usability of expert-elicited probability distributions. In this context, the wider credibility of elicitation will only be improved by better reporting and greater standardisation of approach. Until then, the advantage of eliciting probability distributions from experts may be lost.

Classification of video sequences into chosen generalized use classes of target size and lighting level.

PubMed

Leszczuk, Mikołaj; Dudek, Łukasz; Witkowski, Marcin

The VQiPS (Video Quality in Public Safety) Working Group, supported by the U.S. Department of Homeland Security, has been developing a user guide for public safety video applications. According to VQiPS, five parameters have particular importance influencing the ability to achieve a recognition task. They are: usage time-frame, discrimination level, target size, lighting level, and level of motion. These parameters form what are referred to as Generalized Use Classes (GUCs). The aim of our research was to develop algorithms that would automatically assist classification of input sequences into one of the GUCs. Target size and lighting level parameters were approached. The experiment described reveals the experts' ambiguity and hesitation during the manual target size determination process. However, the automatic methods developed for target size classification make it possible to determine GUC parameters with 70 % compliance to the end-users' opinion. Lighting levels of the entire sequence can be classified with an efficiency reaching 93 %. To make the algorithms available for use, a test application has been developed. It is able to process video files and display classification results, the user interface being very simple and requiring only minimal user interaction.

Development of a Robotic Colonoscopic Manipulation System, Using Haptic Feedback Algorithm.

PubMed

Woo, Jaehong; Choi, Jae Hyuk; Seo, Jong Tae; Kim, Tae Il; Yi, Byung Ju

2017-01-01

Colonoscopy is one of the most effective diagnostic and therapeutic tools for colorectal diseases. We aim to propose a master-slave robotic colonoscopy that is controllable in remote site using conventional colonoscopy. The master and slave robot were developed to use conventional flexible colonoscopy. The robotic colonoscopic procedure was performed using a colonoscope training model by one expert endoscopist and two unexperienced engineers. To provide the haptic sensation, the insertion force and the rotating torque were measured and sent to the master robot. A slave robot was developed to hold the colonoscopy and its knob, and perform insertion, rotation, and two tilting motions of colonoscope. A master robot was designed to teach motions of the slave robot. These measured force and torque were scaled down by one tenth to provide the operator with some reflection force and torque at the haptic device. The haptic sensation and feedback system was successful and helpful to feel the constrained force or torque in colon. The insertion time using robotic system decreased with repeated procedures. This work proposed a robotic approach for colonoscopy using haptic feedback algorithm, and this robotic device would effectively perform colonoscopy with reduced burden and comparable safety for patients in remote site.

Mastering the game of Go with deep neural networks and tree search

NASA Astrophysics Data System (ADS)

Silver, David; Huang, Aja; Maddison, Chris J.; Guez, Arthur; Sifre, Laurent; van den Driessche, George; Schrittwieser, Julian; Antonoglou, Ioannis; Panneershelvam, Veda; Lanctot, Marc; Dieleman, Sander; Grewe, Dominik; Nham, John; Kalchbrenner, Nal; Sutskever, Ilya; Lillicrap, Timothy; Leach, Madeleine; Kavukcuoglu, Koray; Graepel, Thore; Hassabis, Demis

2016-01-01

The game of Go has long been viewed as the most challenging of classic games for artificial intelligence owing to its enormous search space and the difficulty of evaluating board positions and moves. Here we introduce a new approach to computer Go that uses ‘value networks’ to evaluate board positions and ‘policy networks’ to select moves. These deep neural networks are trained by a novel combination of supervised learning from human expert games, and reinforcement learning from games of self-play. Without any lookahead search, the neural networks play Go at the level of state-of-the-art Monte Carlo tree search programs that simulate thousands of random games of self-play. We also introduce a new search algorithm that combines Monte Carlo simulation with value and policy networks. Using this search algorithm, our program AlphaGo achieved a 99.8% winning rate against other Go programs, and defeated the human European Go champion by 5 games to 0. This is the first time that a computer program has defeated a human professional player in the full-sized game of Go, a feat previously thought to be at least a decade away.

Mastering the game of Go with deep neural networks and tree search.

PubMed

Silver, David; Huang, Aja; Maddison, Chris J; Guez, Arthur; Sifre, Laurent; van den Driessche, George; Schrittwieser, Julian; Antonoglou, Ioannis; Panneershelvam, Veda; Lanctot, Marc; Dieleman, Sander; Grewe, Dominik; Nham, John; Kalchbrenner, Nal; Sutskever, Ilya; Lillicrap, Timothy; Leach, Madeleine; Kavukcuoglu, Koray; Graepel, Thore; Hassabis, Demis

2016-01-28

The game of Go has long been viewed as the most challenging of classic games for artificial intelligence owing to its enormous search space and the difficulty of evaluating board positions and moves. Here we introduce a new approach to computer Go that uses 'value networks' to evaluate board positions and 'policy networks' to select moves. These deep neural networks are trained by a novel combination of supervised learning from human expert games, and reinforcement learning from games of self-play. Without any lookahead search, the neural networks play Go at the level of state-of-the-art Monte Carlo tree search programs that simulate thousands of random games of self-play. We also introduce a new search algorithm that combines Monte Carlo simulation with value and policy networks. Using this search algorithm, our program AlphaGo achieved a 99.8% winning rate against other Go programs, and defeated the human European Go champion by 5 games to 0. This is the first time that a computer program has defeated a human professional player in the full-sized game of Go, a feat previously thought to be at least a decade away.

Computer-Aided Diagnosis of Micro-Malignant Melanoma Lesions Applying Support Vector Machines.

PubMed

Jaworek-Korjakowska, Joanna

2016-01-01

Background. One of the fatal disorders causing death is malignant melanoma, the deadliest form of skin cancer. The aim of the modern dermatology is the early detection of skin cancer, which usually results in reducing the mortality rate and less extensive treatment. This paper presents a study on classification of melanoma in the early stage of development using SVMs as a useful technique for data classification. Method. In this paper an automatic algorithm for the classification of melanomas in their early stage, with a diameter under 5 mm, has been presented. The system contains the following steps: image enhancement, lesion segmentation, feature calculation and selection, and classification stage using SVMs. Results. The algorithm has been tested on 200 images including 70 melanomas and 130 benign lesions. The SVM classifier achieved sensitivity of 90% and specificity of 96%. The results indicate that the proposed approach captured most of the malignant cases and could provide reliable information for effective skin mole examination. Conclusions. Micro-melanomas due to the small size and low advancement of development create enormous difficulties during the diagnosis even for experts. The use of advanced equipment and sophisticated computer systems can help in the early diagnosis of skin lesions.

Acquisition of business intelligence from human experience in route planning

NASA Astrophysics Data System (ADS)

Bello Orgaz, Gema; Barrero, David F.; R-Moreno, María D.; Camacho, David

2015-04-01

The logistic sector raises a number of highly challenging problems. Probably one of the most important ones is the shipping planning, i.e. plan the routes that the shippers have to follow to deliver the goods. In this article, we present an artificial intelligence-based solution that has been designed to help a logistic company to improve its routes planning process. In order to achieve this goal, the solution uses the knowledge acquired by the company drivers to propose optimised routes. Hence, the proposed solution gathers the experience of the drivers, processes it and optimises the delivery process. The solution uses data mining to extract knowledge from the company information systems and prepares it for analysis with a case-based reasoning (CBR) algorithm. The CBR obtains critical business intelligence knowledge from the drivers experience that is needed by the planner. The design of the routes is done by a genetic algorithm that, given the processed information, optimises the routes following several objectives, such as minimise the distance or time. Experimentation shows that the proposed approach is able to find routes that improve, on average, the routes made by the human experts.

Automated Classification of Heritage Buildings for As-Built Bim Using Machine Learning Techniques

NASA Astrophysics Data System (ADS)

Bassier, M.; Vergauwen, M.; Van Genechten, B.

2017-08-01

Semantically rich three dimensional models such as Building Information Models (BIMs) are increasingly used in digital heritage. They provide the required information to varying stakeholders during the different stages of the historic buildings life cyle which is crucial in the conservation process. The creation of as-built BIM models is based on point cloud data. However, manually interpreting this data is labour intensive and often leads to misinterpretations. By automatically classifying the point cloud, the information can be proccesed more effeciently. A key aspect in this automated scan-to-BIM process is the classification of building objects. In this research we look to automatically recognise elements in existing buildings to create compact semantic information models. Our algorithm efficiently extracts the main structural components such as floors, ceilings, roofs, walls and beams despite the presence of significant clutter and occlusions. More specifically, Support Vector Machines (SVM) are proposed for the classification. The algorithm is evaluated using real data of a variety of existing buildings. The results prove that the used classifier recognizes the objects with both high precision and recall. As a result, entire data sets are reliably labelled at once. The approach enables experts to better document and process heritage assets.

Practical aspects in the management of statin-associated muscle symptoms (SAMS).

PubMed

Laufs, Ulrich; Filipiak, Krysztof J; Gouni-Berthold, Ioanna; Catapano, Alberico L

2017-04-01

Statin-associated muscle symptoms (SAMS) frequently cause statin non-adherence, switching and discontinuation, contributing to adverse cardiovascular (CV) outcomes. Therefore, the management of SAMS is key in the effective treatment of patients with cardiovascular disease (CVD), through achievement of maximum-tolerated statin dosing and other practical aspects. The aim of this article is to provide practical, focused advice for healthcare professionals on the management of patients with SAMS. An expert working group combined current evidence, published guidelines and experiences surrounding a number of topics concerning SAMS to provide recommendations on how to best assess and manage this condition and reach the highest tolerated dose of statin for each individual patient. The group collaborated to provide guidance on definitions in the SAMS field, psychological issues, re-challenging and switching treatments, as well as interpretation of current guidelines and optimal treatment of SAMS in different patient populations. An algorithm was developed to guide the management of patients with SAMS. In addition, the expert working group considered some of the more complex scenarios in a series of frequently asked questions and suggested answers. The expert working group gave recommendations for healthcare professionals on the management of SAMS but highlighted the importance of tailoring the treatment approach to each individual patient. Evidence supporting the role of nutraceuticals and complementary therapies, such as vitamin D, was lacking, however the majority of the group favoured combination therapy with ezetimibe and the addition of PCSK9 inhibitors in high-risk patients. Copyright © 2017 Elsevier B.V. All rights reserved.

Robot navigation research using the HERMIES mobile robot

DOE Office of Scientific and Technical Information (OSTI.GOV)

Barnett, D.L.

1989-01-01

In recent years robot navigation has attracted much attention from researchers around the world. Not only are theoretical studies being simulated on sophisticated computers, but many mobile robots are now used as test vehicles for these theoretical studies. Various algorithms have been perfected for navigation in a known static environment; but navigation in an unknown and dynamic environment poses a much more challenging problem for researchers. Many different methodologies have been developed for autonomous robot navigation, but each methodology is usually restricted to a particular type of environment. One important research focus of the Center for Engineering Systems Advanced researchmore » (CESAR) at Oak Ridge National Laboratory, is autonomous navigation in unknown and dynamic environments using the series of HERMIES mobile robots. The research uses an expert system for high-level planning interfaced with C-coded routines for implementing the plans, and for quick processing of data requested by the expert system. In using this approach, the navigation is not restricted to one methodology since the expert system can activate a rule module for the methodology best suited for the current situation. Rule modules can be added the rule base as they are developed and tested. Modules are being developed or enhanced for navigating from a map, searching for a target, exploring, artificial potential-field navigation, navigation using edge-detection, etc. This paper will report on the various rule modules and methods of navigation in use, or under development at CESAR, using the HERMIES-IIB robot as a testbed. 13 refs., 5 figs., 1 tab.« less

Clinical practice guidelines for the surgical management of colon cancer: a consensus statement of the Hellenic and Cypriot Colorectal Cancer Study Group by the HeSMO*

PubMed Central

Xynos, Evaghelos; Gouvas, Nikolaos; Triantopoulou, Charina; Tekkis, Paris; Vini, Louiza; Tzardi, Maria; Boukovinas, Ioannis; Androulakis, Nikolaos; Athanasiadis, Athanasios; Christodoulou, Christos; Chrysou, Evangelia; Dervenis, Christos; Emmanouilidis, Christos; Georgiou, Panagiotis; Katopodi, Ourania; Kountourakis, Panteleimon; Makatsoris, Thomas; Papakostas, Pavlos; Papamichael, Demetris; Pentheroudakis, Georgios; Pilpilidis, Ioannis; Sgouros, Joseph; Vassiliou, Vassilios; Xynogalos, Spyridon; Ziras, Nikolaos; Karachaliou, Niki; Zoras, Odysseas; Agalianos, Christos; Souglakos, John

2016-01-01

Despite considerable improvement in the management of colon cancer, there is a great deal of variation in the outcomes among European countries, and in particular among different hospital centers in Greece and Cyprus. Discrepancy in the approach strategies and lack of adherence to guidelines for the management of colon cancer may explain the situation. The aim was to elaborate a consensus on the multidisciplinary management of colon cancer, based on European guidelines (ESMO and EURECCA), and also taking into account local special characteristics of our healthcare system. Following discussion and online communication among members of an executive team, a consensus was developed. Statements entered the Delphi voting system on two rounds to achieve consensus by multidisciplinary international experts. Statements with an agreement rate of ≥80% achieved a large consensus, while those with an agreement rate of 60-80% a moderate consensus. Statements achieving an agreement of <60% after both rounds were rejected and not presented. Sixty statements on the management of colon cancer were subjected to the Delphi methodology. Voting experts were 109. The median rate of abstain per statement was 10% (range: 0-41%). In the end of the voting process, all statements achieved a consensus by more than 80% of the experts. A consensus on the management of colon cancer was developed by applying the Delphi methodology. Guidelines are proposed along with algorithms of diagnosis and treatment. The importance of centralization, care by a multidisciplinary team, and adherence to guidelines is emphasized. PMID:26752945