The application of neural networks to the SSME startup transient

NASA Technical Reports Server (NTRS)

Meyer, Claudia M.; Maul, William A.

1991-01-01

Feedforward neural networks were used to model three parameters during the Space Shuttle Main Engine startup transient. The three parameters were the main combustion chamber pressure, a controlled parameter, the high pressure oxidizer turbine discharge temperature, a redlined parameter, and the high pressure fuel pump discharge pressure, a failure-indicating performance parameter. Network inputs consisted of time windows of data from engine measurements that correlated highly to the modeled parameter. A standard backpropagation algorithm was used to train the feedforward networks on two nominal firings. Each trained network was validated with four additional nominal firings. For all three parameters, the neural networks were able to accurately predict the data in the validation sets as well as the training set.

Weight-elimination neural networks applied to coronary surgery mortality prediction.

PubMed

Ennett, Colleen M; Frize, Monique

2003-06-01

The objective was to assess the effectiveness of the weight-elimination cost function in improving classification performance of artificial neural networks (ANNs) and to observe how changing the a priori distribution of the training set affects network performance. Backpropagation feedforward ANNs with and without weight-elimination estimated mortality for coronary artery surgery patients. The ANNs were trained and tested on cases with 32 input variables describing the patient's medical history; the output variable was in-hospital mortality (mortality rates: training 3.7%, test 3.8%). Artificial training sets with mortality rates of 20%, 50%, and 80% were created to observe the impact of training with a higher-than-normal prevalence. When the results were averaged, weight-elimination networks achieved higher sensitivity rates than those without weight-elimination. Networks trained on higher-than-normal prevalence achieved higher sensitivity rates at the cost of lower specificity and correct classification. The weight-elimination cost function can improve the classification performance when the network is trained with a higher-than-normal prevalence. A network trained with a moderately high artificial mortality rate (artificial mortality rate of 20%) can improve the sensitivity of the model without significantly affecting other aspects of the model's performance. The ANN mortality model achieved comparable performance as additive and statistical models for coronary surgery mortality estimation in the literature.

Optimization of Training Sets For Neural-Net Processing of Characteristic Patterns From Vibrating Solids

NASA Technical Reports Server (NTRS)

Decker, Arthur J. (Inventor)

2006-01-01

An artificial neural network is disclosed that processes holography generated characteristic pattern of vibrating structures along with finite-element models. The present invention provides for a folding operation for conditioning training sets for optimally training forward-neural networks to process characteristic fringe pattern. The folding pattern increases the sensitivity of the feed-forward network for detecting changes in the characteristic pattern The folding routine manipulates input pixels so as to be scaled according to the location in an intensity range rather than the position in the characteristic pattern.

Mapping soil landscape as spatial continua: The Neural Network Approach

NASA Astrophysics Data System (ADS)

Zhu, A.-Xing

2000-03-01

A neural network approach was developed to populate a soil similarity model that was designed to represent soil landscape as spatial continua for hydroecological modeling at watersheds of mesoscale size. The approach employs multilayer feed forward neural networks. The input to the network was data on a set of soil formative environmental factors; the output from the network was a set of similarity values to a set of prescribed soil classes. The network was trained using a conjugate gradient algorithm in combination with a simulated annealing technique to learn the relationships between a set of prescribed soils and their environmental factors. Once trained, the network was used to compute for every location in an area the similarity values of the soil to the set of prescribed soil classes. The similarity values were then used to produce detailed soil spatial information. The approach also included a Geographic Information System procedure for selecting representative training and testing samples and a process of determining the network internal structure. The approach was applied to soil mapping in a watershed, the Lubrecht Experimental Forest, in western Montana. The case study showed that the soil spatial information derived using the neural network approach reveals much greater spatial detail and has a higher quality than that derived from the conventional soil map. Implications of this detailed soil spatial information for hydroecological modeling at the watershed scale are also discussed.

Transfer Learning with Convolutional Neural Networks for Classification of Abdominal Ultrasound Images.

PubMed

Cheng, Phillip M; Malhi, Harshawn S

2017-04-01

The purpose of this study is to evaluate transfer learning with deep convolutional neural networks for the classification of abdominal ultrasound images. Grayscale images from 185 consecutive clinical abdominal ultrasound studies were categorized into 11 categories based on the text annotation specified by the technologist for the image. Cropped images were rescaled to 256 × 256 resolution and randomized, with 4094 images from 136 studies constituting the training set, and 1423 images from 49 studies constituting the test set. The fully connected layers of two convolutional neural networks based on CaffeNet and VGGNet, previously trained on the 2012 Large Scale Visual Recognition Challenge data set, were retrained on the training set. Weights in the convolutional layers of each network were frozen to serve as fixed feature extractors. Accuracy on the test set was evaluated for each network. A radiologist experienced in abdominal ultrasound also independently classified the images in the test set into the same 11 categories. The CaffeNet network classified 77.3% of the test set images accurately (1100/1423 images), with a top-2 accuracy of 90.4% (1287/1423 images). The larger VGGNet network classified 77.9% of the test set accurately (1109/1423 images), with a top-2 accuracy of VGGNet was 89.7% (1276/1423 images). The radiologist classified 71.7% of the test set images correctly (1020/1423 images). The differences in classification accuracies between both neural networks and the radiologist were statistically significant (p < 0.001). The results demonstrate that transfer learning with convolutional neural networks may be used to construct effective classifiers for abdominal ultrasound images.

Teaching artificial neural systems to drive: Manual training techniques for autonomous systems

NASA Technical Reports Server (NTRS)

Shepanski, J. F.; Macy, S. A.

1987-01-01

A methodology was developed for manually training autonomous control systems based on artificial neural systems (ANS). In applications where the rule set governing an expert's decisions is difficult to formulate, ANS can be used to extract rules by associating the information an expert receives with the actions taken. Properly constructed networks imitate rules of behavior that permits them to function autonomously when they are trained on the spanning set of possible situations. This training can be provided manually, either under the direct supervision of a system trainer, or indirectly using a background mode where the networks assimilates training data as the expert performs its day-to-day tasks. To demonstrate these methods, an ANS network was trained to drive a vehicle through simulated freeway traffic.

Classification of breast MRI lesions using small-size training sets: comparison of deep learning approaches

NASA Astrophysics Data System (ADS)

Amit, Guy; Ben-Ari, Rami; Hadad, Omer; Monovich, Einat; Granot, Noa; Hashoul, Sharbell

2017-03-01

Diagnostic interpretation of breast MRI studies requires meticulous work and a high level of expertise. Computerized algorithms can assist radiologists by automatically characterizing the detected lesions. Deep learning approaches have shown promising results in natural image classification, but their applicability to medical imaging is limited by the shortage of large annotated training sets. In this work, we address automatic classification of breast MRI lesions using two different deep learning approaches. We propose a novel image representation for dynamic contrast enhanced (DCE) breast MRI lesions, which combines the morphological and kinetics information in a single multi-channel image. We compare two classification approaches for discriminating between benign and malignant lesions: training a designated convolutional neural network and using a pre-trained deep network to extract features for a shallow classifier. The domain-specific trained network provided higher classification accuracy, compared to the pre-trained model, with an area under the ROC curve of 0.91 versus 0.81, and an accuracy of 0.83 versus 0.71. Similar accuracy was achieved in classifying benign lesions, malignant lesions, and normal tissue images. The trained network was able to improve accuracy by using the multi-channel image representation, and was more robust to reductions in the size of the training set. A small-size convolutional neural network can learn to accurately classify findings in medical images using only a few hundred images from a few dozen patients. With sufficient data augmentation, such a network can be trained to outperform a pre-trained out-of-domain classifier. Developing domain-specific deep-learning models for medical imaging can facilitate technological advancements in computer-aided diagnosis.

Computer interpretation of thallium SPECT studies based on neural network analysis

NASA Astrophysics Data System (ADS)

Wang, David C.; Karvelis, K. C.

1991-06-01

A class of artificial intelligence (Al) programs known as neural networks are well suited to pattern recognition. A neural network is trained rather than programmed to recognize patterns. This differs from "expert system" Al programs in that it is not following an extensive set of rules determined by the programmer, but rather bases its decision on a gestalt interpretation of the image. The "bullseye" images from cardiac stress thallium tests performed on 50 male patients, as well as several simulated images were used to train the network. The network was able to accurately classify all patients in the training set. The network was then tested against 50 unknown patients and was able to correctly categorize 77% of the areas of ischemia and 92% of the areas of infarction. While not yet matching the ability of a trained physician, the neural network shows great promise in this area and has potential application in other areas of medical imaging.

Computing single step operators of logic programming in radial basis function neural networks

NASA Astrophysics Data System (ADS)

Hamadneh, Nawaf; Sathasivam, Saratha; Choon, Ong Hong

2014-07-01

Logic programming is the process that leads from an original formulation of a computing problem to executable programs. A normal logic program consists of a finite set of clauses. A valuation I of logic programming is a mapping from ground atoms to false or true. The single step operator of any logic programming is defined as a function (Tp:I→I). Logic programming is well-suited to building the artificial intelligence systems. In this study, we established a new technique to compute the single step operators of logic programming in the radial basis function neural networks. To do that, we proposed a new technique to generate the training data sets of single step operators. The training data sets are used to build the neural networks. We used the recurrent radial basis function neural networks to get to the steady state (the fixed point of the operators). To improve the performance of the neural networks, we used the particle swarm optimization algorithm to train the networks.

Comparison of Computational-Model and Experimental-Example Trained Neural Networks for Processing Speckled Fringe Patterns

NASA Technical Reports Server (NTRS)

Decker, A. J.; Fite, E. B.; Thorp, S. A.; Mehmed, O.

1998-01-01

The responses of artificial neural networks to experimental and model-generated inputs are compared for detection of damage in twisted fan blades using electronic holography. The training-set inputs, for this work, are experimentally generated characteristic patterns of the vibrating blades. The outputs are damage-flag indicators or second derivatives of the sensitivity-vector-projected displacement vectors from a finite element model. Artificial neural networks have been trained in the past with computational-model-generated training sets. This approach avoids the difficult inverse calculations traditionally used to compare interference fringes with the models. But the high modeling standards are hard to achieve, even with fan-blade finite-element models.

Comparison of Computational, Model and Experimental, Example Trained Neural Networks for Processing Speckled Fringe Patterns

NASA Technical Reports Server (NTRS)

Decker, A. J.; Fite, E. B.; Thorp, S. A.; Mehmed, O.

1998-01-01

The responses of artificial neural networks to experimental and model-generated inputs are compared for detection of damage in twisted fan blades using electronic holography. The training-set inputs, for this work, are experimentally generated characteristic patterns of the vibrating blades. The outputs are damage-flag indicators or second derivatives of the sensitivity-vector-projected displacement vectors from a finite element model. Artificial neural networks have been trained in the past with computational-model- generated training sets. This approach avoids the difficult inverse calculations traditionally used to compare interference fringes with the models. But the high modeling standards are hard to achieve, even with fan-blade finite-element models.

A generalized LSTM-like training algorithm for second-order recurrent neural networks

PubMed Central

Monner, Derek; Reggia, James A.

2011-01-01

The Long Short Term Memory (LSTM) is a second-order recurrent neural network architecture that excels at storing sequential short-term memories and retrieving them many time-steps later. LSTM’s original training algorithm provides the important properties of spatial and temporal locality, which are missing from other training approaches, at the cost of limiting it’s applicability to a small set of network architectures. Here we introduce the Generalized Long Short-Term Memory (LSTM-g) training algorithm, which provides LSTM-like locality while being applicable without modification to a much wider range of second-order network architectures. With LSTM-g, all units have an identical set of operating instructions for both activation and learning, subject only to the configuration of their local environment in the network; this is in contrast to the original LSTM training algorithm, where each type of unit has its own activation and training instructions. When applied to LSTM architectures with peephole connections, LSTM-g takes advantage of an additional source of back-propagated error which can enable better performance than the original algorithm. Enabled by the broad architectural applicability of LSTM-g, we demonstrate that training recurrent networks engineered for specific tasks can produce better results than single-layer networks. We conclude that LSTM-g has the potential to both improve the performance and broaden the applicability of spatially and temporally local gradient-based training algorithms for recurrent neural networks. PMID:21803542

Localization and identification of structural nonlinearities using cascaded optimization and neural networks

NASA Astrophysics Data System (ADS)

Koyuncu, A.; Cigeroglu, E.; Özgüven, H. N.

2017-10-01

In this study, a new approach is proposed for identification of structural nonlinearities by employing cascaded optimization and neural networks. Linear finite element model of the system and frequency response functions measured at arbitrary locations of the system are used in this approach. Using the finite element model, a training data set is created, which appropriately spans the possible nonlinear configurations space of the system. A classification neural network trained on these data sets then localizes and determines the types of all nonlinearities associated with the nonlinear degrees of freedom in the system. A new training data set spanning the parametric space associated with the determined nonlinearities is created to facilitate parametric identification. Utilizing this data set, initially, a feed forward regression neural network is trained, which parametrically identifies the classified nonlinearities. Then, the results obtained are further improved by carrying out an optimization which uses network identified values as starting points. Unlike identification methods available in literature, the proposed approach does not require data collection from the degrees of freedoms where nonlinear elements are attached, and furthermore, it is sufficiently accurate even in the presence of measurement noise. The application of the proposed approach is demonstrated on an example system with nonlinear elements and on a real life experimental setup with a local nonlinearity.

Training set expansion: an approach to improving the reconstruction of biological networks from limited and uneven reliable interactions

PubMed Central

Yip, Kevin Y.; Gerstein, Mark

2009-01-01

Motivation: An important problem in systems biology is reconstructing complete networks of interactions between biological objects by extrapolating from a few known interactions as examples. While there are many computational techniques proposed for this network reconstruction task, their accuracy is consistently limited by the small number of high-confidence examples, and the uneven distribution of these examples across the potential interaction space, with some objects having many known interactions and others few. Results: To address this issue, we propose two computational methods based on the concept of training set expansion. They work particularly effectively in conjunction with kernel approaches, which are a popular class of approaches for fusing together many disparate types of features. Both our methods are based on semi-supervised learning and involve augmenting the limited number of gold-standard training instances with carefully chosen and highly confident auxiliary examples. The first method, prediction propagation, propagates highly confident predictions of one local model to another as the auxiliary examples, thus learning from information-rich regions of the training network to help predict the information-poor regions. The second method, kernel initialization, takes the most similar and most dissimilar objects of each object in a global kernel as the auxiliary examples. Using several sets of experimentally verified protein–protein interactions from yeast, we show that training set expansion gives a measurable performance gain over a number of representative, state-of-the-art network reconstruction methods, and it can correctly identify some interactions that are ranked low by other methods due to the lack of training examples of the involved proteins. Contact: mark.gerstein@yale.edu Availability: The datasets and additional materials can be found at http://networks.gersteinlab.org/tse. PMID:19015141

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

Hamadneh, Nawaf; Sathasivam, Saratha; Choon, Ong Hong

Logic programming is the process that leads from an original formulation of a computing problem to executable programs. A normal logic program consists of a finite set of clauses. A valuation I of logic programming is a mapping from ground atoms to false or true. The single step operator of any logic programming is defined as a function (T{sub p}:I→I). Logic programming is well-suited to building the artificial intelligence systems. In this study, we established a new technique to compute the single step operators of logic programming in the radial basis function neural networks. To do that, we proposed amore » new technique to generate the training data sets of single step operators. The training data sets are used to build the neural networks. We used the recurrent radial basis function neural networks to get to the steady state (the fixed point of the operators). To improve the performance of the neural networks, we used the particle swarm optimization algorithm to train the networks.« less

A neural network approach for image reconstruction in electron magnetic resonance tomography.

PubMed

Durairaj, D Christopher; Krishna, Murali C; Murugesan, Ramachandran

2007-10-01

An object-oriented, artificial neural network (ANN) based, application system for reconstruction of two-dimensional spatial images in electron magnetic resonance (EMR) tomography is presented. The standard back propagation algorithm is utilized to train a three-layer sigmoidal feed-forward, supervised, ANN to perform the image reconstruction. The network learns the relationship between the 'ideal' images that are reconstructed using filtered back projection (FBP) technique and the corresponding projection data (sinograms). The input layer of the network is provided with a training set that contains projection data from various phantoms as well as in vivo objects, acquired from an EMR imager. Twenty five different network configurations are investigated to test the ability of the generalization of the network. The trained ANN then reconstructs two-dimensional temporal spatial images that present the distribution of free radicals in biological systems. Image reconstruction by the trained neural network shows better time complexity than the conventional iterative reconstruction algorithms such as multiplicative algebraic reconstruction technique (MART). The network is further explored for image reconstruction from 'noisy' EMR data and the results show better performance than the FBP method. The network is also tested for its ability to reconstruct from limited-angle EMR data set.

T-wave end detection using neural networks and Support Vector Machines.

PubMed

Suárez-León, Alexander Alexeis; Varon, Carolina; Willems, Rik; Van Huffel, Sabine; Vázquez-Seisdedos, Carlos Román

2018-05-01

In this paper we propose a new approach for detecting the end of the T-wave in the electrocardiogram (ECG) using Neural Networks and Support Vector Machines. Both, Multilayer Perceptron (MLP) neural networks and Fixed-Size Least-Squares Support Vector Machines (FS-LSSVM) were used as regression algorithms to determine the end of the T-wave. Different strategies for selecting the training set such as random selection, k-means, robust clustering and maximum quadratic (Rényi) entropy were evaluated. Individual parameters were tuned for each method during training and the results are given for the evaluation set. A comparison between MLP and FS-LSSVM approaches was performed. Finally, a fair comparison of the FS-LSSVM method with other state-of-the-art algorithms for detecting the end of the T-wave was included. The experimental results show that FS-LSSVM approaches are more suitable as regression algorithms than MLP neural networks. Despite the small training sets used, the FS-LSSVM methods outperformed the state-of-the-art techniques. FS-LSSVM can be successfully used as a T-wave end detection algorithm in ECG even with small training set sizes. Copyright © 2018 Elsevier Ltd. All rights reserved.

Dense modifiable interconnections utilizing photorefractive volume holograms

NASA Astrophysics Data System (ADS)

Psaltis, Demetri; Qiao, Yong

1990-11-01

This report describes an experimental two-layer optical neural network built at Caltech. The system uses photorefractive volume holograms to implement dense, modifiable synaptic interconnections and liquid crystal light valves (LCVS) to perform nonlinear thresholding operations. Kanerva's Sparse, Distributed Memory was implemented using this network and its ability to recognize handwritten character-alphabet (A-Z) has been demonstrated experimentally. According to Kanerva's model, the first layer has fixed, random weights of interconnections and the second layer is trained by sum-of-outer-products rule. After training, the recognition rates of the network on the training set (104 patterns) and test set (520 patterns) are 100 and 50 percent, respectively.

Neural-Network Object-Recognition Program

NASA Technical Reports Server (NTRS)

Spirkovska, L.; Reid, M. B.

1993-01-01

HONTIOR computer program implements third-order neural network exhibiting invariance under translation, change of scale, and in-plane rotation. Invariance incorporated directly into architecture of network. Only one view of each object needed to train network for two-dimensional-translation-invariant recognition of object. Also used for three-dimensional-transformation-invariant recognition by training network on only set of out-of-plane rotated views. Written in C language.

Kohonen and counterpropagation neural networks applied for mapping and interpretation of IR spectra.

PubMed

Novic, Marjana

2008-01-01

The principles of learning strategy of Kohonen and counterpropagation neural networks are introduced. The advantages of unsupervised learning are discussed. The self-organizing maps produced in both methods are suitable for a wide range of applications. Here, we present an example of Kohonen and counterpropagation neural networks used for mapping, interpretation, and simulation of infrared (IR) spectra. The artificial neural network models were trained for prediction of structural fragments of an unknown compound from its infrared spectrum. The training set contained over 3,200 IR spectra of diverse compounds of known chemical structure. The structure-spectra relationship was encompassed by the counterpropagation neural network, which assigned structural fragments to individual compounds within certain probability limits, assessed from the predictions of test compounds. The counterpropagation neural network model for prediction of fragments of chemical structure is reversible, which means that, for a given structural domain, limited to the training data set in the study, it can be used to simulate the IR spectrum of a chemical defined with a set of structural fragments.

Noise-enhanced convolutional neural networks.

PubMed

Audhkhasi, Kartik; Osoba, Osonde; Kosko, Bart

2016-06-01

Injecting carefully chosen noise can speed convergence in the backpropagation training of a convolutional neural network (CNN). The Noisy CNN algorithm speeds training on average because the backpropagation algorithm is a special case of the generalized expectation-maximization (EM) algorithm and because such carefully chosen noise always speeds up the EM algorithm on average. The CNN framework gives a practical way to learn and recognize images because backpropagation scales with training data. It has only linear time complexity in the number of training samples. The Noisy CNN algorithm finds a special separating hyperplane in the network's noise space. The hyperplane arises from the likelihood-based positivity condition that noise-boosts the EM algorithm. The hyperplane cuts through a uniform-noise hypercube or Gaussian ball in the noise space depending on the type of noise used. Noise chosen from above the hyperplane speeds training on average. Noise chosen from below slows it on average. The algorithm can inject noise anywhere in the multilayered network. Adding noise to the output neurons reduced the average per-iteration training-set cross entropy by 39% on a standard MNIST image test set of handwritten digits. It also reduced the average per-iteration training-set classification error by 47%. Adding noise to the hidden layers can also reduce these performance measures. The noise benefit is most pronounced for smaller data sets because the largest EM hill-climbing gains tend to occur in the first few iterations. This noise effect can assist random sampling from large data sets because it allows a smaller random sample to give the same or better performance than a noiseless sample gives. Copyright © 2015 Elsevier Ltd. All rights reserved.

A Structure-Adaptive Hybrid RBF-BP Classifier with an Optimized Learning Strategy

PubMed Central

Wen, Hui; Xie, Weixin; Pei, Jihong

2016-01-01

This paper presents a structure-adaptive hybrid RBF-BP (SAHRBF-BP) classifier with an optimized learning strategy. SAHRBF-BP is composed of a structure-adaptive RBF network and a BP network of cascade, where the number of RBF hidden nodes is adjusted adaptively according to the distribution of sample space, the adaptive RBF network is used for nonlinear kernel mapping and the BP network is used for nonlinear classification. The optimized learning strategy is as follows: firstly, a potential function is introduced into training sample space to adaptively determine the number of initial RBF hidden nodes and node parameters, and a form of heterogeneous samples repulsive force is designed to further optimize each generated RBF hidden node parameters, the optimized structure-adaptive RBF network is used for adaptively nonlinear mapping the sample space; then, according to the number of adaptively generated RBF hidden nodes, the number of subsequent BP input nodes can be determined, and the overall SAHRBF-BP classifier is built up; finally, different training sample sets are used to train the BP network parameters in SAHRBF-BP. Compared with other algorithms applied to different data sets, experiments show the superiority of SAHRBF-BP. Especially on most low dimensional and large number of data sets, the classification performance of SAHRBF-BP outperforms other training SLFNs algorithms. PMID:27792737

Residual Convolutional Neural Network for the Determination of IDH Status in Low- and High-Grade Gliomas from MR Imaging.

PubMed

Chang, Ken; Bai, Harrison X; Zhou, Hao; Su, Chang; Bi, Wenya Linda; Agbodza, Ena; Kavouridis, Vasileios K; Senders, Joeky T; Boaro, Alessandro; Beers, Andrew; Zhang, Biqi; Capellini, Alexandra; Liao, Weihua; Shen, Qin; Li, Xuejun; Xiao, Bo; Cryan, Jane; Ramkissoon, Shakti; Ramkissoon, Lori; Ligon, Keith; Wen, Patrick Y; Bindra, Ranjit S; Woo, John; Arnaout, Omar; Gerstner, Elizabeth R; Zhang, Paul J; Rosen, Bruce R; Yang, Li; Huang, Raymond Y; Kalpathy-Cramer, Jayashree

2018-03-01

Purpose: Isocitrate dehydrogenase ( IDH ) mutations in glioma patients confer longer survival and may guide treatment decision making. We aimed to predict the IDH status of gliomas from MR imaging by applying a residual convolutional neural network to preoperative radiographic data. Experimental Design: Preoperative imaging was acquired for 201 patients from the Hospital of University of Pennsylvania (HUP), 157 patients from Brigham and Women's Hospital (BWH), and 138 patients from The Cancer Imaging Archive (TCIA) and divided into training, validation, and testing sets. We trained a residual convolutional neural network for each MR sequence (FLAIR, T2, T1 precontrast, and T1 postcontrast) and built a predictive model from the outputs. To increase the size of the training set and prevent overfitting, we augmented the training set images by introducing random rotations, translations, flips, shearing, and zooming. Results: With our neural network model, we achieved IDH prediction accuracies of 82.8% (AUC = 0.90), 83.0% (AUC = 0.93), and 85.7% (AUC = 0.94) within training, validation, and testing sets, respectively. When age at diagnosis was incorporated into the model, the training, validation, and testing accuracies increased to 87.3% (AUC = 0.93), 87.6% (AUC = 0.95), and 89.1% (AUC = 0.95), respectively. Conclusions: We developed a deep learning technique to noninvasively predict IDH genotype in grade II-IV glioma using conventional MR imaging using a multi-institutional data set. Clin Cancer Res; 24(5); 1073-81. ©2017 AACR . ©2017 American Association for Cancer Research.

Application of Artificial Neural Network to Optical Fluid Analyzer

NASA Astrophysics Data System (ADS)

Kimura, Makoto; Nishida, Katsuhiko

1994-04-01

A three-layer artificial neural network has been applied to the presentation of optical fluid analyzer (OFA) raw data, and the accuracy of oil fraction determination has been significantly improved compared to previous approaches. To apply the artificial neural network approach to solving a problem, the first step is training to determine the appropriate weight set for calculating the target values. This involves using a series of data sets (each comprising a set of input values and an associated set of output values that the artificial neural network is required to determine) to tune artificial neural network weighting parameters so that the output of the neural network to the given set of input values is as close as possible to the required output. The physical model used to generate the series of learning data sets was the effective flow stream model, developed for OFA data presentation. The effectiveness of the training was verified by reprocessing the same input data as were used to determine the weighting parameters and then by comparing the results of the artificial neural network to the expected output values. The standard deviation of the expected and obtained values was approximately 10% (two sigma).

MATE: Machine Learning for Adaptive Calibration Template Detection

PubMed Central

Donné, Simon; De Vylder, Jonas; Goossens, Bart; Philips, Wilfried

2016-01-01

The problem of camera calibration is two-fold. On the one hand, the parameters are estimated from known correspondences between the captured image and the real world. On the other, these correspondences themselves—typically in the form of chessboard corners—need to be found. Many distinct approaches for this feature template extraction are available, often of large computational and/or implementational complexity. We exploit the generalized nature of deep learning networks to detect checkerboard corners: our proposed method is a convolutional neural network (CNN) trained on a large set of example chessboard images, which generalizes several existing solutions. The network is trained explicitly against noisy inputs, as well as inputs with large degrees of lens distortion. The trained network that we evaluate is as accurate as existing techniques while offering improved execution time and increased adaptability to specific situations with little effort. The proposed method is not only robust against the types of degradation present in the training set (lens distortions, and large amounts of sensor noise), but also to perspective deformations, e.g., resulting from multi-camera set-ups. PMID:27827920

Feed-forward neural network model for hunger and satiety related VAS score prediction.

PubMed

Krishnan, Shaji; Hendriks, Henk F J; Hartvigsen, Merete L; de Graaf, Albert A

2016-07-07

An artificial neural network approach was chosen to model the outcome of the complex signaling pathways in the gastro-intestinal tract and other peripheral organs that eventually produce the satiety feeling in the brain upon feeding. A multilayer feed-forward neural network was trained with sets of experimental data relating concentration-time courses of plasma satiety hormones to Visual Analog Scales (VAS) scores. The network successfully predicted VAS responses from sets of satiety hormone data obtained in experiments using different food compositions. The correlation coefficients for the predicted VAS responses for test sets having i) a full set of three satiety hormones, ii) a set of only two satiety hormones, and iii) a set of only one satiety hormone were 0.96, 0.96, and 0.89, respectively. The predicted VAS responses discriminated the satiety effects of high satiating food types from less satiating food types both in orally fed and ileal infused forms. From this application of artificial neural networks, one may conclude that neural network models are very suitable to describe situations where behavior is complex and incompletely understood. However, training data sets that fit the experimental conditions need to be available.

An Evolutionary Optimization Framework for Neural Networks and Neuromorphic Architectures

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

Schuman, Catherine D; Plank, James; Disney, Adam

2016-01-01

As new neural network and neuromorphic architectures are being developed, new training methods that operate within the constraints of the new architectures are required. Evolutionary optimization (EO) is a convenient training method for new architectures. In this work, we review a spiking neural network architecture and a neuromorphic architecture, and we describe an EO training framework for these architectures. We present the results of this training framework on four classification data sets and compare those results to other neural network and neuromorphic implementations. We also discuss how this EO framework may be extended to other architectures.

Circuity analyses of HSR network and high-speed train paths in China

PubMed Central

Zhao, Shuo; Huang, Jie; Shan, Xinghua

2017-01-01

Circuity, defined as the ratio of the shortest network distance to the Euclidean distance between one origin–destination (O-D) pair, can be adopted as a helpful evaluation method of indirect degrees of train paths. In this paper, the maximum circuity of the paths of operated trains is set to be the threshold value of the circuity of high-speed train paths. For the shortest paths of any node pairs, if their circuity is not higher than the threshold value, the paths can be regarded as the reasonable paths. With the consideration of a certain relative or absolute error, we cluster the reasonable paths on the basis of their inclusion relationship and the center path of each class represents a passenger transit corridor. We take the high-speed rail (HSR) network in China at the end of 2014 as an example, and obtain 51 passenger transit corridors, which are alternative sets of train paths. Furthermore, we analyze the circuity distribution of paths of all node pairs in the network. We find that the high circuity of train paths can be decreased with the construction of a high-speed railway line, which indicates that the structure of the HSR network in China tends to be more complete and the HSR network can make the Chinese railway network more efficient. PMID:28945757

A Multimedia Telematics Network for On-the-Job Training, Tutoring and Assessment.

ERIC Educational Resources Information Center

Ferreira, J. M. Martins; MacKinnon, Lachlan; Desmulliez, Marc; Foulk, Patrick

This paper describes an educational multimedia network developed in Advanced Software for Training and Evaluation of Processes (ASTEP). ASTEP started in February 1998 and was set up by a mixed industry-academia consortium with the objective of meeting the educational/training demands of the highly competitive microelectronics/semiconductor…

Neural network pattern recognition of thermal-signature spectra for chemical defense

NASA Astrophysics Data System (ADS)

Carrieri, Arthur H.; Lim, Pascal I.

1995-05-01

We treat infrared patterns of absorption or emission by nerve and blister agent compounds (and simulants of this chemical group) as features for the training of neural networks to detect the compounds' liquid layers on the ground or their vapor plumes during evaporation by external heating. Training of a four-layer network architecture is composed of a backward-error-propagation algorithm and a gradient-descent paradigm. We conduct testing by feed-forwarding preprocessed spectra through the network in a scaled format consistent with the structure of the training-data-set representation. The best-performance weight matrix (spectral filter) evolved from final network training and testing with software simulation trials is electronically transferred to a set of eight artificial intelligence integrated circuits (ICs') in specific modular form (splitting of weight matrices). This form makes full use of all input-output IC nodes. This neural network computer serves an important real-time detection function when it is integrated into pre-and postprocessing data-handling units of a tactical prototype thermoluminescence sensor now under development at the Edgewood Research, Development, and Engineering Center.

Seismic signal auto-detecing from different features by using Convolutional Neural Network

NASA Astrophysics Data System (ADS)

Huang, Y.; Zhou, Y.; Yue, H.; Zhou, S.

2017-12-01

We try Convolutional Neural Network to detect some features of seismic data and compare their efficience. The features include whether a signal is seismic signal or noise and the arrival time of P and S phase and each feature correspond to a Convolutional Neural Network. We first use traditional STA/LTA to recongnize some events and then use templete matching to find more events as training set for the Neural Network. To make the training set more various, we add some noise to the seismic data and make some synthetic seismic data and noise. The 3-component raw signal and time-frequancy ananlyze are used as the input data for our neural network. Our Training is performed on GPUs to achieve efficient convergence. Our method improved the precision in comparison with STA/LTA and template matching. We will move to recurrent neural network to see if this kind network is better in detect P and S phase.

Deep learning in the small sample size setting: cascaded feed forward neural networks for medical image segmentation

NASA Astrophysics Data System (ADS)

Gaonkar, Bilwaj; Hovda, David; Martin, Neil; Macyszyn, Luke

2016-03-01

Deep Learning, refers to large set of neural network based algorithms, have emerged as promising machine- learning tools in the general imaging and computer vision domains. Convolutional neural networks (CNNs), a specific class of deep learning algorithms, have been extremely effective in object recognition and localization in natural images. A characteristic feature of CNNs, is the use of a locally connected multi layer topology that is inspired by the animal visual cortex (the most powerful vision system in existence). While CNNs, perform admirably in object identification and localization tasks, typically require training on extremely large datasets. Unfortunately, in medical image analysis, large datasets are either unavailable or are extremely expensive to obtain. Further, the primary tasks in medical imaging are organ identification and segmentation from 3D scans, which are different from the standard computer vision tasks of object recognition. Thus, in order to translate the advantages of deep learning to medical image analysis, there is a need to develop deep network topologies and training methodologies, that are geared towards medical imaging related tasks and can work in a setting where dataset sizes are relatively small. In this paper, we present a technique for stacked supervised training of deep feed forward neural networks for segmenting organs from medical scans. Each `neural network layer' in the stack is trained to identify a sub region of the original image, that contains the organ of interest. By layering several such stacks together a very deep neural network is constructed. Such a network can be used to identify extremely small regions of interest in extremely large images, inspite of a lack of clear contrast in the signal or easily identifiable shape characteristics. What is even more intriguing is that the network stack achieves accurate segmentation even when it is trained on a single image with manually labelled ground truth. We validate this approach,using a publicly available head and neck CT dataset. We also show that a deep neural network of similar depth, if trained directly using backpropagation, cannot acheive the tasks achieved using our layer wise training paradigm.

A microsensor array for quantification of lubricant contaminants using a back propagation artificial neural network

NASA Astrophysics Data System (ADS)

Zhu, Xiaoliang; Du, Li; Liu, Bendong; Zhe, Jiang

2016-06-01

We present a method based on an electrochemical sensor array and a back propagation artificial neural network for detection and quantification of four properties of lubrication oil, namely water (0, 500 ppm, 1000 ppm), total acid number (TAN) (13.1, 13.7, 14.4, 15.6 mg KOH g-1), soot (0, 1%, 2%, 3%) and sulfur content (1.3%, 1.37%, 1.44%, 1.51%). The sensor array, consisting of four micromachined electrochemical sensors, detects the four properties with overlapping sensitivities. A total set of 36 oil samples containing mixtures of water, soot, and sulfuric acid with different concentrations were prepared for testing. The sensor array’s responses were then divided to three sets: training sets (80% data), validation sets (10%) and testing sets (10%). Several back propagation artificial neural network architectures were trained with the training and validation sets; one architecture with four input neurons, 50 and 5 neurons in the first and second hidden layer, and four neurons in the output layer was selected. The selected neural network was then tested using the four sets of testing data (10%). Test results demonstrated that the developed artificial neural network is able to quantitatively determine the four lubrication properties (water, TAN, soot, and sulfur content) with a maximum prediction error of 18.8%, 6.0%, 6.7%, and 5.4%, respectively, indicting a good match between the target and predicted values. With the developed network, the sensor array could be potentially used for online lubricant oil condition monitoring.

Vision-based mobile robot navigation through deep convolutional neural networks and end-to-end learning

NASA Astrophysics Data System (ADS)

Zhang, Yachu; Zhao, Yuejin; Liu, Ming; Dong, Liquan; Kong, Lingqin; Liu, Lingling

2017-09-01

In contrast to humans, who use only visual information for navigation, many mobile robots use laser scanners and ultrasonic sensors along with vision cameras to navigate. This work proposes a vision-based robot control algorithm based on deep convolutional neural networks. We create a large 15-layer convolutional neural network learning system and achieve the advanced recognition performance. Our system is trained from end to end to map raw input images to direction in supervised mode. The images of data sets are collected in a wide variety of weather conditions and lighting conditions. Besides, the data sets are augmented by adding Gaussian noise and Salt-and-pepper noise to avoid overfitting. The algorithm is verified by two experiments, which are line tracking and obstacle avoidance. The line tracking experiment is proceeded in order to track the desired path which is composed of straight and curved lines. The goal of obstacle avoidance experiment is to avoid the obstacles indoor. Finally, we get 3.29% error rate on the training set and 5.1% error rate on the test set in the line tracking experiment, 1.8% error rate on the training set and less than 5% error rate on the test set in the obstacle avoidance experiment. During the actual test, the robot can follow the runway centerline outdoor and avoid the obstacle in the room accurately. The result confirms the effectiveness of the algorithm and our improvement in the network structure and train parameters

Automated method for the systematic interpretation of resonance peaks in spectrum data

DOEpatents

Damiano, B.; Wood, R.T.

1997-04-22

A method is described for spectral signature interpretation. The method includes the creation of a mathematical model of a system or process. A neural network training set is then developed based upon the mathematical model. The neural network training set is developed by using the mathematical model to generate measurable phenomena of the system or process based upon model input parameter that correspond to the physical condition of the system or process. The neural network training set is then used to adjust internal parameters of a neural network. The physical condition of an actual system or process represented by the mathematical model is then monitored by extracting spectral features from measured spectra of the actual process or system. The spectral features are then input into said neural network to determine the physical condition of the system or process represented by the mathematical model. More specifically, the neural network correlates the spectral features (i.e. measurable phenomena) of the actual process or system with the corresponding model input parameters. The model input parameters relate to specific components of the system or process, and, consequently, correspond to the physical condition of the process or system. 1 fig.

Automated method for the systematic interpretation of resonance peaks in spectrum data

DOEpatents

Damiano, Brian; Wood, Richard T.

1997-01-01

A method for spectral signature interpretation. The method includes the creation of a mathematical model of a system or process. A neural network training set is then developed based upon the mathematical model. The neural network training set is developed by using the mathematical model to generate measurable phenomena of the system or process based upon model input parameter that correspond to the physical condition of the system or process. The neural network training set is then used to adjust internal parameters of a neural network. The physical condition of an actual system or process represented by the mathematical model is then monitored by extracting spectral features from measured spectra of the actual process or system. The spectral features are then input into said neural network to determine the physical condition of the system or process represented by the mathematical. More specifically, the neural network correlates the spectral features (i.e. measurable phenomena) of the actual process or system with the corresponding model input parameters. The model input parameters relate to specific components of the system or process, and, consequently, correspond to the physical condition of the process or system.

Video Salient Object Detection via Fully Convolutional Networks.

PubMed

Wang, Wenguan; Shen, Jianbing; Shao, Ling

This paper proposes a deep learning model to efficiently detect salient regions in videos. It addresses two important issues: 1) deep video saliency model training with the absence of sufficiently large and pixel-wise annotated video data and 2) fast video saliency training and detection. The proposed deep video saliency network consists of two modules, for capturing the spatial and temporal saliency information, respectively. The dynamic saliency model, explicitly incorporating saliency estimates from the static saliency model, directly produces spatiotemporal saliency inference without time-consuming optical flow computation. We further propose a novel data augmentation technique that simulates video training data from existing annotated image data sets, which enables our network to learn diverse saliency information and prevents overfitting with the limited number of training videos. Leveraging our synthetic video data (150K video sequences) and real videos, our deep video saliency model successfully learns both spatial and temporal saliency cues, thus producing accurate spatiotemporal saliency estimate. We advance the state-of-the-art on the densely annotated video segmentation data set (MAE of .06) and the Freiburg-Berkeley Motion Segmentation data set (MAE of .07), and do so with much improved speed (2 fps with all steps).This paper proposes a deep learning model to efficiently detect salient regions in videos. It addresses two important issues: 1) deep video saliency model training with the absence of sufficiently large and pixel-wise annotated video data and 2) fast video saliency training and detection. The proposed deep video saliency network consists of two modules, for capturing the spatial and temporal saliency information, respectively. The dynamic saliency model, explicitly incorporating saliency estimates from the static saliency model, directly produces spatiotemporal saliency inference without time-consuming optical flow computation. We further propose a novel data augmentation technique that simulates video training data from existing annotated image data sets, which enables our network to learn diverse saliency information and prevents overfitting with the limited number of training videos. Leveraging our synthetic video data (150K video sequences) and real videos, our deep video saliency model successfully learns both spatial and temporal saliency cues, thus producing accurate spatiotemporal saliency estimate. We advance the state-of-the-art on the densely annotated video segmentation data set (MAE of .06) and the Freiburg-Berkeley Motion Segmentation data set (MAE of .07), and do so with much improved speed (2 fps with all steps).

Seismic waveform inversion using neural networks

NASA Astrophysics Data System (ADS)

De Wit, R. W.; Trampert, J.

2012-12-01

Full waveform tomography aims to extract all available information on Earth structure and seismic sources from seismograms. The strongly non-linear nature of this inverse problem is often addressed through simplifying assumptions for the physical theory or data selection, thus potentially neglecting valuable information. Furthermore, the assessment of the quality of the inferred model is often lacking. This calls for the development of methods that fully appreciate the non-linear nature of the inverse problem, whilst providing a quantification of the uncertainties in the final model. We propose to invert seismic waveforms in a fully non-linear way by using artificial neural networks. Neural networks can be viewed as powerful and flexible non-linear filters. They are very common in speech, handwriting and pattern recognition. Mixture Density Networks (MDN) allow us to obtain marginal posterior probability density functions (pdfs) of all model parameters, conditioned on the data. An MDN can approximate an arbitrary conditional pdf as a linear combination of Gaussian kernels. Seismograms serve as input, Earth structure parameters are the so-called targets and network training aims to learn the relationship between input and targets. The network is trained on a large synthetic data set, which we construct by drawing many random Earth models from a prior model pdf and solving the forward problem for each of these models, thus generating synthetic seismograms. As a first step, we aim to construct a 1D Earth model. Training sets are constructed using the Mineos package, which computes synthetic seismograms in a spherically symmetric non-rotating Earth by summing normal modes. We train a network on the body waveforms present in these seismograms. Once the network has been trained, it can be presented with new unseen input data, in our case the body waves in real seismograms. We thus obtain the posterior pdf which represents our final state of knowledge given the information in the training set and the real data.

Student beats the teacher: deep neural networks for lateral ventricles segmentation in brain MR

NASA Astrophysics Data System (ADS)

Ghafoorian, Mohsen; Teuwen, Jonas; Manniesing, Rashindra; Leeuw, Frank-Erik d.; van Ginneken, Bram; Karssemeijer, Nico; Platel, Bram

2018-03-01

Ventricular volume and its progression are known to be linked to several brain diseases such as dementia and schizophrenia. Therefore accurate measurement of ventricle volume is vital for longitudinal studies on these disorders, making automated ventricle segmentation algorithms desirable. In the past few years, deep neural networks have shown to outperform the classical models in many imaging domains. However, the success of deep networks is dependent on manually labeled data sets, which are expensive to acquire especially for higher dimensional data in the medical domain. In this work, we show that deep neural networks can be trained on muchcheaper-to-acquire pseudo-labels (e.g., generated by other automated less accurate methods) and still produce more accurate segmentations compared to the quality of the labels. To show this, we use noisy segmentation labels generated by a conventional region growing algorithm to train a deep network for lateral ventricle segmentation. Then on a large manually annotated test set, we show that the network significantly outperforms the conventional region growing algorithm which was used to produce the training labels for the network. Our experiments report a Dice Similarity Coefficient (DSC) of 0.874 for the trained network compared to 0.754 for the conventional region growing algorithm (p < 0.001).

Android malware detection based on evolutionary super-network

NASA Astrophysics Data System (ADS)

Yan, Haisheng; Peng, Lingling

2018-04-01

In the paper, an android malware detection method based on evolutionary super-network is proposed in order to improve the precision of android malware detection. Chi square statistics method is used for selecting characteristics on the basis of analyzing android authority. Boolean weighting is utilized for calculating characteristic weight. Processed characteristic vector is regarded as the system training set and test set; hyper edge alternative strategy is used for training super-network classification model, thereby classifying test set characteristic vectors, and it is compared with traditional classification algorithm. The results show that the detection method proposed in the paper is close to or better than traditional classification algorithm. The proposed method belongs to an effective Android malware detection means.

Query-based learning for aerospace applications.

PubMed

Saad, E W; Choi, J J; Vian, J L; Wunsch, D C Ii

2003-01-01

Models of real-world applications often include a large number of parameters with a wide dynamic range, which contributes to the difficulties of neural network training. Creating the training data set for such applications becomes costly, if not impossible. In order to overcome the challenge, one can employ an active learning technique known as query-based learning (QBL) to add performance-critical data to the training set during the learning phase, thereby efficiently improving the overall learning/generalization. The performance-critical data can be obtained using an inverse mapping called network inversion (discrete network inversion and continuous network inversion) followed by oracle query. This paper investigates the use of both inversion techniques for QBL learning, and introduces an original heuristic to select the inversion target values for continuous network inversion method. Efficiency and generalization was further enhanced by employing node decoupled extended Kalman filter (NDEKF) training and a causality index (CI) as a means to reduce the input search dimensionality. The benefits of the overall QBL approach are experimentally demonstrated in two aerospace applications: a classification problem with large input space and a control distribution problem.

CONEDEP: COnvolutional Neural network based Earthquake DEtection and Phase Picking

NASA Astrophysics Data System (ADS)

Zhou, Y.; Huang, Y.; Yue, H.; Zhou, S.; An, S.; Yun, N.

2017-12-01

We developed an automatic local earthquake detection and phase picking algorithm based on Fully Convolutional Neural network (FCN). The FCN algorithm detects and segments certain features (phases) in 3 component seismograms to realize efficient picking. We use STA/LTA algorithm and template matching algorithm to construct the training set from seismograms recorded 1 month before and after the Wenchuan earthquake. Precise P and S phases are identified and labeled to construct the training set. Noise data are produced by combining back-ground noise and artificial synthetic noise to form the equivalent scale of noise set as the signal set. Training is performed on GPUs to achieve efficient convergence. Our algorithm has significantly improved performance in terms of the detection rate and precision in comparison with STA/LTA and template matching algorithms.

Assessment of a National Network: The Case of the French Teacher Training Colleges' Health Education Network

ERIC Educational Resources Information Center

Guevel, Marie-Renee; Jourdan, Didier

2009-01-01

The French teacher training colleges' health education (HE) network was set up in 2005 to encourage the inclusion of HE in courses for primary and secondary school teachers. A systematic process of monitoring the activity and the impact of this initiative was implemented. This analysis was systematically compared with the perceptions of teaching…

Application of the clinical matrix to the diagnosis of leukemia

NASA Astrophysics Data System (ADS)

Pakkala, Sampath Y.; Lin, Frank C.

1992-07-01

A system for diagnosing leukemia subtypes has been formulated using neural networks. The statistical data of the symptoms collected by hematologists is fed into a single training set using a neural network, where the network is trained by using fast backpropagation algorithm, which when done can help the general practitioners for making diagnoses on the basis of signs and symptoms alone.

Neural network modelling of planform geometry of headland-bay beaches

NASA Astrophysics Data System (ADS)

Iglesias, G.; López, I.; Castro, A.; Carballo, R.

2009-02-01

The shoreline of beaches in the lee of coastal salients or man-made structures, usually known as headland-bay beaches, has a distinctive curvature; wave fronts curve as a result of wave diffraction at the headland and in turn cause the shoreline to bend. The ensuing curved planform is of great interest both as a peculiar landform and in the context of engineering projects in which it is necessary to predict how a coastal structure will affect the sandy shoreline in its lee. A number of empirical models have been put forward, each based on a specific equation. A novel approach, based on the application of artificial neural networks, is presented in this work. Unlike the conventional method, no particular equation of the planform is embedded in the model. Instead, it is the model itself that learns about the problem from a series of examples of headland-bay beaches (the training set) and thereafter applies this self-acquired knowledge to other cases (the test set) for validation. Twenty-three headland-bay beaches from around the world were selected, of which sixteen and seven make up the training and test sets, respectively. As there is no well-developed theory for deciding upon the most convenient neural network architecture to deal with a particular data set, an experimental study was conducted in which ten different architectures with one and two hidden neuron layers and five training algorithms - 50 different options combining network architecture and training algorithm - were compared. Each of these options was implemented, trained and tested in order to find the best-performing approach for modelling the planform of headland-bay beaches. Finally, the selected neural network model was compared with a state-of-the-art planform model and was shown to outperform it.

Heuristic pattern correction scheme using adaptively trained generalized regression neural networks.

PubMed

Hoya, T; Chambers, J A

2001-01-01

In many pattern classification problems, an intelligent neural system is required which can learn the newly encountered but misclassified patterns incrementally, while keeping a good classification performance over the past patterns stored in the network. In the paper, an heuristic pattern correction scheme is proposed using adaptively trained generalized regression neural networks (GRNNs). The scheme is based upon both network growing and dual-stage shrinking mechanisms. In the network growing phase, a subset of the misclassified patterns in each incoming data set is iteratively added into the network until all the patterns in the incoming data set are classified correctly. Then, the redundancy in the growing phase is removed in the dual-stage network shrinking. Both long- and short-term memory models are considered in the network shrinking, which are motivated from biological study of the brain. The learning capability of the proposed scheme is investigated through extensive simulation studies.

Network-centric decision architecture for financial or 1/f data models

NASA Astrophysics Data System (ADS)

Jaenisch, Holger M.; Handley, James W.; Massey, Stoney; Case, Carl T.; Songy, Claude G.

2002-12-01

This paper presents a decision architecture algorithm for training neural equation based networks to make autonomous multi-goal oriented, multi-class decisions. These architectures make decisions based on their individual goals and draw from the same network centric feature set. Traditionally, these architectures are comprised of neural networks that offer marginal performance due to lack of convergence of the training set. We present an approach for autonomously extracting sample points as I/O exemplars for generation of multi-branch, multi-node decision architectures populated by adaptively derived neural equations. To test the robustness of this architecture, open source data sets in the form of financial time series were used, requiring a three-class decision space analogous to the lethal, non-lethal, and clutter discrimination problem. This algorithm and the results of its application are presented here.

Classification of mineral deposits into types using mineralogy with a probabilistic neural network

USGS Publications Warehouse

Singer, Donald A.; Kouda, Ryoichi

1997-01-01

In order to determine whether it is desirable to quantify mineral-deposit models further, a test of the ability of a probabilistic neural network to classify deposits into types based on mineralogy was conducted. Presence or absence of ore and alteration mineralogy in well-typed deposits were used to train the network. To reduce the number of minerals considered, the analyzed data were restricted to minerals present in at least 20% of at least one deposit type. An advantage of this restriction is that single or rare occurrences of minerals did not dominate the results. Probabilistic neural networks can provide mathematically sound confidence measures based on Bayes theorem and are relatively insensitive to outliers. Founded on Parzen density estimation, they require no assumptions about distributions of random variables used for classification, even handling multimodal distributions. They train quickly and work as well as, or better than, multiple-layer feedforward networks. Tests were performed with a probabilistic neural network employing a Gaussian kernel and separate sigma weights for each class and each variable. The training set was reduced to the presence or absence of 58 reported minerals in eight deposit types. The training set included: 49 Cyprus massive sulfide deposits; 200 kuroko massive sulfide deposits; 59 Comstock epithermal vein gold districts; 17 quartzalunite epithermal gold deposits; 25 Creede epithermal gold deposits; 28 sedimentary-exhalative zinc-lead deposits; 28 Sado epithermal vein gold deposits; and 100 porphyry copper deposits. The most common training problem was the error of classifying about 27% of Cyprus-type deposits in the training set as kuroko. In independent tests with deposits not used in the training set, 88% of 224 kuroko massive sulfide deposits were classed correctly, 92% of 25 porphyry copper deposits, 78% of 9 Comstock epithermal gold-silver districts, and 83% of six quartzalunite epithermal gold deposits were classed correctly. Across all deposit types, 88% of deposits in the validation dataset were correctly classed. Misclassifications were most common if a deposit was characterized by only a few minerals, e.g., pyrite, chalcopyrite,and sphalerite. The success rate jumped to 98% correctly classed deposits when just two rock types were added. Such a high success rate of the probabilistic neural network suggests that not only should this preliminary test be expanded to include other deposit types, but that other deposit features should be added.

A Regularizer Approach for RBF Networks Under the Concurrent Weight Failure Situation.

PubMed

Leung, Chi-Sing; Wan, Wai Yan; Feng, Ruibin

2017-06-01

Many existing results on fault-tolerant algorithms focus on the single fault source situation, where a trained network is affected by one kind of weight failure. In fact, a trained network may be affected by multiple kinds of weight failure. This paper first studies how the open weight fault and the multiplicative weight noise degrade the performance of radial basis function (RBF) networks. Afterward, we define the objective function for training fault-tolerant RBF networks. Based on the objective function, we then develop two learning algorithms, one batch mode and one online mode. Besides, the convergent conditions of our online algorithm are investigated. Finally, we develop a formula to estimate the test set error of faulty networks trained from our approach. This formula helps us to optimize some tuning parameters, such as RBF width.

Detecting central fixation by means of artificial neural networks in a pediatric vision screener using retinal birefringence scanning.

PubMed

Gramatikov, Boris I

2017-04-27

Reliable detection of central fixation and eye alignment is essential in the diagnosis of amblyopia ("lazy eye"), which can lead to blindness. Our lab has developed and reported earlier a pediatric vision screener that performs scanning of the retina around the fovea and analyzes changes in the polarization state of light as the scan progresses. Depending on the direction of gaze and the instrument design, the screener produces several signal frequencies that can be utilized in the detection of central fixation. The objective of this study was to compare artificial neural networks with classical statistical methods, with respect to their ability to detect central fixation reliably. A classical feedforward, pattern recognition, two-layer neural network architecture was used, consisting of one hidden layer and one output layer. The network has four inputs, representing normalized spectral powers at four signal frequencies generated during retinal birefringence scanning. The hidden layer contains four neurons. The output suggests presence or absence of central fixation. Backpropagation was used to train the network, using the gradient descent algorithm and the cross-entropy error as the performance function. The network was trained, validated and tested on a set of controlled calibration data obtained from 600 measurements from ten eyes in a previous study, and was additionally tested on a clinical set of 78 eyes, independently diagnosed by an ophthalmologist. In the first part of this study, a neural network was designed around the calibration set. With a proper architecture and training, the network provided performance that was comparable to classical statistical methods, allowing perfect separation between the central and paracentral fixation data, with both the sensitivity and the specificity of the instrument being 100%. In the second part of the study, the neural network was applied to the clinical data. It allowed reliable separation between normal subjects and affected subjects, its accuracy again matching that of the statistical methods. With a proper choice of a neural network architecture and a good, uncontaminated training data set, the artificial neural network can be an efficient classification tool for detecting central fixation based on retinal birefringence scanning.

Training artificial neural networks directly on the concordance index for censored data using genetic algorithms.

PubMed

Kalderstam, Jonas; Edén, Patrik; Bendahl, Pär-Ola; Strand, Carina; Fernö, Mårten; Ohlsson, Mattias

2013-06-01

The concordance index (c-index) is the standard way of evaluating the performance of prognostic models in the presence of censored data. Constructing prognostic models using artificial neural networks (ANNs) is commonly done by training on error functions which are modified versions of the c-index. Our objective was to demonstrate the capability of training directly on the c-index and to evaluate our approach compared to the Cox proportional hazards model. We constructed a prognostic model using an ensemble of ANNs which were trained using a genetic algorithm. The individual networks were trained on a non-linear artificial data set divided into a training and test set both of size 2000, where 50% of the data was censored. The ANNs were also trained on a data set consisting of 4042 patients treated for breast cancer spread over five different medical studies, 2/3 used for training and 1/3 used as a test set. A Cox model was also constructed on the same data in both cases. The two models' c-indices on the test sets were then compared. The ranking performance of the models is additionally presented visually using modified scatter plots. Cross validation on the cancer training set did not indicate any non-linear effects between the covariates. An ensemble of 30 ANNs with one hidden neuron was therefore used. The ANN model had almost the same c-index score as the Cox model (c-index=0.70 and 0.71, respectively) on the cancer test set. Both models identified similarly sized low risk groups with at most 10% false positives, 49 for the ANN model and 60 for the Cox model, but repeated bootstrap runs indicate that the difference was not significant. A significant difference could however be seen when applied on the non-linear synthetic data set. In that case the ANN ensemble managed to achieve a c-index score of 0.90 whereas the Cox model failed to distinguish itself from the random case (c-index=0.49). We have found empirical evidence that ensembles of ANN models can be optimized directly on the c-index. Comparison with a Cox model indicates that near identical performance is achieved on a real cancer data set while on a non-linear data set the ANN model is clearly superior. Copyright © 2013 Elsevier B.V. All rights reserved.

Three learning phases for radial-basis-function networks.

PubMed

Schwenker, F; Kestler, H A; Palm, G

2001-05-01

In this paper, learning algorithms for radial basis function (RBF) networks are discussed. Whereas multilayer perceptrons (MLP) are typically trained with backpropagation algorithms, starting the training procedure with a random initialization of the MLP's parameters, an RBF network may be trained in many different ways. We categorize these RBF training methods into one-, two-, and three-phase learning schemes. Two-phase RBF learning is a very common learning scheme. The two layers of an RBF network are learnt separately; first the RBF layer is trained, including the adaptation of centers and scaling parameters, and then the weights of the output layer are adapted. RBF centers may be trained by clustering, vector quantization and classification tree algorithms, and the output layer by supervised learning (through gradient descent or pseudo inverse solution). Results from numerical experiments of RBF classifiers trained by two-phase learning are presented in three completely different pattern recognition applications: (a) the classification of 3D visual objects; (b) the recognition hand-written digits (2D objects); and (c) the categorization of high-resolution electrocardiograms given as a time series (ID objects) and as a set of features extracted from these time series. In these applications, it can be observed that the performance of RBF classifiers trained with two-phase learning can be improved through a third backpropagation-like training phase of the RBF network, adapting the whole set of parameters (RBF centers, scaling parameters, and output layer weights) simultaneously. This, we call three-phase learning in RBF networks. A practical advantage of two- and three-phase learning in RBF networks is the possibility to use unlabeled training data for the first training phase. Support vector (SV) learning in RBF networks is a different learning approach. SV learning can be considered, in this context of learning, as a special type of one-phase learning, where only the output layer weights of the RBF network are calculated, and the RBF centers are restricted to be a subset of the training data. Numerical experiments with several classifier schemes including k-nearest-neighbor, learning vector quantization and RBF classifiers trained through two-phase, three-phase and support vector learning are given. The performance of the RBF classifiers trained through SV learning and three-phase learning are superior to the results of two-phase learning, but SV learning often leads to complex network structures, since the number of support vectors is not a small fraction of the total number of data points.

Neural network approach to proximity effect corrections in electron-beam lithography

NASA Astrophysics Data System (ADS)

Frye, Robert C.; Cummings, Kevin D.; Rietman, Edward A.

1990-05-01

The proximity effect, caused by electron beam backscattering during resist exposure, is an important concern in writing submicron features. It can be compensated by appropriate local changes in the incident beam dose, but computation of the optimal correction usually requires a prohibitively long time. We present an example of such a computation on a small test pattern, which we performed by an iterative method. We then used this solution as a training set for an adaptive neural network. After training, the network computed the same correction as the iterative method, but in a much shorter time. Correcting the image with a software based neural network resulted in a decrease in the computation time by a factor of 30, and a hardware based network enhanced the computation speed by more than a factor of 1000. Both methods had an acceptably small error of 0.5% compared to the results of the iterative computation. Additionally, we verified that the neural network correctly generalized the solution of the problem to include patterns not contained in its training set.

Edge detection for optical synthetic aperture based on deep neural network

NASA Astrophysics Data System (ADS)

Tan, Wenjie; Hui, Mei; Liu, Ming; Kong, Lingqin; Dong, Liquan; Zhao, Yuejin

2017-09-01

Synthetic aperture optics systems can meet the demands of the next-generation space telescopes being lighter, larger and foldable. However, the boundaries of segmented aperture systems are much more complex than that of the whole aperture. More edge regions mean more imaging edge pixels, which are often mixed and discretized. In order to achieve high-resolution imaging, it is necessary to identify the gaps between the sub-apertures and the edges of the projected fringes. In this work, we introduced the algorithm of Deep Neural Network into the edge detection of optical synthetic aperture imaging. According to the detection needs, we constructed image sets by experiments and simulations. Based on MatConvNet, a toolbox of MATLAB, we ran the neural network, trained it on training image set and tested its performance on validation set. The training was stopped when the test error on validation set stopped declining. As an input image is given, each intra-neighbor area around the pixel is taken into the network, and scanned pixel by pixel with the trained multi-hidden layers. The network outputs make a judgment on whether the center of the input block is on edge of fringes. We experimented with various pre-processing and post-processing techniques to reveal their influence on edge detection performance. Compared with the traditional algorithms or their improvements, our method makes decision on a much larger intra-neighbor, and is more global and comprehensive. Experiments on more than 2,000 images are also given to prove that our method outperforms classical algorithms in optical images-based edge detection.

A Risk Stratification Model for Lung Cancer Based on Gene Coexpression Network and Deep Learning

PubMed Central

2018-01-01

Risk stratification model for lung cancer with gene expression profile is of great interest. Instead of previous models based on individual prognostic genes, we aimed to develop a novel system-level risk stratification model for lung adenocarcinoma based on gene coexpression network. Using multiple microarray, gene coexpression network analysis was performed to identify survival-related networks. A deep learning based risk stratification model was constructed with representative genes of these networks. The model was validated in two test sets. Survival analysis was performed using the output of the model to evaluate whether it could predict patients' survival independent of clinicopathological variables. Five networks were significantly associated with patients' survival. Considering prognostic significance and representativeness, genes of the two survival-related networks were selected for input of the model. The output of the model was significantly associated with patients' survival in two test sets and training set (p < 0.00001, p < 0.0001 and p = 0.02 for training and test sets 1 and 2, resp.). In multivariate analyses, the model was associated with patients' prognosis independent of other clinicopathological features. Our study presents a new perspective on incorporating gene coexpression networks into the gene expression signature and clinical application of deep learning in genomic data science for prognosis prediction. PMID:29581968

Classification capacity of a modular neural network implementing neurally inspired architecture and training rules.

PubMed

Poirazi, Panayiota; Neocleous, Costas; Pattichis, Costantinos S; Schizas, Christos N

2004-05-01

A three-layer neural network (NN) with novel adaptive architecture has been developed. The hidden layer of the network consists of slabs of single neuron models, where neurons within a slab--but not between slabs--have the same type of activation function. The network activation functions in all three layers have adaptable parameters. The network was trained using a biologically inspired, guided-annealing learning rule on a variety of medical data. Good training/testing classification performance was obtained on all data sets tested. The performance achieved was comparable to that of SVM classifiers. It was shown that the adaptive network architecture, inspired from the modular organization often encountered in the mammalian cerebral cortex, can benefit classification performance.

A Software Package for Neural Network Applications Development

NASA Technical Reports Server (NTRS)

Baran, Robert H.

1993-01-01

Original Backprop (Version 1.2) is an MS-DOS package of four stand-alone C-language programs that enable users to develop neural network solutions to a variety of practical problems. Original Backprop generates three-layer, feed-forward (series-coupled) networks which map fixed-length input vectors into fixed length output vectors through an intermediate (hidden) layer of binary threshold units. Version 1.2 can handle up to 200 input vectors at a time, each having up to 128 real-valued components. The first subprogram, TSET, appends a number (up to 16) of classification bits to each input, thus creating a training set of input output pairs. The second subprogram, BACKPROP, creates a trilayer network to do the prescribed mapping and modifies the weights of its connections incrementally until the training set is leaned. The learning algorithm is the 'back-propagating error correction procedures first described by F. Rosenblatt in 1961. The third subprogram, VIEWNET, lets the trained network be examined, tested, and 'pruned' (by the deletion of unnecessary hidden units). The fourth subprogram, DONET, makes a TSR routine by which the finished product of the neural net design-and-training exercise can be consulted under other MS-DOS applications.

Damage Detection Using Holography and Interferometry

NASA Technical Reports Server (NTRS)

Decker, Arthur J.

2003-01-01

This paper reviews classical approaches to damage detection using laser holography and interferometry. The paper then details the modern uses of electronic holography and neural-net-processed characteristic patterns to detect structural damage. The design of the neural networks and the preparation of the training sets are discussed. The use of a technique to optimize the training sets, called folding, is explained. Then a training procedure is detailed that uses the holography-measured vibration modes of the undamaged structures to impart damage-detection sensitivity to the neural networks. The inspections of an optical strain gauge mounting plate and an International Space Station cold plate are presented as examples.

Sea ice classification using fast learning neural networks

NASA Technical Reports Server (NTRS)

Dawson, M. S.; Fung, A. K.; Manry, M. T.

1992-01-01

A first learning neural network approach to the classification of sea ice is presented. The fast learning (FL) neural network and a multilayer perceptron (MLP) trained with backpropagation learning (BP network) were tested on simulated data sets based on the known dominant scattering characteristics of the target class. Four classes were used in the data simulation: open water, thick lossy saline ice, thin saline ice, and multiyear ice. The BP network was unable to consistently converge to less than 25 percent error while the FL method yielded an average error of approximately 1 percent on the first iteration of training. The fast learning method presented can significantly reduce the CPU time necessary to train a neural network as well as consistently yield higher classification accuracy than BP networks.

Classification of urine sediment based on convolution neural network

NASA Astrophysics Data System (ADS)

Pan, Jingjing; Jiang, Cunbo; Zhu, Tiantian

2018-04-01

By designing a new convolution neural network framework, this paper breaks the constraints of the original convolution neural network framework requiring large training samples and samples of the same size. Move and cropping the input images, generate the same size of the sub-graph. And then, the generated sub-graph uses the method of dropout, increasing the diversity of samples and preventing the fitting generation. Randomly select some proper subset in the sub-graphic set and ensure that the number of elements in the proper subset is same and the proper subset is not the same. The proper subsets are used as input layers for the convolution neural network. Through the convolution layer, the pooling, the full connection layer and output layer, we can obtained the classification loss rate of test set and training set. In the red blood cells, white blood cells, calcium oxalate crystallization classification experiment, the classification accuracy rate of 97% or more.

Beyond Fine Tuning: Adding capacity to leverage few labels

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

Hodas, Nathan O.; Shaffer, Kyle J.; Yankov, Artem

2017-12-09

In this paper we present a technique to train neural network models on small amounts of data. Current methods for training neural networks on small amounts of rich data typically rely on strategies such as fine-tuning a pre-trained neural networks or the use of domain-specific hand-engineered features. Here we take the approach of treating network layers, or entire networks, as modules and combine pre-trained modules with untrained modules, to learn the shift in distributions between data sets. The central impact of using a modular approach comes from adding new representations to a network, as opposed to replacing representations via fine-tuning.more » Using this technique, we are able surpass results using standard fine-tuning transfer learning approaches, and we are also able to significantly increase performance over such approaches when using smaller amounts of data.« less

Deformable image registration using convolutional neural networks

NASA Astrophysics Data System (ADS)

Eppenhof, Koen A. J.; Lafarge, Maxime W.; Moeskops, Pim; Veta, Mitko; Pluim, Josien P. W.

2018-03-01

Deformable image registration can be time-consuming and often needs extensive parameterization to perform well on a specific application. We present a step towards a registration framework based on a three-dimensional convolutional neural network. The network directly learns transformations between pairs of three-dimensional images. The outputs of the network are three maps for the x, y, and z components of a thin plate spline transformation grid. The network is trained on synthetic random transformations, which are applied to a small set of representative images for the desired application. Training therefore does not require manually annotated ground truth deformation information. The methodology is demonstrated on public data sets of inspiration-expiration lung CT image pairs, which come with annotated corresponding landmarks for evaluation of the registration accuracy. Advantages of this methodology are its fast registration times and its minimal parameterization.

A hybrid linear/nonlinear training algorithm for feedforward neural networks.

PubMed

McLoone, S; Brown, M D; Irwin, G; Lightbody, A

1998-01-01

This paper presents a new hybrid optimization strategy for training feedforward neural networks. The algorithm combines gradient-based optimization of nonlinear weights with singular value decomposition (SVD) computation of linear weights in one integrated routine. It is described for the multilayer perceptron (MLP) and radial basis function (RBF) networks and then extended to the local model network (LMN), a new feedforward structure in which a global nonlinear model is constructed from a set of locally valid submodels. Simulation results are presented demonstrating the superiority of the new hybrid training scheme compared to second-order gradient methods. It is particularly effective for the LMN architecture where the linear to nonlinear parameter ratio is large.

Evaluation of a parallel implementation of the learning portion of the backward error propagation neural network: experiments in artifact identification.

PubMed Central

Sittig, D. F.; Orr, J. A.

1991-01-01

Various methods have been proposed in an attempt to solve problems in artifact and/or alarm identification including expert systems, statistical signal processing techniques, and artificial neural networks (ANN). ANNs consist of a large number of simple processing units connected by weighted links. To develop truly robust ANNs, investigators are required to train their networks on huge training data sets, requiring enormous computing power. We implemented a parallel version of the backward error propagation neural network training algorithm in the widely portable parallel programming language C-Linda. A maximum speedup of 4.06 was obtained with six processors. This speedup represents a reduction in total run-time from approximately 6.4 hours to 1.5 hours. We conclude that use of the master-worker model of parallel computation is an excellent method for obtaining speedups in the backward error propagation neural network training algorithm. PMID:1807607

Whole-Genome DNA Methylation Status Associated with Clinical PTSD Measures of OIF/OEF Veterans

DTIC Science & Technology

Emerging knowledge suggests that post -traumatic stress disorder (PTSD) pathophysiology is linked to the patients epigenetic changes, but...promoter-bound CpGIs to identify networks related to PTSD. The identified networks were further validated by an independent test set comprising 31 PTSD /29...set. To improve the statistical power and mitigate the assay bias and batch effects, a union set combining both training and test set was assayed

An analysis of image storage systems for scalable training of deep neural networks

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

Lim, Seung-Hwan; Young, Steven R; Patton, Robert M

This study presents a principled empirical evaluation of image storage systems for training deep neural networks. We employ the Caffe deep learning framework to train neural network models for three different data sets, MNIST, CIFAR-10, and ImageNet. While training the models, we evaluate five different options to retrieve training image data: (1) PNG-formatted image files on local file system; (2) pushing pixel arrays from image files into a single HDF5 file on local file system; (3) in-memory arrays to hold the pixel arrays in Python and C++; (4) loading the training data into LevelDB, a log-structured merge tree based key-valuemore » storage; and (5) loading the training data into LMDB, a B+tree based key-value storage. The experimental results quantitatively highlight the disadvantage of using normal image files on local file systems to train deep neural networks and demonstrate reliable performance with key-value storage based storage systems. When training a model on the ImageNet dataset, the image file option was more than 17 times slower than the key-value storage option. Along with measurements on training time, this study provides in-depth analysis on the cause of performance advantages/disadvantages of each back-end to train deep neural networks. We envision the provided measurements and analysis will shed light on the optimal way to architect systems for training neural networks in a scalable manner.« less

MABAL: a Novel Deep-Learning Architecture for Machine-Assisted Bone Age Labeling.

PubMed

Mutasa, Simukayi; Chang, Peter D; Ruzal-Shapiro, Carrie; Ayyala, Rama

2018-02-05

Bone age assessment (BAA) is a commonly performed diagnostic study in pediatric radiology to assess skeletal maturity. The most commonly utilized method for assessment of BAA is the Greulich and Pyle method (Pediatr Radiol 46.9:1269-1274, 2016; Arch Dis Child 81.2:172-173, 1999) atlas. The evaluation of BAA can be a tedious and time-consuming process for the radiologist. As such, several computer-assisted detection/diagnosis (CAD) methods have been proposed for automation of BAA. Classical CAD tools have traditionally relied on hard-coded algorithmic features for BAA which suffer from a variety of drawbacks. Recently, the advent and proliferation of convolutional neural networks (CNNs) has shown promise in a variety of medical imaging applications. There have been at least two published applications of using deep learning for evaluation of bone age (Med Image Anal 36:41-51, 2017; JDI 1-5, 2017). However, current implementations are limited by a combination of both architecture design and relatively small datasets. The purpose of this study is to demonstrate the benefits of a customized neural network algorithm carefully calibrated to the evaluation of bone age utilizing a relatively large institutional dataset. In doing so, this study will aim to show that advanced architectures can be successfully trained from scratch in the medical imaging domain and can generate results that outperform any existing proposed algorithm. The training data consisted of 10,289 images of different skeletal age examinations, 8909 from the hospital Picture Archiving and Communication System at our institution and 1383 from the public Digital Hand Atlas Database. The data was separated into four cohorts, one each for male and female children above the age of 8, and one each for male and female children below the age of 10. The testing set consisted of 20 radiographs of each 1-year-age cohort from 0 to 1 years to 14-15+ years, half male and half female. The testing set included left-hand radiographs done for bone age assessment, trauma evaluation without significant findings, and skeletal surveys. A 14 hidden layer-customized neural network was designed for this study. The network included several state of the art techniques including residual-style connections, inception layers, and spatial transformer layers. Data augmentation was applied to the network inputs to prevent overfitting. A linear regression output was utilized. Mean square error was used as the network loss function and mean absolute error (MAE) was utilized as the primary performance metric. MAE accuracies on the validation and test sets for young females were 0.654 and 0.561 respectively. For older females, validation and test accuracies were 0.662 and 0.497 respectively. For young males, validation and test accuracies were 0.649 and 0.585 respectively. Finally, for older males, validation and test set accuracies were 0.581 and 0.501 respectively. The female cohorts were trained for 900 epochs each and the male cohorts were trained for 600 epochs. An eightfold cross-validation set was employed for hyperparameter tuning. Test error was obtained after training on a full data set with the selected hyperparameters. Using our proposed customized neural network architecture on our large available data, we achieved an aggregate validation and test set mean absolute errors of 0.637 and 0.536 respectively. To date, this is the best published performance on utilizing deep learning for bone age assessment. Our results support our initial hypothesis that customized, purpose-built neural networks provide improved performance over networks derived from pre-trained imaging data sets. We build on that initial work by showing that the addition of state-of-the-art techniques such as residual connections and inception architecture further improves prediction accuracy. This is important because the current assumption for use of residual and/or inception architectures is that a large pre-trained network is required for successful implementation given the relatively small datasets in medical imaging. Instead we show that a small, customized architecture incorporating advanced CNN strategies can indeed be trained from scratch, yielding significant improvements in algorithm accuracy. It should be noted that for all four cohorts, testing error outperformed validation error. One reason for this is that our ground truth for our test set was obtained by averaging two pediatric radiologist reads compared to our training data for which only a single read was used. This suggests that despite relatively noisy training data, the algorithm could successfully model the variation between observers and generate estimates that are close to the expected ground truth.

Interacting neural networks.

PubMed

Metzler, R; Kinzel, W; Kanter, I

2000-08-01

Several scenarios of interacting neural networks which are trained either in an identical or in a competitive way are solved analytically. In the case of identical training each perceptron receives the output of its neighbor. The symmetry of the stationary state as well as the sensitivity to the used training algorithm are investigated. Two competitive perceptrons trained on mutually exclusive learning aims and a perceptron which is trained on the opposite of its own output are examined analytically. An ensemble of competitive perceptrons is used as decision-making algorithms in a model of a closed market (El Farol Bar problem or the Minority Game. In this game, a set of agents who have to make a binary decision is considered.); each network is trained on the history of minority decisions. This ensemble of perceptrons relaxes to a stationary state whose performance can be better than random.

Interacting neural networks

NASA Astrophysics Data System (ADS)

Metzler, R.; Kinzel, W.; Kanter, I.

2000-08-01

Several scenarios of interacting neural networks which are trained either in an identical or in a competitive way are solved analytically. In the case of identical training each perceptron receives the output of its neighbor. The symmetry of the stationary state as well as the sensitivity to the used training algorithm are investigated. Two competitive perceptrons trained on mutually exclusive learning aims and a perceptron which is trained on the opposite of its own output are examined analytically. An ensemble of competitive perceptrons is used as decision-making algorithms in a model of a closed market (El Farol Bar problem or the Minority Game. In this game, a set of agents who have to make a binary decision is considered.); each network is trained on the history of minority decisions. This ensemble of perceptrons relaxes to a stationary state whose performance can be better than random.

A frequency-domain approach to improve ANNs generalization quality via proper initialization.

PubMed

Chaari, Majdi; Fekih, Afef; Seibi, Abdennour C; Hmida, Jalel Ben

2018-08-01

The ability to train a network without memorizing the input/output data, thereby allowing a good predictive performance when applied to unseen data, is paramount in ANN applications. In this paper, we propose a frequency-domain approach to evaluate the network initialization in terms of quality of training, i.e., generalization capabilities. As an alternative to the conventional time-domain methods, the proposed approach eliminates the approximate nature of network validation using an excess of unseen data. The benefits of the proposed approach are demonstrated using two numerical examples, where two trained networks performed similarly on the training and the validation data sets, yet they revealed a significant difference in prediction accuracy when tested using a different data set. This observation is of utmost importance in modeling applications requiring a high degree of accuracy. The efficiency of the proposed approach is further demonstrated on a real-world problem, where unlike other initialization methods, a more conclusive assessment of generalization is achieved. On the practical front, subtle methodological and implementational facets are addressed to ensure reproducibility and pinpoint the limitations of the proposed approach. Copyright © 2018 Elsevier Ltd. All rights reserved.

An Event-Driven Classifier for Spiking Neural Networks Fed with Synthetic or Dynamic Vision Sensor Data.

PubMed

Stromatias, Evangelos; Soto, Miguel; Serrano-Gotarredona, Teresa; Linares-Barranco, Bernabé

2017-01-01

This paper introduces a novel methodology for training an event-driven classifier within a Spiking Neural Network (SNN) System capable of yielding good classification results when using both synthetic input data and real data captured from Dynamic Vision Sensor (DVS) chips. The proposed supervised method uses the spiking activity provided by an arbitrary topology of prior SNN layers to build histograms and train the classifier in the frame domain using the stochastic gradient descent algorithm. In addition, this approach can cope with leaky integrate-and-fire neuron models within the SNN, a desirable feature for real-world SNN applications, where neural activation must fade away after some time in the absence of inputs. Consequently, this way of building histograms captures the dynamics of spikes immediately before the classifier. We tested our method on the MNIST data set using different synthetic encodings and real DVS sensory data sets such as N-MNIST, MNIST-DVS, and Poker-DVS using the same network topology and feature maps. We demonstrate the effectiveness of our approach by achieving the highest classification accuracy reported on the N-MNIST (97.77%) and Poker-DVS (100%) real DVS data sets to date with a spiking convolutional network. Moreover, by using the proposed method we were able to retrain the output layer of a previously reported spiking neural network and increase its performance by 2%, suggesting that the proposed classifier can be used as the output layer in works where features are extracted using unsupervised spike-based learning methods. In addition, we also analyze SNN performance figures such as total event activity and network latencies, which are relevant for eventual hardware implementations. In summary, the paper aggregates unsupervised-trained SNNs with a supervised-trained SNN classifier, combining and applying them to heterogeneous sets of benchmarks, both synthetic and from real DVS chips.

Neural Network Prediction of New Aircraft Design Coefficients

NASA Technical Reports Server (NTRS)

Norgaard, Magnus; Jorgensen, Charles C.; Ross, James C.

1997-01-01

This paper discusses a neural network tool for more effective aircraft design evaluations during wind tunnel tests. Using a hybrid neural network optimization method, we have produced fast and reliable predictions of aerodynamical coefficients, found optimal flap settings, and flap schedules. For validation, the tool was tested on a 55% scale model of the USAF/NASA Subsonic High Alpha Research Concept aircraft (SHARC). Four different networks were trained to predict coefficients of lift, drag, moment of inertia, and lift drag ratio (C(sub L), C(sub D), C(sub M), and L/D) from angle of attack and flap settings. The latter network was then used to determine an overall optimal flap setting and for finding optimal flap schedules.

Digital image classification with the help of artificial neural network by simple histogram.

PubMed

Dey, Pranab; Banerjee, Nirmalya; Kaur, Rajwant

2016-01-01

Visual image classification is a great challenge to the cytopathologist in routine day-to-day work. Artificial neural network (ANN) may be helpful in this matter. In this study, we have tried to classify digital images of malignant and benign cells in effusion cytology smear with the help of simple histogram data and ANN. A total of 404 digital images consisting of 168 benign cells and 236 malignant cells were selected for this study. The simple histogram data was extracted from these digital images and an ANN was constructed with the help of Neurointelligence software [Alyuda Neurointelligence 2.2 (577), Cupertino, California, USA]. The network architecture was 6-3-1. The images were classified as training set (281), validation set (63), and test set (60). The on-line backpropagation training algorithm was used for this study. A total of 10,000 iterations were done to train the ANN system with the speed of 609.81/s. After the adequate training of this ANN model, the system was able to identify all 34 malignant cell images and 24 out of 26 benign cells. The ANN model can be used for the identification of the individual malignant cells with the help of simple histogram data. This study will be helpful in the future to identify malignant cells in unknown situations.

Artificial Neural Networks and Instructional Technology.

ERIC Educational Resources Information Center

Carlson, Patricia A.

1991-01-01

Artificial neural networks (ANN), part of artificial intelligence, are discussed. Such networks are fed sample cases (training sets), learn how to recognize patterns in the sample data, and use this experience in handling new cases. Two cognitive roles for ANNs (intelligent filters and spreading, associative memories) are examined. Prototypes…

Coarse-coded higher-order neural networks for PSRI object recognition. [position, scale, and rotation invariant

NASA Technical Reports Server (NTRS)

Spirkovska, Lilly; Reid, Max B.

1993-01-01

A higher-order neural network (HONN) can be designed to be invariant to changes in scale, translation, and inplane rotation. Invariances are built directly into the architecture of a HONN and do not need to be learned. Consequently, fewer training passes and a smaller training set are required to learn to distinguish between objects. The size of the input field is limited, however, because of the memory required for the large number of interconnections in a fully connected HONN. By coarse coding the input image, the input field size can be increased to allow the larger input scenes required for practical object recognition problems. We describe a coarse coding technique and present simulation results illustrating its usefulness and its limitations. Our simulations show that a third-order neural network can be trained to distinguish between two objects in a 4096 x 4096 pixel input field independent of transformations in translation, in-plane rotation, and scale in less than ten passes through the training set. Furthermore, we empirically determine the limits of the coarse coding technique in the object recognition domain.

Emulation of reionization simulations for Bayesian inference of astrophysics parameters using neural networks

NASA Astrophysics Data System (ADS)

Schmit, C. J.; Pritchard, J. R.

2018-03-01

Next generation radio experiments such as LOFAR, HERA, and SKA are expected to probe the Epoch of Reionization (EoR) and claim a first direct detection of the cosmic 21cm signal within the next decade. Data volumes will be enormous and can thus potentially revolutionize our understanding of the early Universe and galaxy formation. However, numerical modelling of the EoR can be prohibitively expensive for Bayesian parameter inference and how to optimally extract information from incoming data is currently unclear. Emulation techniques for fast model evaluations have recently been proposed as a way to bypass costly simulations. We consider the use of artificial neural networks as a blind emulation technique. We study the impact of training duration and training set size on the quality of the network prediction and the resulting best-fitting values of a parameter search. A direct comparison is drawn between our emulation technique and an equivalent analysis using 21CMMC. We find good predictive capabilities of our network using training sets of as low as 100 model evaluations, which is within the capabilities of fully numerical radiative transfer codes.

Space shuttle main engine fault detection using neural networks

NASA Technical Reports Server (NTRS)

Bishop, Thomas; Greenwood, Dan; Shew, Kenneth; Stevenson, Fareed

1991-01-01

A method for on-line Space Shuttle Main Engine (SSME) anomaly detection and fault typing using a feedback neural network is described. The method involves the computation of features representing time-variance of SSME sensor parameters, using historical test case data. The network is trained, using backpropagation, to recognize a set of fault cases. The network is then able to diagnose new fault cases correctly. An essential element of the training technique is the inclusion of randomly generated data along with the real data, in order to span the entire input space of potential non-nominal data.

Application of neural networks to prediction of fish diversity and salmonid production in the Lake Ontario basin

USGS Publications Warehouse

McKenna, James E.

2005-01-01

Diversity and fish productivity are important measures of the health and status of aquatic systems. Being able to predict the values of these indices as a function of environmental variables would be valuable to management. Diversity and productivity have been related to environmental conditions by multiple linear regression and discriminant analysis, but such methods have several shortcomings. In an effort to predict fish species diversity and estimate salmonid production for streams in the eastern basin of Lake Ontario, I constructed neural networks and trained them on a data set containing abiotic information and either fish diversity or juvenile salmonid abundance. Twenty percent of the original data were retained as a test data set and used in the training. The ability to extend these neural networks to conditions throughout the streams was tested with data not involved in the network training. The resulting neural networks were able to predict the number of salmonids with more than 84% accuracy and diversity with more than 73% accuracy, which was far superior to the performance of multiple regression. The networks also identified the environmental variables with the greatest predictive power, namely, those describing water movement, stream size, and water chemistry. Thirteen input variables were used to predict diversity and 17 to predict salmonid abundance.

Reconstruction of sub-surface archaeological remains from magnetic data using neural computing.

NASA Astrophysics Data System (ADS)

Bescoby, D. J.; Cawley, G. C.; Chroston, P. N.

2003-04-01

The remains of a former Roman colonial settlement, once part of the classical city of Butrint in southern Albania have been the subject of a high resolution magnetic survey using a caesium-vapour magnetometer. The survey revealed the surviving remains of an extensive planned settlement and a number of outlying buildings, today buried beneath over 0.5 m of alluvial deposits. The aim of the current research is to derive a sub-surface model from the magnetic survey measurements, allowing an enhanced archaeological interpretation of the data. Neural computing techniques are used to perform the non-linear mapping between magnetic data and corresponding sub-surface model parameters. The adoption of neural computing paradigms potentially holds several advantages over other modelling techniques, allowing fast solutions for complex data, while having a high tolerance to noise. A multi-layer perceptron network with a feed-forward architecture is trained to estimate the shape and burial depth of wall foundations using a series of representative models as training data. Parameters used to forward model the training data sets are derived from a number of trial trench excavations targeted over features identified by the magnetic survey. The training of the network was optimized by first applying it to synthetic test data of known source parameters. Pre-processing of the network input data, including the use of a rotationally invariant transform, enhanced network performance and the efficiency of the training data. The approach provides good results when applied to real magnetic data, accurately predicting the depths and layout of wall foundations within the former settlement, verified by subsequent excavation. The resulting sub-surface model is derived from the averaged outputs of a ‘committee’ of five networks, trained with individualized training sets. Fuzzy logic inference has also been used to combine individual network outputs through correlation with data from a second geophysical technique, allowing the integration of data from a separate series of measurements.

Functional Inference of Complex Anatomical Tendinous Networks at a Macroscopic Scale via Sparse Experimentation

PubMed Central

Saxena, Anupam; Lipson, Hod; Valero-Cuevas, Francisco J.

2012-01-01

In systems and computational biology, much effort is devoted to functional identification of systems and networks at the molecular-or cellular scale. However, similarly important networks exist at anatomical scales such as the tendon network of human fingers: the complex array of collagen fibers that transmits and distributes muscle forces to finger joints. This network is critical to the versatility of the human hand, and its function has been debated since at least the 16th century. Here, we experimentally infer the structure (both topology and parameter values) of this network through sparse interrogation with force inputs. A population of models representing this structure co-evolves in simulation with a population of informative future force inputs via the predator-prey estimation-exploration algorithm. Model fitness depends on their ability to explain experimental data, while the fitness of future force inputs depends on causing maximal functional discrepancy among current models. We validate our approach by inferring two known synthetic Latex networks, and one anatomical tendon network harvested from a cadaver's middle finger. We find that functionally similar but structurally diverse models can exist within a narrow range of the training set and cross-validation errors. For the Latex networks, models with low training set error [<4%] and resembling the known network have the smallest cross-validation errors [∼5%]. The low training set [<4%] and cross validation [<7.2%] errors for models for the cadaveric specimen demonstrate what, to our knowledge, is the first experimental inference of the functional structure of complex anatomical networks. This work expands current bioinformatics inference approaches by demonstrating that sparse, yet informative interrogation of biological specimens holds significant computational advantages in accurate and efficient inference over random testing, or assuming model topology and only inferring parameters values. These findings also hold clues to both our evolutionary history and the development of versatile machines. PMID:23144601

Functional inference of complex anatomical tendinous networks at a macroscopic scale via sparse experimentation.

PubMed

Saxena, Anupam; Lipson, Hod; Valero-Cuevas, Francisco J

2012-01-01

In systems and computational biology, much effort is devoted to functional identification of systems and networks at the molecular-or cellular scale. However, similarly important networks exist at anatomical scales such as the tendon network of human fingers: the complex array of collagen fibers that transmits and distributes muscle forces to finger joints. This network is critical to the versatility of the human hand, and its function has been debated since at least the 16(th) century. Here, we experimentally infer the structure (both topology and parameter values) of this network through sparse interrogation with force inputs. A population of models representing this structure co-evolves in simulation with a population of informative future force inputs via the predator-prey estimation-exploration algorithm. Model fitness depends on their ability to explain experimental data, while the fitness of future force inputs depends on causing maximal functional discrepancy among current models. We validate our approach by inferring two known synthetic Latex networks, and one anatomical tendon network harvested from a cadaver's middle finger. We find that functionally similar but structurally diverse models can exist within a narrow range of the training set and cross-validation errors. For the Latex networks, models with low training set error [<4%] and resembling the known network have the smallest cross-validation errors [∼5%]. The low training set [<4%] and cross validation [<7.2%] errors for models for the cadaveric specimen demonstrate what, to our knowledge, is the first experimental inference of the functional structure of complex anatomical networks. This work expands current bioinformatics inference approaches by demonstrating that sparse, yet informative interrogation of biological specimens holds significant computational advantages in accurate and efficient inference over random testing, or assuming model topology and only inferring parameters values. These findings also hold clues to both our evolutionary history and the development of versatile machines.

Convolutional networks for vehicle track segmentation

NASA Astrophysics Data System (ADS)

Quach, Tu-Thach

2017-10-01

Existing methods to detect vehicle tracks in coherent change detection images, a product of combining two synthetic aperture radar images taken at different times of the same scene, rely on simple and fast models to label track pixels. These models, however, are unable to capture natural track features, such as continuity and parallelism. More powerful but computationally expensive models can be used in offline settings. We present an approach that uses dilated convolutional networks consisting of a series of 3×3 convolutions to segment vehicle tracks. The design of our networks considers the fact that remote sensing applications tend to operate in low power and have limited training data. As a result, we aim for small and efficient networks that can be trained end-to-end to learn natural track features entirely from limited training data. We demonstrate that our six-layer network, trained on just 90 images, is computationally efficient and improves the F-score on a standard dataset to 0.992, up from 0.959 obtained by the current state-of-the-art method.

Use of qualitative and quantitative information in neural networks for assessing agricultural chemical contamination of domestic wells

USGS Publications Warehouse

Mishra, A.; Ray, C.; Kolpin, D.W.

2004-01-01

A neural network analysis of agrichemical occurrence in groundwater was conducted using data from a pilot study of 192 small-diameter drilled and driven wells and 115 dug and bored wells in Illinois, a regional reconnaissance network of 303 wells across 12 Midwestern states, and a study of 687 domestic wells across Iowa. Potential factors contributing to well contamination (e.g., depth to aquifer material, well depth, and distance to cropland) were investigated. These contributing factors were available in either numeric (actual or categorical) or descriptive (yes or no) format. A method was devised to use the numeric and descriptive values simultaneously. Training of the network was conducted using a standard backpropagation algorithm. Approximately 15% of the data was used for testing. Analysis indicated that training error was quite low for most data. Testing results indicated that it was possible to predict the contamination potential of a well with pesticides. However, predicting the actual level of contamination was more difficult. For pesticide occurrence in drilled and driven wells, the network predictions were good. The performance of the network was poorer for predicting nitrate occurrence in dug and bored wells. Although the data set for Iowa was large, the prediction ability of the trained network was poor, due to descriptive or categorical input parameters, compared with smaller data sets such as that for Illinois, which contained more numeric information.

Cloud Classification in Polar and Desert Regions and Smoke Classification from Biomass Burning Using a Hierarchical Neural Network

NASA Technical Reports Server (NTRS)

Alexander, June; Corwin, Edward; Lloyd, David; Logar, Antonette; Welch, Ronald

1996-01-01

This research focuses on a new neural network scene classification technique. The task is to identify scene elements in Advanced Very High Resolution Radiometry (AVHRR) data from three scene types: polar, desert and smoke from biomass burning in South America (smoke). The ultimate goal of this research is to design and implement a computer system which will identify the clouds present on a whole-Earth satellite view as a means of tracking global climate changes. Previous research has reported results for rule-based systems (Tovinkere et at 1992, 1993) for standard back propagation (Watters et at. 1993) and for a hierarchical approach (Corwin et al 1994) for polar data. This research uses a hierarchical neural network with don't care conditions and applies this technique to complex scenes. A hierarchical neural network consists of a switching network and a collection of leaf networks. The idea of the hierarchical neural network is that it is a simpler task to classify a certain pattern from a subset of patterns than it is to classify a pattern from the entire set. Therefore, the first task is to cluster the classes into groups. The switching, or decision network, performs an initial classification by selecting a leaf network. The leaf networks contain a reduced set of similar classes, and it is in the various leaf networks that the actual classification takes place. The grouping of classes in the various leaf networks is determined by applying an iterative clustering algorithm. Several clustering algorithms were investigated, but due to the size of the data sets, the exhaustive search algorithms were eliminated. A heuristic approach using a confusion matrix from a lightly trained neural network provided the basis for the clustering algorithm. Once the clusters have been identified, the hierarchical network can be trained. The approach of using don't care nodes results from the difficulty in generating extremely complex surfaces in order to separate one class from all of the others. This approach finds pairwise separating surfaces and forms the more complex separating surface from combinations of simpler surfaces. This technique both reduces training time and improves accuracy over the previously reported results. Accuracies of 97.47%, 95.70%, and 99.05% were achieved for the polar, desert and smoke data sets.

Protein contact prediction using patterns of correlation.

PubMed

Hamilton, Nicholas; Burrage, Kevin; Ragan, Mark A; Huber, Thomas

2004-09-01

We describe a new method for using neural networks to predict residue contact pairs in a protein. The main inputs to the neural network are a set of 25 measures of correlated mutation between all pairs of residues in two "windows" of size 5 centered on the residues of interest. While the individual pair-wise correlations are a relatively weak predictor of contact, by training the network on windows of correlation the accuracy of prediction is significantly improved. The neural network is trained on a set of 100 proteins and then tested on a disjoint set of 1033 proteins of known structure. An average predictive accuracy of 21.7% is obtained taking the best L/2 predictions for each protein, where L is the sequence length. Taking the best L/10 predictions gives an average accuracy of 30.7%. The predictor is also tested on a set of 59 proteins from the CASP5 experiment. The accuracy is found to be relatively consistent across different sequence lengths, but to vary widely according to the secondary structure. Predictive accuracy is also found to improve by using multiple sequence alignments containing many sequences to calculate the correlations. Copyright 2004 Wiley-Liss, Inc.

Application of the Kohonen artificial neural network in the identification of proteinaceous binders in samples of panel painting using gas chromatography-mass spectrometry.

PubMed

Lletí, R; Sarabia, L A; Ortiz, M C; Todeschini, R; Colombini, M P

2003-03-01

Historically, three types of proteinaceous matter--casein, egg and animal glue--were used as binders for pigments or as adhesives in easel and wall painting. The relative percentage content of alanine, glycine, valine, leucine, isoleucine, serine, tyrosine, phenylalanine, aspartic acid, glutamic acid, lysine, methionine, proline and hydroxyproline, as determined by GC-MS, is used for binder identification. In this paper we analyse the viability of a multivariate modelling using Kohonen's neural network to characterise the wood adhesive in 16 old samples from Italian panel paintings of the 12-16th centuries. As a training set we use the amino acid composition of 141 samples contributed by the Opificio delle Pietre Dure of Florence (Cultural Heritage Ministry, Italy). Of the 141 samples, 113 were used to train the Kohonen neural network and the remaining 28 as the evaluation set. A specificity and sensitivity of 100% was achieved in training and 92-100% in prediction depending on the assignation criteria employed. The neural network thus trained and evaluated was applied to the old samples, achieving identification of all of them. In addition, the map obtained for each amino acid provides relevant information as to its importance in the characterisation of the sample.

Analysing the 21 cm signal from the epoch of reionization with artificial neural networks

NASA Astrophysics Data System (ADS)

Shimabukuro, Hayato; Semelin, Benoit

2017-07-01

The 21 cm signal from the epoch of reionization should be observed within the next decade. While a simple statistical detection is expected with Square Kilometre Array (SKA) pathfinders, the SKA will hopefully produce a full 3D mapping of the signal. To extract from the observed data constraints on the parameters describing the underlying astrophysical processes, inversion methods must be developed. For example, the Markov Chain Monte Carlo method has been successfully applied. Here, we test another possible inversion method: artificial neural networks (ANNs). We produce a training set that consists of 70 individual samples. Each sample is made of the 21 cm power spectrum at different redshifts produced with the 21cmFast code plus the value of three parameters used in the seminumerical simulations that describe astrophysical processes. Using this set, we train the network to minimize the error between the parameter values it produces as an output and the true values. We explore the impact of the architecture of the network on the quality of the training. Then we test the trained network on the new set of 54 test samples with different values of the parameters. We find that the quality of the parameter reconstruction depends on the sensitivity of the power spectrum to the different parameters at a given redshift, that including thermal noise and sample variance decreases the quality of the reconstruction and that using the power spectrum at several redshifts as an input to the ANN improves the quality of the reconstruction. We conclude that ANNs are a viable inversion method whose main strength is that they require a sparse exploration of the parameter space and thus should be usable with full numerical simulations.

A Neural Network Aero Design System for Advanced Turbo-Engines

NASA Technical Reports Server (NTRS)

Sanz, Jose M.

1999-01-01

An inverse design method calculates the blade shape that produces a prescribed input pressure distribution. By controlling this input pressure distribution the aerodynamic design objectives can easily be met. Because of the intrinsic relationship between pressure distribution and airfoil physical properties, a Neural Network can be trained to choose the optimal pressure distribution that would meet a set of physical requirements. Neural network systems have been attempted in the context of direct design methods. From properties ascribed to a set of blades the neural network is trained to infer the properties of an 'interpolated' blade shape. The problem is that, especially in transonic regimes where we deal with intrinsically non linear and ill posed problems, small perturbations of the blade shape can produce very large variations of the flow parameters. It is very unlikely that, under these circumstances, a neural network will be able to find the proper solution. The unique situation in the present method is that the neural network can be trained to extract the required input pressure distribution from a database of pressure distributions while the inverse method will still compute the exact blade shape that corresponds to this 'interpolated' input pressure distribution. In other words, the interpolation process is transferred to a smoother problem, namely, finding what pressure distribution would produce the required flow conditions and, once this is done, the inverse method will compute the exact solution for this problem. The use of neural network is, in this context, highly related to the use of proper optimization techniques. The optimization is used essentially as an automation procedure to force the input pressure distributions to achieve the required aero and structural design parameters. A multilayered feed forward network with back-propagation is used to train the system for pattern association and classification.

Improvement of experimental testing and network training conditions with genome-wide microarrays for more accurate predictions of drug gene targets

PubMed Central

2014-01-01

Background Genome-wide microarrays have been useful for predicting chemical-genetic interactions at the gene level. However, interpreting genome-wide microarray results can be overwhelming due to the vast output of gene expression data combined with off-target transcriptional responses many times induced by a drug treatment. This study demonstrates how experimental and computational methods can interact with each other, to arrive at more accurate predictions of drug-induced perturbations. We present a two-stage strategy that links microarray experimental testing and network training conditions to predict gene perturbations for a drug with a known mechanism of action in a well-studied organism. Results S. cerevisiae cells were treated with the antifungal, fluconazole, and expression profiling was conducted under different biological conditions using Affymetrix genome-wide microarrays. Transcripts were filtered with a formal network-based method, sparse simultaneous equation models and Lasso regression (SSEM-Lasso), under different network training conditions. Gene expression results were evaluated using both gene set and single gene target analyses, and the drug’s transcriptional effects were narrowed first by pathway and then by individual genes. Variables included: (i) Testing conditions – exposure time and concentration and (ii) Network training conditions – training compendium modifications. Two analyses of SSEM-Lasso output – gene set and single gene – were conducted to gain a better understanding of how SSEM-Lasso predicts perturbation targets. Conclusions This study demonstrates that genome-wide microarrays can be optimized using a two-stage strategy for a more in-depth understanding of how a cell manifests biological reactions to a drug treatment at the transcription level. Additionally, a more detailed understanding of how the statistical model, SSEM-Lasso, propagates perturbations through a network of gene regulatory interactions is achieved. PMID:24444313

Recursive least-squares learning algorithms for neural networks

NASA Astrophysics Data System (ADS)

Lewis, Paul S.; Hwang, Jenq N.

1990-11-01

This paper presents the development of a pair of recursive least squares (ItLS) algorithms for online training of multilayer perceptrons which are a class of feedforward artificial neural networks. These algorithms incorporate second order information about the training error surface in order to achieve faster learning rates than are possible using first order gradient descent algorithms such as the generalized delta rule. A least squares formulation is derived from a linearization of the training error function. Individual training pattern errors are linearized about the network parameters that were in effect when the pattern was presented. This permits the recursive solution of the least squares approximation either via conventional RLS recursions or by recursive QR decomposition-based techniques. The computational complexity of the update is 0(N2) where N is the number of network parameters. This is due to the estimation of the N x N inverse Hessian matrix. Less computationally intensive approximations of the ilLS algorithms can be easily derived by using only block diagonal elements of this matrix thereby partitioning the learning into independent sets. A simulation example is presented in which a neural network is trained to approximate a two dimensional Gaussian bump. In this example RLS training required an order of magnitude fewer iterations on average (527) than did training with the generalized delta rule (6 1 BACKGROUND Artificial neural networks (ANNs) offer an interesting and potentially useful paradigm for signal processing and pattern recognition. The majority of ANN applications employ the feed-forward multilayer perceptron (MLP) network architecture in which network parameters are " trained" by a supervised learning algorithm employing the generalized delta rule (GDIt) [1 2]. The GDR algorithm approximates a fixed step steepest descent algorithm using derivatives computed by error backpropagatiori. The GDII algorithm is sometimes referred to as the backpropagation algorithm. However in this paper we will use the term backpropagation to refer only to the process of computing error derivatives. While multilayer perceptrons provide a very powerful nonlinear modeling capability GDR training can be very slow and inefficient. In linear adaptive filtering the analog of the GDR algorithm is the leastmean- squares (LMS) algorithm. Steepest descent-based algorithms such as GDR or LMS are first order because they use only first derivative or gradient information about the training error to be minimized. To speed up the training process second order algorithms may be employed that take advantage of second derivative or Hessian matrix information. Second order information can be incorporated into MLP training in different ways. In many applications especially in the area of pattern recognition the training set is finite. In these cases block learning can be applied using standard nonlinear optimization techniques [3 4 5].

Fuzzy logic and neural network technologies

NASA Technical Reports Server (NTRS)

Villarreal, James A.; Lea, Robert N.; Savely, Robert T.

1992-01-01

Applications of fuzzy logic technologies in NASA projects are reviewed to examine their advantages in the development of neural networks for aerospace and commercial expert systems and control. Examples of fuzzy-logic applications include a 6-DOF spacecraft controller, collision-avoidance systems, and reinforcement-learning techniques. The commercial applications examined include a fuzzy autofocusing system, an air conditioning system, and an automobile transmission application. The practical use of fuzzy logic is set in the theoretical context of artificial neural systems (ANSs) to give the background for an overview of ANS research programs at NASA. The research and application programs include the Network Execution and Training Simulator and faster training algorithms such as the Difference Optimized Training Scheme. The networks are well suited for pattern-recognition applications such as predicting sunspots, controlling posture maintenance, and conducting adaptive diagnoses.

Improved artificial neural networks in prediction of malignancy of lesions in contrast-enhanced MR-mammography.

PubMed

Vomweg, T W; Buscema, M; Kauczor, H U; Teifke, A; Intraligi, M; Terzi, S; Heussel, C P; Achenbach, T; Rieker, O; Mayer, D; Thelen, M

2003-09-01

The aim of this study was to evaluate the capability of improved artificial neural networks (ANN) and additional novel training methods in distinguishing between benign and malignant breast lesions in contrast-enhanced magnetic resonance-mammography (MRM). A total of 604 histologically proven cases of contrast-enhanced lesions of the female breast at MRI were analyzed. Morphological, dynamic and clinical parameters were collected and stored in a database. The data set was divided into several groups using random or experimental methods [Training & Testing (T&T) algorithm] to train and test different ANNs. An additional novel computer program for input variable selection was applied. Sensitivity and specificity were calculated and compared with a statistical method and an expert radiologist. After optimization of the distribution of cases among the training and testing sets by the T & T algorithm and the reduction of input variables by the Input Selection procedure a highly sophisticated ANN achieved a sensitivity of 93.6% and a specificity of 91.9% in predicting malignancy of lesions within an independent prediction sample set. The best statistical method reached a sensitivity of 90.5% and a specificity of 68.9%. An expert radiologist performed better than the statistical method but worse than the ANN (sensitivity 92.1%, specificity 85.6%). Features extracted out of dynamic contrast-enhanced MRM and additional clinical data can be successfully analyzed by advanced ANNs. The quality of the resulting network strongly depends on the training methods, which are improved by the use of novel training tools. The best results of an improved ANN outperform expert radiologists.

SU-E-T-206: Improving Radiotherapy Toxicity Based On Artificial Neural Network (ANN) for Head and Neck Cancer Patients

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

Cho, Daniel D; Wernicke, A Gabriella; Nori, Dattatreyudu

Purpose/Objective(s): The aim of this study is to build the estimator of toxicity using artificial neural network (ANN) for head and neck cancer patients Materials/Methods: An ANN can combine variables into a predictive model during training and considered all possible correlations of variables. We constructed an ANN based on the data from 73 patients with advanced H and N cancer treated with external beam radiotherapy and/or chemotherapy at our institution. For the toxicity estimator we defined input data including age, sex, site, stage, pathology, status of chemo, technique of external beam radiation therapy (EBRT), length of treatment, dose of EBRT,more » status of post operation, length of follow-up, the status of local recurrences and distant metastasis. These data were digitized based on the significance and fed to the ANN as input nodes. We used 20 hidden nodes (for the 13 input nodes) to take care of the correlations of input nodes. For training ANN, we divided data into three subsets such as training set, validation set and test set. Finally, we built the estimator for the toxicity from ANN output. Results: We used 13 input variables including the status of local recurrences and distant metastasis and 20 hidden nodes for correlations. 59 patients for training set, 7 patients for validation set and 7 patients for test set and fed the inputs to Matlab neural network fitting tool. We trained the data within 15% of errors of outcome. In the end we have the toxicity estimation with 74% of accuracy. Conclusion: We proved in principle that ANN can be a very useful tool for predicting the RT outcomes for high risk H and N patients. Currently we are improving the results using cross validation.« less

System and Method for Modeling the Flow Performance Features of an Object

NASA Technical Reports Server (NTRS)

Jorgensen, Charles (Inventor); Ross, James (Inventor)

1997-01-01

The method and apparatus includes a neural network for generating a model of an object in a wind tunnel from performance data on the object. The network is trained from test input signals (e.g., leading edge flap position, trailing edge flap position, angle of attack, and other geometric configurations, and power settings) and test output signals (e.g., lift, drag, pitching moment, or other performance features). In one embodiment, the neural network training method employs a modified Levenberg-Marquardt optimization technique. The model can be generated 'real time' as wind tunnel testing proceeds. Once trained, the model is used to estimate performance features associated with the aircraft given geometric configuration and/or power setting input. The invention can also be applied in other similar static flow modeling applications in aerodynamics, hydrodynamics, fluid dynamics, and other such disciplines. For example, the static testing of cars, sails, and foils, propellers, keels, rudders, turbines, fins, and the like, in a wind tunnel, water trough, or other flowing medium.

A Study of the Solar Wind-Magnetosphere Coupling Using Neural Networks

NASA Astrophysics Data System (ADS)

Wu, Jian-Guo; Lundstedt, Henrik

1996-12-01

The interaction between solar wind plasma and interplanetary magnetic field (IMF) and Earth's magnetosphere induces geomagnetic activity. Geomagnetic storms can cause many adverse effects on technical systems in space and on the Earth. It is therefore of great significance to accurately predict geomagnetic activity so as to minimize the amount of disruption to these operational systems and to allow them to work as efficiently as possible. Dynamic neural networks are powerful in modeling the dynamics encoded in time series of data. In this study, we use partially recurrent neural networks to study the solar wind-magnetosphere coupling by predicting geomagnetic storms (as measured by the Dstindex) from solar wind measurements. The solar wind, the IMF and the geomagnetic index Dst data are hourly averaged and read from the National Space Science Data Center's OMNI database. We selected these data from the period 1963 to 1992, which cover 10552h and contain storm time periods 9552h and quiet time periods 1000h. The data are then categorized into three data sets: a training set (6634h), across-validation set (1962h), and a test set (1956h). The validation set is used to determine where the training should be stopped whereas the test set is used for neural networks to get the generalization capability (the out-of-sample performance). Based on the correlation analysis between the Dst index and various solar wind parameters (including various combinations of solar wind parameters), the best coupling functions can be found from the out-of-sample performance of trained neural networks. The coupling functions found are then used to forecast geomagnetic storms one to several hours in advance. The comparisons are made on iterating the single-step prediction several times and on making a non iterated, direct prediction. Thus, we will present the best solar wind-magnetosphere coupling functions and the corresponding prediction results. Interesting Links: Lund Space Weather and AI Center

Artificial intelligence in fracture detection: transfer learning from deep convolutional neural networks.

PubMed

Kim, D H; MacKinnon, T

2018-05-01

To identify the extent to which transfer learning from deep convolutional neural networks (CNNs), pre-trained on non-medical images, can be used for automated fracture detection on plain radiographs. The top layer of the Inception v3 network was re-trained using lateral wrist radiographs to produce a model for the classification of new studies as either "fracture" or "no fracture". The model was trained on a total of 11,112 images, after an eightfold data augmentation technique, from an initial set of 1,389 radiographs (695 "fracture" and 694 "no fracture"). The training data set was split 80:10:10 into training, validation, and test groups, respectively. An additional 100 wrist radiographs, comprising 50 "fracture" and 50 "no fracture" images, were used for final testing and statistical analysis. The area under the receiver operator characteristic curve (AUC) for this test was 0.954. Setting the diagnostic cut-off at a threshold designed to maximise both sensitivity and specificity resulted in values of 0.9 and 0.88, respectively. The AUC scores for this test were comparable to state-of-the-art providing proof of concept for transfer learning from CNNs in fracture detection on plain radiographs. This was achieved using only a moderate sample size. This technique is largely transferable, and therefore, has many potential applications in medical imaging, which may lead to significant improvements in workflow productivity and in clinical risk reduction. Copyright © 2017 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.

Development of Artificial Neural Network Model for Diesel Fuel Properties Prediction using Vibrational Spectroscopy.

PubMed

Bolanča, Tomislav; Marinović, Slavica; Ukić, Sime; Jukić, Ante; Rukavina, Vinko

2012-06-01

This paper describes development of artificial neural network models which can be used to correlate and predict diesel fuel properties from several FTIR-ATR absorbances and Raman intensities as input variables. Multilayer feed forward and radial basis function neural networks have been used to rapid and simultaneous prediction of cetane number, cetane index, density, viscosity, distillation temperatures at 10% (T10), 50% (T50) and 90% (T90) recovery, contents of total aromatics and polycyclic aromatic hydrocarbons of commercial diesel fuels. In this study two-phase training procedures for multilayer feed forward networks were applied. While first phase training algorithm was constantly the back propagation one, two second phase training algorithms were varied and compared, namely: conjugate gradient and quasi Newton. In case of radial basis function network, radial layer was trained using K-means radial assignment algorithm and three different radial spread algorithms: explicit, isotropic and K-nearest neighbour. The number of hidden layer neurons and experimental data points used for the training set have been optimized for both neural networks in order to insure good predictive ability by reducing unnecessary experimental work. This work shows that developed artificial neural network models can determine main properties of diesel fuels simultaneously based on a single and fast IR or Raman measurement.

Prediction of Weld Penetration in FCAW of HSLA steel using Artificial Neural Networks

NASA Astrophysics Data System (ADS)

Asl, Y. Dadgar; Mostafa, N. B.; Panahizadeh R., V.; Seyedkashi, S. M. H.

2011-01-01

Flux-cored arc welding (FCAW) is a semiautomatic or automatic arc welding process that requires a continuously-fed consumable tubular electrode containing a flux. The main FCAW process parameters affecting the depth of penetration are welding current, arc voltage, nozzle-to-work distance, torch angle and welding speed. Shallow depth of penetration may contribute to failure of a welded structure since penetration determines the stress-carrying capacity of a welded joint. To avoid such occurrences; the welding process parameters influencing the weld penetration must be properly selected to obtain an acceptable weld penetration and hence a high quality joint. Artificial neural networks (ANN), also called neural networks (NN), are computational models used to express complex non-linear relationships between input and output data. In this paper, artificial neural network (ANN) method is used to predict the effects of welding current, arc voltage, nozzle-to-work distance, torch angle and welding speed on weld penetration depth in gas shielded FCAW of a grade of high strength low alloy steel. 32 experimental runs were carried out using the bead-on-plate welding technique. Weld penetrations were measured and on the basis of these 32 sets of experimental data, a feed-forward back-propagation neural network was created. 28 sets of the experiments were used as the training data and the remaining 4 sets were used for the testing phase of the network. The ANN has one hidden layer with eight neurons and is trained after 840 iterations. The comparison between the experimental results and ANN results showed that the trained network could predict the effects of the FCAW process parameters on weld penetration adequately.

Application of the Intuitionistic Fuzzy InterCriteria Analysis Method with Triples to a Neural Network Preprocessing Procedure

PubMed Central

Atanassova, Vassia; Sotirova, Evdokia; Doukovska, Lyubka; Bureva, Veselina; Mavrov, Deyan; Tomov, Jivko

2017-01-01

The approach of InterCriteria Analysis (ICA) was applied for the aim of reducing the set of variables on the input of a neural network, taking into account the fact that their large number increases the number of neurons in the network, thus making them unusable for hardware implementation. Here, for the first time, with the help of the ICA method, correlations between triples of the input parameters for training of the neural networks were obtained. In this case, we use the approach of ICA for data preprocessing, which may yield reduction of the total time for training the neural networks, hence, the time for the network's processing of data and images. PMID:28874908

Emotional intelligence skills for maintaining social networks in healthcare organizations.

PubMed

Freshman, Brenda; Rubino, Louis

2004-01-01

For healthcare organizations to survive in these increasingly challenging times, leadership and management must face mounting interpersonal concerns. The authors present the boundaries of internal and external social networks with respect to leadership and managerial functions: Social networks within the organization are stretched by reductions in available resources and structural ambiguity, whereas external social networks are stressed by interorganizational competitive pressures. The authors present the development of emotional intelligence skills in employees as a strategic training objective that can strengthen the internal and external social networks of healthcare organizations. The authors delineate the unique functions of leadership and management with respect to the application of emotional intelligence skills and discuss training and future research implications for emotional intelligence skill sets and social networks.

Data classification using metaheuristic Cuckoo Search technique for Levenberg Marquardt back propagation (CSLM) algorithm

NASA Astrophysics Data System (ADS)

Nawi, Nazri Mohd.; Khan, Abdullah; Rehman, M. Z.

2015-05-01

A nature inspired behavior metaheuristic techniques which provide derivative-free solutions to solve complex problems. One of the latest additions to the group of nature inspired optimization procedure is Cuckoo Search (CS) algorithm. Artificial Neural Network (ANN) training is an optimization task since it is desired to find optimal weight set of a neural network in training process. Traditional training algorithms have some limitation such as getting trapped in local minima and slow convergence rate. This study proposed a new technique CSLM by combining the best features of two known algorithms back-propagation (BP) and Levenberg Marquardt algorithm (LM) for improving the convergence speed of ANN training and avoiding local minima problem by training this network. Some selected benchmark classification datasets are used for simulation. The experiment result show that the proposed cuckoo search with Levenberg Marquardt algorithm has better performance than other algorithm used in this study.

Detection of prostate cancer on multiparametric MRI

NASA Astrophysics Data System (ADS)

Seah, Jarrel C. Y.; Tang, Jennifer S. N.; Kitchen, Andy

2017-03-01

In this manuscript, we describe our approach and methods to the ProstateX challenge, which achieved an overall AUC of 0.84 and the runner-up position. We train a deep convolutional neural network to classify lesions marked on multiparametric MRI of the prostate as clinically significant or not. We implement a novel addition to the standard convolutional architecture described as auto-windowing which is clinically inspired and designed to overcome some of the difficulties faced in MRI interpretation, where high dynamic ranges and low contrast edges may cause difficulty for traditional convolutional neural networks trained on high contrast natural imagery. We demonstrate that this system can be trained end to end and outperforms a similar architecture without such additions. Although a relatively small training set was provided, we use extensive data augmentation to prevent overfitting and transfer learning to improve convergence speed, showing that deep convolutional neural networks can be feasibly trained on small datasets.

Neural networks in data analysis and modeling for detecting littoral oil-spills by airborne laser fluorosensor remote sensing

NASA Astrophysics Data System (ADS)

Lin, Bin; An, Jubai; Brown, Carl E.; Chen, Weiwei

2003-05-01

In this paper an artificial neural network (ANN) approach, which is based on flexible nonlinear models for a very broad class of transfer functions, is applied for multi-spectral data analysis and modeling of airborne laser fluorosensor in order to differentiate between classes of oil on water surface. We use three types of algorithm: Perceptron Network, Back-Propagation (B-P) Network and Self-Organizing feature Maps (SOM) Network. Using the data in form of 64-channel spectra as inputs, the ANN presents the analysis and estimation results of the oil type on the basis of the type of background materials as outputs. The ANN is trained and tested using sample data set to the network. The results of the above 3 types of network are compared in this paper. It is proved that the training has developed a network that not only fits the training data, but also fits real-world data that the network will process operationally. The ANN model would play a significant role in the ocean oil-spill identification in the future.

Neural network computer simulation of medical aerosols.

PubMed

Richardson, C J; Barlow, D J

1996-06-01

Preliminary investigations have been conducted to assess the potential for using artificial neural networks to simulate aerosol behaviour, with a view to employing this type of methodology in the evaluation and design of pulmonary drug-delivery systems. Details are presented of the general purpose software developed for these tasks; it implements a feed-forward back-propagation algorithm with weight decay and connection pruning, the user having complete run-time control of the network architecture and mode of training. A series of exploratory investigations is then reported in which different network structures and training strategies are assessed in terms of their ability to simulate known patterns of fluid flow in simple model systems. The first of these involves simulations of cellular automata-generated data for fluid flow through a partially obstructed two-dimensional pipe. The artificial neural networks are shown to be highly successful in simulating the behaviour of this simple linear system, but with important provisos relating to the information content of the training data and the criteria used to judge when the network is properly trained. A second set of investigations is then reported in which similar networks are used to simulate patterns of fluid flow through aerosol generation devices, using training data furnished through rigorous computational fluid dynamics modelling. These more complex three-dimensional systems are modelled with equal success. It is concluded that carefully tailored, well trained networks could provide valuable tools not just for predicting but also for analysing the spatial dynamics of pharmaceutical aerosols.

Identifying apple surface defects using principal components analysis and artifical neural networks

USDA-ARS?s Scientific Manuscript database

Artificial neural networks and principal components were used to detect surface defects on apples in near-infrared images. Neural networks were trained and tested on sets of principal components derived from columns of pixels from images of apples acquired at two wavelengths (740 nm and 950 nm). I...

Carbon Nanotube Growth Rate Regression using Support Vector Machines and Artificial Neural Networks

DTIC Science & Technology

2014-03-27

intensity D peak. Reprinted with permission from [38]. The SVM classifier is trained using custom written Java code leveraging the Sequential Minimal...Society Encog is a machine learning framework for Java , C++ and .Net applications that supports Bayesian Networks, Hidden Markov Models, SVMs and ANNs [13...SVM classifiers are trained using Weka libraries and leveraging custom written Java code. The data set is created as an Attribute Relationship File

Generative adversarial networks for brain lesion detection

NASA Astrophysics Data System (ADS)

Alex, Varghese; Safwan, K. P. Mohammed; Chennamsetty, Sai Saketh; Krishnamurthi, Ganapathy

2017-02-01

Manual segmentation of brain lesions from Magnetic Resonance Images (MRI) is cumbersome and introduces errors due to inter-rater variability. This paper introduces a semi-supervised technique for detection of brain lesion from MRI using Generative Adversarial Networks (GANs). GANs comprises of a Generator network and a Discriminator network which are trained simultaneously with the objective of one bettering the other. The networks were trained using non lesion patches (n=13,000) from 4 different MR sequences. The network was trained on BraTS dataset and patches were extracted from regions excluding tumor region. The Generator network generates data by modeling the underlying probability distribution of the training data, (PData). The Discriminator learns the posterior probability P (Label Data) by classifying training data and generated data as "Real" or "Fake" respectively. The Generator upon learning the joint distribution, produces images/patches such that the performance of the Discriminator on them are random, i.e. P (Label Data = GeneratedData) = 0.5. During testing, the Discriminator assigns posterior probability values close to 0.5 for patches from non lesion regions, while patches centered on lesion arise from a different distribution (PLesion) and hence are assigned lower posterior probability value by the Discriminator. On the test set (n=14), the proposed technique achieves whole tumor dice score of 0.69, sensitivity of 91% and specificity of 59%. Additionally the generator network was capable of generating non lesion patches from various MR sequences.

Digital image classification with the help of artificial neural network by simple histogram

PubMed Central

Dey, Pranab; Banerjee, Nirmalya; Kaur, Rajwant

2016-01-01

Background: Visual image classification is a great challenge to the cytopathologist in routine day-to-day work. Artificial neural network (ANN) may be helpful in this matter. Aims and Objectives: In this study, we have tried to classify digital images of malignant and benign cells in effusion cytology smear with the help of simple histogram data and ANN. Materials and Methods: A total of 404 digital images consisting of 168 benign cells and 236 malignant cells were selected for this study. The simple histogram data was extracted from these digital images and an ANN was constructed with the help of Neurointelligence software [Alyuda Neurointelligence 2.2 (577), Cupertino, California, USA]. The network architecture was 6-3-1. The images were classified as training set (281), validation set (63), and test set (60). The on-line backpropagation training algorithm was used for this study. Result: A total of 10,000 iterations were done to train the ANN system with the speed of 609.81/s. After the adequate training of this ANN model, the system was able to identify all 34 malignant cell images and 24 out of 26 benign cells. Conclusion: The ANN model can be used for the identification of the individual malignant cells with the help of simple histogram data. This study will be helpful in the future to identify malignant cells in unknown situations. PMID:27279679

Building Extraction Based on an Optimized Stacked Sparse Autoencoder of Structure and Training Samples Using LIDAR DSM and Optical Images.

PubMed

Yan, Yiming; Tan, Zhichao; Su, Nan; Zhao, Chunhui

2017-08-24

In this paper, a building extraction method is proposed based on a stacked sparse autoencoder with an optimized structure and training samples. Building extraction plays an important role in urban construction and planning. However, some negative effects will reduce the accuracy of extraction, such as exceeding resolution, bad correction and terrain influence. Data collected by multiple sensors, as light detection and ranging (LIDAR), optical sensor etc., are used to improve the extraction. Using digital surface model (DSM) obtained from LIDAR data and optical images, traditional method can improve the extraction effect to a certain extent, but there are some defects in feature extraction. Since stacked sparse autoencoder (SSAE) neural network can learn the essential characteristics of the data in depth, SSAE was employed to extract buildings from the combined DSM data and optical image. A better setting strategy of SSAE network structure is given, and an idea of setting the number and proportion of training samples for better training of SSAE was presented. The optical data and DSM were combined as input of the optimized SSAE, and after training by an optimized samples, the appropriate network structure can extract buildings with great accuracy and has good robustness.

Neural Network Burst Pressure Prediction in Graphite/Epoxy Pressure Vessels from Acoustic Emission Amplitude Data

NASA Technical Reports Server (NTRS)

Hill, Eric v. K.; Walker, James L., II; Rowell, Ginger H.

1995-01-01

Acoustic emission (AE) data were taken during hydroproof for three sets of ASTM standard 5.75 inch diameter filament wound graphite/epoxy bottles. All three sets of bottles had the same design and were wound from the same graphite fiber; the only difference was in the epoxies used. Two of the epoxies had similar mechanical properties, and because the acoustic properties of materials are a function of their stiffnesses, it was thought that the AE data from the two sets might also be similar; however, this was not the case. Therefore, the three resin types were categorized using dummy variables, which allowed the prediction of burst pressures all three sets of bottles using a single neural network. Three bottles from each set were used to train the network. The resin category, the AE amplitude distribution data taken up to 25 % of the expected burst pressure, and the actual burst pressures were used as inputs. Architecturally, the network consisted of a forty-three neuron input layer (a single categorical variable defining the resin type plus forty-two continuous variables for the AE amplitude frequencies), a fifteen neuron hidden layer for mapping, and a single output neuron for burst pressure prediction. The network trained on all three bottle sets was able to predict burst pressures in the remaining bottles with a worst case error of + 6.59%, slightly greater than the desired goal of + 5%. This larger than desired error was due to poor resolution in the amplitude data for the third bottle set. When the third set of bottles was eliminated from consideration, only four hidden layer neurons were necessary to generate a worst case prediction error of - 3.43%, well within the desired goal.

Convolutional neural networks based on augmented training samples for synthetic aperture radar target recognition

NASA Astrophysics Data System (ADS)

Yan, Yue

2018-03-01

A synthetic aperture radar (SAR) automatic target recognition (ATR) method based on the convolutional neural networks (CNN) trained by augmented training samples is proposed. To enhance the robustness of CNN to various extended operating conditions (EOCs), the original training images are used to generate the noisy samples at different signal-to-noise ratios (SNRs), multiresolution representations, and partially occluded images. Then, the generated images together with the original ones are used to train a designed CNN for target recognition. The augmented training samples can contrapuntally improve the robustness of the trained CNN to the covered EOCs, i.e., the noise corruption, resolution variance, and partial occlusion. Moreover, the significantly larger training set effectively enhances the representation capability for other conditions, e.g., the standard operating condition (SOC), as well as the stability of the network. Therefore, better performance can be achieved by the proposed method for SAR ATR. For experimental evaluation, extensive experiments are conducted on the Moving and Stationary Target Acquisition and Recognition dataset under SOC and several typical EOCs.

Evolving optimised decision rules for intrusion detection using particle swarm paradigm

NASA Astrophysics Data System (ADS)

Sivatha Sindhu, Siva S.; Geetha, S.; Kannan, A.

2012-12-01

The aim of this article is to construct a practical intrusion detection system (IDS) that properly analyses the statistics of network traffic pattern and classify them as normal or anomalous class. The objective of this article is to prove that the choice of effective network traffic features and a proficient machine-learning paradigm enhances the detection accuracy of IDS. In this article, a rule-based approach with a family of six decision tree classifiers, namely Decision Stump, C4.5, Naive Baye's Tree, Random Forest, Random Tree and Representative Tree model to perform the detection of anomalous network pattern is introduced. In particular, the proposed swarm optimisation-based approach selects instances that compose training set and optimised decision tree operate over this trained set producing classification rules with improved coverage, classification capability and generalisation ability. Experiment with the Knowledge Discovery and Data mining (KDD) data set which have information on traffic pattern, during normal and intrusive behaviour shows that the proposed algorithm produces optimised decision rules and outperforms other machine-learning algorithm.

A Matlab Program for Textural Classification Using Neural Networks

NASA Astrophysics Data System (ADS)

Leite, E. P.; de Souza, C.

2008-12-01

A new MATLAB code that provides tools to perform classification of textural images for applications in the Geosciences is presented. The program, here coined TEXTNN, comprises the computation of variogram maps in the frequency domain for specific lag distances in the neighborhood of a pixel. The result is then converted back to spatial domain, where directional or ominidirectional semivariograms are extracted. Feature vectors are built with textural information composed of the semivariance values at these lag distances and, moreover, with histogram measures of mean, standard deviation and weighted fill-ratio. This procedure is applied to a selected group of pixels or to all pixels in an image using a moving window. A feed- forward back-propagation Neural Network can then be designed and trained on feature vectors of predefined classes (training set). The training phase minimizes the mean-squared error on the training set. Additionally, at each iteration, the mean-squared error for every validation is assessed and a test set is evaluated. The program also calculates contingency matrices, global accuracy and kappa coefficient for the three data sets, allowing a quantitative appraisal of the predictive power of the Neural Network models. The interpreter is able to select the best model obtained from a k-fold cross-validation or to use a unique split-sample data set for classification of all pixels in a given textural image. The code is opened to the geoscientific community and is very flexible, allowing the experienced user to modify it as necessary. The performance of the algorithms and the end-user program were tested using synthetic images, orbital SAR (RADARSAT) imagery for oil seepage detection, and airborne, multi-polarimetric SAR imagery for geologic mapping. The overall results proved very promising.

Emulating RRTMG Radiation with Deep Neural Networks for the Accelerated Model for Climate and Energy

NASA Astrophysics Data System (ADS)

Pal, A.; Norman, M. R.

2017-12-01

The RRTMG radiation scheme in the Accelerated Model for Climate and Energy Multi-scale Model Framework (ACME-MMF), is a bottleneck and consumes approximately 50% of the computational time. To simulate a case using RRTMG radiation scheme in ACME-MMF with high throughput and high resolution will therefore require a speed-up of this calculation while retaining physical fidelity. In this study, RRTMG radiation is emulated with Deep Neural Networks (DNNs). The first step towards this goal is to run a case with ACME-MMF and generate input data sets for the DNNs. A principal component analysis of these input data sets are carried out. Artificial data sets are created using the previous data sets to cover a wider space. These artificial data sets are used in a standalone RRTMG radiation scheme to generate outputs in a cost effective manner. These input-output pairs are used to train multiple architectures DNNs(1). Another DNN(2) is trained using the inputs to predict the error. A reverse emulation is trained to map the output to input. An error controlled code is developed with the two DNNs (1 and 2) and will determine when/if the original parameterization needs to be used.

Neural Network Emulation of Reionization Simulations

NASA Astrophysics Data System (ADS)

Schmit, Claude J.; Pritchard, Jonathan R.

2018-05-01

Next generation radio experiments such as LOFAR, HERA and SKA are expected to probe the Epoch of Reionization and claim a first direct detection of the cosmic 21cm signal within the next decade. One of the major challenges for these experiments will be dealing with enormous incoming data volumes. Machine learning is key to increasing our data analysis efficiency. We consider the use of an artificial neural network to emulate 21cmFAST simulations and use it in a Bayesian parameter inference study. We then compare the network predictions to a direct evaluation of the EoR simulations and analyse the dependence of the results on the training set size. We find that the use of a training set of size 100 samples can recover the error contours of a full scale MCMC analysis which evaluates the model at each step.

Membership generation using multilayer neural network

NASA Technical Reports Server (NTRS)

Kim, Jaeseok

1992-01-01

There has been intensive research in neural network applications to pattern recognition problems. Particularly, the back-propagation network has attracted many researchers because of its outstanding performance in pattern recognition applications. In this section, we describe a new method to generate membership functions from training data using a multilayer neural network. The basic idea behind the approach is as follows. The output values of a sigmoid activation function of a neuron bear remarkable resemblance to membership values. Therefore, we can regard the sigmoid activation values as the membership values in fuzzy set theory. Thus, in order to generate class membership values, we first train a suitable multilayer network using a training algorithm such as the back-propagation algorithm. After the training procedure converges, the resulting network can be treated as a membership generation network, where the inputs are feature values and the outputs are membership values in the different classes. This method allows fairly complex membership functions to be generated because the network is highly nonlinear in general. Also, it is to be noted that the membership functions are generated from a classification point of view. For pattern recognition applications, this is highly desirable, although the membership values may not be indicative of the degree of typicality of a feature value in a particular class.

Detection of flow limitation in obstructive sleep apnea with an artificial neural network.

PubMed

Norman, Robert G; Rapoport, David M; Ayappa, Indu

2007-09-01

During sleep, the development of a plateau on the inspiratory airflow/time contour provides a non-invasive indicator of airway collapsibility. Humans recognize this abnormal contour easily, and this study replicates this with an artificial neural network (ANN) using a normalized shape. Five 10 min segments were selected from each of 18 sleep records (respiratory airflow measured with a nasal cannula) with varying degrees of sleep disordered breathing. Each breath was visually scored for shape, and breaths split randomly into a training and test set. Equally spaced, peak amplitude normalized flow values (representing breath shape) formed the only input to a back propagation ANN. Following training, breath-by-breath agreement of the ANN with the manual classification was tabulated for the training and test sets separately. Agreement of the ANN was 89% in the training set and 70.6% in the test set. When the categories of 'probably normal' and 'normal', and 'probably flow limited' and 'flow limited' were combined, the agreement increased to 92.7% and 89.4% respectively, similar to the intra- and inter-rater agreements obtained by a visual classification of these breaths. On a naive dataset, the agreement of the ANN to visual classification was 57.7% overall and 82.4% when the categories were collapsed. A neural network based only on the shape of inspiratory airflow succeeded in classifying breaths as to the presence/absence of flow limitation. This approach could be used to provide a standardized, reproducible and automated means of detecting elevated upper airway resistance.

Application of artificial neural networks to establish a predictive mortality risk model in children admitted to a paediatric intensive care unit.

PubMed

Chan, C H; Chan, E Y; Ng, D K; Chow, P Y; Kwok, K L

2006-11-01

Paediatric risk of mortality and paediatric index of mortality (PIM) are the commonly-used mortality prediction models (MPM) in children admitted to paediatric intensive care unit (PICU). The current study was undertaken to develop a better MPM using artificial neural network, a domain of artificial intelligence. The purpose of this retrospective case series was to compare an artificial neural network (ANN) model and PIM with the observed mortality in a cohort of patients admitted to a five-bed PICU in a Hong Kong non-teaching general hospital. The patients were under the age of 17 years and admitted to our PICU from April 2001 to December 2004. Data were collected from each patient admitted to our PICU. All data were randomly allocated to either the training or validation set. The data from the training set were used to construct a series of ANN models. The data from the validation set were used to validate the ANN and PIM models. The accuracy of ANN models and PIM was assessed by area under the receiver operator characteristics (ROC) curve and calibration. All data were randomly allocated to either the training (n=274) or validation set (n=273). Three ANN models were developed using the data from the training set, namely ANN8 (trained with variables required for PIM), ANN9 (trained with variables required for PIM and pre-ICU intubation) and ANN23 (trained with variables required for ANN9 and 14 principal ICU diagnoses). Three ANN models and PIM were used to predict mortality in the validation set. We found that PIM and ANN9 had a high ROC curve (PIM: 0.808, 95 percent confidence interval 0.552 to 1.000, ANN9: 0.957, 95 percent confidence interval 0.915 to 1.000), whereas ANN8 and ANN23 gave a suboptimal area under the ROC curve. ANN8 required only five variables for the calculation of risk, compared with eight for PIM. The current study demonstrated the process of predictive mortality risk model development using ANN. Further multicentre studies are required to produce a representative ANN-based mortality prediction model for use in different PICUs.

A decentralized training algorithm for Echo State Networks in distributed big data applications.

PubMed

Scardapane, Simone; Wang, Dianhui; Panella, Massimo

2016-06-01

The current big data deluge requires innovative solutions for performing efficient inference on large, heterogeneous amounts of information. Apart from the known challenges deriving from high volume and velocity, real-world big data applications may impose additional technological constraints, including the need for a fully decentralized training architecture. While several alternatives exist for training feed-forward neural networks in such a distributed setting, less attention has been devoted to the case of decentralized training of recurrent neural networks (RNNs). In this paper, we propose such an algorithm for a class of RNNs known as Echo State Networks. The algorithm is based on the well-known Alternating Direction Method of Multipliers optimization procedure. It is formulated only in terms of local exchanges between neighboring agents, without reliance on a coordinating node. Additionally, it does not require the communication of training patterns, which is a crucial component in realistic big data implementations. Experimental results on large scale artificial datasets show that it compares favorably with a fully centralized implementation, in terms of speed, efficiency and generalization accuracy. Copyright © 2015 Elsevier Ltd. All rights reserved.

Convolutional networks for vehicle track segmentation

DOE PAGES

Quach, Tu-Thach

2017-08-19

Existing methods to detect vehicle tracks in coherent change detection images, a product of combining two synthetic aperture radar images taken at different times of the same scene, rely on simple, fast models to label track pixels. These models, however, are unable to capture natural track features such as continuity and parallelism. More powerful, but computationally expensive models can be used in offline settings. We present an approach that uses dilated convolutional networks consisting of a series of 3-by-3 convolutions to segment vehicle tracks. The design of our networks considers the fact that remote sensing applications tend to operate inmore » low power and have limited training data. As a result, we aim for small, efficient networks that can be trained end-to-end to learn natural track features entirely from limited training data. We demonstrate that our 6-layer network, trained on just 90 images, is computationally efficient and improves the F-score on a standard dataset to 0.992, up from 0.959 obtained by the current state-of-the-art method.« less

Convolutional networks for vehicle track segmentation

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

Quach, Tu-Thach

Existing methods to detect vehicle tracks in coherent change detection images, a product of combining two synthetic aperture radar images taken at different times of the same scene, rely on simple, fast models to label track pixels. These models, however, are unable to capture natural track features such as continuity and parallelism. More powerful, but computationally expensive models can be used in offline settings. We present an approach that uses dilated convolutional networks consisting of a series of 3-by-3 convolutions to segment vehicle tracks. The design of our networks considers the fact that remote sensing applications tend to operate inmore » low power and have limited training data. As a result, we aim for small, efficient networks that can be trained end-to-end to learn natural track features entirely from limited training data. We demonstrate that our 6-layer network, trained on just 90 images, is computationally efficient and improves the F-score on a standard dataset to 0.992, up from 0.959 obtained by the current state-of-the-art method.« less

Extracting Dynamic Evidence Networks

DTIC Science & Technology

2004-12-01

on the performance of the three models as a function of training set size, and on experiments showing the viability of using active learning techniques...potential relation instances, which include 28K actual relations. 2.3.2 Active Learning We also ran a set of experiments designed to explore the...viability of using active learning techniques to maximize the usefulness of the training data annotated for use by the system. The idea is to

Building the team for team science

USGS Publications Warehouse

Read, Emily K.; O'Rourke, M.; Hong, G. S.; Hanson, P. C.; Winslow, Luke A.; Crowley, S.; Brewer, C. A.; Weathers, K. C.

2016-01-01

The ability to effectively exchange information and develop trusting, collaborative relationships across disciplinary boundaries is essential for 21st century scientists charged with solving complex and large-scale societal and environmental challenges, yet these communication skills are rarely taught. Here, we describe an adaptable training program designed to increase the capacity of scientists to engage in information exchange and relationship development in team science settings. A pilot of the program, developed by a leader in ecological network science, the Global Lake Ecological Observatory Network (GLEON), indicates that the training program resulted in improvement in early career scientists’ confidence in team-based network science collaborations within and outside of the program. Fellows in the program navigated human-network challenges, expanded communication skills, and improved their ability to build professional relationships, all in the context of producing collaborative scientific outcomes. Here, we describe the rationale for key communication training elements and provide evidence that such training is effective in building essential team science skills.

Speech reconstruction using a deep partially supervised neural network.

PubMed

McLoughlin, Ian; Li, Jingjie; Song, Yan; Sharifzadeh, Hamid R

2017-08-01

Statistical speech reconstruction for larynx-related dysphonia has achieved good performance using Gaussian mixture models and, more recently, restricted Boltzmann machine arrays; however, deep neural network (DNN)-based systems have been hampered by the limited amount of training data available from individual voice-loss patients. The authors propose a novel DNN structure that allows a partially supervised training approach on spectral features from smaller data sets, yielding very good results compared with the current state-of-the-art.

Pattern recognition and classification of vibrational spectra by artificial neural networks

NASA Astrophysics Data System (ADS)

Yang, Husheng

1999-10-01

A drawback of current open-path Fourier transform infrared (OP/FT-IR) systems is that they need a human expert to determine those compounds that may be quantified from a given spectrum. In this study, three types of artificial neural networks were used to alleviate this problem. Firstly, multi-layer feed-forward neural networks were used to automatically recognize compounds in an OP/FT-IR spectrum. Each neural network was trained to recognize one compound in the presence of up to ten interferents in an OP/FT-IR spectrum. The networks were successfully used to recognize five alcohols and two chlorinated compounds in field-measured controlled-release OP/FT-IR spectra of mixtures of these compounds. It has also been demonstrated that a neural network could correctly identify a spectrum in the presence of an interferent that was not included in the training set and could also reject interferents it has not seen before. Secondly, the possibility of using one- and two- dimensional Kohonen self-organizing maps (SOMs) to recognize similarities in low-resolution vapor-phase infrared spectra without any additional information has been investigated. Both full-range reference spectra and open-path window reference spectra were used to train the networks and the trained networks were then used to classify the reference spectra into several groups. The results showed that the SOMs obtained from the two different training sets were quite different, and it is more appropriate to use the second SOM in OP/FT-IR spectrometry. Thirdly, vapor-phase FT-IR reference spectra of five alcohols along with four baseline spectra were encoded as prototype vectors for a Hopfield network. Inclusion of the baseline spectra allowed the network to classify spectra as unknowns, when the reference spectra of these compounds were not stored as prototype vectors in the network. The network could identify each of the 5 alcohols correctly even in the presence of noise and interfering compounds. Finally, one- and two-dimensional Kohonen SOMs were also successfully used for the unsupervised differentiation of the Fourier transform Raman spectra of hardwoods from softwoods. A semi-quantitative method that is based on the Euclidean distances of the weight matrix has been developed to assist the automatic clustering of the neurons in a two-dimensional SOM.

Comparison of Multiple Linear Regressions and Neural Networks based QSAR models for the design of new antitubercular compounds.

PubMed

Ventura, Cristina; Latino, Diogo A R S; Martins, Filomena

2013-01-01

The performance of two QSAR methodologies, namely Multiple Linear Regressions (MLR) and Neural Networks (NN), towards the modeling and prediction of antitubercular activity was evaluated and compared. A data set of 173 potentially active compounds belonging to the hydrazide family and represented by 96 descriptors was analyzed. Models were built with Multiple Linear Regressions (MLR), single Feed-Forward Neural Networks (FFNNs), ensembles of FFNNs and Associative Neural Networks (AsNNs) using four different data sets and different types of descriptors. The predictive ability of the different techniques used were assessed and discussed on the basis of different validation criteria and results show in general a better performance of AsNNs in terms of learning ability and prediction of antitubercular behaviors when compared with all other methods. MLR have, however, the advantage of pinpointing the most relevant molecular characteristics responsible for the behavior of these compounds against Mycobacterium tuberculosis. The best results for the larger data set (94 compounds in training set and 18 in test set) were obtained with AsNNs using seven descriptors (R(2) of 0.874 and RMSE of 0.437 against R(2) of 0.845 and RMSE of 0.472 in MLRs, for test set). Counter-Propagation Neural Networks (CPNNs) were trained with the same data sets and descriptors. From the scrutiny of the weight levels in each CPNN and the information retrieved from MLRs, a rational design of potentially active compounds was attempted. Two new compounds were synthesized and tested against M. tuberculosis showing an activity close to that predicted by the majority of the models. Copyright © 2013 Elsevier Masson SAS. All rights reserved.

Thermal non-equilibrium in porous medium adjacent to vertical plate: ANN approach

NASA Astrophysics Data System (ADS)

Ahmed, N. J. Salman; Ahamed, K. S. Nazim; Al-Rashed, Abdullah A. A. A.; Kamangar, Sarfaraz; Athani, Abdulgaphur

2018-05-01

Thermal non-equilibrium in porous medium is a condition that refers to temperature discrepancy in solid matrix and fluid of porous medium. This type of flow is complex flow requiring complex set of partial differential equations that govern the flow behavior. The current work is undertaken to predict the thermal non-equilibrium behavior of porous medium adjacent to vertical plate using artificial neural network. A set of neurons in 3 layers are trained to predict the heat transfer characteristics. It is found that the thermal non-equilibrium heat transfer behavior in terms of Nusselt number of fluid as well as solid phase can be predicted accurately by using well-trained neural network.

The Emergence of Selective Attention through Probabilistic Associations between Stimuli and Actions.

PubMed

Simione, Luca; Nolfi, Stefano

2016-01-01

In this paper we show how a multilayer neural network trained to master a context-dependent task in which the action co-varies with a certain stimulus in a first context and with a second stimulus in an alternative context exhibits selective attention, i.e. filtering out of irrelevant information. This effect is rather robust and it is observed in several variations of the experiment in which the characteristics of the network as well as of the training procedure have been varied. Our result demonstrates how the filtering out of irrelevant information can originate spontaneously as a consequence of the regularities present in context-dependent training set and therefore does not necessarily depend on specific architectural constraints. The post-evaluation of the network in an instructed-delay experimental scenario shows how the behaviour of the network is consistent with the data collected in neuropsychological studies. The analysis of the network at the end of the training process indicates how selective attention originates as a result of the effects caused by relevant and irrelevant stimuli mediated by context-dependent and context-independent bidirectional associations between stimuli and actions that are extracted by the network during the learning.

Machine-printed Arabic OCR

NASA Astrophysics Data System (ADS)

Hassibi, Khosrow M.

1994-02-01

This paper presents a brief overview of our research in the development of an OCR system for recognition of machine-printed texts in languages that use the Arabic alphabet. The cursive nature of machine-printed Arabic makes the segmentation of words into letters a challenging problem. In our approach, through a novel preliminary segmentation technique, a word is broken into pieces where each piece may not represent a valid letter in general. Neural networks trained on a training sample set of about 500 Arabic text images are used for recognition of these pieces. The rules governing the alphabet and character-level contextual information are used for recombining these pieces into valid letters. Higher-level contextual analysis schemes including the use of an Arabic lexicon and n-grams is also under development and are expected to improve the word recognition accuracy. The segmentation, recognition, and contextual analysis processes are closely integrated using a feedback scheme. The details of preparation of the training set and some recent results on training of the networks will be presented.

Feature-based RNN target recognition

NASA Astrophysics Data System (ADS)

Bakircioglu, Hakan; Gelenbe, Erol

1998-09-01

Detection and recognition of target signatures in sensory data obtained by synthetic aperture radar (SAR), forward- looking infrared, or laser radar, have received considerable attention in the literature. In this paper, we propose a feature based target classification methodology to detect and classify targets in cluttered SAR images, that makes use of selective signature data from sensory data, together with a neural network technique which uses a set of trained networks based on the Random Neural Network (RNN) model (Gelenbe 89, 90, 91, 93) which is trained to act as a matched filter. We propose and investigate radial features of target shapes that are invariant to rotation, translation, and scale, to characterize target and clutter signatures. These features are then used to train a set of learning RNNs which can be used to detect targets within clutter with high accuracy, and to classify the targets or man-made objects from natural clutter. Experimental data from SAR imagery is used to illustrate and validate the proposed method, and to calculate Receiver Operating Characteristics which illustrate the performance of the proposed algorithm.

Testing of information condensation in a model reverberating spiking neural network.

PubMed

Vidybida, Alexander

2011-06-01

Information about external world is delivered to the brain in the form of structured in time spike trains. During further processing in higher areas, information is subjected to a certain condensation process, which results in formation of abstract conceptual images of external world, apparently, represented as certain uniform spiking activity partially independent on the input spike trains details. Possible physical mechanism of condensation at the level of individual neuron was discussed recently. In a reverberating spiking neural network, due to this mechanism the dynamics should settle down to the same uniform/ periodic activity in response to a set of various inputs. Since the same periodic activity may correspond to different input spike trains, we interpret this as possible candidate for information condensation mechanism in a network. Our purpose is to test this possibility in a network model consisting of five fully connected neurons, particularly, the influence of geometric size of the network, on its ability to condense information. Dynamics of 20 spiking neural networks of different geometric sizes are modelled by means of computer simulation. Each network was propelled into reverberating dynamics by applying various initial input spike trains. We run the dynamics until it becomes periodic. The Shannon's formula is used to calculate the amount of information in any input spike train and in any periodic state found. As a result, we obtain explicit estimate of the degree of information condensation in the networks, and conclude that it depends strongly on the net's geometric size.

Using Ensemble Decisions and Active Selection to Improve Low-Cost Labeling for Multi-View Data

NASA Technical Reports Server (NTRS)

Rebbapragada, Umaa; Wagstaff, Kiri L.

2011-01-01

This paper seeks to improve low-cost labeling in terms of training set reliability (the fraction of correctly labeled training items) and test set performance for multi-view learning methods. Co-training is a popular multiview learning method that combines high-confidence example selection with low-cost (self) labeling. However, co-training with certain base learning algorithms significantly reduces training set reliability, causing an associated drop in prediction accuracy. We propose the use of ensemble labeling to improve reliability in such cases. We also discuss and show promising results on combining low-cost ensemble labeling with active (low-confidence) example selection. We unify these example selection and labeling strategies under collaborative learning, a family of techniques for multi-view learning that we are developing for distributed, sensor-network environments.

An artificial neural network improves prediction of observed survival in patients with laryngeal squamous carcinoma.

PubMed

Jones, Andrew S; Taktak, Azzam G F; Helliwell, Timothy R; Fenton, John E; Birchall, Martin A; Husband, David J; Fisher, Anthony C

2006-06-01

The accepted method of modelling and predicting failure/survival, Cox's proportional hazards model, is theoretically inferior to neural network derived models for analysing highly complex systems with large datasets. A blinded comparison of the neural network versus the Cox's model in predicting survival utilising data from 873 treated patients with laryngeal cancer. These were divided randomly and equally into a training set and a study set and Cox's and neural network models applied in turn. Data were then divided into seven sets of binary covariates and the analysis repeated. Overall survival was not significantly different on Kaplan-Meier plot, or with either test model. Although the network produced qualitatively similar results to Cox's model it was significantly more sensitive to differences in survival curves for age and N stage. We propose that neural networks are capable of prediction in systems involving complex interactions between variables and non-linearity.

Development of a Deep Learning Algorithm for Automatic Diagnosis of Diabetic Retinopathy.

PubMed

Raju, Manoj; Pagidimarri, Venkatesh; Barreto, Ryan; Kadam, Amrit; Kasivajjala, Vamsichandra; Aswath, Arun

2017-01-01

This paper mainly focuses on the deep learning application in classifying the stage of diabetic retinopathy and detecting the laterality of the eye using funduscopic images. Diabetic retinopathy is a chronic, progressive, sight-threatening disease of the retinal blood vessels. Ophthalmologists diagnose diabetic retinopathy through early funduscopic screening. Normally, there is a time delay in reporting and intervention, apart from the financial cost and risk of blindness associated with it. Using a convolutional neural network based approach for automatic diagnosis of diabetic retinopathy, we trained the prediction network on the publicly available Kaggle dataset. Approximately 35,000 images were used to train the network, which observed a sensitivity of 80.28% and a specificity of 92.29% on the validation dataset of ~53,000 images. Using 8,810 images, the network was trained for detecting the laterality of the eye and observed an accuracy of 93.28% on the validation set of 8,816 images.

Improved Autoassociative Neural Networks

NASA Technical Reports Server (NTRS)

Hand, Charles

2003-01-01

Improved autoassociative neural networks, denoted nexi, have been proposed for use in controlling autonomous robots, including mobile exploratory robots of the biomorphic type. In comparison with conventional autoassociative neural networks, nexi would be more complex but more capable in that they could be trained to do more complex tasks. A nexus would use bit weights and simple arithmetic in a manner that would enable training and operation without a central processing unit, programs, weight registers, or large amounts of memory. Only a relatively small amount of memory (to hold the bit weights) and a simple logic application- specific integrated circuit would be needed. A description of autoassociative neural networks is prerequisite to a meaningful description of a nexus. An autoassociative network is a set of neurons that are completely connected in the sense that each neuron receives input from, and sends output to, all the other neurons. (In some instantiations, a neuron could also send output back to its own input terminal.) The state of a neuron is completely determined by the inner product of its inputs with weights associated with its input channel. Setting the weights sets the behavior of the network. The neurons of an autoassociative network are usually regarded as comprising a row or vector. Time is a quantized phenomenon for most autoassociative networks in the sense that time proceeds in discrete steps. At each time step, the row of neurons forms a pattern: some neurons are firing, some are not. Hence, the current state of an autoassociative network can be described with a single binary vector. As time goes by, the network changes the vector. Autoassociative networks move vectors over hyperspace landscapes of possibilities.

Multi-task transfer learning deep convolutional neural network: application to computer-aided diagnosis of breast cancer on mammograms

NASA Astrophysics Data System (ADS)

Samala, Ravi K.; Chan, Heang-Ping; Hadjiiski, Lubomir M.; Helvie, Mark A.; Cha, Kenny H.; Richter, Caleb D.

2017-12-01

Transfer learning in deep convolutional neural networks (DCNNs) is an important step in its application to medical imaging tasks. We propose a multi-task transfer learning DCNN with the aim of translating the ‘knowledge’ learned from non-medical images to medical diagnostic tasks through supervised training and increasing the generalization capabilities of DCNNs by simultaneously learning auxiliary tasks. We studied this approach in an important application: classification of malignant and benign breast masses. With Institutional Review Board (IRB) approval, digitized screen-film mammograms (SFMs) and digital mammograms (DMs) were collected from our patient files and additional SFMs were obtained from the Digital Database for Screening Mammography. The data set consisted of 2242 views with 2454 masses (1057 malignant, 1397 benign). In single-task transfer learning, the DCNN was trained and tested on SFMs. In multi-task transfer learning, SFMs and DMs were used to train the DCNN, which was then tested on SFMs. N-fold cross-validation with the training set was used for training and parameter optimization. On the independent test set, the multi-task transfer learning DCNN was found to have significantly (p  =  0.007) higher performance compared to the single-task transfer learning DCNN. This study demonstrates that multi-task transfer learning may be an effective approach for training DCNN in medical imaging applications when training samples from a single modality are limited.

Prediction of near-surface soil moisture at large scale by digital terrain modeling and neural networks.

PubMed

Lavado Contador, J F; Maneta, M; Schnabel, S

2006-10-01

The capability of Artificial Neural Network models to forecast near-surface soil moisture at fine spatial scale resolution has been tested for a 99.5 ha watershed located in SW Spain using several easy to achieve digital models of topographic and land cover variables as inputs and a series of soil moisture measurements as training data set. The study methods were designed in order to determining the potentials of the neural network model as a tool to gain insight into soil moisture distribution factors and also in order to optimize the data sampling scheme finding the optimum size of the training data set. Results suggest the efficiency of the methods in forecasting soil moisture, as a tool to assess the optimum number of field samples, and the importance of the variables selected in explaining the final map obtained.

Supervised local error estimation for nonlinear image registration using convolutional neural networks

NASA Astrophysics Data System (ADS)

Eppenhof, Koen A. J.; Pluim, Josien P. W.

2017-02-01

Error estimation in medical image registration is valuable when validating, comparing, or combining registration methods. To validate a nonlinear image registration method, ideally the registration error should be known for the entire image domain. We propose a supervised method for the estimation of a registration error map for nonlinear image registration. The method is based on a convolutional neural network that estimates the norm of the residual deformation from patches around each pixel in two registered images. This norm is interpreted as the registration error, and is defined for every pixel in the image domain. The network is trained using a set of artificially deformed images. Each training example is a pair of images: the original image, and a random deformation of that image. No manually labeled ground truth error is required. At test time, only the two registered images are required as input. We train and validate the network on registrations in a set of 2D digital subtraction angiography sequences, such that errors up to eight pixels can be estimated. We show that for this range of errors the convolutional network is able to learn the registration error in pairs of 2D registered images at subpixel precision. Finally, we present a proof of principle for the extension to 3D registration problems in chest CTs, showing that the method has the potential to estimate errors in 3D registration problems.

Real-time object-to-features vectorisation via Siamese neural networks

NASA Astrophysics Data System (ADS)

Fedorenko, Fedor; Usilin, Sergey

2017-03-01

Object-to-features vectorisation is a hard problem to solve for objects that can be hard to distinguish. Siamese and Triplet neural networks are one of the more recent tools used for such task. However, most networks used are very deep networks that prove to be hard to compute in the Internet of Things setting. In this paper, a computationally efficient neural network is proposed for real-time object-to-features vectorisation into a Euclidean metric space. We use L2 distance to reflect feature vector similarity during both training and testing. In this way, feature vectors we develop can be easily classified using K-Nearest Neighbours classifier. Such approach can be used to train networks to vectorise such "problematic" objects like images of human faces, keypoint image patches, like keypoints on Arctic maps and surrounding marine areas.

Looking for underlying features in automatic and reviewed seismic bulletins through a neural network

NASA Astrophysics Data System (ADS)

Carluccio, R.; Console, R.; Chiappini, M.; Chiappini, S.

2009-12-01

SEL1 bulletins are, among all IDC products, a fundamental tool for NDCs in their task of national assessment of compliance with the CTBT. This is because SEL1s are expected to be disseminated within 2 hours from the occurrence of any detected waveform event, and the National Authorities are supposed to take a political decision in nearly real time, especially in the case when the event could triggers the request for an on site inspection. In this context not only the rapidity, but also the reliability of the SEL1 is a fundamental requirement. Our last years experience gained in the comparison between SEL1 and Italian Seismic Bulletin events has shown that SEL1s usually contain a big fraction of bogus events (sometimes close to 50%). This is due to many factors, all related to the availability of processing data and to the fast automatic algorithms involved. On the other hand, REBs are much more reliable as proved by our experience. Therefore, in spite of their relevant time delay by which they are distributed, which prevents their real-time use, REBs can be still useful in a retrospective way as reference information for comparison with SEL1s. This study tries to set up a sort of logical filter on the SEL1s that, while maintaining the rapidity requirements, improves their reliability. Our idea is based on the assumption that the SEL1s are produced by systematic algorithm of phase association and therefore some patterns among the input and output data could exist and be recognized. Our approach was initially based on a set of rules suggested by human experts on their personal experience, and its application on large datasets on a global scale. Other approaches not involving human interaction (data mining techniques) do exist. This study refers specifically to a semi-automatic approach: fitting of multi-parametric relationships hidden in the data set, through the application of neural networks by an algorithm of supervised learning. Full SEL1 and REB bulletins from Jan 2005 to Oct 2008 have been inserted in a database, together with IMS stations availability information. Part of these data have been used to create two sets of independent data (learning and verifying) used to train a "feed-forward" supervised neural network. A network supervised training algorithm using "confirmation flag" values has been used. In order to optimize network training input a significant, not redundant subset of input parameters has been looked for with the help of a genetic algorithm search tool. A suitable 12 input subset has been found and a network architecture of 12-20-1 has thus been chosen and trained on a 15094 records data set. Different runs of training sequences have been conducted, all showing CCR (Correct Classification Rate) values of the order of 75% - 80%. The trained network behavior is shown in term of ROC curve and input-out success-error matrices. The results of the analysis on our testing and validating data groups appear promising.

Neural Network-Based Sensor Validation for Turboshaft Engines

NASA Technical Reports Server (NTRS)

Moller, James C.; Litt, Jonathan S.; Guo, Ten-Huei

1998-01-01

Sensor failure detection, isolation, and accommodation using a neural network approach is described. An auto-associative neural network is configured to perform dimensionality reduction on the sensor measurement vector and provide estimated sensor values. The sensor validation scheme is applied in a simulation of the T700 turboshaft engine in closed loop operation. Performance is evaluated based on the ability to detect faults correctly and maintain stable and responsive engine operation. The set of sensor outputs used for engine control forms the network input vector. Analytical redundancy is verified by training networks of successively smaller bottleneck layer sizes. Training data generation and strategy are discussed. The engine maintained stable behavior in the presence of sensor hard failures. With proper selection of fault determination thresholds, stability was maintained in the presence of sensor soft failures.

An artificial neural network prediction model of congenital heart disease based on risk factors: A hospital-based case-control study.

PubMed

Li, Huixia; Luo, Miyang; Zheng, Jianfei; Luo, Jiayou; Zeng, Rong; Feng, Na; Du, Qiyun; Fang, Junqun

2017-02-01

An artificial neural network (ANN) model was developed to predict the risks of congenital heart disease (CHD) in pregnant women.This hospital-based case-control study involved 119 CHD cases and 239 controls all recruited from birth defect surveillance hospitals in Hunan Province between July 2013 and June 2014. All subjects were interviewed face-to-face to fill in a questionnaire that covered 36 CHD-related variables. The 358 subjects were randomly divided into a training set and a testing set at the ratio of 85:15. The training set was used to identify the significant predictors of CHD by univariate logistic regression analyses and develop a standard feed-forward back-propagation neural network (BPNN) model for the prediction of CHD. The testing set was used to test and evaluate the performance of the ANN model. Univariate logistic regression analyses were performed on SPSS 18.0. The ANN models were developed on Matlab 7.1.The univariate logistic regression identified 15 predictors that were significantly associated with CHD, including education level (odds ratio  = 0.55), gravidity (1.95), parity (2.01), history of abnormal reproduction (2.49), family history of CHD (5.23), maternal chronic disease (4.19), maternal upper respiratory tract infection (2.08), environmental pollution around maternal dwelling place (3.63), maternal exposure to occupational hazards (3.53), maternal mental stress (2.48), paternal chronic disease (4.87), paternal exposure to occupational hazards (2.51), intake of vegetable/fruit (0.45), intake of fish/shrimp/meat/egg (0.59), and intake of milk/soymilk (0.55). After many trials, we selected a 3-layer BPNN model with 15, 12, and 1 neuron in the input, hidden, and output layers, respectively, as the best prediction model. The prediction model has accuracies of 0.91 and 0.86 on the training and testing sets, respectively. The sensitivity, specificity, and Yuden Index on the testing set (training set) are 0.78 (0.83), 0.90 (0.95), and 0.68 (0.78), respectively. The areas under the receiver operating curve on the testing and training sets are 0.87 and 0.97, respectively.This study suggests that the BPNN model could be used to predict the risk of CHD in individuals. This model should be further improved by large-sample-size research.

Multimedia Technologies for Training: An Introduction.

ERIC Educational Resources Information Center

Barron, Ann E.; Orwig, Gary W.

This guide introduces trainers, managers, and educators to a variety of new multimedia technologies now being used for presentation and training in business, military, and academic settings. The text describes advances in and implementation of technologies that range from wireless local area networks (LANs) and high definition television (HDTV) to…

Training transfer: a systematic review of the impact of inner setting factors.

PubMed

Jackson, Carrie B; Brabson, Laurel A; Quetsch, Lauren B; Herschell, Amy D

2018-06-19

Consistent with Baldwin and Ford's model (Pers Psychol 41(1):63-105, 1988), training transfer is defined as the generalization of learning from a training to everyday practice in the workplace. The purpose of this review was to examine the influence of work-environment factors, one component of the model hypothesized to influence training transfer within behavioral health. An electronic literature search guided by the Consolidated Framework for Implementation Research's inner setting domain was conducted was conducted on Medline OVID, Medline EMBASE, and PsycINFO databases. Of 9184 unique articles, 169 full-text versions of articles were screened for eligibility, yielding 26 articles meeting inclusion criteria. Results from the 26 studies revealed that overall, having more positive networks and communication, culture, implementation climate, and readiness for implementation can facilitate training transfer. Although few studies have examined the impact of inner setting factors on training transfer, these results suggest organizational context is important to consider with training efforts. These findings have important implications for individuals in the broader health professions educational field.

Predicting Positive and Negative Relationships in Large Social Networks.

PubMed

Wang, Guan-Nan; Gao, Hui; Chen, Lian; Mensah, Dennis N A; Fu, Yan

2015-01-01

In a social network, users hold and express positive and negative attitudes (e.g. support/opposition) towards other users. Those attitudes exhibit some kind of binary relationships among the users, which play an important role in social network analysis. However, some of those binary relationships are likely to be latent as the scale of social network increases. The essence of predicting latent binary relationships have recently began to draw researchers' attention. In this paper, we propose a machine learning algorithm for predicting positive and negative relationships in social networks inspired by structural balance theory and social status theory. More specifically, we show that when two users in the network have fewer common neighbors, the prediction accuracy of the relationship between them deteriorates. Accordingly, in the training phase, we propose a segment-based training framework to divide the training data into two subsets according to the number of common neighbors between users, and build a prediction model for each subset based on support vector machine (SVM). Moreover, to deal with large-scale social network data, we employ a sampling strategy that selects small amount of training data while maintaining high accuracy of prediction. We compare our algorithm with traditional algorithms and adaptive boosting of them. Experimental results of typical data sets show that our algorithm can deal with large social networks and consistently outperforms other methods.

Artificial Neural Network Approach in Laboratory Test Reporting:  Learning Algorithms.

PubMed

Demirci, Ferhat; Akan, Pinar; Kume, Tuncay; Sisman, Ali Riza; Erbayraktar, Zubeyde; Sevinc, Suleyman

2016-08-01

In the field of laboratory medicine, minimizing errors and establishing standardization is only possible by predefined processes. The aim of this study was to build an experimental decision algorithm model open to improvement that would efficiently and rapidly evaluate the results of biochemical tests with critical values by evaluating multiple factors concurrently. The experimental model was built by Weka software (Weka, Waikato, New Zealand) based on the artificial neural network method. Data were received from Dokuz Eylül University Central Laboratory. "Training sets" were developed for our experimental model to teach the evaluation criteria. After training the system, "test sets" developed for different conditions were used to statistically assess the validity of the model. After developing the decision algorithm with three iterations of training, no result was verified that was refused by the laboratory specialist. The sensitivity of the model was 91% and specificity was 100%. The estimated κ score was 0.950. This is the first study based on an artificial neural network to build an experimental assessment and decision algorithm model. By integrating our trained algorithm model into a laboratory information system, it may be possible to reduce employees' workload without compromising patient safety. © American Society for Clinical Pathology, 2016. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

A neural network for noise correlation classification

NASA Astrophysics Data System (ADS)

Paitz, Patrick; Gokhberg, Alexey; Fichtner, Andreas

2018-02-01

We present an artificial neural network (ANN) for the classification of ambient seismic noise correlations into two categories, suitable and unsuitable for noise tomography. By using only a small manually classified data subset for network training, the ANN allows us to classify large data volumes with low human effort and to encode the valuable subjective experience of data analysts that cannot be captured by a deterministic algorithm. Based on a new feature extraction procedure that exploits the wavelet-like nature of seismic time-series, we efficiently reduce the dimensionality of noise correlation data, still keeping relevant features needed for automated classification. Using global- and regional-scale data sets, we show that classification errors of 20 per cent or less can be achieved when the network training is performed with as little as 3.5 per cent and 16 per cent of the data sets, respectively. Furthermore, the ANN trained on the regional data can be applied to the global data, and vice versa, without a significant increase of the classification error. An experiment where four students manually classified the data, revealed that the classification error they would assign to each other is substantially larger than the classification error of the ANN (>35 per cent). This indicates that reproducibility would be hampered more by human subjectivity than by imperfections of the ANN.

An Emotional ANN (EANN) approach to modeling rainfall-runoff process

NASA Astrophysics Data System (ADS)

Nourani, Vahid

2017-01-01

This paper presents the first hydrological implementation of Emotional Artificial Neural Network (EANN), as a new generation of Artificial Intelligence-based models for daily rainfall-runoff (r-r) modeling of the watersheds. Inspired by neurophysiological form of brain, in addition to conventional weights and bias, an EANN includes simulated emotional parameters aimed at improving the network learning process. EANN trained by a modified version of back-propagation (BP) algorithm was applied to single and multi-step-ahead runoff forecasting of two watersheds with two distinct climatic conditions. Also to evaluate the ability of EANN trained by smaller training data set, three data division strategies with different number of training samples were considered for the training purpose. The overall comparison of the obtained results of the r-r modeling indicates that the EANN could outperform the conventional feed forward neural network (FFNN) model up to 13% and 34% in terms of training and verification efficiency criteria, respectively. The superiority of EANN over classic ANN is due to its ability to recognize and distinguish dry (rainless days) and wet (rainy days) situations using hormonal parameters of the artificial emotional system.

Reformulated Neural Network (ReNN): a New Alternative for Data-driven Modelling in Hydrology and Water Resources Engineering

NASA Astrophysics Data System (ADS)

Razavi, S.; Tolson, B.; Burn, D.; Seglenieks, F.

2012-04-01

Reformulated Neural Network (ReNN) has been recently developed as an efficient and more effective alternative to feedforward multi-layer perceptron (MLP) neural networks [Razavi, S., and Tolson, B. A. (2011). "A new formulation for feedforward neural networks." IEEE Transactions on Neural Networks, 22(10), 1588-1598, DOI: 1510.1109/TNN.2011.2163169]. This presentation initially aims to introduce the ReNN to the water resources community and then demonstrates ReNN applications to water resources related problems. ReNN is essentially equivalent to a single-hidden-layer MLP neural network but defined on a new set of network variables which is more effective than the traditional set of network weights and biases. The main features of the new network variables are that they are geometrically interpretable and each variable has a distinct role in forming the network response. ReNN is more efficiently trained as it has a less complex error response surface. In addition to the ReNN training efficiency, the interpretability of the ReNN variables enables the users to monitor and understand the internal behaviour of the network while training. Regularization in the ReNN response can be also directly measured and controlled. This feature improves the generalization ability of the network. The appeal of the ReNN is demonstrated with two ReNN applications to water resources engineering problems. In the first application, the ReNN is used to model the rainfall-runoff relationships in multiple watersheds in the Great Lakes basin located in northeastern North America. Modelling inflows to the Great Lakes are of great importance to the management of the Great Lakes system. Due to the lack of some detailed physical data about existing control structures in many subwatersheds of this huge basin, the data-driven approach to modelling such as the ReNN are required to replace predictions from a physically-based rainfall runoff model. Unlike traditional MLPs, the ReNN does not necessarily require an independent set of data for validation as the ReNN has the capability to control and verify the network degree of regularization. As such, the ReNN can be very beneficial in this case study as the data available in this case study is limited. In the second application, ReNN is fitted on the response function of the SWAT hydrologic model to act as a fast-to-run response surface surrogate (i.e., metamodel) of the original computationally intensive SWAT model. Besides the training efficiency gains, the ReNN applications demonstrate how the ReNN interpretability could help users develop more reliable networks which perform predictably better in terms of generalization.

Application of artificial neural networks to the design optimization of aerospace structural components

NASA Technical Reports Server (NTRS)

Berke, Laszlo; Patnaik, Surya N.; Murthy, Pappu L. N.

1993-01-01

The application of artificial neural networks to capture structural design expertise is demonstrated. The principal advantage of a trained neural network is that it requires trivial computational effort to produce an acceptable new design. For the class of problems addressed, the development of a conventional expert system would be extremely difficult. In the present effort, a structural optimization code with multiple nonlinear programming algorithms and an artificial neural network code NETS were used. A set of optimum designs for a ring and two aircraft wings for static and dynamic constraints were generated by using the optimization codes. The optimum design data were processed to obtain input and output pairs, which were used to develop a trained artificial neural network with the code NETS. Optimum designs for new design conditions were predicted by using the trained network. Neural net prediction of optimum designs was found to be satisfactory for most of the output design parameters. However, results from the present study indicate that caution must be exercised to ensure that all design variables are within selected error bounds.

Optimum Design of Aerospace Structural Components Using Neural Networks

NASA Technical Reports Server (NTRS)

Berke, L.; Patnaik, S. N.; Murthy, P. L. N.

1993-01-01

The application of artificial neural networks to capture structural design expertise is demonstrated. The principal advantage of a trained neural network is that it requires a trivial computational effort to produce an acceptable new design. For the class of problems addressed, the development of a conventional expert system would be extremely difficult. In the present effort, a structural optimization code with multiple nonlinear programming algorithms and an artificial neural network code NETS were used. A set of optimum designs for a ring and two aircraft wings for static and dynamic constraints were generated using the optimization codes. The optimum design data were processed to obtain input and output pairs, which were used to develop a trained artificial neural network using the code NETS. Optimum designs for new design conditions were predicted using the trained network. Neural net prediction of optimum designs was found to be satisfactory for the majority of the output design parameters. However, results from the present study indicate that caution must be exercised to ensure that all design variables are within selected error bounds.

Case Study: Does training of private networks of Family Planning clinicians in urban Pakistan affect service utilization?

PubMed Central

2010-01-01

Background To determine whether training of providers participating in franchise clinic networks is associated with increased Family Planning service use among low-income urban families in Pakistan. Methods The study uses 2001 survey data consisting of interviews with 1113 clinical and non-clinical providers working in public and private hospitals/clinics. Data analysis excludes non-clinical providers reducing sample size to 822. Variables for the analysis are divided into client volume, and training in family planning. Regression models are used to compute the association between training and service use in franchise versus private non-franchise clinics. Results In franchise clinic networks, staff are 6.5 times more likely to receive family planning training (P = 0.00) relative to private non-franchises. Service use was significantly associated with training (P = 0.00), franchise affiliation (P = 0.01), providers' years of family planning experience (P = 0.02) and the number of trained staff working at government owned clinics (P = 0.00). In this setting, nurses are significantly less likely to receive training compared to doctors (P = 0.00). Conclusions These findings suggest that franchises recruit and train various cadres of health workers and training maybe associated with increased service use through improvement in quality of services. PMID:21062460

Case Study: Does training of private networks of Family Planning clinicians in urban Pakistan affect service utilization?

PubMed

Qureshi, Asma M

2010-11-09

To determine whether training of providers participating in franchise clinic networks is associated with increased Family Planning service use among low-income urban families in Pakistan. The study uses 2001 survey data consisting of interviews with 1113 clinical and non-clinical providers working in public and private hospitals/clinics. Data analysis excludes non-clinical providers reducing sample size to 822. Variables for the analysis are divided into client volume, and training in family planning. Regression models are used to compute the association between training and service use in franchise versus private non-franchise clinics. In franchise clinic networks, staff are 6.5 times more likely to receive family planning training (P = 0.00) relative to private non-franchises. Service use was significantly associated with training (P = 0.00), franchise affiliation (P = 0.01), providers' years of family planning experience (P = 0.02) and the number of trained staff working at government owned clinics (P = 0.00). In this setting, nurses are significantly less likely to receive training compared to doctors (P = 0.00). These findings suggest that franchises recruit and train various cadres of health workers and training maybe associated with increased service use through improvement in quality of services.

Applying Deep Learning in Medical Images: The Case of Bone Age Estimation.

PubMed

Lee, Jang Hyung; Kim, Kwang Gi

2018-01-01

A diagnostic need often arises to estimate bone age from X-ray images of the hand of a subject during the growth period. Together with measured physical height, such information may be used as indicators for the height growth prognosis of the subject. We present a way to apply the deep learning technique to medical image analysis using hand bone age estimation as an example. Age estimation was formulated as a regression problem with hand X-ray images as input and estimated age as output. A set of hand X-ray images was used to form a training set with which a regression model was trained. An image preprocessing procedure is described which reduces image variations across data instances that are unrelated to age-wise variation. The use of Caffe, a deep learning tool is demonstrated. A rather simple deep learning network was adopted and trained for tutorial purpose. A test set distinct from the training set was formed to assess the validity of the approach. The measured mean absolute difference value was 18.9 months, and the concordance correlation coefficient was 0.78. It is shown that the proposed deep learning-based neural network can be used to estimate a subject's age from hand X-ray images, which eliminates the need for tedious atlas look-ups in clinical environments and should improve the time and cost efficiency of the estimation process.

Parametric motion control of robotic arms: A biologically based approach using neural networks

NASA Technical Reports Server (NTRS)

Bock, O.; D'Eleuterio, G. M. T.; Lipitkas, J.; Grodski, J. J.

1993-01-01

A neural network based system is presented which is able to generate point-to-point movements of robotic manipulators. The foundation of this approach is the use of prototypical control torque signals which are defined by a set of parameters. The parameter set is used for scaling and shaping of these prototypical torque signals to effect a desired outcome of the system. This approach is based on neurophysiological findings that the central nervous system stores generalized cognitive representations of movements called synergies, schemas, or motor programs. It has been proposed that these motor programs may be stored as torque-time functions in central pattern generators which can be scaled with appropriate time and magnitude parameters. The central pattern generators use these parameters to generate stereotypical torque-time profiles, which are then sent to the joint actuators. Hence, only a small number of parameters need to be determined for each point-to-point movement instead of the entire torque-time trajectory. This same principle is implemented for controlling the joint torques of robotic manipulators where a neural network is used to identify the relationship between the task requirements and the torque parameters. Movements are specified by the initial robot position in joint coordinates and the desired final end-effector position in Cartesian coordinates. This information is provided to the neural network which calculates six torque parameters for a two-link system. The prototypical torque profiles (one per joint) are then scaled by those parameters. After appropriate training of the network, our parametric control design allowed the reproduction of a trained set of movements with relatively high accuracy, and the production of previously untrained movements with comparable accuracy. We conclude that our approach was successful in discriminating between trained movements and in generalizing to untrained movements.

Creating Diverse Ensemble Classifiers to Reduce Supervision

DTIC Science & Technology

2005-12-01

artificial examples. Quite often training with noise improves network generalization (Bishop, 1995; Raviv & Intrator, 1996). Adding noise to training...full training set, as seen by comparing to the to- tal dataset sizes. Hence, improving on the data utilization of DECORATE is a fairly difficult task...prohibitively expensive, except (perhaps) with an incremen- tal learner such as Naive Bayes. Our AFA framework is significantly more efficient because

Vector Quantization Algorithm Based on Associative Memories

NASA Astrophysics Data System (ADS)

Guzmán, Enrique; Pogrebnyak, Oleksiy; Yáñez, Cornelio; Manrique, Pablo

This paper presents a vector quantization algorithm for image compression based on extended associative memories. The proposed algorithm is divided in two stages. First, an associative network is generated applying the learning phase of the extended associative memories between a codebook generated by the LBG algorithm and a training set. This associative network is named EAM-codebook and represents a new codebook which is used in the next stage. The EAM-codebook establishes a relation between training set and the LBG codebook. Second, the vector quantization process is performed by means of the recalling stage of EAM using as associative memory the EAM-codebook. This process generates a set of the class indices to which each input vector belongs. With respect to the LBG algorithm, the main advantages offered by the proposed algorithm is high processing speed and low demand of resources (system memory); results of image compression and quality are presented.

Convolutional neural networks for an automatic classification of prostate tissue slides with high-grade Gleason score

NASA Astrophysics Data System (ADS)

Jiménez del Toro, Oscar; Atzori, Manfredo; Otálora, Sebastian; Andersson, Mats; Eurén, Kristian; Hedlund, Martin; Rönnquist, Peter; Müller, Henning

2017-03-01

The Gleason grading system was developed for assessing prostate histopathology slides. It is correlated to the outcome and incidence of relapse in prostate cancer. Although this grading is part of a standard protocol performed by pathologists, visual inspection of whole slide images (WSIs) has an inherent subjectivity when evaluated by different pathologists. Computer aided pathology has been proposed to generate an objective and reproducible assessment that can help pathologists in their evaluation of new tissue samples. Deep convolutional neural networks are a promising approach for the automatic classification of histopathology images and can hierarchically learn subtle visual features from the data. However, a large number of manual annotations from pathologists are commonly required to obtain sufficient statistical generalization when training new models that can evaluate the daily generated large amounts of pathology data. A fully automatic approach that detects prostatectomy WSIs with high-grade Gleason score is proposed. We evaluate the performance of various deep learning architectures training them with patches extracted from automatically generated regions-of-interest rather than from manually segmented ones. Relevant parameters for training the deep learning model such as size and number of patches as well as the inclusion or not of data augmentation are compared between the tested deep learning architectures. 235 prostate tissue WSIs with their pathology report from the publicly available TCGA data set were used. An accuracy of 78% was obtained in a balanced set of 46 unseen test images with different Gleason grades in a 2-class decision: high vs. low Gleason grade. Grades 7-8, which represent the boundary decision of the proposed task, were particularly well classified. The method is scalable to larger data sets with straightforward re-training of the model to include data from multiple sources, scanners and acquisition techniques. Automatically generated heatmaps for theWSIs could be useful for improving the selection of patches when training networks for big data sets and to guide the visual inspection of these images.

Deep neural nets as a method for quantitative structure-activity relationships.

PubMed

Ma, Junshui; Sheridan, Robert P; Liaw, Andy; Dahl, George E; Svetnik, Vladimir

2015-02-23

Neural networks were widely used for quantitative structure-activity relationships (QSAR) in the 1990s. Because of various practical issues (e.g., slow on large problems, difficult to train, prone to overfitting, etc.), they were superseded by more robust methods like support vector machine (SVM) and random forest (RF), which arose in the early 2000s. The last 10 years has witnessed a revival of neural networks in the machine learning community thanks to new methods for preventing overfitting, more efficient training algorithms, and advancements in computer hardware. In particular, deep neural nets (DNNs), i.e. neural nets with more than one hidden layer, have found great successes in many applications, such as computer vision and natural language processing. Here we show that DNNs can routinely make better prospective predictions than RF on a set of large diverse QSAR data sets that are taken from Merck's drug discovery effort. The number of adjustable parameters needed for DNNs is fairly large, but our results show that it is not necessary to optimize them for individual data sets, and a single set of recommended parameters can achieve better performance than RF for most of the data sets we studied. The usefulness of the parameters is demonstrated on additional data sets not used in the calibration. Although training DNNs is still computationally intensive, using graphical processing units (GPUs) can make this issue manageable.

Neural-network Kohn-Sham exchange-correlation potential and its out-of-training transferability

NASA Astrophysics Data System (ADS)

Nagai, Ryo; Akashi, Ryosuke; Sasaki, Shu; Tsuneyuki, Shinji

2018-06-01

We incorporate in the Kohn-Sham self-consistent equation a trained neural-network projection from the charge density distribution to the Hartree-exchange-correlation potential n → VHxc for a possible numerical approach to the exact Kohn-Sham scheme. The potential trained through a newly developed scheme enables us to evaluate the total energy without explicitly treating the formula of the exchange-correlation energy. With a case study of a simple model, we show that the well-trained neural-network VHxc achieves accuracy for the charge density and total energy out of the model parameter range used for the training, indicating that the property of the elusive ideal functional form of VHxc can approximately be encapsulated by the machine-learning construction. We also exemplify a factor that crucially limits the transferability—the boundary in the model parameter space where the number of the one-particle bound states changes—and see that this is cured by setting the training parameter range across that boundary. The training scheme and insights from the model study apply to more general systems, opening a novel path to numerically efficient Kohn-Sham potential.

An Adaptive Resonance Theory account of the implicit learning of orthographic word forms.

PubMed

Glotin, H; Warnier, P; Dandurand, F; Dufau, S; Lété, B; Touzet, C; Ziegler, J C; Grainger, J

2010-01-01

An Adaptive Resonance Theory (ART) network was trained to identify unique orthographic word forms. Each word input to the model was represented as an unordered set of ordered letter pairs (open bigrams) that implement a flexible prelexical orthographic code. The network learned to map this prelexical orthographic code onto unique word representations (orthographic word forms). The network was trained on a realistic corpus of reading textbooks used in French primary schools. The amount of training was strictly identical to children's exposure to reading material from grade 1 to grade 5. Network performance was examined at each grade level. Adjustment of the learning and vigilance parameters of the network allowed us to reproduce the developmental growth of word identification performance seen in children. The network exhibited a word frequency effect and was found to be sensitive to the order of presentation of word inputs, particularly with low frequency words. These words were better learned with a randomized presentation order compared with the order of presentation in the school books. These results open up interesting perspectives for the application of ART networks in the study of the dynamics of learning to read. 2009 Elsevier Ltd. All rights reserved.

Emotion-independent face recognition

NASA Astrophysics Data System (ADS)

De Silva, Liyanage C.; Esther, Kho G. P.

2000-12-01

Current face recognition techniques tend to work well when recognizing faces under small variations in lighting, facial expression and pose, but deteriorate under more extreme conditions. In this paper, a face recognition system to recognize faces of known individuals, despite variations in facial expression due to different emotions, is developed. The eigenface approach is used for feature extraction. Classification methods include Euclidean distance, back propagation neural network and generalized regression neural network. These methods yield 100% recognition accuracy when the training database is representative, containing one image representing the peak expression for each emotion of each person apart from the neutral expression. The feature vectors used for comparison in the Euclidean distance method and for training the neural network must be all the feature vectors of the training set. These results are obtained for a face database consisting of only four persons.

Radial basis function network learns ceramic processing and predicts related strength and density

NASA Technical Reports Server (NTRS)

Cios, Krzysztof J.; Baaklini, George Y.; Vary, Alex; Tjia, Robert E.

1993-01-01

Radial basis function (RBF) neural networks were trained using the data from 273 Si3N4 modulus of rupture (MOR) bars which were tested at room temperature and 135 MOR bars which were tested at 1370 C. Milling time, sintering time, and sintering gas pressure were the processing parameters used as the input features. Flexural strength and density were the outputs by which the RBF networks were assessed. The 'nodes-at-data-points' method was used to set the hidden layer centers and output layer training used the gradient descent method. The RBF network predicted strength with an average error of less than 12 percent and density with an average error of less than 2 percent. Further, the RBF network demonstrated a potential for optimizing and accelerating the development and processing of ceramic materials.

Error estimation of deformable image registration of pulmonary CT scans using convolutional neural networks.

PubMed

Eppenhof, Koen A J; Pluim, Josien P W

2018-04-01

Error estimation in nonlinear medical image registration is a nontrivial problem that is important for validation of registration methods. We propose a supervised method for estimation of registration errors in nonlinear registration of three-dimensional (3-D) images. The method is based on a 3-D convolutional neural network that learns to estimate registration errors from a pair of image patches. By applying the network to patches centered around every voxel, we construct registration error maps. The network is trained using a set of representative images that have been synthetically transformed to construct a set of image pairs with known deformations. The method is evaluated on deformable registrations of inhale-exhale pairs of thoracic CT scans. Using ground truth target registration errors on manually annotated landmarks, we evaluate the method's ability to estimate local registration errors. Estimation of full domain error maps is evaluated using a gold standard approach. The two evaluation approaches show that we can train the network to robustly estimate registration errors in a predetermined range, with subvoxel accuracy. We achieved a root-mean-square deviation of 0.51 mm from gold standard registration errors and of 0.66 mm from ground truth landmark registration errors.

California Health Services/Educational Activities. Consortium Network.

ERIC Educational Resources Information Center

White, Charles H.

Profiles are presented of each of the 10 consortia that make up the California Health Services/Education Activities (HS/EA) network (new relationships between educational facilities where health care manpower is trained in the community settings where they practice). The first part of the booklet is a comparative analysis of (1) Area Health…

Validating the Use of Deep Learning Neural Networks for Correction of Large Hydrometric Datasets

NASA Astrophysics Data System (ADS)

Frazier, N.; Ogden, F. L.; Regina, J. A.; Cheng, Y.

2017-12-01

Collection and validation of Earth systems data can be time consuming and labor intensive. In particular, high resolution hydrometric data, including rainfall and streamflow measurements, are difficult to obtain due to a multitude of complicating factors. Measurement equipment is subject to clogs, environmental disturbances, and sensor drift. Manual intervention is typically required to identify, correct, and validate these data. Weirs can become clogged and the pressure transducer may float or drift over time. We typically employ a graphical tool called Time Series Editor to manually remove clogs and sensor drift from the data. However, this process is highly subjective and requires hydrological expertise. Two different people may produce two different data sets. To use this data for scientific discovery and model validation, a more consistent method is needed to processes this field data. Deep learning neural networks have proved to be excellent mechanisms for recognizing patterns in data. We explore the use of Recurrent Neural Networks (RNN) to capture the patterns in the data over time using various gating mechanisms (LSTM and GRU), network architectures, and hyper-parameters to build an automated data correction model. We also explore the required amount of manually corrected training data required to train the network for reasonable accuracy. The benefits of this approach are that the time to process a data set is significantly reduced, and the results are 100% reproducible after training is complete. Additionally, we train the RNN and calibrate a physically-based hydrological model against the same portion of data. Both the RNN and the model are applied to the remaining data using a split-sample methodology. Performance of the machine learning is evaluated for plausibility by comparing with the output of the hydrological model, and this analysis identifies potential periods where additional investigation is warranted.

Inversion of surface parameters using fast learning neural networks

NASA Technical Reports Server (NTRS)

Dawson, M. S.; Olvera, J.; Fung, A. K.; Manry, M. T.

1992-01-01

A neural network approach to the inversion of surface scattering parameters is presented. Simulated data sets based on a surface scattering model are used so that the data may be viewed as taken from a completely known randomly rough surface. The fast learning (FL) neural network and a multilayer perceptron (MLP) trained with backpropagation learning (BP network) are tested on the simulated backscattering data. The RMS error of training the FL network is found to be less than one half the error of the BP network while requiring one to two orders of magnitude less CPU time. When applied to inversion of parameters from a statistically rough surface, the FL method is successful at recovering the surface permittivity, the surface correlation length, and the RMS surface height in less time and with less error than the BP network. Further applications of the FL neural network to the inversion of parameters from backscatter measurements of an inhomogeneous layer above a half space are shown.

Neonatal Seizure Detection Using Deep Convolutional Neural Networks.

PubMed

Ansari, Amir H; Cherian, Perumpillichira J; Caicedo, Alexander; Naulaers, Gunnar; De Vos, Maarten; Van Huffel, Sabine

2018-04-02

Identifying a core set of features is one of the most important steps in the development of an automated seizure detector. In most of the published studies describing features and seizure classifiers, the features were hand-engineered, which may not be optimal. The main goal of the present paper is using deep convolutional neural networks (CNNs) and random forest to automatically optimize feature selection and classification. The input of the proposed classifier is raw multi-channel EEG and the output is the class label: seizure/nonseizure. By training this network, the required features are optimized, while fitting a nonlinear classifier on the features. After training the network with EEG recordings of 26 neonates, five end layers performing the classification were replaced with a random forest classifier in order to improve the performance. This resulted in a false alarm rate of 0.9 per hour and seizure detection rate of 77% using a test set of EEG recordings of 22 neonates that also included dubious seizures. The newly proposed CNN classifier outperformed three data-driven feature-based approaches and performed similar to a previously developed heuristic method.

High accuracy operon prediction method based on STRING database scores.

PubMed

Taboada, Blanca; Verde, Cristina; Merino, Enrique

2010-07-01

We present a simple and highly accurate computational method for operon prediction, based on intergenic distances and functional relationships between the protein products of contiguous genes, as defined by STRING database (Jensen,L.J., Kuhn,M., Stark,M., Chaffron,S., Creevey,C., Muller,J., Doerks,T., Julien,P., Roth,A., Simonovic,M. et al. (2009) STRING 8-a global view on proteins and their functional interactions in 630 organisms. Nucleic Acids Res., 37, D412-D416). These two parameters were used to train a neural network on a subset of experimentally characterized Escherichia coli and Bacillus subtilis operons. Our predictive model was successfully tested on the set of experimentally defined operons in E. coli and B. subtilis, with accuracies of 94.6 and 93.3%, respectively. As far as we know, these are the highest accuracies ever obtained for predicting bacterial operons. Furthermore, in order to evaluate the predictable accuracy of our model when using an organism's data set for the training procedure, and a different organism's data set for testing, we repeated the E. coli operon prediction analysis using a neural network trained with B. subtilis data, and a B. subtilis analysis using a neural network trained with E. coli data. Even for these cases, the accuracies reached with our method were outstandingly high, 91.5 and 93%, respectively. These results show the potential use of our method for accurately predicting the operons of any other organism. Our operon predictions for fully-sequenced genomes are available at http://operons.ibt.unam.mx/OperonPredictor/.

An Artificial Neural Network Based Analysis of Factors Controlling Particle Size in a Virgin Coconut Oil-Based Nanoemulsion System Containing Copper Peptide

PubMed Central

Samson, Shazwani; Basri, Mahiran; Fard Masoumi, Hamid Reza; Abdul Malek, Emilia; Abedi Karjiban, Roghayeh

2016-01-01

A predictive model of a virgin coconut oil (VCO) nanoemulsion system for the topical delivery of copper peptide (an anti-aging compound) was developed using an artificial neural network (ANN) to investigate the factors that influence particle size. Four independent variables including the amount of VCO, Tween 80: Pluronic F68 (T80:PF68), xanthan gum and water were the inputs whereas particle size was taken as the response for the trained network. Genetic algorithms (GA) were used to model the data which were divided into training sets, testing sets and validation sets. The model obtained indicated the high quality performance of the neural network and its capability to identify the critical composition factors for the VCO nanoemulsion. The main factor controlling the particle size was found out to be xanthan gum (28.56%) followed by T80:PF68 (26.9%), VCO (22.8%) and water (21.74%). The formulation containing copper peptide was then successfully prepared using optimum conditions and particle sizes of 120.7 nm were obtained. The final formulation exhibited a zeta potential lower than -25 mV and showed good physical stability towards centrifugation test, freeze-thaw cycle test and storage at temperature 25°C and 45°C. PMID:27383135

An Artificial Neural Network Based Analysis of Factors Controlling Particle Size in a Virgin Coconut Oil-Based Nanoemulsion System Containing Copper Peptide.

PubMed

Samson, Shazwani; Basri, Mahiran; Fard Masoumi, Hamid Reza; Abdul Malek, Emilia; Abedi Karjiban, Roghayeh

2016-01-01

A predictive model of a virgin coconut oil (VCO) nanoemulsion system for the topical delivery of copper peptide (an anti-aging compound) was developed using an artificial neural network (ANN) to investigate the factors that influence particle size. Four independent variables including the amount of VCO, Tween 80: Pluronic F68 (T80:PF68), xanthan gum and water were the inputs whereas particle size was taken as the response for the trained network. Genetic algorithms (GA) were used to model the data which were divided into training sets, testing sets and validation sets. The model obtained indicated the high quality performance of the neural network and its capability to identify the critical composition factors for the VCO nanoemulsion. The main factor controlling the particle size was found out to be xanthan gum (28.56%) followed by T80:PF68 (26.9%), VCO (22.8%) and water (21.74%). The formulation containing copper peptide was then successfully prepared using optimum conditions and particle sizes of 120.7 nm were obtained. The final formulation exhibited a zeta potential lower than -25 mV and showed good physical stability towards centrifugation test, freeze-thaw cycle test and storage at temperature 25°C and 45°C.

Evaluation of prediction capability, robustness, and sensitivity in non-linear landslide susceptibility models, Guantánamo, Cuba

NASA Astrophysics Data System (ADS)

Melchiorre, C.; Castellanos Abella, E. A.; van Westen, C. J.; Matteucci, M.

2011-04-01

This paper describes a procedure for landslide susceptibility assessment based on artificial neural networks, and focuses on the estimation of the prediction capability, robustness, and sensitivity of susceptibility models. The study is carried out in the Guantanamo Province of Cuba, where 186 landslides were mapped using photo-interpretation. Twelve conditioning factors were mapped including geomorphology, geology, soils, landuse, slope angle, slope direction, internal relief, drainage density, distance from roads and faults, rainfall intensity, and ground peak acceleration. A methodology was used that subdivided the database in 3 subsets. A training set was used for updating the weights. A validation set was used to stop the training procedure when the network started losing generalization capability, and a test set was used to calculate the performance of the network. A 10-fold cross-validation was performed in order to show that the results are repeatable. The prediction capability, the robustness analysis, and the sensitivity analysis were tested on 10 mutually exclusive datasets. The results show that by means of artificial neural networks it is possible to obtain models with high prediction capability and high robustness, and that an exploration of the effect of the individual variables is possible, even if they are considered as a black-box model.

Effect of missing data on multitask prediction methods.

PubMed

de la Vega de León, Antonio; Chen, Beining; Gillet, Valerie J

2018-05-22

There has been a growing interest in multitask prediction in chemoinformatics, helped by the increasing use of deep neural networks in this field. This technique is applied to multitarget data sets, where compounds have been tested against different targets, with the aim of developing models to predict a profile of biological activities for a given compound. However, multitarget data sets tend to be sparse; i.e., not all compound-target combinations have experimental values. There has been little research on the effect of missing data on the performance of multitask methods. We have used two complete data sets to simulate sparseness by removing data from the training set. Different models to remove the data were compared. These sparse sets were used to train two different multitask methods, deep neural networks and Macau, which is a Bayesian probabilistic matrix factorization technique. Results from both methods were remarkably similar and showed that the performance decrease because of missing data is at first small before accelerating after large amounts of data are removed. This work provides a first approximation to assess how much data is required to produce good performance in multitask prediction exercises.

Neural Network and Response Surface Methodology for Rocket Engine Component Optimization

NASA Technical Reports Server (NTRS)

Vaidyanathan, Rajkumar; Papita, Nilay; Shyy, Wei; Tucker, P. Kevin; Griffin, Lisa W.; Haftka, Raphael; Fitz-Coy, Norman; McConnaughey, Helen (Technical Monitor)

2000-01-01

The goal of this work is to compare the performance of response surface methodology (RSM) and two types of neural networks (NN) to aid preliminary design of two rocket engine components. A data set of 45 training points and 20 test points obtained from a semi-empirical model based on three design variables is used for a shear coaxial injector element. Data for supersonic turbine design is based on six design variables, 76 training, data and 18 test data obtained from simplified aerodynamic analysis. Several RS and NN are first constructed using the training data. The test data are then employed to select the best RS or NN. Quadratic and cubic response surfaces. radial basis neural network (RBNN) and back-propagation neural network (BPNN) are compared. Two-layered RBNN are generated using two different training algorithms, namely solverbe and solverb. A two layered BPNN is generated with Tan-Sigmoid transfer function. Various issues related to the training of the neural networks are addressed including number of neurons, error goals, spread constants and the accuracy of different models in representing the design space. A search for the optimum design is carried out using a standard gradient-based optimization algorithm over the response surfaces represented by the polynomials and trained neural networks. Usually a cubic polynominal performs better than the quadratic polynomial but exceptions have been noticed. Among the NN choices, the RBNN designed using solverb yields more consistent performance for both engine components considered. The training of RBNN is easier as it requires linear regression. This coupled with the consistency in performance promise the possibility of it being used as an optimization strategy for engineering design problems.

Combining two open source tools for neural computation (BioPatRec and Netlab) improves movement classification for prosthetic control.

PubMed

Prahm, Cosima; Eckstein, Korbinian; Ortiz-Catalan, Max; Dorffner, Georg; Kaniusas, Eugenijus; Aszmann, Oskar C

2016-08-31

Controlling a myoelectric prosthesis for upper limbs is increasingly challenging for the user as more electrodes and joints become available. Motion classification based on pattern recognition with a multi-electrode array allows multiple joints to be controlled simultaneously. Previous pattern recognition studies are difficult to compare, because individual research groups use their own data sets. To resolve this shortcoming and to facilitate comparisons, open access data sets were analysed using components of BioPatRec and Netlab pattern recognition models. Performances of the artificial neural networks, linear models, and training program components were compared. Evaluation took place within the BioPatRec environment, a Matlab-based open source platform that provides feature extraction, processing and motion classification algorithms for prosthetic control. The algorithms were applied to myoelectric signals for individual and simultaneous classification of movements, with the aim of finding the best performing algorithm and network model. Evaluation criteria included classification accuracy and training time. Results in both the linear and the artificial neural network models demonstrated that Netlab's implementation using scaled conjugate training algorithm reached significantly higher accuracies than BioPatRec. It is concluded that the best movement classification performance would be achieved through integrating Netlab training algorithms in the BioPatRec environment so that future prosthesis training can be shortened and control made more reliable. Netlab was therefore included into the newest release of BioPatRec (v4.0).

Convolutional Neural Networks for Medical Image Analysis: Full Training or Fine Tuning?

PubMed

Tajbakhsh, Nima; Shin, Jae Y; Gurudu, Suryakanth R; Hurst, R Todd; Kendall, Christopher B; Gotway, Michael B; Jianming Liang

2016-05-01

Training a deep convolutional neural network (CNN) from scratch is difficult because it requires a large amount of labeled training data and a great deal of expertise to ensure proper convergence. A promising alternative is to fine-tune a CNN that has been pre-trained using, for instance, a large set of labeled natural images. However, the substantial differences between natural and medical images may advise against such knowledge transfer. In this paper, we seek to answer the following central question in the context of medical image analysis: Can the use of pre-trained deep CNNs with sufficient fine-tuning eliminate the need for training a deep CNN from scratch? To address this question, we considered four distinct medical imaging applications in three specialties (radiology, cardiology, and gastroenterology) involving classification, detection, and segmentation from three different imaging modalities, and investigated how the performance of deep CNNs trained from scratch compared with the pre-trained CNNs fine-tuned in a layer-wise manner. Our experiments consistently demonstrated that 1) the use of a pre-trained CNN with adequate fine-tuning outperformed or, in the worst case, performed as well as a CNN trained from scratch; 2) fine-tuned CNNs were more robust to the size of training sets than CNNs trained from scratch; 3) neither shallow tuning nor deep tuning was the optimal choice for a particular application; and 4) our layer-wise fine-tuning scheme could offer a practical way to reach the best performance for the application at hand based on the amount of available data.

SKYNET: an efficient and robust neural network training tool for machine learning in astronomy

NASA Astrophysics Data System (ADS)

Graff, Philip; Feroz, Farhan; Hobson, Michael P.; Lasenby, Anthony

2014-06-01

We present the first public release of our generic neural network training algorithm, called SKYNET. This efficient and robust machine learning tool is able to train large and deep feed-forward neural networks, including autoencoders, for use in a wide range of supervised and unsupervised learning applications, such as regression, classification, density estimation, clustering and dimensionality reduction. SKYNET uses a `pre-training' method to obtain a set of network parameters that has empirically been shown to be close to a good solution, followed by further optimization using a regularized variant of Newton's method, where the level of regularization is determined and adjusted automatically; the latter uses second-order derivative information to improve convergence, but without the need to evaluate or store the full Hessian matrix, by using a fast approximate method to calculate Hessian-vector products. This combination of methods allows for the training of complicated networks that are difficult to optimize using standard backpropagation techniques. SKYNET employs convergence criteria that naturally prevent overfitting, and also includes a fast algorithm for estimating the accuracy of network outputs. The utility and flexibility of SKYNET are demonstrated by application to a number of toy problems, and to astronomical problems focusing on the recovery of structure from blurred and noisy images, the identification of gamma-ray bursters, and the compression and denoising of galaxy images. The SKYNET software, which is implemented in standard ANSI C and fully parallelized using MPI, is available at http://www.mrao.cam.ac.uk/software/skynet/.

Shakeout: A New Approach to Regularized Deep Neural Network Training.

PubMed

Kang, Guoliang; Li, Jun; Tao, Dacheng

2018-05-01

Recent years have witnessed the success of deep neural networks in dealing with a plenty of practical problems. Dropout has played an essential role in many successful deep neural networks, by inducing regularization in the model training. In this paper, we present a new regularized training approach: Shakeout. Instead of randomly discarding units as Dropout does at the training stage, Shakeout randomly chooses to enhance or reverse each unit's contribution to the next layer. This minor modification of Dropout has the statistical trait: the regularizer induced by Shakeout adaptively combines , and regularization terms. Our classification experiments with representative deep architectures on image datasets MNIST, CIFAR-10 and ImageNet show that Shakeout deals with over-fitting effectively and outperforms Dropout. We empirically demonstrate that Shakeout leads to sparser weights under both unsupervised and supervised settings. Shakeout also leads to the grouping effect of the input units in a layer. Considering the weights in reflecting the importance of connections, Shakeout is superior to Dropout, which is valuable for the deep model compression. Moreover, we demonstrate that Shakeout can effectively reduce the instability of the training process of the deep architecture.

Neural network approach for the calculation of potential coefficients in quantum mechanics

NASA Astrophysics Data System (ADS)

Ossandón, Sebastián; Reyes, Camilo; Cumsille, Patricio; Reyes, Carlos M.

2017-05-01

A numerical method based on artificial neural networks is used to solve the inverse Schrödinger equation for a multi-parameter class of potentials. First, the finite element method was used to solve repeatedly the direct problem for different parametrizations of the chosen potential function. Then, using the attainable eigenvalues as a training set of the direct radial basis neural network a map of new eigenvalues was obtained. This relationship was later inverted and refined by training an inverse radial basis neural network, allowing the calculation of the unknown parameters and therefore estimating the potential function. Three numerical examples are presented in order to prove the effectiveness of the method. The results show that the method proposed has the advantage to use less computational resources without a significant accuracy loss.

Composability-Centered Convolutional Neural Network Pruning

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

Shen, Xipeng; Guan, Hui; Lim, Seung-Hwan

This work studies the composability of the building blocks ofstructural CNN models (e.g., GoogleLeNet and Residual Networks) in thecontext of network pruning. We empirically validate that a networkcomposed of pre-trained building blocks (e.g. residual blocks andInception modules) not only gives a better initial setting fortraining, but also allows the training process to converge at asignificantly higher accuracy in much less time. Based on thatinsight, we propose a {\\em composability-centered} design for CNNnetwork pruning. Experiments show that this new scheme shortens theconfiguration process in CNN network pruning by up to 186.8X forResNet-50 and up to 30.2X for Inception-V3, and meanwhile, themore » modelsit finds that meet the accuracy requirement are significantly morecompact than those found by default schemes.« less

Finite-horizon control-constrained nonlinear optimal control using single network adaptive critics.

PubMed

Heydari, Ali; Balakrishnan, Sivasubramanya N

2013-01-01

To synthesize fixed-final-time control-constrained optimal controllers for discrete-time nonlinear control-affine systems, a single neural network (NN)-based controller called the Finite-horizon Single Network Adaptive Critic is developed in this paper. Inputs to the NN are the current system states and the time-to-go, and the network outputs are the costates that are used to compute optimal feedback control. Control constraints are handled through a nonquadratic cost function. Convergence proofs of: 1) the reinforcement learning-based training method to the optimal solution; 2) the training error; and 3) the network weights are provided. The resulting controller is shown to solve the associated time-varying Hamilton-Jacobi-Bellman equation and provide the fixed-final-time optimal solution. Performance of the new synthesis technique is demonstrated through different examples including an attitude control problem wherein a rigid spacecraft performs a finite-time attitude maneuver subject to control bounds. The new formulation has great potential for implementation since it consists of only one NN with single set of weights and it provides comprehensive feedback solutions online, though it is trained offline.

Sample selection via angular distance in the space of the arguments of an artificial neural network

NASA Astrophysics Data System (ADS)

Fernández Jaramillo, J. M.; Mayerle, R.

2018-05-01

In the construction of an artificial neural network (ANN) a proper data splitting of the available samples plays a major role in the training process. This selection of subsets for training, testing and validation affects the generalization ability of the neural network. Also the number of samples has an impact in the time required for the design of the ANN and the training. This paper introduces an efficient and simple method for reducing the set of samples used for training a neural network. The method reduces the required time to calculate the network coefficients, while keeping the diversity and avoiding overtraining the ANN due the presence of similar samples. The proposed method is based on the calculation of the angle between two vectors, each one representing one input of the neural network. When the angle formed among samples is smaller than a defined threshold only one input is accepted for the training. The accepted inputs are scattered throughout the sample space. Tidal records are used to demonstrate the proposed method. The results of a cross-validation show that with few inputs the quality of the outputs is not accurate and depends on the selection of the first sample, but as the number of inputs increases the accuracy is improved and differences among the scenarios with a different starting sample have and important reduction. A comparison with the K-means clustering algorithm shows that for this application the proposed method with a smaller number of samples is producing a more accurate network.

A Deep Learning Network Approach to ab initio Protein Secondary Structure Prediction

PubMed Central

Spencer, Matt; Eickholt, Jesse; Cheng, Jianlin

2014-01-01

Ab initio protein secondary structure (SS) predictions are utilized to generate tertiary structure predictions, which are increasingly demanded due to the rapid discovery of proteins. Although recent developments have slightly exceeded previous methods of SS prediction, accuracy has stagnated around 80% and many wonder if prediction cannot be advanced beyond this ceiling. Disciplines that have traditionally employed neural networks are experimenting with novel deep learning techniques in attempts to stimulate progress. Since neural networks have historically played an important role in SS prediction, we wanted to determine whether deep learning could contribute to the advancement of this field as well. We developed an SS predictor that makes use of the position-specific scoring matrix generated by PSI-BLAST and deep learning network architectures, which we call DNSS. Graphical processing units and CUDA software optimize the deep network architecture and efficiently train the deep networks. Optimal parameters for the training process were determined, and a workflow comprising three separately trained deep networks was constructed in order to make refined predictions. This deep learning network approach was used to predict SS for a fully independent test data set of 198 proteins, achieving a Q3 accuracy of 80.7% and a Sov accuracy of 74.2%. PMID:25750595

Surgical education and training in an outer metropolitan hospital: a qualitative study of surgical trainers and trainees.

PubMed

Nestel, Debra; Harlim, Jennifer; Bryant, Melanie; Rampersad, Rajay; Hunter-Smith, David; Spychal, Bob

2017-08-01

The landscape of surgical training is changing. The anticipated increase in the numbers of surgical trainees and the shift to competency-based surgical training places pressures on an already stretched health service. With these pressures in mind, we explored trainers' and trainees' experiences of surgical training in a less traditional rotation, an outer metropolitan hospital. We considered practice-based learning theories to make meaning of surgical training in this setting, in particular Actor-network theory. We adopted a qualitative approach and purposively sampled surgical trainers and trainees to participate in individual interviews and focus groups respectively. Transcripts were made and thematically analysed. Institutional human research ethics approval was obtained. Four surgical trainers and fourteen trainees participated. Almost without exception, participants' report training needs to be well met. Emergent inter-related themes were: learning as social activity; learning and programmatic factors; learning and physical infrastructure; and, learning and organizational structure. This outer metropolitan hospital is suited to the provision of surgical training with the current rotational system for trainees. The setting offers experiences that enable consolidation of learning providing a rich and varied overall surgical training program. Although relational elements of learning were paramount they occurred within a complex environment. Actor-network theory was used to give meaning to emergent themes acknowledging that actors (both people and objects) and their interactions combine to influence training quality, shifting the focus of responsibility for learning away from individuals to the complex interactions in which they work and learn.

Control of autonomous robot using neural networks

NASA Astrophysics Data System (ADS)

Barton, Adam; Volna, Eva

2017-07-01

The aim of the article is to design a method of control of an autonomous robot using artificial neural networks. The introductory part describes control issues from the perspective of autonomous robot navigation and the current mobile robots controlled by neural networks. The core of the article is the design of the controlling neural network, and generation and filtration of the training set using ART1 (Adaptive Resonance Theory). The outcome of the practical part is an assembled Lego Mindstorms EV3 robot solving the problem of avoiding obstacles in space. To verify models of an autonomous robot behavior, a set of experiments was created as well as evaluation criteria. The speed of each motor was adjusted by the controlling neural network with respect to the situation in which the robot was found.

Artificial neural systems for interpretation and inversion of seismic data

NASA Astrophysics Data System (ADS)

Calderon-Macias, Carlos

The goal of this work is to investigate the feasibility of using neural network (NN) models for solving geophysical exploration problems. First, a feedforward neural network (FNN) is used to solve inverse problems. The operational characteristics of a FNN are primarily controlled by a set of weights and a nonlinear function that performs a mapping between two sets of data. In a process known as training, the FNN weights are iteratively adjusted to perform the mapping. After training, the computed weights encode important features of the data that enable one pattern to be distinguished from another. Synthetic data computed from an ensemble of earth models and the corresponding models provide the training data. Two training methods are studied: the backpropagation method which is a gradient scheme, and a global optimization method called very fast simulated annealing (VFSA). A trained network is then used to predict models from new data (e.g., data from a new location) in a one-step procedure. The application of this method to the problems of obtaining formation resistivities and layer thicknesses from resistivity sounding data and 1D velocity models from seismic data shows that trained FNNs produce reasonably accurate earth models when observed data are input to the FNNs. In a second application, a FNN is used for automating the NMO correction process of seismic reflection data. The task of the FNN is to map CMP data at control locations along a seismic line into subsurface velocities. The network is trained while the velocity analyses are performed at the control locations. Once trained, the computed weights are used as an operator that acts on the remaining CMP data as a velocity interpolator, resulting in a fast method for NMO correction. The second part of this dissertation describes the application of a Hopfield neural network (HNN) to the problems of deconvolution and multiple attenuation. In these applications, the unknown parameters (reflection coefficients and source wavelet in the first problem and an operator in the second) are mapped as neurons of the HNN. The proposed deconvolution method attempts to reproduce the data with a limited number of events. The multiple attenuation method resembles the predictive deconvolution method. Results of this method are compared with a multiple elimination method based on estimating the source wavelet from the seismic data.

Sentence alignment using feed forward neural network.

PubMed

Fattah, Mohamed Abdel; Ren, Fuji; Kuroiwa, Shingo

2006-12-01

Parallel corpora have become an essential resource for work in multi lingual natural language processing. However, sentence aligned parallel corpora are more efficient than non-aligned parallel corpora for cross language information retrieval and machine translation applications. In this paper, we present a new approach to align sentences in bilingual parallel corpora based on feed forward neural network classifier. A feature parameter vector is extracted from the text pair under consideration. This vector contains text features such as length, punctuate score, and cognate score values. A set of manually prepared training data has been assigned to train the feed forward neural network. Another set of data was used for testing. Using this new approach, we could achieve an error reduction of 60% over length based approach when applied on English-Arabic parallel documents. Moreover this new approach is valid for any language pair and it is quite flexible approach since the feature parameter vector may contain more/less or different features than that we used in our system such as lexical match feature.

Novel maximum-margin training algorithms for supervised neural networks.

PubMed

Ludwig, Oswaldo; Nunes, Urbano

2010-06-01

This paper proposes three novel training methods, two of them based on the backpropagation approach and a third one based on information theory for multilayer perceptron (MLP) binary classifiers. Both backpropagation methods are based on the maximal-margin (MM) principle. The first one, based on the gradient descent with adaptive learning rate algorithm (GDX) and named maximum-margin GDX (MMGDX), directly increases the margin of the MLP output-layer hyperplane. The proposed method jointly optimizes both MLP layers in a single process, backpropagating the gradient of an MM-based objective function, through the output and hidden layers, in order to create a hidden-layer space that enables a higher margin for the output-layer hyperplane, avoiding the testing of many arbitrary kernels, as occurs in case of support vector machine (SVM) training. The proposed MM-based objective function aims to stretch out the margin to its limit. An objective function based on Lp-norm is also proposed in order to take into account the idea of support vectors, however, overcoming the complexity involved in solving a constrained optimization problem, usually in SVM training. In fact, all the training methods proposed in this paper have time and space complexities O(N) while usual SVM training methods have time complexity O(N (3)) and space complexity O(N (2)) , where N is the training-data-set size. The second approach, named minimization of interclass interference (MICI), has an objective function inspired on the Fisher discriminant analysis. Such algorithm aims to create an MLP hidden output where the patterns have a desirable statistical distribution. In both training methods, the maximum area under ROC curve (AUC) is applied as stop criterion. The third approach offers a robust training framework able to take the best of each proposed training method. The main idea is to compose a neural model by using neurons extracted from three other neural networks, each one previously trained by MICI, MMGDX, and Levenberg-Marquard (LM), respectively. The resulting neural network was named assembled neural network (ASNN). Benchmark data sets of real-world problems have been used in experiments that enable a comparison with other state-of-the-art classifiers. The results provide evidence of the effectiveness of our methods regarding accuracy, AUC, and balanced error rate.

Dynamic security contingency screening and ranking using neural networks.

PubMed

Mansour, Y; Vaahedi, E; El-Sharkawi, M A

1997-01-01

This paper summarizes BC Hydro's experience in applying neural networks to dynamic security contingency screening and ranking. The idea is to use the information on the prevailing operating condition and directly provide contingency screening and ranking using a trained neural network. To train the two neural networks for the large scale systems of BC Hydro and Hydro Quebec, in total 1691 detailed transient stability simulation were conducted, 1158 for BC Hydro system and 533 for the Hydro Quebec system. The simulation program was equipped with the energy margin calculation module (second kick) to measure the energy margin in each run. The first set of results showed poor performance for the neural networks in assessing the dynamic security. However a number of corrective measures improved the results significantly. These corrective measures included: 1) the effectiveness of output; 2) the number of outputs; 3) the type of features (static versus dynamic); 4) the number of features; 5) system partitioning; and 6) the ratio of training samples to features. The final results obtained using the large scale systems of BC Hydro and Hydro Quebec demonstrates a good potential for neural network in dynamic security assessment contingency screening and ranking.

Demystifying Multitask Deep Neural Networks for Quantitative Structure-Activity Relationships.

PubMed

Xu, Yuting; Ma, Junshui; Liaw, Andy; Sheridan, Robert P; Svetnik, Vladimir

2017-10-23

Deep neural networks (DNNs) are complex computational models that have found great success in many artificial intelligence applications, such as computer vision1,2 and natural language processing.3,4 In the past four years, DNNs have also generated promising results for quantitative structure-activity relationship (QSAR) tasks.5,6 Previous work showed that DNNs can routinely make better predictions than traditional methods, such as random forests, on a diverse collection of QSAR data sets. It was also found that multitask DNN models-those trained on and predicting multiple QSAR properties simultaneously-outperform DNNs trained separately on the individual data sets in many, but not all, tasks. To date there has been no satisfactory explanation of why the QSAR of one task embedded in a multitask DNN can borrow information from other unrelated QSAR tasks. Thus, using multitask DNNs in a way that consistently provides a predictive advantage becomes a challenge. In this work, we explored why multitask DNNs make a difference in predictive performance. Our results show that during prediction a multitask DNN does borrow "signal" from molecules with similar structures in the training sets of the other tasks. However, whether this borrowing leads to better or worse predictive performance depends on whether the activities are correlated. On the basis of this, we have developed a strategy to use multitask DNNs that incorporate prior domain knowledge to select training sets with correlated activities, and we demonstrate its effectiveness on several examples.

Generating Seismograms with Deep Neural Networks

NASA Astrophysics Data System (ADS)

Krischer, L.; Fichtner, A.

2017-12-01

The recent surge of successful uses of deep neural networks in computer vision, speech recognition, and natural language processing, mainly enabled by the availability of fast GPUs and extremely large data sets, is starting to see many applications across all natural sciences. In seismology these are largely confined to classification and discrimination tasks. In this contribution we explore the use of deep neural networks for another class of problems: so called generative models.Generative modelling is a branch of statistics concerned with generating new observed data samples, usually by drawing from some underlying probability distribution. Samples with specific attributes can be generated by conditioning on input variables. In this work we condition on seismic source (mechanism and location) and receiver (location) parameters to generate multi-component seismograms.The deep neural networks are trained on synthetic data calculated with Instaseis (http://instaseis.net, van Driel et al. (2015)) and waveforms from the global ShakeMovie project (http://global.shakemovie.princeton.edu, Tromp et al. (2010)). The underlying radially symmetric or smoothly three dimensional Earth structures result in comparatively small waveform differences from similar events or at close receivers and the networks learn to interpolate between training data samples.Of particular importance is the chosen misfit functional. Generative adversarial networks (Goodfellow et al. (2014)) implement a system in which two networks compete: the generator network creates samples and the discriminator network distinguishes these from the true training examples. Both are trained in an adversarial fashion until the discriminator can no longer distinguish between generated and real samples. We show how this can be applied to seismograms and in particular how it compares to networks trained with more conventional misfit metrics. Last but not least we attempt to shed some light on the black-box nature of neural networks by estimating the quality and uncertainties of the generated seismograms.

Application of convolutional artificial neural networks to echocardiograms for differentiating congenital heart diseases in a pediatric population

NASA Astrophysics Data System (ADS)

Perrin, Douglas P.; Bueno, Alejandra; Rodriguez, Andrea; Marx, Gerald R.; del Nido, Pedro J.

2017-03-01

In this paper we describe a pilot study, where machine learning methods are used to differentiate between congenital heart diseases. Our approach was to apply convolutional neural networks (CNNs) to echocardiographic images from five different pediatric populations: normal, coarctation of the aorta (CoA), hypoplastic left heart syndrome (HLHS), transposition of the great arteries (TGA), and single ventricle (SV). We used a single network topology that was trained in a pairwise fashion in order to evaluate the potential to differentiate between patient populations. In total we used 59,151 echo frames drawn from 1,666 clinical sequences. Approximately 80% of the data was used for training, and the remainder for validation. Data was split at sequence boundaries to avoid having related images in the training and validation sets. While training was done with echo images/frames, evaluation was performed for both single frame discrimination as well as sequence discrimination (by majority voting). In total 10 networks were generated and evaluated. Unlike other domains where this network topology has been used, in ultrasound there is low visual variation between classes. This work shows the potential for CNNs to be applied to this low-variation domain of medical imaging for disease discrimination.

Applications of neural networks in training science.

PubMed

Pfeiffer, Mark; Hohmann, Andreas

2012-04-01

Training science views itself as an integrated and applied science, developing practical measures founded on scientific method. Therefore, it demands consideration of a wide spectrum of approaches and methods. Especially in the field of competitive sports, research questions are usually located in complex environments, so that mainly field studies are drawn upon to obtain broad external validity. Here, the interrelations between different variables or variable sets are mostly of a nonlinear character. In these cases, methods like neural networks, e.g., the pattern recognizing methods of Self-Organizing Kohonen Feature Maps or similar instruments to identify interactions might be successfully applied to analyze data. Following on from a classification of data analysis methods in training-science research, the aim of the contribution is to give examples of varied sports in which network approaches can be effectually used in training science. First, two examples are given in which neural networks are employed for pattern recognition. While one investigation deals with the detection of sporting talent in swimming, the other is located in game sports research, identifying tactical patterns in team handball. The third and last example shows how an artificial neural network can be used to predict competitive performance in swimming. Copyright © 2011 Elsevier B.V. All rights reserved.

Investigating the effect of traditional Persian music on ECG signals in young women using wavelet transform and neural networks.

PubMed

Abedi, Behzad; Abbasi, Ataollah; Goshvarpour, Atefeh

2017-05-01

In the past few decades, several studies have reported the physiological effects of listening to music. The physiological effects of different music types on different people are different. In the present study, we aimed to examine the effects of listening to traditional Persian music on electrocardiogram (ECG) signals in young women. Twenty-two healthy females participated in this study. ECG signals were recorded under two conditions: rest and music. For each ECG signal, 20 morphological and wavelet-based features were selected. Artificial neural network (ANN) and probabilistic neural network (PNN) classifiers were used for the classification of ECG signals during and before listening to music. Collected data were separated into two data sets: train and test. Classification accuracies of 88% and 97% were achieved in train data sets using ANN and PNN, respectively. In addition, the test data set was employed for evaluating the classifiers, and classification rates of 84% and 93% were obtained using ANN and PNN, respectively. The present study investigated the effect of music on ECG signals based on wavelet transform and morphological features. The results obtained here can provide a good understanding on the effects of music on ECG signals to researchers.

Optimization of multilayer neural network parameters for speaker recognition

NASA Astrophysics Data System (ADS)

Tovarek, Jaromir; Partila, Pavol; Rozhon, Jan; Voznak, Miroslav; Skapa, Jan; Uhrin, Dominik; Chmelikova, Zdenka

2016-05-01

This article discusses the impact of multilayer neural network parameters for speaker identification. The main task of speaker identification is to find a specific person in the known set of speakers. It means that the voice of an unknown speaker (wanted person) belongs to a group of reference speakers from the voice database. One of the requests was to develop the text-independent system, which means to classify wanted person regardless of content and language. Multilayer neural network has been used for speaker identification in this research. Artificial neural network (ANN) needs to set parameters like activation function of neurons, steepness of activation functions, learning rate, the maximum number of iterations and a number of neurons in the hidden and output layers. ANN accuracy and validation time are directly influenced by the parameter settings. Different roles require different settings. Identification accuracy and ANN validation time were evaluated with the same input data but different parameter settings. The goal was to find parameters for the neural network with the highest precision and shortest validation time. Input data of neural networks are a Mel-frequency cepstral coefficients (MFCC). These parameters describe the properties of the vocal tract. Audio samples were recorded for all speakers in a laboratory environment. Training, testing and validation data set were split into 70, 15 and 15 %. The result of the research described in this article is different parameter setting for the multilayer neural network for four speakers.

Artificial neural network and classical least-squares methods for neurotransmitter mixture analysis.

PubMed

Schulze, H G; Greek, L S; Gorzalka, B B; Bree, A V; Blades, M W; Turner, R F

1995-02-01

Identification of individual components in biological mixtures can be a difficult problem regardless of the analytical method employed. In this work, Raman spectroscopy was chosen as a prototype analytical method due to its inherent versatility and applicability to aqueous media, making it useful for the study of biological samples. Artificial neural networks (ANNs) and the classical least-squares (CLS) method were used to identify and quantify the Raman spectra of the small-molecule neurotransmitters and mixtures of such molecules. The transfer functions used by a network, as well as the architecture of a network, played an important role in the ability of the network to identify the Raman spectra of individual neurotransmitters and the Raman spectra of neurotransmitter mixtures. Specifically, networks using sigmoid and hyperbolic tangent transfer functions generalized better from the mixtures in the training data set to those in the testing data sets than networks using sine functions. Networks with connections that permit the local processing of inputs generally performed better than other networks on all the testing data sets. and better than the CLS method of curve fitting, on novel spectra of some neurotransmitters. The CLS method was found to perform well on noisy, shifted, and difference spectra.

Invariant-feature-based adaptive automatic target recognition in obscured 3D point clouds

NASA Astrophysics Data System (ADS)

Khuon, Timothy; Kershner, Charles; Mattei, Enrico; Alverio, Arnel; Rand, Robert

2014-06-01

Target recognition and classification in a 3D point cloud is a non-trivial process due to the nature of the data collected from a sensor system. The signal can be corrupted by noise from the environment, electronic system, A/D converter, etc. Therefore, an adaptive system with a desired tolerance is required to perform classification and recognition optimally. The feature-based pattern recognition algorithm architecture as described below is particularly devised for solving a single-sensor classification non-parametrically. Feature set is extracted from an input point cloud, normalized, and classifier a neural network classifier. For instance, automatic target recognition in an urban area would require different feature sets from one in a dense foliage area. The figure above (see manuscript) illustrates the architecture of the feature based adaptive signature extraction of 3D point cloud including LIDAR, RADAR, and electro-optical data. This network takes a 3D cluster and classifies it into a specific class. The algorithm is a supervised and adaptive classifier with two modes: the training mode and the performing mode. For the training mode, a number of novel patterns are selected from actual or artificial data. A particular 3D cluster is input to the network as shown above for the decision class output. The network consists of three sequential functional modules. The first module is for feature extraction that extracts the input cluster into a set of singular value features or feature vector. Then the feature vector is input into the feature normalization module to normalize and balance it before being fed to the neural net classifier for the classification. The neural net can be trained by actual or artificial novel data until each trained output reaches the declared output within the defined tolerance. In case new novel data is added after the neural net has been learned, the training is then resumed until the neural net has incrementally learned with the new novel data. The associative memory capability of the neural net enables the incremental learning. The back propagation algorithm or support vector machine can be utilized for the classification and recognition.

HONTIOR - HIGHER-ORDER NEURAL NETWORK FOR TRANSFORMATION INVARIANT OBJECT RECOGNITION

NASA Technical Reports Server (NTRS)

Spirkovska, L.

1994-01-01

Neural networks have been applied in numerous fields, including transformation invariant object recognition, wherein an object is recognized despite changes in the object's position in the input field, size, or rotation. One of the more successful neural network methods used in invariant object recognition is the higher-order neural network (HONN) method. With a HONN, known relationships are exploited and the desired invariances are built directly into the architecture of the network, eliminating the need for the network to learn invariance to transformations. This results in a significant reduction in the training time required, since the network needs to be trained on only one view of each object, not on numerous transformed views. Moreover, one hundred percent accuracy is guaranteed for images characterized by the built-in distortions, providing noise is not introduced through pixelation. The program HONTIOR implements a third-order neural network having invariance to translation, scale, and in-plane rotation built directly into the architecture, Thus, for 2-D transformation invariance, the network needs only to be trained on just one view of each object. HONTIOR can also be used for 3-D transformation invariant object recognition by training the network only on a set of out-of-plane rotated views. Historically, the major drawback of HONNs has been that the size of the input field was limited to the memory required for the large number of interconnections in a fully connected network. HONTIOR solves this problem by coarse coding the input images (coding an image as a set of overlapping but offset coarser images). Using this scheme, large input fields (4096 x 4096 pixels) can easily be represented using very little virtual memory (30Mb). The HONTIOR distribution consists of three main programs. The first program contains the training and testing routines for a third-order neural network. The second program contains the same training and testing procedures as the first, but it also contains a number of functions to display and edit training and test images. Finally, the third program is an auxiliary program which calculates the included angles for a given input field size. HONTIOR is written in C language, and was originally developed for Sun3 and Sun4 series computers. Both graphic and command line versions of the program are provided. The command line version has been successfully compiled and executed both on computers running the UNIX operating system and on DEC VAX series computer running VMS. The graphic version requires the SunTools windowing environment, and therefore runs only on Sun series computers. The executable for the graphics version of HONTIOR requires 1Mb of RAM. The standard distribution medium for HONTIOR is a .25 inch streaming magnetic tape cartridge in UNIX tar format. It is also available on a 3.5 inch diskette in UNIX tar format. The package includes sample input and output data. HONTIOR was developed in 1991. Sun, Sun3 and Sun4 are trademarks of Sun Microsystems, Inc. UNIX is a registered trademark of AT&T Bell Laboratories. DEC, VAX, and VMS are trademarks of Digital Equipment Corporation.

Artificial neural networks predict the incidence of portosplenomesenteric venous thrombosis in patients with acute pancreatitis.

PubMed

Fei, Y; Hu, J; Li, W-Q; Wang, W; Zong, G-Q

2017-03-01

Essentials Predicting the occurrence of portosplenomesenteric vein thrombosis (PSMVT) is difficult. We studied 72 patients with acute pancreatitis. Artificial neural networks modeling was more accurate than logistic regression in predicting PSMVT. Additional predictive factors may be incorporated into artificial neural networks. Objective To construct and validate artificial neural networks (ANNs) for predicting the occurrence of portosplenomesenteric venous thrombosis (PSMVT) and compare the predictive ability of the ANNs with that of logistic regression. Methods The ANNs and logistic regression modeling were constructed using simple clinical and laboratory data of 72 acute pancreatitis (AP) patients. The ANNs and logistic modeling were first trained on 48 randomly chosen patients and validated on the remaining 24 patients. The accuracy and the performance characteristics were compared between these two approaches by SPSS17.0 software. Results The training set and validation set did not differ on any of the 11 variables. After training, the back propagation network training error converged to 1 × 10 -20 , and it retained excellent pattern recognition ability. When the ANNs model was applied to the validation set, it revealed a sensitivity of 80%, specificity of 85.7%, a positive predictive value of 77.6% and negative predictive value of 90.7%. The accuracy was 83.3%. Differences could be found between ANNs modeling and logistic regression modeling in these parameters (10.0% [95% CI, -14.3 to 34.3%], 14.3% [95% CI, -8.6 to 37.2%], 15.7% [95% CI, -9.9 to 41.3%], 11.8% [95% CI, -8.2 to 31.8%], 22.6% [95% CI, -1.9 to 47.1%], respectively). When ANNs modeling was used to identify PSMVT, the area under receiver operating characteristic curve was 0.849 (95% CI, 0.807-0.901), which demonstrated better overall properties than logistic regression modeling (AUC = 0.716) (95% CI, 0.679-0.761). Conclusions ANNs modeling was a more accurate tool than logistic regression in predicting the occurrence of PSMVT following AP. More clinical factors or biomarkers may be incorporated into ANNs modeling to improve its predictive ability. © 2016 International Society on Thrombosis and Haemostasis.

Shallow Transits—Deep Learning. I. Feasibility Study of Deep Learning to Detect Periodic Transits of Exoplanets

NASA Astrophysics Data System (ADS)

Zucker, Shay; Giryes, Raja

2018-04-01

Transits of habitable planets around solar-like stars are expected to be shallow, and to have long periods, which means low information content. The current bottleneck in the detection of such transits is caused in large part by the presence of red (correlated) noise in the light curves obtained from the dedicated space telescopes. Based on the groundbreaking results deep learning achieves in many signal and image processing applications, we propose to use deep neural networks to solve this problem. We present a feasibility study, in which we applied a convolutional neural network on a simulated training set. The training set comprised light curves received from a hypothetical high-cadence space-based telescope. We simulated the red noise by using Gaussian Processes with a wide variety of hyper-parameters. We then tested the network on a completely different test set simulated in the same way. Our study proves that very difficult cases can indeed be detected. Furthermore, we show how detection trends can be studied and detection biases quantified. We have also checked the robustness of the neural-network performance against practical artifacts such as outliers and discontinuities, which are known to affect space-based high-cadence light curves. Future work will allow us to use the neural networks to characterize the transit model and identify individual transits. This new approach will certainly be an indispensable tool for the detection of habitable planets in the future planet-detection space missions such as PLATO.

Outreach in natural settings: the use of peer leaders for HIV prevention among injecting drug users' networks.

PubMed Central

Latkin, C A

1998-01-01

OBJECTIVE: Guided by a social influence and empowerment framework, peer leaders in the injecting drug user (IDU) community were trained to promote human immunodeficiency virus (HIV) prevention among their contacts within and beyond their sex and drug networks. METHODS: From 1994 to 1995 in Baltimore, Maryland, 36 peer leaders who participated in the 10-session training program were administered pretest and posttest surveys. Evaluation included leaders' self-reported HIV-related behaviors and outreach activities. Survey data also were collected from 78 of the leaders' risk network members. RESULTS: Peer leaders reported a significant increase in condom use and in cleaning used needles with bleach. The leaders' risk network members, compared with controls, were significantly more likely to report greater needle hygiene. In an assessment of diffusion of information, the majority of risk network members who were current injectors reported receiving needle-cleaning materials from the leaders, and the majority of risk network members were able to correctly identify the HIV prevention slogans that had been taught to the leaders. The leaders documented 2165 HIV prevention interactions, of which 84% were with active drug users. CONCLUSIONS: The results from this study suggest that, in the IDU community, training peer leaders as HIV educators may promote HIV prevention among the leaders' risk network members and others at risk of acquiring and transmitting HIV. This training also may provide the leaders with effective prosocial roles. PMID:9722820

Radial basis function neural networks in non-destructive determination of compound aspirin tablets on NIR spectroscopy.

PubMed

Dou, Ying; Mi, Hong; Zhao, Lingzhi; Ren, Yuqiu; Ren, Yulin

2006-09-01

The application of the second most popular artificial neural networks (ANNs), namely, the radial basis function (RBF) networks, has been developed for quantitative analysis of drugs during the last decade. In this paper, the two components (aspirin and phenacetin) were simultaneously determined in compound aspirin tablets by using near-infrared (NIR) spectroscopy and RBF networks. The total database was randomly divided into a training set (50) and a testing set (17). Different preprocessing methods (standard normal variate (SNV), multiplicative scatter correction (MSC), first-derivative and second-derivative) were applied to two sets of NIR spectra of compound aspirin tablets with different concentrations of two active components and compared each other. After that, the performance of RBF learning algorithm adopted the nearest neighbor clustering algorithm (NNCA) and the criterion for selection used a cross-validation technique. Results show that using RBF networks to quantificationally analyze tablets is reliable, and the best RBF model was obtained by first-derivative spectra.

A new algorithm to detect earthquakes outside the seismic network: preliminary results

NASA Astrophysics Data System (ADS)

Giudicepietro, Flora; Esposito, Antonietta Maria; Ricciolino, Patrizia

2017-04-01

In this text we are going to present a new technique for detecting earthquakes outside the seismic network, which are often the cause of fault of automatic analysis system. Our goal is to develop a robust method that provides the discrimination result as quickly as possible. We discriminate local earthquakes from regional earthquakes, both recorded at SGG station, equipped with short period sensors, operated by Osservatorio Vesuviano (INGV) in the Southern Apennines (Italy). The technique uses a Multi Layer Perceptron (MLP) neural network with an architecture composed by an input layer, a hidden layer and a single node output layer. We pre-processed the data using the Linear Predictive Coding (LPC) technique to extract the spectral features of the signals in a compact form. We performed several experiments by shortening the signal window length. In particular, we used windows of 4, 2 and 1 seconds containing the onset of the local and the regional earthquakes. We used a dataset of 103 local earthquakes and 79 regional earthquakes, most of which occurred in Greece, Albania and Crete. We split the dataset into a training set, for the network training, and a testing set to evaluate the network's capacity of discrimination. In order to assess the network stability, we repeated this procedure six times, randomly changing the data composition of the training and testing set and the initial weights of the net. We estimated the performance of this method by calculating the average of correct detection percentages obtained for each of the six permutations. The average performances are 99.02%, 98.04% and 98.53%, which concern respectively the experiments carried out on 4, 2 and 1 seconds signal windows. The results show that our method is able to recognize the earthquakes outside the seismic network using only the first second of the seismic records, with a suitable percentage of correct detection. Therefore, this algorithm can be profitably used to make earthquake automatic analyses more robust and reliable. Finally, with appropriate tuning, it can be integrated in multi-parametric systems for monitoring high natural risk areas.

Spatial prediction of ground subsidence susceptibility using an artificial neural network.

PubMed

Lee, Saro; Park, Inhye; Choi, Jong-Kuk

2012-02-01

Ground subsidence in abandoned underground coal mine areas can result in loss of life and property. We analyzed ground subsidence susceptibility (GSS) around abandoned coal mines in Jeong-am, Gangwon-do, South Korea, using artificial neural network (ANN) and geographic information system approaches. Spatial data of subsidence area, topography, and geology, as well as various ground-engineering data, were collected and used to create a raster database of relevant factors for a GSS map. Eight major factors causing ground subsidence were extracted from the existing ground subsidence area: slope, depth of coal mine, distance from pit, groundwater depth, rock-mass rating, distance from fault, geology, and land use. Areas of ground subsidence were randomly divided into a training set to analyze GSS using the ANN and a test set to validate the predicted GSS map. Weights of each factor's relative importance were determined by the back-propagation training algorithms and applied to the input factor. The GSS was then calculated using the weights, and GSS maps were created. The process was repeated ten times to check the stability of analysis model using a different training data set. The map was validated using area-under-the-curve analysis with the ground subsidence areas that had not been used to train the model. The validation showed prediction accuracies between 94.84 and 95.98%, representing overall satisfactory agreement. Among the input factors, "distance from fault" had the highest average weight (i.e., 1.5477), indicating that this factor was most important. The generated maps can be used to estimate hazards to people, property, and existing infrastructure, such as the transportation network, and as part of land-use and infrastructure planning.

Automated Melanoma Recognition in Dermoscopy Images via Very Deep Residual Networks.

PubMed

Yu, Lequan; Chen, Hao; Dou, Qi; Qin, Jing; Heng, Pheng-Ann

2017-04-01

Automated melanoma recognition in dermoscopy images is a very challenging task due to the low contrast of skin lesions, the huge intraclass variation of melanomas, the high degree of visual similarity between melanoma and non-melanoma lesions, and the existence of many artifacts in the image. In order to meet these challenges, we propose a novel method for melanoma recognition by leveraging very deep convolutional neural networks (CNNs). Compared with existing methods employing either low-level hand-crafted features or CNNs with shallower architectures, our substantially deeper networks (more than 50 layers) can acquire richer and more discriminative features for more accurate recognition. To take full advantage of very deep networks, we propose a set of schemes to ensure effective training and learning under limited training data. First, we apply the residual learning to cope with the degradation and overfitting problems when a network goes deeper. This technique can ensure that our networks benefit from the performance gains achieved by increasing network depth. Then, we construct a fully convolutional residual network (FCRN) for accurate skin lesion segmentation, and further enhance its capability by incorporating a multi-scale contextual information integration scheme. Finally, we seamlessly integrate the proposed FCRN (for segmentation) and other very deep residual networks (for classification) to form a two-stage framework. This framework enables the classification network to extract more representative and specific features based on segmented results instead of the whole dermoscopy images, further alleviating the insufficiency of training data. The proposed framework is extensively evaluated on ISBI 2016 Skin Lesion Analysis Towards Melanoma Detection Challenge dataset. Experimental results demonstrate the significant performance gains of the proposed framework, ranking the first in classification and the second in segmentation among 25 teams and 28 teams, respectively. This study corroborates that very deep CNNs with effective training mechanisms can be employed to solve complicated medical image analysis tasks, even with limited training data.

Competitive STDP Learning of Overlapping Spatial Patterns.

PubMed

Krunglevicius, Dalius

2015-08-01

Spike-timing-dependent plasticity (STDP) is a set of Hebbian learning rules firmly based on biological evidence. It has been demonstrated that one of the STDP learning rules is suited for learning spatiotemporal patterns. When multiple neurons are organized in a simple competitive spiking neural network, this network is capable of learning multiple distinct patterns. If patterns overlap significantly (i.e., patterns are mutually inclusive), however, competition would not preclude trained neuron's responding to a new pattern and adjusting synaptic weights accordingly. This letter presents a simple neural network that combines vertical inhibition and Euclidean distance-dependent synaptic strength factor. This approach helps to solve the problem of pattern size-dependent parameter optimality and significantly reduces the probability of a neuron's forgetting an already learned pattern. For demonstration purposes, the network was trained for the first ten letters of the Braille alphabet.

Supporting the Integration of HIV Testing Into Primary Care Settings

PubMed Central

Bradley-Springer, Lucy; Kang Dufour, Mi-Suk; Koester, Kimberly A.; Beane, Stephanie; Warren, Nancy; Beal, Jeffrey; Frank, Linda Rose

2012-01-01

Objectives. We examined the efforts of the US network of AIDS Education and Training Centers (AETCs) to increase HIV testing capacity across a variety of clinical settings. Methods. We used quantitative process data from 8 regional AETCs for July 1, 2008, to June 30, 2009, and qualitative program descriptions to demonstrate how AETC education helped providers integrate HIV testing into routine clinical care with the goals of early diagnosis and treatment. Results. Compared with other AETC training, HIV testing training was longer and used a broader variety of strategies to educate more providers per training. During education, providers were able to understand their primary care responsibility to address public health concerns through HIV testing. Conclusions. AETC efforts illustrate how integration of the principles of primary care and public health can be promoted through professional training. PMID:22515867

Informal Training in Staff Networks to Support Dissemination of Health Promotion Programs

PubMed Central

Ramanadhan, Shoba; Wiecha, Jean L.; Gortmaker, Steven L.; Emmons, Karen M.; Viswanath, Kasisomayajula

2011-01-01

Purpose To study informal skill transfer via staff networks as a complement to formal training among afterschool childcare providers implementing a health promotion program. Design Cross-sectional, sociometric network analysis. Setting Boston Young Men’s Christian Association (YMCA) afterschool programs implementing the iPLAY program. Participants All 91 staff members at 20 sites were eligible; 80 completed the survey (88% response rate). Measures At the network level, network density measured system-level connectedness. At the staff level, the independent variable was out degree, the number of individuals to whom respondents noted a program-related connection. The dependent variable was skill gains, the number of key implementation skills gained from the network. Analysis We mapped the staff program-related social network. We utilized multiple linear regression to estimate the relationship between out degree and skill gains, and we adjusted for clustering of staff in sites. Results Most staff (77%) reported gaining at least one skill from the network, but only 2% of potential network connections were established. The regression model showed that out degree (i.e., number of program-related contacts) was significantly associated with skill gains (β = .48, p < .01) independent of other variables. Conclusion Informal skill transfer in staff networks may be a useful complement to formal training for implementation of health promotion programs, but informal skill transfer was likely underutilized in this network. Future research employing longitudinal and/or multisite data should examine these findings in greater detail. PMID:20809826

Exploring QSARs of the interaction of flavonoids with GABA (A) receptor using MLR, ANN and SVM techniques.

PubMed

Deeb, Omar; Shaik, Basheerulla; Agrawal, Vijay K

2014-10-01

Quantitative Structure-Activity Relationship (QSAR) models for binding affinity constants (log Ki) of 78 flavonoid ligands towards the benzodiazepine site of GABA (A) receptor complex were calculated using the machine learning methods: artificial neural network (ANN) and support vector machine (SVM) techniques. The models obtained were compared with those obtained using multiple linear regression (MLR) analysis. The descriptor selection and model building were performed with 10-fold cross-validation using the training data set. The SVM and MLR coefficient of determination values are 0.944 and 0.879, respectively, for the training set and are higher than those of ANN models. Though the SVM model shows improvement of training set fitting, the ANN model was superior to SVM and MLR in predicting the test set. Randomization test is employed to check the suitability of the models.

Communications and control for electric power systems: Power flow classification for static security assessment

NASA Technical Reports Server (NTRS)

Niebur, D.; Germond, A.

1993-01-01

This report investigates the classification of power system states using an artificial neural network model, Kohonen's self-organizing feature map. The ultimate goal of this classification is to assess power system static security in real-time. Kohonen's self-organizing feature map is an unsupervised neural network which maps N-dimensional input vectors to an array of M neurons. After learning, the synaptic weight vectors exhibit a topological organization which represents the relationship between the vectors of the training set. This learning is unsupervised, which means that the number and size of the classes are not specified beforehand. In the application developed in this report, the input vectors used as the training set are generated by off-line load-flow simulations. The learning algorithm and the results of the organization are discussed.

In-situ trainable intrusion detection system

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

Symons, Christopher T.; Beaver, Justin M.; Gillen, Rob

A computer implemented method detects intrusions using a computer by analyzing network traffic. The method includes a semi-supervised learning module connected to a network node. The learning module uses labeled and unlabeled data to train a semi-supervised machine learning sensor. The method records events that include a feature set made up of unauthorized intrusions and benign computer requests. The method identifies at least some of the benign computer requests that occur during the recording of the events while treating the remainder of the data as unlabeled. The method trains the semi-supervised learning module at the network node in-situ, such thatmore » the semi-supervised learning modules may identify malicious traffic without relying on specific rules, signatures, or anomaly detection.« less

Use of probabilistic neural networks for emitter correlation

NASA Astrophysics Data System (ADS)

Maloney, P. S.

1990-08-01

The Probabilistic Neural Network (PNN) as described by Specht''3 has been successfully applied to a number of emitter correlation problems involving operational data for training and testing of the neural net work. The PNN has been found to be a reliable classification tool for determining emitter type or even identifying specific emitter platforms given appropriate representative data sets for training con sisting only of parametric data from electronic intelligence (ELINT) reports. Four separate feasibility studies have been conducted to prove the usefulness of PNN in this application area: . Hull-to-emitter correlation (HULTEC) for identification of seagoing emitter platforms . Identification of landbased emitters from airborne sensors . Pulse sorting according to emitter of origin . Emitter typing based on a dynamically learning neural network. 1 .

Vertical integration in medical settings: A brief introduction to its potential effects on professional psychology.

PubMed

Sumerall, S W; Oehlert, M E; Trent, D D

1995-12-01

Vertical integration in medical settings typically involves the merging of independent physicians, physician groups, and hospitals to render an organized health care network. Such systems are considered to be vertical, as they may allow for a seamless continuation of services throughout the range of needs a patient may require. Mergers often result in the redefining of professional services offered in the acquired facility or across the network. As such, mergers have the potential of adversely impacting psychological practices. Professional psychology needs to take a proactive stance in this changing health care landscape. Research regarding empirically validated treatments and effects of psychological interventions on overall health-care costs needs to be properly disseminated to health care administrators to assure their knowledge of the utility of psychological services in the medical setting. Training psychologists to assume leadership positions in health-care institutions, gaining representation on hospital staff boards, and linking psychologists and physicians through collaborative training, to provide improved care, may allow for psychology to influence health care delivery.

Differential recruitment of the sensorimotor putamen and frontoparietal cortex during motor chunking in humans

PubMed Central

Wymbs, Nicholas F.; Bassett, Danielle S.; Mucha, Peter J.; Porter, Mason A.; Grafton, Scott T.

2012-01-01

Motor chunking facilitates movement production by combining motor elements into integrated units of behavior. Previous research suggests that chunking involves two processes: concatenation, aimed at the formation of motor-motor associations between elements or sets of elements; and segmentation, aimed at the parsing of multiple contiguous elements into shorter action sets. We used fMRI to measure the trial-wise recruitment of brain regions associated with these chunking processes as healthy subjects performed a cued sequence production task. A novel dynamic network analysis identified chunking structure for a set of motor sequences acquired during fMRI and collected on three days of training. Activity in the bilateral sensorimotor putamen positively correlated with chunk concatenation, whereas a left hemisphere frontoparietal network was correlated with chunk segmentation. Across subjects, there was an aggregate increase in chunk strength (concatenation) with training, suggesting that subcortical circuits play a direct role in the creation of fluid transitions across chunks. PMID:22681696

Differential recruitment of the sensorimotor putamen and frontoparietal cortex during motor chunking in humans.

PubMed

Wymbs, Nicholas F; Bassett, Danielle S; Mucha, Peter J; Porter, Mason A; Grafton, Scott T

2012-06-07

Motor chunking facilitates movement production by combining motor elements into integrated units of behavior. Previous research suggests that chunking involves two processes: concatenation, aimed at the formation of motor-motor associations between elements or sets of elements, and segmentation, aimed at the parsing of multiple contiguous elements into shorter action sets. We used fMRI to measure the trial-wise recruitment of brain regions associated with these chunking processes as healthy subjects performed a cued-sequence production task. A dynamic network analysis identified chunking structure for a set of motor sequences acquired during fMRI and collected over 3 days of training. Activity in the bilateral sensorimotor putamen positively correlated with chunk concatenation, whereas a left-hemisphere frontoparietal network was correlated with chunk segmentation. Across subjects, there was an aggregate increase in chunk strength (concatenation) with training, suggesting that subcortical circuits play a direct role in the creation of fluid transitions across chunks. Copyright © 2012 Elsevier Inc. All rights reserved.

A fully automatic microcalcification detection approach based on deep convolution neural network

NASA Astrophysics Data System (ADS)

Cai, Guanxiong; Guo, Yanhui; Zhang, Yaqin; Qin, Genggeng; Zhou, Yuanpin; Lu, Yao

2018-02-01

Breast cancer is one of the most common cancers and has high morbidity and mortality worldwide, posing a serious threat to the health of human beings. The emergence of microcalcifications (MCs) is an important signal of early breast cancer. However, it is still challenging and time consuming for radiologists to identify some tiny and subtle individual MCs in mammograms. This study proposed a novel computer-aided MC detection algorithm on the full field digital mammograms (FFDMs) using deep convolution neural network (DCNN). Firstly, a MC candidate detection system was used to obtain potential MC candidates. Then a DCNN was trained using a novel adaptive learning strategy, neutrosophic reinforcement sample learning (NRSL) strategy to speed up the learning process. The trained DCNN served to recognize true MCs. After been classified by DCNN, a density-based regional clustering method was imposed to form MC clusters. The accuracy of the DCNN with our proposed NRSL strategy converges faster and goes higher than the traditional DCNN at same epochs, and the obtained an accuracy of 99.87% on training set, 95.12% on validation set, and 93.68% on testing set at epoch 40. For cluster-based MC cluster detection evaluation, a sensitivity of 90% was achieved at 0.13 false positives (FPs) per image. The obtained results demonstrate that the designed DCNN plays a significant role in the MC detection after being prior trained.

Generalization error analysis: deep convolutional neural network in mammography

NASA Astrophysics Data System (ADS)

Richter, Caleb D.; Samala, Ravi K.; Chan, Heang-Ping; Hadjiiski, Lubomir; Cha, Kenny

2018-02-01

We conducted a study to gain understanding of the generalizability of deep convolutional neural networks (DCNNs) given their inherent capability to memorize data. We examined empirically a specific DCNN trained for classification of masses on mammograms. Using a data set of 2,454 lesions from 2,242 mammographic views, a DCNN was trained to classify masses into malignant and benign classes using transfer learning from ImageNet LSVRC-2010. We performed experiments with varying amounts of label corruption and types of pixel randomization to analyze the generalization error for the DCNN. Performance was evaluated using the area under the receiver operating characteristic curve (AUC) with an N-fold cross validation. Comparisons were made between the convergence times, the inference AUCs for both the training set and the test set of the original image patches without corruption, and the root-mean-squared difference (RMSD) in the layer weights of the DCNN trained with different amounts and methods of corruption. Our experiments observed trends which revealed that the DCNN overfitted by memorizing corrupted data. More importantly, this study improved our understanding of DCNN weight updates when learning new patterns or new labels. Although we used a specific classification task with the ImageNet as example, similar methods may be useful for analysis of the DCNN learning processes, especially those that employ transfer learning for medical image analysis where sample size is limited and overfitting risk is high.

Vehicle classification in WAMI imagery using deep network

NASA Astrophysics Data System (ADS)

Yi, Meng; Yang, Fan; Blasch, Erik; Sheaff, Carolyn; Liu, Kui; Chen, Genshe; Ling, Haibin

2016-05-01

Humans have always had a keen interest in understanding activities and the surrounding environment for mobility, communication, and survival. Thanks to recent progress in photography and breakthroughs in aviation, we are now able to capture tens of megapixels of ground imagery, namely Wide Area Motion Imagery (WAMI), at multiple frames per second from unmanned aerial vehicles (UAVs). WAMI serves as a great source for many applications, including security, urban planning and route planning. These applications require fast and accurate image understanding which is time consuming for humans, due to the large data volume and city-scale area coverage. Therefore, automatic processing and understanding of WAMI imagery has been gaining attention in both industry and the research community. This paper focuses on an essential step in WAMI imagery analysis, namely vehicle classification. That is, deciding whether a certain image patch contains a vehicle or not. We collect a set of positive and negative sample image patches, for training and testing the detector. Positive samples are 64 × 64 image patches centered on annotated vehicles. We generate two sets of negative images. The first set is generated from positive images with some location shift. The second set of negative patches is generated from randomly sampled patches. We also discard those patches if a vehicle accidentally locates at the center. Both positive and negative samples are randomly divided into 9000 training images and 3000 testing images. We propose to train a deep convolution network for classifying these patches. The classifier is based on a pre-trained AlexNet Model in the Caffe library, with an adapted loss function for vehicle classification. The performance of our classifier is compared to several traditional image classifier methods using Support Vector Machine (SVM) and Histogram of Oriented Gradient (HOG) features. While the SVM+HOG method achieves an accuracy of 91.2%, the accuracy of our deep network-based classifier reaches 97.9%.

Calculation of Crystallographic Texture of BCC Steels During Cold Rolling

NASA Astrophysics Data System (ADS)

Das, Arpan

2017-05-01

BCC alloys commonly tend to develop strong fibre textures and often represent as isointensity diagrams in φ 1 sections or by fibre diagrams. Alpha fibre in bcc steels is generally characterised by <110> crystallographic axis parallel to the rolling direction. The objective of present research is to correlate carbon content, carbide dispersion, rolling reduction, Euler angles (ϕ) (when φ 1 = 0° and φ 2 = 45° along alpha fibre) and the resulting alpha fibre texture orientation intensity. In the present research, Bayesian neural computation has been employed to correlate these and compare with the existing feed-forward neural network model comprehensively. Excellent match to the measured texture data within the bounding box of texture training data set has been already predicted through the feed-forward neural network model by other researchers. Feed-forward neural network prediction outside the bounds of training texture data showed deviations from the expected values. Currently, Bayesian computation has been similarly applied to confirm that the predictions are reasonable in the context of basic metallurgical principles, and matched better outside the bounds of training texture data set than the reported feed-forward neural network. Bayesian computation puts error bars on predicted values and allows significance of each individual parameters to be estimated. Additionally, it is also possible by Bayesian computation to estimate the isolated influence of particular variable such as carbon concentration, which exactly cannot in practice be varied independently. This shows the ability of the Bayesian neural network to examine the new phenomenon in situations where the data cannot be accessed through experiments.

Actividad solar del ciclo 23. Predicción del máximo y fase decreciente utilizando redes neuronales

NASA Astrophysics Data System (ADS)

Parodi, M. A.; Ceccatto, H. A.; Piacentini, R. D.; García, P. J.

Different methods have been proposed in order to predict the maximum amplitude of solar cycles, either as a consequence of the intrinsic importance of this event and because of its relation with solar storms and possible effects upon satellites, communication systems, etc. In this work, a neural network solar activity prediction is presented, measured through the sunspot number (SSN). The 16-units neural network, with a 12:3:1 architecture, was trained in a ``feed-forward" propagation way and learning by the so called ``back propagation rule". The annual mean SSN data in the 1700-1975 and 1987-1998 periods were used as the training set. The solar cycle 21 (1976-1986) was taken as the cross-validation data set. After performing the network training we obtained a prediction of the maximum annual mean for the current solar cycle 23, SSNmax= 135 ±17 at the year 2000, which is 13% smaller than the International Consensus Commitee's mean maximum prediction obtained through ``precursor techniques". On the other hand, our prediction is only about 4% smaller than the Consensus's neural network mean prediction. A ``multiple step" prediction technique was also performed and SSN annual mean predicted values for the near-maximum (from the present year 1999 to beyond the maximum) and the declining activity of solar cycle 23 are presented in this work. The sensibility of predictions is also tested. To do so, we changed the interval width and comparated our results with those of a previous neural network prediction and those of others authors using differents methods.

Signal or noise: brain network interactions underlying the experience and training of mindfulness.

PubMed

Mooneyham, Benjamin W; Mrazek, Michael D; Mrazek, Alissa J; Schooler, Jonathan W

2016-04-01

A broad set of brain regions has been associated with the experience and training of mindfulness. Many of these regions lie within key intrinsic brain networks, including the executive control, salience, and default networks. In this paper, we review the existing literature on the cognitive neuroscience of mindfulness through the lens of network science. We describe the characteristics of the intrinsic brain networks implicated in mindfulness and summarize the relevant findings pertaining to changes in functional connectivity (FC) within and between these networks. Convergence across these findings suggests that mindfulness may be associated with increased FC between two regions within the default network: the posterior cingulate cortex and the ventromedial prefrontal cortex. Additionally, extensive meditation experience may be associated with increased FC between the insula and the dorsolateral prefrontal cortex. However, little consensus has emerged within the existing literature owing to the diversity of operational definitions of mindfulness, neuroimaging methods, and network characterizations. We describe several challenges to develop a coherent cognitive neuroscience of mindfulness and to provide detailed recommendations for future research. © 2016 New York Academy of Sciences.

Multiclass classification of obstructive sleep apnea/hypopnea based on a convolutional neural network from a single-lead electrocardiogram.

PubMed

Urtnasan, Erdenebayar; Park, Jong-Uk; Lee, Kyoung-Joung

2018-05-24

In this paper, we propose a convolutional neural network (CNN)-based deep learning architecture for multiclass classification of obstructive sleep apnea and hypopnea (OSAH) using single-lead electrocardiogram (ECG) recordings. OSAH is the most common sleep-related breathing disorder. Many subjects who suffer from OSAH remain undiagnosed; thus, early detection of OSAH is important. In this study, automatic classification of three classes-normal, hypopnea, and apnea-based on a CNN is performed. An optimal six-layer CNN model is trained on a training dataset (45,096 events) and evaluated on a test dataset (11,274 events). The training set (69 subjects) and test set (17 subjects) were collected from 86 subjects with length of approximately 6 h and segmented into 10 s durations. The proposed CNN model reaches a mean -score of 93.0 for the training dataset and 87.0 for the test dataset. Thus, proposed deep learning architecture achieved a high performance for multiclass classification of OSAH using single-lead ECG recordings. The proposed method can be employed in screening of patients suspected of having OSAH. © 2018 Institute of Physics and Engineering in Medicine.

Introduction to the EC's Marie Curie Initial Training Network (MC-ITN) project: Particle Training Network for European Radiotherapy (PARTNER).

PubMed

Dosanjh, Manjit; Magrin, Giulio

2013-07-01

PARTNER (Particle Training Network for European Radiotherapy) is a project funded by the European Commission's Marie Curie-ITN funding scheme through the ENLIGHT Platform for 5.6 million Euro. PARTNER has brought together academic institutes, research centres and leading European companies, focusing in particular on a specialized radiotherapy (RT) called hadron therapy (HT), interchangeably referred to as particle therapy (PT). The ultimate goal of HT is to deliver more effective treatment to cancer patients leading to major improvement in the health of citizens. In Europe, several hundred million Euro have been invested, since the beginning of this century, in PT. In this decade, the use of HT is rapidly growing across Europe, and there is an urgent need for qualified researchers from a range of disciplines to work on its translational research. In response to this need, the European community of HT, and in particular 10 leading academic institutes, research centres, companies and small and medium-sized enterprises, joined together to form the PARTNER consortium. All partners have international reputations in the diverse but complementary fields associated with PT: clinical, radiobiological and technological. Thus the network incorporates a unique set of competencies, expertise, infrastructures and training possibilities. This paper describes the status and needs of PT research in Europe, the importance of and challenges associated with the creation of a training network, the objectives, the initial results, and the expected long-term benefits of the PARTNER initiative.

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

Nogueira, C. P. S. M.; Guimarães, J. G.

In this paper, an auto-associative neural network using single-electron tunneling (SET) devices is proposed and simulated at low temperature. The nanoelectronic auto-associative network is able to converge to a stable state, previously stored during training. The recognition of the pattern involves decreasing the energy of the input state until it achieves a point of local minimum energy, which corresponds to one of the stored patterns.

Pruning Neural Networks with Distribution Estimation Algorithms

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

Cantu-Paz, E

2003-01-15

This paper describes the application of four evolutionary algorithms to the pruning of neural networks used in classification problems. Besides of a simple genetic algorithm (GA), the paper considers three distribution estimation algorithms (DEAs): a compact GA, an extended compact GA, and the Bayesian Optimization Algorithm. The objective is to determine if the DEAs present advantages over the simple GA in terms of accuracy or speed in this problem. The experiments used a feed forward neural network trained with standard back propagation and public-domain and artificial data sets. The pruned networks seemed to have better or equal accuracy than themore » original fully-connected networks. Only in a few cases, pruning resulted in less accurate networks. We found few differences in the accuracy of the networks pruned by the four EAs, but found important differences in the execution time. The results suggest that a simple GA with a small population might be the best algorithm for pruning networks on the data sets we tested.« less

Neural network modeling of drying of rice in BAU-STR dryer

NASA Astrophysics Data System (ADS)

Alam, Md. Ashraful; Saha, Chayan Kumer; Alam, Md. Monjurul; Ashraf, Md. Ali; Bala, Bilash Kanti; Harvey, Jagger

2018-05-01

The experimental performance and artificial neural network modeling of rice drying in BAU-STR dryer is presented in this paper. The dryer consists of a biomass stove as a heat source, a perforated inner bin and a perforated outer bin with annular space for grains, and a blower (1 hp) to supply heated air. The dryer capacity was 500 kg of freshly harvested rice. Twenty experimental runs were conducted to investigate the experimental performance of the dryer for drying of rice. An independent multilayer neural network approach was used to predict the performance of the BAU-STR dryer for drying of rice. Ten sets of experimental data were used for training using back propagation algorithm and another ten sets of data were used for testing the artificial neural network model. The prediction of the performance of the dryer was found to be excellent after it was adequately trained. The statistical analysis showed that the errors (MSE and RMSE) were within and acceptable range of ±5% with a coefficient of determination (R2) of 99%. The model can be used to predict the potential of the dryer for different locations, and can also be used in a predictive optimal control algorithm.

Modeling of steam distillation mechanism during steam injection process using artificial intelligence.

PubMed

Daryasafar, Amin; Ahadi, Arash; Kharrat, Riyaz

2014-01-01

Steam distillation as one of the important mechanisms has a great role in oil recovery in thermal methods and so it is important to simulate this process experimentally and theoretically. In this work, the simulation of steam distillation is performed on sixteen sets of crude oil data found in the literature. Artificial intelligence (AI) tools such as artificial neural network (ANN) and also adaptive neurofuzzy interference system (ANFIS) are used in this study as effective methods to simulate the distillate recoveries of these sets of data. Thirteen sets of data were used to train the models and three sets were used to test the models. The developed models are highly compatible with respect to input oil properties and can predict the distillate yield with minimum entry. For showing the performance of the proposed models, simulation of steam distillation is also done using modified Peng-Robinson equation of state. Comparison between the calculated distillates by ANFIS and neural network models and also equation of state-based method indicates that the errors of the ANFIS model for training data and test data sets are lower than those of other methods.

Modeling of Steam Distillation Mechanism during Steam Injection Process Using Artificial Intelligence

PubMed Central

Ahadi, Arash; Kharrat, Riyaz

2014-01-01

Steam distillation as one of the important mechanisms has a great role in oil recovery in thermal methods and so it is important to simulate this process experimentally and theoretically. In this work, the simulation of steam distillation is performed on sixteen sets of crude oil data found in the literature. Artificial intelligence (AI) tools such as artificial neural network (ANN) and also adaptive neurofuzzy interference system (ANFIS) are used in this study as effective methods to simulate the distillate recoveries of these sets of data. Thirteen sets of data were used to train the models and three sets were used to test the models. The developed models are highly compatible with respect to input oil properties and can predict the distillate yield with minimum entry. For showing the performance of the proposed models, simulation of steam distillation is also done using modified Peng-Robinson equation of state. Comparison between the calculated distillates by ANFIS and neural network models and also equation of state-based method indicates that the errors of the ANFIS model for training data and test data sets are lower than those of other methods. PMID:24883365

Performance of an artificial neural network for vertical root fracture detection: an ex vivo study.

PubMed

Kositbowornchai, Suwadee; Plermkamon, Supattra; Tangkosol, Tawan

2013-04-01

To develop an artificial neural network for vertical root fracture detection. A probabilistic neural network design was used to clarify whether a tooth root was sound or had a vertical root fracture. Two hundred images (50 sound and 150 vertical root fractures) derived from digital radiography--used to train and test the artificial neural network--were divided into three groups according to the number of training and test data sets: 80/120,105/95 and 130/70, respectively. Either training or tested data were evaluated using grey-scale data per line passing through the root. These data were normalized to reduce the grey-scale variance and fed as input data of the neural network. The variance of function in recognition data was calculated between 0 and 1 to select the best performance of neural network. The performance of the neural network was evaluated using a diagnostic test. After testing data under several variances of function, we found the highest sensitivity (98%), specificity (90.5%) and accuracy (95.7%) occurred in Group three, for which the variance of function in recognition data was between 0.025 and 0.005. The neural network designed in this study has sufficient sensitivity, specificity and accuracy to be a model for vertical root fracture detection. © 2012 John Wiley & Sons A/S.

Training and Validating a Deep Convolutional Neural Network for Computer-Aided Detection and Classification of Abnormalities on Frontal Chest Radiographs.

PubMed

Cicero, Mark; Bilbily, Alexander; Colak, Errol; Dowdell, Tim; Gray, Bruce; Perampaladas, Kuhan; Barfett, Joseph

2017-05-01

Convolutional neural networks (CNNs) are a subtype of artificial neural network that have shown strong performance in computer vision tasks including image classification. To date, there has been limited application of CNNs to chest radiographs, the most frequently performed medical imaging study. We hypothesize CNNs can learn to classify frontal chest radiographs according to common findings from a sufficiently large data set. Our institution's research ethics board approved a single-center retrospective review of 35,038 adult posterior-anterior chest radiographs and final reports performed between 2005 and 2015 (56% men, average age of 56, patient type: 24% inpatient, 39% outpatient, 37% emergency department) with a waiver for informed consent. The GoogLeNet CNN was trained using 3 graphics processing units to automatically classify radiographs as normal (n = 11,702) or into 1 or more of cardiomegaly (n = 9240), consolidation (n = 6788), pleural effusion (n = 7786), pulmonary edema (n = 1286), or pneumothorax (n = 1299). The network's performance was evaluated using receiver operating curve analysis on a test set of 2443 radiographs with the criterion standard being board-certified radiologist interpretation. Using 256 × 256-pixel images as input, the network achieved an overall sensitivity and specificity of 91% with an area under the curve of 0.964 for classifying a study as normal (n = 1203). For the abnormal categories, the sensitivity, specificity, and area under the curve, respectively, were 91%, 91%, and 0.962 for pleural effusion (n = 782), 82%, 82%, and 0.868 for pulmonary edema (n = 356), 74%, 75%, and 0.850 for consolidation (n = 214), 81%, 80%, and 0.875 for cardiomegaly (n = 482), and 78%, 78%, and 0.861 for pneumothorax (n = 167). Current deep CNN architectures can be trained with modest-sized medical data sets to achieve clinically useful performance at detecting and excluding common pathology on chest radiographs.

Implementation and Outcomes of a Collaborative Multi-Center Network Aimed at Web-Based Cognitive Training - COGWEB Network.

PubMed

Tedim Cruz, Vítor; Pais, Joana; Ruano, Luis; Mateus, Cátia; Colunas, Márcio; Alves, Ivânia; Barreto, Rui; Conde, Eduardo; Sousa, Andreia; Araújo, Isabel; Bento, Virgílio; Coutinho, Paula; Rocha, Nelson

2014-01-01

Cognitive care for the most prevalent neurologic and psychiatric conditions will only improve through the implementation of new sustainable approaches. Innovative cognitive training methodologies and collaborative professional networks are necessary evolutions in the mental health sector. The objective of the study was to describe the implementation process and early outcomes of a nationwide multi-organizational network supported on a Web-based cognitive training system (COGWEB). The setting for network implementation was the Portuguese mental health system and the hospital-, academic-, community-based institutions and professionals providing cognitive training. The network started in August 2012, with 16 centers, and was monitored until September 2013 (inclusions were open). After onsite training, all were allowed to use COGWEB in their clinical or research activities. For supervision and maintenance were implemented newsletters, questionnaires, visits and webinars. The following outcomes were prospectively measured: (1) number, (2) type, (3) time to start, and (4) activity state of centers; age, gender, level of education, and medical diagnosis of patients enrolled. The network included 68 professionals from 41 centers, (33/41) 80% clinical, (8/41) 19% nonclinical. A total of 298 patients received cognitive training; 45.3% (n=135) female, mean age 54.4 years (SD 18.7), mean educational level 9.8 years (SD 4.8). The number enrolled each month increased significantly (r=0.6; P=.031). At 12 months, 205 remained on treatment. The major causes of cognitive impairment were: (1) neurodegenerative (115/298, 38.6%), (2) structural brain lesions (63/298, 21.1%), (3) autoimmune (40/298, 13.4%), (4) schizophrenia (30/298, 10.1%), and (5) others (50/298, 16.8%). The comparison of the patient profiles, promoter versus all other clinical centers, showed significant increases in the diversity of causes and spectrums of ages and education. Over its first year, there was a major increase in the number of new centers and professionals, as well as of the clinical diversity of patients treated. The consolidation of such a national collaborative network represents an innovative step in mental health care evolution. Furthermore, it may contribute to translational processes in the field of cognitive training and reduce disease burden.

Neural network based system for equipment surveillance

DOEpatents

Vilim, Richard B.; Gross, Kenneth C.; Wegerich, Stephan W.

1998-01-01

A method and system for performing surveillance of transient signals of an industrial device to ascertain the operating state. The method and system involves the steps of reading into a memory training data, determining neural network weighting values until achieving target outputs close to the neural network output. If the target outputs are inadequate, wavelet parameters are determined to yield neural network outputs close to the desired set of target outputs and then providing signals characteristic of an industrial process and comparing the neural network output to the industrial process signals to evaluate the operating state of the industrial process.

Neural network based system for equipment surveillance

DOEpatents

Vilim, R.B.; Gross, K.C.; Wegerich, S.W.

1998-04-28

A method and system are disclosed for performing surveillance of transient signals of an industrial device to ascertain the operating state. The method and system involves the steps of reading into a memory training data, determining neural network weighting values until achieving target outputs close to the neural network output. If the target outputs are inadequate, wavelet parameters are determined to yield neural network outputs close to the desired set of target outputs and then providing signals characteristic of an industrial process and comparing the neural network output to the industrial process signals to evaluate the operating state of the industrial process. 33 figs.

Particle swarm optimization-based automatic parameter selection for deep neural networks and its applications in large-scale and high-dimensional data

PubMed Central

2017-01-01

In this paper, we propose a new automatic hyperparameter selection approach for determining the optimal network configuration (network structure and hyperparameters) for deep neural networks using particle swarm optimization (PSO) in combination with a steepest gradient descent algorithm. In the proposed approach, network configurations were coded as a set of real-number m-dimensional vectors as the individuals of the PSO algorithm in the search procedure. During the search procedure, the PSO algorithm is employed to search for optimal network configurations via the particles moving in a finite search space, and the steepest gradient descent algorithm is used to train the DNN classifier with a few training epochs (to find a local optimal solution) during the population evaluation of PSO. After the optimization scheme, the steepest gradient descent algorithm is performed with more epochs and the final solutions (pbest and gbest) of the PSO algorithm to train a final ensemble model and individual DNN classifiers, respectively. The local search ability of the steepest gradient descent algorithm and the global search capabilities of the PSO algorithm are exploited to determine an optimal solution that is close to the global optimum. We constructed several experiments on hand-written characters and biological activity prediction datasets to show that the DNN classifiers trained by the network configurations expressed by the final solutions of the PSO algorithm, employed to construct an ensemble model and individual classifier, outperform the random approach in terms of the generalization performance. Therefore, the proposed approach can be regarded an alternative tool for automatic network structure and parameter selection for deep neural networks. PMID:29236718

Prediction of strain values in reinforcements and concrete of a RC frame using neural networks

NASA Astrophysics Data System (ADS)

Vafaei, Mohammadreza; Alih, Sophia C.; Shad, Hossein; Falah, Ali; Halim, Nur Hajarul Falahi Abdul

2018-03-01

The level of strain in structural elements is an important indicator for the presence of damage and its intensity. Considering this fact, often structural health monitoring systems employ strain gauges to measure strains in critical elements. However, because of their sensitivity to the magnetic fields, inadequate long-term durability especially in harsh environments, difficulties in installation on existing structures, and maintenance cost, installation of strain gauges is not always possible for all structural components. Therefore, a reliable method that can accurately estimate strain values in critical structural elements is necessary for damage identification. In this study, a full-scale test was conducted on a planar RC frame to investigate the capability of neural networks for predicting the strain values. Two neural networks each of which having a single hidden layer was trained to relate the measured rotations and vertical displacements of the frame to the strain values measured at different locations of the frame. Results of trained neural networks indicated that they accurately estimated the strain values both in reinforcements and concrete. In addition, the trained neural networks were capable of predicting strains for the unseen input data set.

An Automatic Diagnosis Method of Facial Acne Vulgaris Based on Convolutional Neural Network.

PubMed

Shen, Xiaolei; Zhang, Jiachi; Yan, Chenjun; Zhou, Hong

2018-04-11

In this paper, we present a new automatic diagnosis method for facial acne vulgaris which is based on convolutional neural networks (CNNs). To overcome the shortcomings of previous methods which were the inability to classify enough types of acne vulgaris. The core of our method is to extract features of images based on CNNs and achieve classification by classifier. A binary-classifier of skin-and-non-skin is used to detect skin area and a seven-classifier is used to achieve the classification task of facial acne vulgaris and healthy skin. In the experiments, we compare the effectiveness of our CNN and the VGG16 neural network which is pre-trained on the ImageNet data set. We use a ROC curve to evaluate the performance of binary-classifier and use a normalized confusion matrix to evaluate the performance of seven-classifier. The results of our experiments show that the pre-trained VGG16 neural network is effective in extracting features from facial acne vulgaris images. And the features are very useful for the follow-up classifiers. Finally, we try applying the classifiers both based on the pre-trained VGG16 neural network to assist doctors in facial acne vulgaris diagnosis.

Defect detection and classification of galvanized stamping parts based on fully convolution neural network

NASA Astrophysics Data System (ADS)

Xiao, Zhitao; Leng, Yanyi; Geng, Lei; Xi, Jiangtao

2018-04-01

In this paper, a new convolution neural network method is proposed for the inspection and classification of galvanized stamping parts. Firstly, all workpieces are divided into normal and defective by image processing, and then the defective workpieces extracted from the region of interest (ROI) area are input to the trained fully convolutional networks (FCN). The network utilizes an end-to-end and pixel-to-pixel training convolution network that is currently the most advanced technology in semantic segmentation, predicts result of each pixel. Secondly, we mark the different pixel values of the workpiece, defect and background for the training image, and use the pixel value and the number of pixels to realize the recognition of the defects of the output picture. Finally, the defect area's threshold depended on the needs of the project is set to achieve the specific classification of the workpiece. The experiment results show that the proposed method can successfully achieve defect detection and classification of galvanized stamping parts under ordinary camera and illumination conditions, and its accuracy can reach 99.6%. Moreover, it overcomes the problem of complex image preprocessing and difficult feature extraction and performs better adaptability.

Computer aided detection of clusters of microcalcifications on full field digital mammograms

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

Ge Jun; Sahiner, Berkman; Hadjiiski, Lubomir M.

2006-08-15

We are developing a computer-aided detection (CAD) system to identify microcalcification clusters (MCCs) automatically on full field digital mammograms (FFDMs). The CAD system includes six stages: preprocessing; image enhancement; segmentation of microcalcification candidates; false positive (FP) reduction for individual microcalcifications; regional clustering; and FP reduction for clustered microcalcifications. At the stage of FP reduction for individual microcalcifications, a truncated sum-of-squares error function was used to improve the efficiency and robustness of the training of an artificial neural network in our CAD system for FFDMs. At the stage of FP reduction for clustered microcalcifications, morphological features and features derived from themore » artificial neural network outputs were extracted from each cluster. Stepwise linear discriminant analysis (LDA) was used to select the features. An LDA classifier was then used to differentiate clustered microcalcifications from FPs. A data set of 96 cases with 192 images was collected at the University of Michigan. This data set contained 96 MCCs, of which 28 clusters were proven by biopsy to be malignant and 68 were proven to be benign. The data set was separated into two independent data sets for training and testing of the CAD system in a cross-validation scheme. When one data set was used to train and validate the convolution neural network (CNN) in our CAD system, the other data set was used to evaluate the detection performance. With the use of a truncated error metric, the training of CNN could be accelerated and the classification performance was improved. The CNN in combination with an LDA classifier could substantially reduce FPs with a small tradeoff in sensitivity. By using the free-response receiver operating characteristic methodology, it was found that our CAD system can achieve a cluster-based sensitivity of 70, 80, and 90 % at 0.21, 0.61, and 1.49 FPs/image, respectively. For case-based performance evaluation, a sensitivity of 70, 80, and 90 % can be achieved at 0.07, 0.17, and 0.65 FPs/image, respectively. We also used a data set of 216 mammograms negative for clustered microcalcifications to further estimate the FP rate of our CAD system. The corresponding FP rates were 0.15, 0.31, and 0.86 FPs/image for cluster-based detection when negative mammograms were used for estimation of FP rates.« less

Modification Of Learning Rate With Lvq Model Improvement In Learning Backpropagation

NASA Astrophysics Data System (ADS)

Tata Hardinata, Jaya; Zarlis, Muhammad; Budhiarti Nababan, Erna; Hartama, Dedy; Sembiring, Rahmat W.

2017-12-01

One type of artificial neural network is a backpropagation, This algorithm trained with the network architecture used during the training as well as providing the correct output to insert a similar but not the same with the architecture in use at training.The selection of appropriate parameters also affects the outcome, value of learning rate is one of the parameters which influence the process of training, Learning rate affects the speed of learning process on the network architecture.If the learning rate is set too large, then the algorithm will become unstable and otherwise the algorithm will converge in a very long period of time.So this study was made to determine the value of learning rate on the backpropagation algorithm. LVQ models of learning rate is one of the models used in the determination of the value of the learning rate of the algorithm LVQ.By modifying this LVQ model to be applied to the backpropagation algorithm. From the experimental results known to modify the learning rate LVQ models were applied to the backpropagation algorithm learning process becomes faster (epoch less).

A long non-coding RNA expression profile can predict early recurrence in hepatocellular carcinoma after curative resection.

PubMed

Lv, Yufeng; Wei, Wenhao; Huang, Zhong; Chen, Zhichao; Fang, Yuan; Pan, Lili; Han, Xueqiong; Xu, Zihai

2018-06-20

The aim of this study was to develop a novel long non-coding RNA (lncRNA) expression signature to accurately predict early recurrence for patients with hepatocellular carcinoma (HCC) after curative resection. Using expression profiles downloaded from The Cancer Genome Atlas database, we identified multiple lncRNAs with differential expression between early recurrence (ER) group and non-early recurrence (non-ER) group of HCC. Least absolute shrinkage and selection operator (LASSO) for logistic regression models were used to develop a lncRNA-based classifier for predicting ER in the training set. An independent test set was used to validated the predictive value of this classifier. Futhermore, a co-expression network based on these lncRNAs and its highly related genes was constructed and Gene Ontology and Kyoto Encyclopedia of Genes and Genomes pathway enrichment analyses of genes in the network were performed. We identified 10 differentially expressed lncRNAs, including 3 that were upregulated and 7 that were downregulated in ER group. The lncRNA-based classifier was constructed based on 7 lncRNAs (AL035661.1, PART1, AC011632.1, AC109588.1, AL365361.1, LINC00861 and LINC02084), and its accuracy was 0.83 in training set, 0.87 in test set and 0.84 in total set. And ROC curve analysis showed the AUROC was 0.741 in training set, 0.824 in the test set and 0.765 in total set. A functional enrichment analysis suggested that the genes of which is highly related to 4 lncRNAs were involved in immune system. This 7-lncRNA expression profile can effectively predict the early recurrence after surgical resection for HCC. This article is protected by copyright. All rights reserved.

Artificial neural network modeling of dissolved oxygen in the Heihe River, Northwestern China.

PubMed

Wen, Xiaohu; Fang, Jing; Diao, Meina; Zhang, Chuanqi

2013-05-01

Identification and quantification of dissolved oxygen (DO) profiles of river is one of the primary concerns for water resources managers. In this research, an artificial neural network (ANN) was developed to simulate the DO concentrations in the Heihe River, Northwestern China. A three-layer back-propagation ANN was used with the Bayesian regularization training algorithm. The input variables of the neural network were pH, electrical conductivity, chloride (Cl(-)), calcium (Ca(2+)), total alkalinity, total hardness, nitrate nitrogen (NO3-N), and ammonical nitrogen (NH4-N). The ANN structure with 14 hidden neurons obtained the best selection. By making comparison between the results of the ANN model and the measured data on the basis of correlation coefficient (r) and root mean square error (RMSE), a good model-fitting DO values indicated the effectiveness of neural network model. It is found that the coefficient of correlation (r) values for the training, validation, and test sets were 0.9654, 0.9841, and 0.9680, respectively, and the respective values of RMSE for the training, validation, and test sets were 0.4272, 0.3667, and 0.4570, respectively. Sensitivity analysis was used to determine the influence of input variables on the dependent variable. The most effective inputs were determined as pH, NO3-N, NH4-N, and Ca(2+). Cl(-) was found to be least effective variables on the proposed model. The identified ANN model can be used to simulate the water quality parameters.

LavaNet—Neural network development environment in a general mine planning package

NASA Astrophysics Data System (ADS)

Kapageridis, Ioannis Konstantinou; Triantafyllou, A. G.

2011-04-01

LavaNet is a series of scripts written in Perl that gives access to a neural network simulation environment inside a general mine planning package. A well known and a very popular neural network development environment, the Stuttgart Neural Network Simulator, is used as the base for the development of neural networks. LavaNet runs inside VULCAN™—a complete mine planning package with advanced database, modelling and visualisation capabilities. LavaNet is taking advantage of VULCAN's Perl based scripting environment, Lava, to bring all the benefits of neural network development and application to geologists, mining engineers and other users of the specific mine planning package. LavaNet enables easy development of neural network training data sets using information from any of the data and model structures available, such as block models and drillhole databases. Neural networks can be trained inside VULCAN™ and the results be used to generate new models that can be visualised in 3D. Direct comparison of developed neural network models with conventional and geostatistical techniques is now possible within the same mine planning software package. LavaNet supports Radial Basis Function networks, Multi-Layer Perceptrons and Self-Organised Maps.

An automatic microseismic or acoustic emission arrival identification scheme with deep recurrent neural networks

NASA Astrophysics Data System (ADS)

Zheng, Jing; Lu, Jiren; Peng, Suping; Jiang, Tianqi

2018-02-01

The conventional arrival pick-up algorithms cannot avoid the manual modification of the parameters for the simultaneous identification of multiple events under different signal-to-noise ratios (SNRs). Therefore, in order to automatically obtain the arrivals of multiple events with high precision under different SNRs, in this study an algorithm was proposed which had the ability to pick up the arrival of microseismic or acoustic emission events based on deep recurrent neural networks. The arrival identification was performed using two important steps, which included a training phase and a testing phase. The training process was mathematically modelled by deep recurrent neural networks using Long Short-Term Memory architecture. During the testing phase, the learned weights were utilized to identify the arrivals through the microseismic/acoustic emission data sets. The data sets were obtained by rock physics experiments of the acoustic emission. In order to obtain the data sets under different SNRs, this study added random noise to the raw experiments' data sets. The results showed that the outcome of the proposed method was able to attain an above 80 per cent hit-rate at SNR 0 dB, and an approximately 70 per cent hit-rate at SNR -5 dB, with an absolute error in 10 sampling points. These results indicated that the proposed method had high selection precision and robustness.

Training the emotional brain: improving affective control through emotional working memory training.

PubMed

Schweizer, Susanne; Grahn, Jessica; Hampshire, Adam; Mobbs, Dean; Dalgleish, Tim

2013-03-20

Affective cognitive control capacity (e.g., the ability to regulate emotions or manipulate emotional material in the service of task goals) is associated with professional and interpersonal success. Impoverished affective control, by contrast, characterizes many neuropsychiatric disorders. Insights from neuroscience indicate that affective cognitive control relies on the same frontoparietal neural circuitry as working memory (WM) tasks, which suggests that systematic WM training, performed in an emotional context, has the potential to augment affective control. Here we show, using behavioral and fMRI measures, that 20 d of training on a novel emotional WM protocol successfully enhanced the efficiency of this frontoparietal demand network. Critically, compared with placebo training, emotional WM training also accrued transfer benefits to a "gold standard" measure of affective cognitive control-emotion regulation. These emotion regulation gains were associated with greater activity in the targeted frontoparietal demand network along with other brain regions implicated in affective control, notably the subgenual anterior cingulate cortex. The results have important implications for the utility of WM training in clinical, prevention, and occupational settings.

Topsoil pollution forecasting using artificial neural networks on the example of the abnormally distributed heavy metal at Russian subarctic

NASA Astrophysics Data System (ADS)

Tarasov, D. A.; Buevich, A. G.; Sergeev, A. P.; Shichkin, A. V.; Baglaeva, E. M.

2017-06-01

Forecasting the soil pollution is a considerable field of study in the light of the general concern of environmental protection issues. Due to the variation of content and spatial heterogeneity of pollutants distribution at urban areas, the conventional spatial interpolation models implemented in many GIS packages mostly cannot provide appreciate interpolation accuracy. Moreover, the problem of prediction the distribution of the element with high variability in the concentration at the study site is particularly difficult. The work presents two neural networks models forecasting a spatial content of the abnormally distributed soil pollutant (Cr) at a particular location of the subarctic Novy Urengoy, Russia. A method of generalized regression neural network (GRNN) was compared to a common multilayer perceptron (MLP) model. The proposed techniques have been built, implemented and tested using ArcGIS and MATLAB. To verify the models performances, 150 scattered input data points (pollutant concentrations) have been selected from 8.5 km2 area and then split into independent training data set (105 points) and validation data set (45 points). The training data set was generated for the interpolation using ordinary kriging while the validation data set was used to test their accuracies. The networks structures have been chosen during a computer simulation based on the minimization of the RMSE. The predictive accuracy of both models was confirmed to be significantly higher than those achieved by the geostatistical approach (kriging). It is shown that MLP could achieve better accuracy than both kriging and even GRNN for interpolating surfaces.

Protocols for Handling Messages Between Simulation Computers

NASA Technical Reports Server (NTRS)

Balcerowski, John P.; Dunnam, Milton

2006-01-01

Practical Simulator Network (PSimNet) is a set of data-communication protocols designed especially for use in handling messages between computers that are engaging cooperatively in real-time or nearly-real-time training simulations. In a typical application, computers that provide individualized training at widely dispersed locations would communicate, by use of PSimNet, with a central host computer that would provide a common computational- simulation environment and common data. Originally intended for use in supporting interfaces between training computers and computers that simulate the responses of spacecraft scientific payloads, PSimNet could be especially well suited for a variety of other applications -- for example, group automobile-driver training in a classroom. Another potential application might lie in networking of automobile-diagnostic computers at repair facilities to a central computer that would compile the expertise of numerous technicians and engineers and act as an expert consulting technician.

A Hybrid Spectral Clustering and Deep Neural Network Ensemble Algorithm for Intrusion Detection in Sensor Networks

PubMed Central

Ma, Tao; Wang, Fen; Cheng, Jianjun; Yu, Yang; Chen, Xiaoyun

2016-01-01

The development of intrusion detection systems (IDS) that are adapted to allow routers and network defence systems to detect malicious network traffic disguised as network protocols or normal access is a critical challenge. This paper proposes a novel approach called SCDNN, which combines spectral clustering (SC) and deep neural network (DNN) algorithms. First, the dataset is divided into k subsets based on sample similarity using cluster centres, as in SC. Next, the distance between data points in a testing set and the training set is measured based on similarity features and is fed into the deep neural network algorithm for intrusion detection. Six KDD-Cup99 and NSL-KDD datasets and a sensor network dataset were employed to test the performance of the model. These experimental results indicate that the SCDNN classifier not only performs better than backpropagation neural network (BPNN), support vector machine (SVM), random forest (RF) and Bayes tree models in detection accuracy and the types of abnormal attacks found. It also provides an effective tool of study and analysis of intrusion detection in large networks. PMID:27754380

A Hybrid Spectral Clustering and Deep Neural Network Ensemble Algorithm for Intrusion Detection in Sensor Networks.

PubMed

Ma, Tao; Wang, Fen; Cheng, Jianjun; Yu, Yang; Chen, Xiaoyun

2016-10-13

The development of intrusion detection systems (IDS) that are adapted to allow routers and network defence systems to detect malicious network traffic disguised as network protocols or normal access is a critical challenge. This paper proposes a novel approach called SCDNN, which combines spectral clustering (SC) and deep neural network (DNN) algorithms. First, the dataset is divided into k subsets based on sample similarity using cluster centres, as in SC. Next, the distance between data points in a testing set and the training set is measured based on similarity features and is fed into the deep neural network algorithm for intrusion detection. Six KDD-Cup99 and NSL-KDD datasets and a sensor network dataset were employed to test the performance of the model. These experimental results indicate that the SCDNN classifier not only performs better than backpropagation neural network (BPNN), support vector machine (SVM), random forest (RF) and Bayes tree models in detection accuracy and the types of abnormal attacks found. It also provides an effective tool of study and analysis of intrusion detection in large networks.

Higher-order neural network software for distortion invariant object recognition

NASA Technical Reports Server (NTRS)

Reid, Max B.; Spirkovska, Lilly

1991-01-01

The state-of-the-art in pattern recognition for such applications as automatic target recognition and industrial robotic vision relies on digital image processing. We present a higher-order neural network model and software which performs the complete feature extraction-pattern classification paradigm required for automatic pattern recognition. Using a third-order neural network, we demonstrate complete, 100 percent accurate invariance to distortions of scale, position, and in-plate rotation. In a higher-order neural network, feature extraction is built into the network, and does not have to be learned. Only the relatively simple classification step must be learned. This is key to achieving very rapid training. The training set is much smaller than with standard neural network software because the higher-order network only has to be shown one view of each object to be learned, not every possible view. The software and graphical user interface run on any Sun workstation. Results of the use of the neural software in autonomous robotic vision systems are presented. Such a system could have extensive application in robotic manufacturing.

Temporal coding in a silicon network of integrate-and-fire neurons.

PubMed

Liu, Shih-Chii; Douglas, Rodney

2004-09-01

Spatio-temporal processing of spike trains by neuronal networks depends on a variety of mechanisms distributed across synapses, dendrites, and somata. In natural systems, the spike trains and the processing mechanisms cohere though their common physical instantiation. This coherence is lost when the natural system is encoded for simulation on a general purpose computer. By contrast, analog VLSI circuits are, like neurons, inherently related by their real-time physics, and so, could provide a useful substrate for exploring neuronlike event-based processing. Here, we describe a hybrid analog-digital VLSI chip comprising a set of integrate-and-fire neurons and short-term dynamical synapses that can be configured into simple network architectures with some properties of neocortical neuronal circuits. We show that, despite considerable fabrication variance in the properties of individual neurons, the chip offers a viable substrate for exploring real-time spike-based processing in networks of neurons.

Hacia la predicción del Número R de Wolf de manchas solares utilizando Redes Neuronales con retardos temporales

NASA Astrophysics Data System (ADS)

Francile, C.; Luoni, M. L.

We present a prediction of the time series of the Wolf number R of sunspots using "time lagged feed forward neural networks". We use two types of networks: the focused and distributed ones which were trained with the back propagation of errors algorithm and the temporal back propagation algorithm respectively. As inputs to neural networks we use the time series of the number R averaged annually and monthly with the method IR5. As data sets for training and test we choose certain intervals of the time series similar to other works, in order to compare the results. Finally we discuss the topology of the networks used, the number of delays used, the number of neurons per layer, the number of hidden layers and the results in the prediction of the series between one and six steps ahead. FULL TEXT IN SPANISH

Neural network evaluation of reflectometry density profiles for control purposes

NASA Astrophysics Data System (ADS)

Santos, J.; Nunes, F.; Manso, M.; Nunes, I.

1999-01-01

Broadband reflectometry is a diagnostic that is able to measure the density profile with high spatial and temporal resolutions, therefore it can be used to improve the performance of advanced tokamak operation modes and to supplement or correct the magnetics for plasma position control. To perform these tasks real-time processing is needed. Here we present a method that uses a neural network to make a fast evaluation of radial positions for selected density layers. Typical ASDEX Upgrade density profiles were used to generate the simulated network training and test sets. It is shown that the method has the potential to meet the tight timing requirements of control applications with the required accuracy. The network is also able to provide an accurate estimation of the position of density layers below the first density layer which is probed by an O-mode reflectometer, provided that it is trained with a realistic density profile model.

"Geo-statistics methods and neural networks in geophysical applications: A case study"

NASA Astrophysics Data System (ADS)

Rodriguez Sandoval, R.; Urrutia Fucugauchi, J.; Ramirez Cruz, L. C.

2008-12-01

The study is focus in the Ebano-Panuco basin of northeastern Mexico, which is being explored for hydrocarbon reservoirs. These reservoirs are in limestones and there is interest in determining porosity and permeability in the carbonate sequences. The porosity maps presented in this study are estimated from application of multiattribute and neural networks techniques, which combine geophysics logs and 3-D seismic data by means of statistical relationships. The multiattribute analysis is a process to predict a volume of any underground petrophysical measurement from well-log and seismic data. The data consist of a series of target logs from wells which tie a 3-D seismic volume. The target logs are neutron porosity logs. From the 3-D seismic volume a series of sample attributes is calculated. The objective of this study is to derive a set of attributes and the target log values. The selected set is determined by a process of forward stepwise regression. The analysis can be linear or nonlinear. In the linear mode the method consists of a series of weights derived by least-square minimization. In the nonlinear mode, a neural network is trained using the select attributes as inputs. In this case we used a probabilistic neural network PNN. The method is applied to a real data set from PEMEX. For better reservoir characterization the porosity distribution was estimated using both techniques. The case shown a continues improvement in the prediction of the porosity from the multiattribute to the neural network analysis. The improvement is in the training and the validation, which are important indicators of the reliability of the results. The neural network showed an improvement in resolution over the multiattribute analysis. The final maps provide more realistic results of the porosity distribution.

Combined Ozone Retrieval From METOP Sensors Using META-Training Of Deep Neural Networks

NASA Astrophysics Data System (ADS)

Felder, Martin; Sehnke, Frank; Kaifel, Anton

2013-12-01

The newest installment of our well-proven Neural Net- work Ozone Retrieval System (NNORSY) combines the METOP sensors GOME-2 and IASI with cloud information from AVHRR. Through the use of advanced meta- learning techniques like automatic feature selection and automatic architecture search applied to a set of deep neural networks, having at least two or three hidden layers, we have been able to avoid many technical issues normally encountered during the construction of such a joint retrieval system. This has been made possible by harnessing the processing power of modern consumer graphics cards with high performance graphic processors (GPU), which decreases training times by about two orders of magnitude. The system was trained on data from 2009 and 2010, including target ozone profiles from ozone sondes, ACE- FTS and MLS-AURA. To make maximum use of tropospheric information in the spectra, the data were partitioned into several sets of different cloud fraction ranges with the GOME-2 FOV, on which specialized retrieval networks are being trained. For the final ozone retrieval processing the different specialized networks are combined. The resulting retrieval system is very stable and does not show any systematic dependence on solar zenith angle, scan angle or sensor degradation. We present several sensitivity studies with regard to cloud fraction and target sensor type, as well as the performance in several latitude bands and with respect to independent validation stations. A visual cross-comparison against high-resolution ozone profiles from the KNMI EUMETSAT Ozone SAF product has also been performed and shows some distinctive features which we will briefly discuss. Overall, we demonstrate that a complex retrieval system can now be constructed with a minimum of ma- chine learning knowledge, using automated algorithms for many design decisions previously requiring expert knowledge. Provided sufficient training data and computation power of GPUs is available, the method can be applied to almost any kind of retrieval or, more generally, regression problem.

Image aesthetic quality evaluation using convolution neural network embedded learning

NASA Astrophysics Data System (ADS)

Li, Yu-xin; Pu, Yuan-yuan; Xu, Dan; Qian, Wen-hua; Wang, Li-peng

2017-11-01

A way of embedded learning convolution neural network (ELCNN) based on the image content is proposed to evaluate the image aesthetic quality in this paper. Our approach can not only solve the problem of small-scale data but also score the image aesthetic quality. First, we chose Alexnet and VGG_S to compare for confirming which is more suitable for this image aesthetic quality evaluation task. Second, to further boost the image aesthetic quality classification performance, we employ the image content to train aesthetic quality classification models. But the training samples become smaller and only using once fine-tuning cannot make full use of the small-scale data set. Third, to solve the problem in second step, a way of using twice fine-tuning continually based on the aesthetic quality label and content label respective is proposed, the classification probability of the trained CNN models is used to evaluate the image aesthetic quality. The experiments are carried on the small-scale data set of Photo Quality. The experiment results show that the classification accuracy rates of our approach are higher than the existing image aesthetic quality evaluation approaches.

Artificial neural network study on organ-targeting peptides

NASA Astrophysics Data System (ADS)

Jung, Eunkyoung; Kim, Junhyoung; Choi, Seung-Hoon; Kim, Minkyoung; Rhee, Hokyoung; Shin, Jae-Min; Choi, Kihang; Kang, Sang-Kee; Lee, Nam Kyung; Choi, Yun-Jaie; Jung, Dong Hyun

2010-01-01

We report a new approach to studying organ targeting of peptides on the basis of peptide sequence information. The positive control data sets consist of organ-targeting peptide sequences identified by the peroral phage-display technique for four organs, and the negative control data are prepared from random sequences. The capacity of our models to make appropriate predictions is validated by statistical indicators including sensitivity, specificity, enrichment curve, and the area under the receiver operating characteristic (ROC) curve (the ROC score). VHSE descriptor produces statistically significant training models and the models with simple neural network architectures show slightly greater predictive power than those with complex ones. The training and test set statistics indicate that our models could discriminate between organ-targeting and random sequences. We anticipate that our models will be applicable to the selection of organ-targeting peptides for generating peptide drugs or peptidomimetics.

[Rapid Identification of Epicarpium Citri Grandis via Infrared Spectroscopy and Fluorescence Spectrum Imaging Technology Combined with Neural Network].

PubMed

Pan, Sha-sha; Huang, Fu-rong; Xiao, Chi; Xian, Rui-yi; Ma, Zhi-guo

2015-10-01

To explore rapid reliable methods for detection of Epicarpium citri grandis (ECG), the experiment using Fourier Transform Attenuated Total Reflection Infrared Spectroscopy (FTIR/ATR) and Fluorescence Spectrum Imaging Technology combined with Multilayer Perceptron (MLP) Neural Network pattern recognition, for the identification of ECG, and the two methods are compared. Infrared spectra and fluorescence spectral images of 118 samples, 81 ECG and 37 other kinds of ECG, are collected. According to the differences in tspectrum, the spectra data in the 550-1 800 cm(-1) wavenumber range and 400-720 nm wavelength are regarded as the study objects of discriminant analysis. Then principal component analysis (PCA) is applied to reduce the dimension of spectroscopic data of ECG and MLP Neural Network is used in combination to classify them. During the experiment were compared the effects of different methods of data preprocessing on the model: multiplicative scatter correction (MSC), standard normal variable correction (SNV), first-order derivative(FD), second-order derivative(SD) and Savitzky-Golay (SG). The results showed that: after the infrared spectra data via the Savitzky-Golay (SG) pretreatment through the MLP Neural Network with the hidden layer function as sigmoid, we can get the best discrimination of ECG, the correct percent of training set and testing set are both 100%. Using fluorescence spectral imaging technology, corrected by the multiple scattering (MSC) results in the pretreatment is the most ideal. After data preprocessing, the three layers of the MLP Neural Network of the hidden layer function as sigmoid function can get 100% correct percent of training set and 96.7% correct percent of testing set. It was shown that the FTIR/ATR and fluorescent spectral imaging technology combined with MLP Neural Network can be used for the identification study of ECG and has the advantages of rapid, reliable effect.

Prediction of Aerodynamic Coefficients for Wind Tunnel Data using a Genetic Algorithm Optimized Neural Network

NASA Technical Reports Server (NTRS)

Rajkumar, T.; Aragon, Cecilia; Bardina, Jorge; Britten, Roy

2002-01-01

A fast, reliable way of predicting aerodynamic coefficients is produced using a neural network optimized by a genetic algorithm. Basic aerodynamic coefficients (e.g. lift, drag, pitching moment) are modelled as functions of angle of attack and Mach number. The neural network is first trained on a relatively rich set of data from wind tunnel tests of numerical simulations to learn an overall model. Most of the aerodynamic parameters can be well-fitted using polynomial functions. A new set of data, which can be relatively sparse, is then supplied to the network to produce a new model consistent with the previous model and the new data. Because the new model interpolates realistically between the sparse test data points, it is suitable for use in piloted simulations. The genetic algorithm is used to choose a neural network architecture to give best results, avoiding over-and under-fitting of the test data.

Deep learning based state recognition of substation switches

NASA Astrophysics Data System (ADS)

Wang, Jin

2018-06-01

Different from the traditional method which recognize the state of substation switches based on the running rules of electrical power system, this work proposes a novel convolutional neuron network-based state recognition approach of substation switches. Inspired by the theory of transfer learning, we first establish a convolutional neuron network model trained on the large-scale image set ILSVRC2012, then the restricted Boltzmann machine is employed to replace the full connected layer of the convolutional neuron network and trained on our small image dataset of 110kV substation switches to get a stronger model. Experiments conducted on our image dataset of 110kV substation switches show that, the proposed approach can be applicable to the substation to reduce the running cost and implement the real unattended operation.

Recursive Bayesian recurrent neural networks for time-series modeling.

PubMed

Mirikitani, Derrick T; Nikolaev, Nikolay

2010-02-01

This paper develops a probabilistic approach to recursive second-order training of recurrent neural networks (RNNs) for improved time-series modeling. A general recursive Bayesian Levenberg-Marquardt algorithm is derived to sequentially update the weights and the covariance (Hessian) matrix. The main strengths of the approach are a principled handling of the regularization hyperparameters that leads to better generalization, and stable numerical performance. The framework involves the adaptation of a noise hyperparameter and local weight prior hyperparameters, which represent the noise in the data and the uncertainties in the model parameters. Experimental investigations using artificial and real-world data sets show that RNNs equipped with the proposed approach outperform standard real-time recurrent learning and extended Kalman training algorithms for recurrent networks, as well as other contemporary nonlinear neural models, on time-series modeling.

Forecasting Flare Activity Using Deep Convolutional Neural Networks

NASA Astrophysics Data System (ADS)

Hernandez, T.

2017-12-01

Current operational flare forecasting relies on human morphological analysis of active regions and the persistence of solar flare activity through time (i.e. that the Sun will continue to do what it is doing right now: flaring or remaining calm). In this talk we present the results of applying deep Convolutional Neural Networks (CNNs) to the problem of solar flare forecasting. CNNs operate by training a set of tunable spatial filters that, in combination with neural layer interconnectivity, allow CNNs to automatically identify significant spatial structures predictive for classification and regression problems. We will start by discussing the applicability and success rate of the approach, the advantages it has over non-automated forecasts, and how mining our trained neural network provides a fresh look into the mechanisms behind magnetic energy storage and release.

Layered Ensemble Architecture for Time Series Forecasting.

PubMed

Rahman, Md Mustafizur; Islam, Md Monirul; Murase, Kazuyuki; Yao, Xin

2016-01-01

Time series forecasting (TSF) has been widely used in many application areas such as science, engineering, and finance. The phenomena generating time series are usually unknown and information available for forecasting is only limited to the past values of the series. It is, therefore, necessary to use an appropriate number of past values, termed lag, for forecasting. This paper proposes a layered ensemble architecture (LEA) for TSF problems. Our LEA consists of two layers, each of which uses an ensemble of multilayer perceptron (MLP) networks. While the first ensemble layer tries to find an appropriate lag, the second ensemble layer employs the obtained lag for forecasting. Unlike most previous work on TSF, the proposed architecture considers both accuracy and diversity of the individual networks in constructing an ensemble. LEA trains different networks in the ensemble by using different training sets with an aim of maintaining diversity among the networks. However, it uses the appropriate lag and combines the best trained networks to construct the ensemble. This indicates LEAs emphasis on accuracy of the networks. The proposed architecture has been tested extensively on time series data of neural network (NN)3 and NN5 competitions. It has also been tested on several standard benchmark time series data. In terms of forecasting accuracy, our experimental results have revealed clearly that LEA is better than other ensemble and nonensemble methods.

Automatic detection of diabetic retinopathy using an artificial neural network: a screening tool.

PubMed

Gardner, G G; Keating, D; Williamson, T H; Elliott, A T

1996-11-01

To determine if neural networks can detect diabetic features in fundus images and compare the network against an ophthalmologist screening a set of fundus images. 147 diabetic and 32 normal images were captured from a fundus camera, stored on computer, and analysed using a back propagation neural network. The network was trained to recognise features in the retinal image. The effects of digital filtering techniques and different network variables were assessed. 200 diabetic and 101 normal images were then randomised and used to evaluate the network's performance for the detection of diabetic retinopathy against an ophthalmologist. Detection rates for the recognition of vessels, exudates, and haemorrhages were 91.7%, 93.1%, and 73.8% respectively. When compared with the results of the ophthalmologist, the network achieved a sensitivity of 88.4% and a specificity of 83.5% for the detection of diabetic retinopathy. Detection of vessels, exudates, and haemorrhages was possible, with success rates dependent upon preprocessing and the number of images used in training. When compared with the ophthalmologist, the network achieved good accuracy for the detection of diabetic retinopathy. The system could be used as an aid to the screening of diabetic patients for retinopathy.

An improved multi-domain convolution tracking algorithm

NASA Astrophysics Data System (ADS)

Sun, Xin; Wang, Haiying; Zeng, Yingsen

2018-04-01

Along with the wide application of the Deep Learning in the field of Computer vision, Deep learning has become a mainstream direction in the field of object tracking. The tracking algorithm in this paper is based on the improved multidomain convolution neural network, and the VOT video set is pre-trained on the network by multi-domain training strategy. In the process of online tracking, the network evaluates candidate targets sampled from vicinity of the prediction target in the previous with Gaussian distribution, and the candidate target with the highest score is recognized as the prediction target of this frame. The Bounding Box Regression model is introduced to make the prediction target closer to the ground-truths target box of the test set. Grouping-update strategy is involved to extract and select useful update samples in each frame, which can effectively prevent over fitting. And adapt to changes in both target and environment. To improve the speed of the algorithm while maintaining the performance, the number of candidate target succeed in adjusting dynamically with the help of Self-adaption parameter Strategy. Finally, the algorithm is tested by OTB set, compared with other high-performance tracking algorithms, and the plot of success rate and the accuracy are drawn. which illustrates outstanding performance of the tracking algorithm in this paper.

Application of adaptive boosting to EP-derived multilayer feed-forward neural networks (MLFN) to improve benign/malignant breast cancer classification

NASA Astrophysics Data System (ADS)

Land, Walker H., Jr.; Masters, Timothy D.; Lo, Joseph Y.; McKee, Dan

2001-07-01

A new neural network technology was developed for improving the benign/malignant diagnosis of breast cancer using mammogram findings. A new paradigm, Adaptive Boosting (AB), uses a markedly different theory in solutioning Computational Intelligence (CI) problems. AB, a new machine learning paradigm, focuses on finding weak learning algorithm(s) that initially need to provide slightly better than random performance (i.e., approximately 55%) when processing a mammogram training set. Then, by successive development of additional architectures (using the mammogram training set), the adaptive boosting process improves the performance of the basic Evolutionary Programming derived neural network architectures. The results of these several EP-derived hybrid architectures are then intelligently combined and tested using a similar validation mammogram data set. Optimization focused on improving specificity and positive predictive value at very high sensitivities, where an analysis of the performance of the hybrid would be most meaningful. Using the DUKE mammogram database of 500 biopsy proven samples, on average this hybrid was able to achieve (under statistical 5-fold cross-validation) a specificity of 48.3% and a positive predictive value (PPV) of 51.8% while maintaining 100% sensitivity. At 97% sensitivity, a specificity of 56.6% and a PPV of 55.8% were obtained.

Retrieval of ice thickness from polarimetric SAR data

NASA Technical Reports Server (NTRS)

Kwok, R.; Yueh, S. H.; Nghiem, S. V.; Huynh, D. D.

1993-01-01

We describe a potential procedure for retrieving ice thickness from multi-frequency polarimetric SAR data for thin ice. This procedure includes first masking out the thicker ice types with a simple classifier and then deriving the thickness of the remaining pixels using a model-inversion technique. The technique used to derive ice thickness from polarimetric observations is provided by a numerical estimator or neural network. A three-layer perceptron implemented with the backpropagation algorithm is used in this investigation with several improved aspects for a faster convergence rate and a better accuracy of the neural network. These improvements include weight initialization, normalization of the output range, the selection of offset constant, and a heuristic learning algorithm. The performance of the neural network is demonstrated by using training data generated by a theoretical scattering model for sea ice matched to the database of interest. The training data are comprised of the polarimetric backscattering coefficients of thin ice and the corresponding input ice parameters to the scattering model. The retrieved ice thickness from the theoretical backscattering coefficients is compare with the input ice thickness to the scattering model to illustrate the accuracy of the inversion method. Results indicate that the network convergence rate and accuracy are higher when multi-frequency training sets are presented. In addition, the dominant backscattering coefficients in retrieving ice thickness are found by comparing the behavior of the network trained backscattering data at various incidence angels. After the neural network is trained with the theoretical backscattering data at various incidence anges, the interconnection weights between nodes are saved and applied to the experimental data to be investigated. In this paper, we illustrate the effectiveness of this technique using polarimetric SAR data collected by the JPL DC-8 radar over a sea ice scene.

Neural-network-designed pulse sequences for robust control of singlet-triplet qubits

NASA Astrophysics Data System (ADS)

Yang, Xu-Chen; Yung, Man-Hong; Wang, Xin

2018-04-01

Composite pulses are essential for universal manipulation of singlet-triplet spin qubits. In the absence of noise, they are required to perform arbitrary single-qubit operations due to the special control constraint of a singlet-triplet qubit, while in a noisy environment, more complicated sequences have been developed to dynamically correct the error. Tailoring these sequences typically requires numerically solving a set of nonlinear equations. Here we demonstrate that these pulse sequences can be generated by a well-trained, double-layer neural network. For sequences designed for the noise-free case, the trained neural network is capable of producing almost exactly the same pulses known in the literature. For more complicated noise-correcting sequences, the neural network produces pulses with slightly different line shapes, but the robustness against noises remains comparable. These results indicate that the neural network can be a judicious and powerful alternative to existing techniques in developing pulse sequences for universal fault-tolerant quantum computation.

QSRR using evolved artificial neural network for 52 common pharmaceuticals and drugs of abuse in hair from UPLC-TOF-MS.

PubMed

Noorizadeh, Hadi; Farmany, Abbas; Narimani, Hojat; Noorizadeh, Mehrab

2013-05-01

A quantitative structure-retention relationship (QSRR) study based on an artificial neural network (ANN) was carried out for the prediction of the ultra-performance liquid chromatography-Time-of-Flight mass spectrometry (UPLC-TOF-MS) retention time (RT) of a set of 52 pharmaceuticals and drugs of abuse in hair. The genetic algorithm was used as a variable selection tool. A partial least squares (PLS) method was used to select the best descriptors which were used as input neurons in neural network model. For choosing the best predictive model from among comparable models, square correlation coefficient R(2) for the whole set calculated based on leave-group-out predicted values of the training set and model-derived predicted values for the test set compounds is suggested to be a good criterion. Finally, to improve the results, structure-retention relationships were followed by a non-linear approach using artificial neural networks and consequently better results were obtained. This also demonstrates the advantages of ANN. Copyright © 2011 John Wiley & Sons, Ltd.

Collaborative Supervised Learning for Sensor Networks

NASA Technical Reports Server (NTRS)

Wagstaff, Kiri L.; Rebbapragada, Umaa; Lane, Terran

2011-01-01

Collaboration methods for distributed machine-learning algorithms involve the specification of communication protocols for the learners, which can query other learners and/or broadcast their findings preemptively. Each learner incorporates information from its neighbors into its own training set, and they are thereby able to bootstrap each other to higher performance. Each learner resides at a different node in the sensor network and makes observations (collects data) independently of the other learners. After being seeded with an initial labeled training set, each learner proceeds to learn in an iterative fashion. New data is collected and classified. The learner can then either broadcast its most confident classifications for use by other learners, or can query neighbors for their classifications of its least confident items. As such, collaborative learning combines elements of both passive (broadcast) and active (query) learning. It also uses ideas from ensemble learning to combine the multiple responses to a given query into a single useful label. This approach has been evaluated against current non-collaborative alternatives, including training a single classifier and deploying it at all nodes with no further learning possible, and permitting learners to learn from their own most confident judgments, absent interaction with their neighbors. On several data sets, it has been consistently found that active collaboration is the best strategy for a distributed learner network. The main advantages include the ability for learning to take place autonomously by collaboration rather than by requiring intervention from an oracle (usually human), and also the ability to learn in a distributed environment, permitting decisions to be made in situ and to yield faster response time.

Classification of asteroid spectra using a neural network

NASA Technical Reports Server (NTRS)

Howell, E. S.; Merenyi, E.; Lebofsky, L. A.

1994-01-01

The 52-color asteroid survey (Bell et al., 1988) together with the 8-color asteroid survey (Zellner et al., 1985) provide a data set of asteroid spectra spanning 0.3-2.5 micrometers. An artificial neural network clusters these asteroid spectra based on their similarity to each other. We have also trained the neural network with a categorization learning output layer in a supervised mode to associate the established clusters with taxonomic classes. Results of our classification agree with Tholen's classification based on the 8-color data alone. When extending the spectral range using the 52-color survey data, we find that some modification of the Tholen classes is indicated to produce a cleaner, self-consistent set of taxonomic classes. After supervised training using our modified classes, the network correctly classifies both the training examples, and additional spectra into the correct class with an average of 90% accuracy. Our classification supports the separation of the K class from the S class, as suggested by Bell et al. (1987), based on the near-infrared spectrum. We define two end-member subclasses which seem to have compositional significance within the S class: the So class, which is olivine-rich and red, and the Sp class, which is pyroxene-rich and less red. The remaining S-class asteroids have intermediate compositions of both olivine and pyroxene and moderately red continua. The network clustering suggests some additional structure within the E-, M-, and P-class asteroids, even in the absence of albedo information, which is the only discriminant between these in the Tholen classification. New relationships are seen between the C class and related G, B, and F classes. However, in both cases, the number of spectra is too small to interpret or determine the significance of these separations.

Automated segmentation of lesions including subretinal hyperreflective material in neovascular age-related macular degeneration.

PubMed

Lee, Hyungwoo; Kang, Kyung Eun; Chung, Hyewon; Kim, Hyung Chan

2018-04-12

To evaluate an automated segmentation algorithm with a convolutional neural network (CNN) to quantify and detect intraretinal fluid (IRF), subretinal fluid (SRF), pigment epithelial detachment (PED), and subretinal hyperreflective material (SHRM) through analyses of spectral domain optical coherence tomography (SD-OCT) images from patients with neovascular age-related macular degeneration (nAMD). Reliability and validity analysis of a diagnostic tool. We constructed a dataset including 930 B-scans from 93 eyes of 93 patients with nAMD. A CNN-based deep neural network was trained using 11550 augmented images derived from 550 B-scans. The performance of the trained network was evaluated using a validation set including 140 B-scans and a test set of 240 B-scans. The Dice coefficient, positive predictive value (PPV), sensitivity, relative area difference (RAD), and intraclass correlation coefficient (ICC) were used to evaluate segmentation and detection performance. Good agreement was observed for both segmentation and detection of lesions between the trained network and clinicians. The Dice coefficients for segmentation of IRF, SRF, SHRM, and PED were 0.78, 0.82, 0.75, and 0.80, respectively; the PPVs were 0.79, 0.80, 0.75, and 0.80, respectively; and the sensitivities were 0.77, 0.84, 0.73, and 0.81, respectively. The RADs were -4.32%, -10.29%, 4.13%, and 0.34%, respectively, and the ICCs were 0.98, 0.98, 0.97, and 0.98, respectively. All lesions were detected with high PPVs (range 0.94-0.99) and sensitivities (range 0.97-0.99). A CNN-based network provides clinicians with quantitative data regarding nAMD through automatic segmentation and detection of pathological lesions, including IRF, SRF, PED, and SHRM. Copyright © 2018 Elsevier Inc. All rights reserved.

Digital implementation of a neural network for imaging

NASA Astrophysics Data System (ADS)

Wood, Richard; McGlashan, Alex; Yatulis, Jay; Mascher, Peter; Bruce, Ian

2012-10-01

This paper outlines the design and testing of a digital imaging system that utilizes an artificial neural network with unsupervised and supervised learning to convert streaming input (real time) image space into parameter space. The primary objective of this work is to investigate the effectiveness of using a neural network to significantly reduce the information density of streaming images so that objects can be readily identified by a limited set of primary parameters and act as an enhanced human machine interface (HMI). Many applications are envisioned including use in biomedical imaging, anomaly detection and as an assistive device for the visually impaired. A digital circuit was designed and tested using a Field Programmable Gate Array (FPGA) and an off the shelf digital camera. Our results indicate that the networks can be readily trained when subject to limited sets of objects such as the alphabet. We can also separate limited object sets with rotational and positional invariance. The results also show that limited visual fields form with only local connectivity.

A system of recurrent neural networks for modularising, parameterising and dynamic analysis of cell signalling networks.

PubMed

Samarasinghe, S; Ling, H

In this paper, we show how to extend our previously proposed novel continuous time Recurrent Neural Networks (RNN) approach that retains the advantage of continuous dynamics offered by Ordinary Differential Equations (ODE) while enabling parameter estimation through adaptation, to larger signalling networks using a modular approach. Specifically, the signalling network is decomposed into several sub-models based on important temporal events in the network. Each sub-model is represented by the proposed RNN and trained using data generated from the corresponding ODE model. Trained sub-models are assembled into a whole system RNN which is then subjected to systems dynamics and sensitivity analyses. The concept is illustrated by application to G1/S transition in cell cycle using Iwamoto et al. (2008) ODE model. We decomposed the G1/S network into 3 sub-models: (i) E2F transcription factor release; (ii) E2F and CycE positive feedback loop for elevating cyclin levels; and (iii) E2F and CycA negative feedback to degrade E2F. The trained sub-models accurately represented system dynamics and parameters were in good agreement with the ODE model. The whole system RNN however revealed couple of parameters contributing to compounding errors due to feedback and required refinement to sub-model 2. These related to the reversible reaction between CycE/CDK2 and p27, its inhibitor. The revised whole system RNN model very accurately matched dynamics of the ODE system. Local sensitivity analysis of the whole system model further revealed the most dominant influence of the above two parameters in perturbing G1/S transition, giving support to a recent hypothesis that the release of inhibitor p27 from Cyc/CDK complex triggers cell cycle stage transition. To make the model useful in a practical setting, we modified each RNN sub-model with a time relay switch to facilitate larger interval input data (≈20min) (original model used data for 30s or less) and retrained them that produced parameters and protein concentrations similar to the original RNN system. Results thus demonstrated the reliability of the proposed RNN method for modelling relatively large networks by modularisation for practical settings. Advantages of the method are its ability to represent accurate continuous system dynamics and ease of: parameter estimation through training with data from a practical setting, model analysis (40% faster than ODE), fine tuning parameters when more data are available, sub-model extension when new elements and/or interactions come to light and model expansion with addition of sub-models. Copyright © 2017 Elsevier B.V. All rights reserved.

Automated Wildfire Detection Through Artificial Neural Networks

NASA Technical Reports Server (NTRS)

Miller, Jerry; Borne, Kirk; Thomas, Brian; Huang, Zhenping; Chi, Yuechen

2005-01-01

We have tested and deployed Artificial Neural Network (ANN) data mining techniques to analyze remotely sensed multi-channel imaging data from MODIS, GOES, and AVHRR. The goal is to train the ANN to learn the signatures of wildfires in remotely sensed data in order to automate the detection process. We train the ANN using the set of human-detected wildfires in the U.S., which are provided by the Hazard Mapping System (HMS) wildfire detection group at NOAA/NESDIS. The ANN is trained to mimic the behavior of fire detection algorithms and the subjective decision- making by N O M HMS Fire Analysts. We use a local extremum search in order to isolate fire pixels, and then we extract a 7x7 pixel array around that location in 3 spectral channels. The corresponding 147 pixel values are used to populate a 147-dimensional input vector that is fed into the ANN. The ANN accuracy is tested and overfitting is avoided by using a subset of the training data that is set aside as a test data set. We have achieved an automated fire detection accuracy of 80-92%, depending on a variety of ANN parameters and for different instrument channels among the 3 satellites. We believe that this system can be deployed worldwide or for any region to detect wildfires automatically in satellite imagery of those regions. These detections can ultimately be used to provide thermal inputs to climate models.

Fixed versus mixed RSA: Explaining visual representations by fixed and mixed feature sets from shallow and deep computational models.

PubMed

Khaligh-Razavi, Seyed-Mahdi; Henriksson, Linda; Kay, Kendrick; Kriegeskorte, Nikolaus

2017-02-01

Studies of the primate visual system have begun to test a wide range of complex computational object-vision models. Realistic models have many parameters, which in practice cannot be fitted using the limited amounts of brain-activity data typically available. Task performance optimization (e.g. using backpropagation to train neural networks) provides major constraints for fitting parameters and discovering nonlinear representational features appropriate for the task (e.g. object classification). Model representations can be compared to brain representations in terms of the representational dissimilarities they predict for an image set. This method, called representational similarity analysis (RSA), enables us to test the representational feature space as is (fixed RSA) or to fit a linear transformation that mixes the nonlinear model features so as to best explain a cortical area's representational space (mixed RSA). Like voxel/population-receptive-field modelling, mixed RSA uses a training set (different stimuli) to fit one weight per model feature and response channel (voxels here), so as to best predict the response profile across images for each response channel. We analysed response patterns elicited by natural images, which were measured with functional magnetic resonance imaging (fMRI). We found that early visual areas were best accounted for by shallow models, such as a Gabor wavelet pyramid (GWP). The GWP model performed similarly with and without mixing, suggesting that the original features already approximated the representational space, obviating the need for mixing. However, a higher ventral-stream visual representation (lateral occipital region) was best explained by the higher layers of a deep convolutional network and mixing of its feature set was essential for this model to explain the representation. We suspect that mixing was essential because the convolutional network had been trained to discriminate a set of 1000 categories, whose frequencies in the training set did not match their frequencies in natural experience or their behavioural importance. The latter factors might determine the representational prominence of semantic dimensions in higher-level ventral-stream areas. Our results demonstrate the benefits of testing both the specific representational hypothesis expressed by a model's original feature space and the hypothesis space generated by linear transformations of that feature space.

Scalable learning method for feedforward neural networks using minimal-enclosing-ball approximation.

PubMed

Wang, Jun; Deng, Zhaohong; Luo, Xiaoqing; Jiang, Yizhang; Wang, Shitong

2016-06-01

Training feedforward neural networks (FNNs) is one of the most critical issues in FNNs studies. However, most FNNs training methods cannot be directly applied for very large datasets because they have high computational and space complexity. In order to tackle this problem, the CCMEB (Center-Constrained Minimum Enclosing Ball) problem in hidden feature space of FNN is discussed and a novel learning algorithm called HFSR-GCVM (hidden-feature-space regression using generalized core vector machine) is developed accordingly. In HFSR-GCVM, a novel learning criterion using L2-norm penalty-based ε-insensitive function is formulated and the parameters in the hidden nodes are generated randomly independent of the training sets. Moreover, the learning of parameters in its output layer is proved equivalent to a special CCMEB problem in FNN hidden feature space. As most CCMEB approximation based machine learning algorithms, the proposed HFSR-GCVM training algorithm has the following merits: The maximal training time of the HFSR-GCVM training is linear with the size of training datasets and the maximal space consumption is independent of the size of training datasets. The experiments on regression tasks confirm the above conclusions. Copyright © 2016 Elsevier Ltd. All rights reserved.

Self-organizing neural networks--an alternative way of cluster analysis in clinical chemistry.

PubMed

Reibnegger, G; Wachter, H

1996-04-15

Supervised learning schemes have been employed by several workers for training neural networks designed to solve clinical problems. We demonstrate that unsupervised techniques can also produce interesting and meaningful results. Using a data set on the chemical composition of milk from 22 different mammals, we demonstrate that self-organizing feature maps (Kohonen networks) as well as a modified version of error backpropagation technique yield results mimicking conventional cluster analysis. Both techniques are able to project a potentially multi-dimensional input vector onto a two-dimensional space whereby neighborhood relationships remain conserved. Thus, these techniques can be used for reducing dimensionality of complicated data sets and for enhancing comprehensibility of features hidden in the data matrix.

Application of Multilayer Feedforward Neural Networks to Precipitation Cell-Top Altitude Estimation

NASA Technical Reports Server (NTRS)

Spina, Michelle S.; Schwartz, Michael J.; Staelin, David H.; Gasiewski, Albin J.

1998-01-01

The use of passive 118-GHz O2 observations of rain cells for precipitation cell-top altitude estimation is demonstrated by using a multilayer feed forward neural network retrieval system. Rain cell observations at 118 GHz were compared with estimates of the cell-top altitude obtained by optical stereoscopy. The observations were made with 2 4 km horizontal spatial resolution by using the Millimeter-wave Temperature Sounder (MTS) scanning spectrometer aboard the NASA ER-2 research aircraft during the Genesis of Atlantic Lows Experiment (GALE) and the COoperative Huntsville Meteorological EXperiment (COHMEX) in 1986. The neural network estimator applied to MTS spectral differences between clouds, and nearby clear air yielded an rms discrepancy of 1.76 km for a combined cumulus, mature, and dissipating cell set and 1.44 km for the cumulus-only set. An improvement in rms discrepancy to 1.36 km was achieved by including additional MTS information on the absolute atmospheric temperature profile. An incremental method for training neural networks was developed that yielded robust results, despite the use of as few as 56 training spectra. Comparison of these results with a nonlinear statistical estimator shows that superior results can be obtained with a neural network retrieval system. Imagery of estimated cell-top altitudes was created from 118-GHz spectral imagery gathered from CAMEX, September through October 1993, and from cyclone Oliver, February 7, 1993.

Web based scoring is useful for validation and harmonisation of scoring criteria within RENEB.

PubMed

Romm, Horst; Ainsbury, Elizabeth A; Barquinero, Joan Francesc; Barrios, Leonardo; Beinke, Christina; Cucu, Alexandra; Domene, Mercedes Moreno; Filippi, Silvia; Monteiro Gil, Octávia; Gregoire, Eric; Hadjidekova, Valeria; Hatzi, Vasia; Lindholm, Carita; M Kacher, Radhia; Montoro, Alegria; Moquet, Jayne; Noditi, Mihaela; Oestreicher, Ursula; Palitti, Fabrizio; Pantelias, Gabriel; Prieto, María Jesús; Popescu, Irina; Rothkamm, Kai; Sebastià, Natividad; Sommer, Sylwester; Terzoudi, Georgia; Testa, Antonella; Wojcik, Andrzej

2017-01-01

To establish a training data set of digital images and to investigate the scoring criteria and dose assessment of the dicentric assay within the European network of biodosimetry (RENEB), a web based scoring inter-comparison was undertaken by 17 RENEB partners. Two sets of 50 high resolution images were uploaded onto the RENEB website. One set included metaphases after a moderate exposure (1.3 Gy) and the other set consisted of metaphases after a high dose exposure (3.5 Gy). The laboratories used their own calibration curves for estimating doses based on observed aberration frequencies. The dose estimations and 95% confidence limits were compared to the actual doses and the corresponding z-values were satisfactory for the majority; only the dose estimations from two laboratories were too low or too high. The coefficients of variation were 17.6% for the moderate and 11.2% for the high dose. Metaphases with controversial results could be identified for training purposes. Overall, the web based scoring of the two galleries by the 17 laboratories produced very good results. Application of web based scoring for the dicentric assay may therefore be a relevant strategy for an operational biodosimetry assistance network.

Using Neural Networks for Sensor Validation

NASA Technical Reports Server (NTRS)

Mattern, Duane L.; Jaw, Link C.; Guo, Ten-Huei; Graham, Ronald; McCoy, William

1998-01-01

This paper presents the results of applying two different types of neural networks in two different approaches to the sensor validation problem. The first approach uses a functional approximation neural network as part of a nonlinear observer in a model-based approach to analytical redundancy. The second approach uses an auto-associative neural network to perform nonlinear principal component analysis on a set of redundant sensors to provide an estimate for a single failed sensor. The approaches are demonstrated using a nonlinear simulation of a turbofan engine. The fault detection and sensor estimation results are presented and the training of the auto-associative neural network to provide sensor estimates is discussed.

PREVENT: a program of the National Training Initiative on Injury and Violence Prevention.

PubMed

Runyan, Carol W; Gunther-Mohr, Carol; Orton, Stephen; Umble, Karl; Martin, Sandra L; Coyne-Beasley, Tamera

2005-12-01

Training practitioners to use evidence-based approaches to the primary prevention of violence is challenging as a result of the dearth of well-evaluated intervention programs and the lack of familiarity of some practitioners in drawing critically on existing literature. An element of the National Training Initiative in Injury and Violence Prevention, the PREVENT (Preventing Violence Through Education, Networking, and Technical Assistance) program began in late 2003 to train practitioners to address multiple types of violence by encouraging more widespread use of evidence-based approaches to primary prevention. It is intended to reach practitioners involved in addressing violence against women, sexual violence, child maltreatment, youth violence, and suicide in varied community settings. The program uses a combination of varied types of face-to-face training and distance learning coupled with opportunities for networking and technical assistance. Ultimately the program intends to stimulate and facilitate changes in individual, organizational, and cultural awareness and practices fostering primary prevention of violence. The project employs formative, process, and impact evaluation techniques aimed at improving delivery of the training as well as tracking changes in individual and organizations.

Neural network decoder for quantum error correcting codes

NASA Astrophysics Data System (ADS)

Krastanov, Stefan; Jiang, Liang

Artificial neural networks form a family of extremely powerful - albeit still poorly understood - tools used in anything from image and sound recognition through text generation to, in our case, decoding. We present a straightforward Recurrent Neural Network architecture capable of deducing the correcting procedure for a quantum error-correcting code from a set of repeated stabilizer measurements. We discuss the fault-tolerance of our scheme and the cost of training the neural network for a system of a realistic size. Such decoders are especially interesting when applied to codes, like the quantum LDPC codes, that lack known efficient decoding schemes.

Application of neural network for real-time measurement of electrical resistivity in cold crucible

NASA Astrophysics Data System (ADS)

Votava, Pavel; Poznyak, Igor

2017-08-01

The article describes use of an Induction furnace with cold crucible as a tool for real-time measurement of a melted material electrical resistivity. The measurement is based on an inverse problem solution of a 2D mathematical model, possibly implementable in a microcontroller or a FPGA in a form of a neural network. The 2D mathematical model results has been provided as a training set for the neural network. At the end, the implementation results are discussed together with uncertainty of measurement, which is done by the neural network implementation itself.

Pornographic image detection with Gabor filters

NASA Astrophysics Data System (ADS)

Durrell, Kevan; Murray, Daniel J. C.

2002-04-01

As Internet-enabled computers become ubiquitous in homes, schools, and other publicly accessible locations, there are more people 'surfing the net' who would prefer not to be exposed to offensive material. There is a lot of material freely available on the Internet that we, as a responsible and caring society, would like to keep our children from viewing. Pornographic image content is one category of material over which we would like some control. We have been conducting experiments to determine the effectiveness of using characteristic feature vectors and neural networks to identify semantic image content. This paper will describe our approach to identifying pornographic images using Gabor filters, Principal Component Analysis (PCA), Correllograms, and Neural Networks. In brief, we used a set of 5,000 typical images available from the Internet, 20% of which were judged to be pornographic, to train a neural network. We then apply the trained neural network to feature vectors from images that had not been used in training. We measure our performance as Recall, how many of the verification images labeled pornographic were correctly identified, and Precision, how many images deemed pornographic by the neural network are in fact pornographic. The set of images that were used in the experiment described in this paper for its training and validation sets are freely available on the Internet. Freely available is an important qualifier, since high-resolution, studio-quality pornographic images are often protected by portals that charge members a fee to gain access to their material. Although this is not a hard and fast rule, many of the pornographic images that are available easily and without charge on the Internet are of low image quality. Some of these images are collages or contain textual elements or have had their resolution intentionally lowered to reduce their file size. These are the offensive images that a user, without a credit card, might inadvertently come across on the Internet. Identifying this type of pornographic pictures of low image quality poses particular challenges for any detection software. This paper will address some of the challenges and hurdles we faced in designing and carrying out our experiments. The paper will also discuss the main results of our experiments, as well as some confounds that, at present, limit the effectiveness of our approach to identifying pornographic images, and some directions that may be taken in future research.

Research on artificial neural network intrusion detection photochemistry based on the improved wavelet analysis and transformation

NASA Astrophysics Data System (ADS)

Li, Hong; Ding, Xue

2017-03-01

This paper combines wavelet analysis and wavelet transform theory with artificial neural network, through the pretreatment on point feature attributes before in intrusion detection, to make them suitable for improvement of wavelet neural network. The whole intrusion classification model gets the better adaptability, self-learning ability, greatly enhances the wavelet neural network for solving the problem of field detection invasion, reduces storage space, contributes to improve the performance of the constructed neural network, and reduces the training time. Finally the results of the KDDCup99 data set simulation experiment shows that, this method reduces the complexity of constructing wavelet neural network, but also ensures the accuracy of the intrusion classification.

VINE: A Variational Inference -Based Bayesian Neural Network Engine

DTIC Science & Technology

2018-01-01

networks are trained using the same dataset and hyper parameter settings as discussed. Table 1 Performance evaluation of the proposed transfer learning...multiplication/addition/subtraction. These operations can be implemented using nested loops in which various iterations of a loop are independent of...each other. This introduces an opportunity for optimization where a loop may be unrolled fully or partially to increase parallelism at the cost of

On the sample complexity of learning for networks of spiking neurons with nonlinear synaptic interactions.

PubMed

Schmitt, Michael

2004-09-01

We study networks of spiking neurons that use the timing of pulses to encode information. Nonlinear interactions model the spatial groupings of synapses on the neural dendrites and describe the computations performed at local branches. Within a theoretical framework of learning we analyze the question of how many training examples these networks must receive to be able to generalize well. Bounds for this sample complexity of learning can be obtained in terms of a combinatorial parameter known as the pseudodimension. This dimension characterizes the computational richness of a neural network and is given in terms of the number of network parameters. Two types of feedforward architectures are considered: constant-depth networks and networks of unconstrained depth. We derive asymptotically tight bounds for each of these network types. Constant depth networks are shown to have an almost linear pseudodimension, whereas the pseudodimension of general networks is quadratic. Networks of spiking neurons that use temporal coding are becoming increasingly more important in practical tasks such as computer vision, speech recognition, and motor control. The question of how well these networks generalize from a given set of training examples is a central issue for their successful application as adaptive systems. The results show that, although coding and computation in these networks is quite different and in many cases more powerful, their generalization capabilities are at least as good as those of traditional neural network models.

Distributed multisensory integration in a recurrent network model through supervised learning

NASA Astrophysics Data System (ADS)

Wang, He; Wong, K. Y. Michael

Sensory integration between different modalities has been extensively studied. It is suggested that the brain integrates signals from different modalities in a Bayesian optimal way. However, how the Bayesian rule is implemented in a neural network remains under debate. In this work we propose a biologically plausible recurrent network model, which can perform Bayesian multisensory integration after trained by supervised learning. Our model is composed of two modules, each for one modality. We assume that each module is a recurrent network, whose activity represents the posterior distribution of each stimulus. The feedforward input on each module is the likelihood of each modality. Two modules are integrated through cross-links, which are feedforward connections from the other modality, and reciprocal connections, which are recurrent connections between different modules. By stochastic gradient descent, we successfully trained the feedforward and recurrent coupling matrices simultaneously, both of which resembles the Mexican-hat. We also find that there are more than one set of coupling matrices that can approximate the Bayesian theorem well. Specifically, reciprocal connections and cross-links will compensate each other if one of them is removed. Even though trained with two inputs, the network's performance with only one input is in good accordance with what is predicted by the Bayesian theorem.

High-Lift Optimization Design Using Neural Networks on a Multi-Element Airfoil

NASA Technical Reports Server (NTRS)

Greenman, Roxana M.; Roth, Karlin R.; Smith, Charles A. (Technical Monitor)

1998-01-01

The high-lift performance of a multi-element airfoil was optimized by using neural-net predictions that were trained using a computational data set. The numerical data was generated using a two-dimensional, incompressible, Navier-Stokes algorithm with the Spalart-Allmaras turbulence model. Because it is difficult to predict maximum lift for high-lift systems, an empirically-based maximum lift criteria was used in this study to determine both the maximum lift and the angle at which it occurs. Multiple input, single output networks were trained using the NASA Ames variation of the Levenberg-Marquardt algorithm for each of the aerodynamic coefficients (lift, drag, and moment). The artificial neural networks were integrated with a gradient-based optimizer. Using independent numerical simulations and experimental data for this high-lift configuration, it was shown that this design process successfully optimized flap deflection, gap, overlap, and angle of attack to maximize lift. Once the neural networks were trained and integrated with the optimizer, minimal additional computer resources were required to perform optimization runs with different initial conditions and parameters. Applying the neural networks within the high-lift rigging optimization process reduced the amount of computational time and resources by 83% compared with traditional gradient-based optimization procedures for multiple optimization runs.

Artificial Neural Network with Hardware Training and Hardware Refresh

NASA Technical Reports Server (NTRS)

Duong, Tuan A. (Inventor)

2003-01-01

A neural network circuit is provided having a plurality of circuits capable of charge storage. Also provided is a plurality of circuits each coupled to at least one of the plurality of charge storage circuits and constructed to generate an output in accordance with a neuron transfer function. Each of a plurality of circuits is coupled to one of the plurality of neuron transfer function circuits and constructed to generate a derivative of the output. A weight update circuit updates the charge storage circuits based upon output from the plurality of transfer function circuits and output from the plurality of derivative circuits. In preferred embodiments, separate training and validation networks share the same set of charge storage circuits and may operate concurrently. The validation network has a separate transfer function circuits each being coupled to the charge storage circuits so as to replicate the training network s coupling of the plurality of charge storage to the plurality of transfer function circuits. The plurality of transfer function circuits may be constructed each having a transconductance amplifier providing differential currents combined to provide an output in accordance with a transfer function. The derivative circuits may have a circuit constructed to generate a biased differential currents combined so as to provide the derivative of the transfer function.

NeuroRecovery Network provides standardization of locomotor training for persons with incomplete spinal cord injury.

PubMed

Morrison, Sarah A; Forrest, Gail F; VanHiel, Leslie R; Davé, Michele; D'Urso, Denise

2012-09-01

To illustrate the continuity of care afforded by a standardized locomotor training program across a multisite network setting within the Christopher and Dana Reeve Foundation NeuroRecovery Network (NRN). Single patient case study. Two geographically different hospital-based outpatient facilities. This case highlights a 25-year-old man diagnosed with C4 motor incomplete spinal cord injury with American Spinal Injury Association Impairment Scale grade D. Standardized locomotor training program 5 sessions per week for 1.5 hours per session, for a total of 100 treatment sessions, with 40 sessions at 1 center and 60 at another. Ten-meter walk test and 6-minute walk test were assessed at admission and discharge across both facilities. For each of the 100 treatment sessions percent body weight support, average, and maximum treadmill speed were evaluated. Locomotor endurance, as measured by the 6-minute walk test, and overground gait speed showed consistent improvement from admission to discharge. Throughout training, the patient decreased the need for body weight support and was able to tolerate faster treadmill speeds. Data indicate that the patient continued to improve on both treatment parameters and walking function. Standardization across the NRN centers provided a mechanism for delivering consistent and reproducible locomotor training programs across 2 facilities without disrupting training or recovery progression. Copyright © 2012 American Congress of Rehabilitation Medicine. Published by Elsevier Inc. All rights reserved.

Cross-Participant EEG-Based Assessment of Cognitive Workload Using Multi-Path Convolutional Recurrent Neural Networks.

PubMed

Hefron, Ryan; Borghetti, Brett; Schubert Kabban, Christine; Christensen, James; Estepp, Justin

2018-04-26

Applying deep learning methods to electroencephalograph (EEG) data for cognitive state assessment has yielded improvements over previous modeling methods. However, research focused on cross-participant cognitive workload modeling using these techniques is underrepresented. We study the problem of cross-participant state estimation in a non-stimulus-locked task environment, where a trained model is used to make workload estimates on a new participant who is not represented in the training set. Using experimental data from the Multi-Attribute Task Battery (MATB) environment, a variety of deep neural network models are evaluated in the trade-space of computational efficiency, model accuracy, variance and temporal specificity yielding three important contributions: (1) The performance of ensembles of individually-trained models is statistically indistinguishable from group-trained methods at most sequence lengths. These ensembles can be trained for a fraction of the computational cost compared to group-trained methods and enable simpler model updates. (2) While increasing temporal sequence length improves mean accuracy, it is not sufficient to overcome distributional dissimilarities between individuals’ EEG data, as it results in statistically significant increases in cross-participant variance. (3) Compared to all other networks evaluated, a novel convolutional-recurrent model using multi-path subnetworks and bi-directional, residual recurrent layers resulted in statistically significant increases in predictive accuracy and decreases in cross-participant variance.

Cross-Participant EEG-Based Assessment of Cognitive Workload Using Multi-Path Convolutional Recurrent Neural Networks

PubMed Central

Hefron, Ryan; Borghetti, Brett; Schubert Kabban, Christine; Christensen, James; Estepp, Justin

2018-01-01

Applying deep learning methods to electroencephalograph (EEG) data for cognitive state assessment has yielded improvements over previous modeling methods. However, research focused on cross-participant cognitive workload modeling using these techniques is underrepresented. We study the problem of cross-participant state estimation in a non-stimulus-locked task environment, where a trained model is used to make workload estimates on a new participant who is not represented in the training set. Using experimental data from the Multi-Attribute Task Battery (MATB) environment, a variety of deep neural network models are evaluated in the trade-space of computational efficiency, model accuracy, variance and temporal specificity yielding three important contributions: (1) The performance of ensembles of individually-trained models is statistically indistinguishable from group-trained methods at most sequence lengths. These ensembles can be trained for a fraction of the computational cost compared to group-trained methods and enable simpler model updates. (2) While increasing temporal sequence length improves mean accuracy, it is not sufficient to overcome distributional dissimilarities between individuals’ EEG data, as it results in statistically significant increases in cross-participant variance. (3) Compared to all other networks evaluated, a novel convolutional-recurrent model using multi-path subnetworks and bi-directional, residual recurrent layers resulted in statistically significant increases in predictive accuracy and decreases in cross-participant variance. PMID:29701668

Hierarchical Control Using Networks Trained with Higher-Level Forward Models

PubMed Central

Wayne, Greg; Abbott, L.F.

2015-01-01

We propose and develop a hierarchical approach to network control of complex tasks. In this approach, a low-level controller directs the activity of a “plant,” the system that performs the task. However, the low-level controller may only be able to solve fairly simple problems involving the plant. To accomplish more complex tasks, we introduce a higher-level controller that controls the lower-level controller. We use this system to direct an articulated truck to a specified location through an environment filled with static or moving obstacles. The final system consists of networks that have memorized associations between the sensory data they receive and the commands they issue. These networks are trained on a set of optimal associations that are generated by minimizing cost functions. Cost function minimization requires predicting the consequences of sequences of commands, which is achieved by constructing forward models, including a model of the lower-level controller. The forward models and cost minimization are only used during training, allowing the trained networks to respond rapidly. In general, the hierarchical approach can be extended to larger numbers of levels, dividing complex tasks into more manageable sub-tasks. The optimization procedure and the construction of the forward models and controllers can be performed in similar ways at each level of the hierarchy, which allows the system to be modified to perform other tasks, or to be extended for more complex tasks without retraining lower-levels. PMID:25058706

KNT-artificial neural network model for flux prediction of ultrafiltration membrane producing drinking water.

PubMed

Oh, H K; Yu, M J; Gwon, E M; Koo, J Y; Kim, S G; Koizumi, A

2004-01-01

This paper describes the prediction of flux behavior in an ultrafiltration (UF) membrane system using a Kalman neuro training (KNT) network model. The experimental data was obtained from operating a pilot plant of hollow fiber UF membrane with groundwater for 7 months. The network was trained using operating conditions such as inlet pressure, filtration duration, and feed water quality parameters including turbidity, temperature and UV254. Pre-processing of raw data allowed the normalized input data to be used in sigmoid activation functions. A neural network architecture was structured by modifying the number of hidden layers, neurons and learning iterations. The structure of KNT-neural network with 3 layers and 5 neurons allowed a good prediction of permeate flux by 0.997 of correlation coefficient during the learning phase. Also the validity of the designed model was evaluated with other experimental data not used during the training phase and nonlinear flux behavior was accurately estimated with 0.999 of correlation coefficient and a lower error of prediction in the testing phase. This good flux prediction can provide preliminary criteria in membrane design and set up the proper cleaning cycle in membrane operation. The KNT-artificial neural network is also expected to predict the variation of transmembrane pressure during filtration cycles and can be applied to automation and control of full scale treatment plants.

Neural network evaluation of tokamak current profiles for real time control

NASA Astrophysics Data System (ADS)

Wróblewski, Dariusz

1997-02-01

Active feedback control of the current profile, requiring real-time determination of the current profile parameters, is envisioned for tokamaks operating in enhanced confinement regimes. The distribution of toroidal current in a tokamak is now routinely evaluated based on external (magnetic probes, flux loops) and internal (motional Stark effect) measurements of the poloidal magnetic field. However, the analysis involves reconstruction of magnetohydrodynamic equilibrium and is too intensive computationally to be performed in real time. In the present study, a neural network is used to provide a mapping from the magnetic measurements (internal and external) to selected parameters of the safety factor profile. The single-pass, feedforward calculation of output of a trained neural network is very fast, making this approach particularly suitable for real-time applications. The network was trained on a large set of simulated equilibrium data for the DIII-D tokamak. The database encompasses a large variety of current profiles including the hollow current profiles important for reversed central shear operation. The parameters of safety factor profile (a quantity related to the current profile through the magnetic field tilt angle) estimated by the neural network include central safety factor, q0, minimum value of q, qmin, and the location of qmin. Very good performance of the trained neural network both for simulated test data and for experimental datais demonstrated.

Neural network evaluation of tokamak current profiles for real time control (abstract)

NASA Astrophysics Data System (ADS)

Wróblewski, Dariusz

1997-01-01

Active feedback control of the current profile, requiring real-time determination of the current profile parameters, is envisioned for tokamaks operating in enhanced confinement regimes. The distribution of toroidal current in a tokamak is now routinely evaluated based on external (magnetic probes, flux loops) and internal (motional Stark effect) measurements of the poloidal magnetic field. However, the analysis involves reconstruction of magnetohydrodynamic equilibrium and is too intensive computationally to be performed in real time. In the present study, a neural network is used to provide a mapping from the magnetic measurements (internal and external) to selected parameters of the safety factor profile. The single-pass, feedforward calculation of output of a trained neural network is very fast, making this approach particularly suitable for real-time applications. The network was trained on a large set of simulated equilibrium data for the DIII-D tokamak. The database encompasses a large variety of current profiles including the hollow current profiles important for reversed central shear operation. The parameters of safety factor profile (a quantity related to the current profile through the magnetic field tilt angle) estimated by the neural network include central safety factor, q0, minimum value of q, qmin, and the location of qmin. Very good performance of the trained neural network both for simulated test data and for experimental data is demonstrated.

An Ensemble of Neural Networks for Stock Trading Decision Making

NASA Astrophysics Data System (ADS)

Chang, Pei-Chann; Liu, Chen-Hao; Fan, Chin-Yuan; Lin, Jun-Lin; Lai, Chih-Ming

Stock turning signals detection are very interesting subject arising in numerous financial and economic planning problems. In this paper, Ensemble Neural Network system with Intelligent Piecewise Linear Representation for stock turning points detection is presented. The Intelligent piecewise linear representation method is able to generate numerous stocks turning signals from the historic data base, then Ensemble Neural Network system will be applied to train the pattern and retrieve similar stock price patterns from historic data for training. These turning signals represent short-term and long-term trading signals for selling or buying stocks from the market which are applied to forecast the future turning points from the set of test data. Experimental results demonstrate that the hybrid system can make a significant and constant amount of profit when compared with other approaches using stock data available in the market.

Neural Network and Letter Recognition.

NASA Astrophysics Data System (ADS)

Lee, Hue Yeon

Neural net architectures and learning algorithms that recognize hand written 36 alphanumeric characters are studied. The thin line input patterns written in 32 x 32 binary array are used. The system is comprised of two major components, viz. a preprocessing unit and a Recognition unit. The preprocessing unit in turn consists of three layers of neurons; the U-layer, the V-layer, and the C -layer. The functions of the U-layer is to extract local features by template matching. The correlation between the detected local features are considered. Through correlating neurons in a plane with their neighboring neurons, the V-layer would thicken the on-cells or lines that are groups of on-cells of the previous layer. These two correlations would yield some deformation tolerance and some of the rotational tolerance of the system. The C-layer then compresses data through the 'Gabor' transform. Pattern dependent choice of center and wavelengths of 'Gabor' filters is the cause of shift and scale tolerance of the system. Three different learning schemes had been investigated in the recognition unit, namely; the error back propagation learning with hidden units, a simple perceptron learning, and a competitive learning. Their performances were analyzed and compared. Since sometimes the network fails to distinguish between two letters that are inherently similar, additional ambiguity resolving neural nets are introduced on top of the above main neural net. The two dimensional Fourier transform is used as the preprocessing and the perceptron is used as the recognition unit of the ambiguity resolver. One hundred different person's handwriting sets are collected. Some of these are used as the training sets and the remainders are used as the test sets. The correct recognition rate of the system increases with the number of training sets and eventually saturates at a certain value. Similar recognition rates are obtained for the above three different learning algorithms. The minimum error rate, 4.9% is achieved for alphanumeric sets when 50 sets are trained. With the ambiguity resolver, it is reduced to 2.5%. In case that only numeral sets are trained and tested, 2.0% error rate is achieved. When only alphabet sets are considered, the error rate is reduced to 1.1%.

Mining big data sets of plankton images: a zero-shot learning approach to retrieve labels without training data

NASA Astrophysics Data System (ADS)

Orenstein, E. C.; Morgado, P. M.; Peacock, E.; Sosik, H. M.; Jaffe, J. S.

2016-02-01

Technological advances in instrumentation and computing have allowed oceanographers to develop imaging systems capable of collecting extremely large data sets. With the advent of in situ plankton imaging systems, scientists must now commonly deal with "big data" sets containing tens of millions of samples spanning hundreds of classes, making manual classification untenable. Automated annotation methods are now considered to be the bottleneck between collection and interpretation. Typically, such classifiers learn to approximate a function that predicts a predefined set of classes for which a considerable amount of labeled training data is available. The requirement that the training data span all the classes of concern is problematic for plankton imaging systems since they sample such diverse, rapidly changing populations. These data sets may contain relatively rare, sparsely distributed, taxa that will not have associated training data; a classifier trained on a limited set of classes will miss these samples. The computer vision community, leveraging advances in Convolutional Neural Networks (CNNs), has recently attempted to tackle such problems using "zero-shot" object categorization methods. Under a zero-shot framework, a classifier is trained to map samples onto a set of attributes rather than a class label. These attributes can include visual and non-visual information such as what an organism is made out of, where it is distributed globally, or how it reproduces. A second stage classifier is then used to extrapolate a class. In this work, we demonstrate a zero-shot classifier, implemented with a CNN, to retrieve out-of-training-set labels from images. This method is applied to data from two continuously imaging, moored instruments: the Scripps Plankton Camera System (SPCS) and the Imaging FlowCytobot (IFCB). Results from simulated deployment scenarios indicate zero-shot classifiers could be successful at recovering samples of rare taxa in image sets. This capability will allow ecologists to identify trends in the distribution of difficult to sample organisms in their data.

Static facial expression recognition with convolution neural networks

NASA Astrophysics Data System (ADS)

Zhang, Feng; Chen, Zhong; Ouyang, Chao; Zhang, Yifei

2018-03-01

Facial expression recognition is a currently active research topic in the fields of computer vision, pattern recognition and artificial intelligence. In this paper, we have developed a convolutional neural networks (CNN) for classifying human emotions from static facial expression into one of the seven facial emotion categories. We pre-train our CNN model on the combined FER2013 dataset formed by train, validation and test set and fine-tune on the extended Cohn-Kanade database. In order to reduce the overfitting of the models, we utilized different techniques including dropout and batch normalization in addition to data augmentation. According to the experimental result, our CNN model has excellent classification performance and robustness for facial expression recognition.

A convolutional neural network-based screening tool for X-ray serial crystallography

PubMed Central

Ke, Tsung-Wei; Brewster, Aaron S.; Yu, Stella X.; Ushizima, Daniela; Yang, Chao; Sauter, Nicholas K.

2018-01-01

A new tool is introduced for screening macromolecular X-ray crystallography diffraction images produced at an X-ray free-electron laser light source. Based on a data-driven deep learning approach, the proposed tool executes a convolutional neural network to detect Bragg spots. Automatic image processing algorithms described can enable the classification of large data sets, acquired under realistic conditions consisting of noisy data with experimental artifacts. Outcomes are compared for different data regimes, including samples from multiple instruments and differing amounts of training data for neural network optimization. PMID:29714177

A convolutional neural network-based screening tool for X-ray serial crystallography.

PubMed

Ke, Tsung Wei; Brewster, Aaron S; Yu, Stella X; Ushizima, Daniela; Yang, Chao; Sauter, Nicholas K

2018-05-01

A new tool is introduced for screening macromolecular X-ray crystallography diffraction images produced at an X-ray free-electron laser light source. Based on a data-driven deep learning approach, the proposed tool executes a convolutional neural network to detect Bragg spots. Automatic image processing algorithms described can enable the classification of large data sets, acquired under realistic conditions consisting of noisy data with experimental artifacts. Outcomes are compared for different data regimes, including samples from multiple instruments and differing amounts of training data for neural network optimization. open access.

A convolutional neural network-based screening tool for X-ray serial crystallography

DOE PAGES

Ke, Tsung-Wei; Brewster, Aaron S.; Yu, Stella X.; ...

2018-04-24

A new tool is introduced for screening macromolecular X-ray crystallography diffraction images produced at an X-ray free-electron laser light source. Based on a data-driven deep learning approach, the proposed tool executes a convolutional neural network to detect Bragg spots. Automatic image processing algorithms described can enable the classification of large data sets, acquired under realistic conditions consisting of noisy data with experimental artifacts. Outcomes are compared for different data regimes, including samples from multiple instruments and differing amounts of training data for neural network optimization.

A convolutional neural network-based screening tool for X-ray serial crystallography

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

Ke, Tsung-Wei; Brewster, Aaron S.; Yu, Stella X.

A new tool is introduced for screening macromolecular X-ray crystallography diffraction images produced at an X-ray free-electron laser light source. Based on a data-driven deep learning approach, the proposed tool executes a convolutional neural network to detect Bragg spots. Automatic image processing algorithms described can enable the classification of large data sets, acquired under realistic conditions consisting of noisy data with experimental artifacts. Outcomes are compared for different data regimes, including samples from multiple instruments and differing amounts of training data for neural network optimization.

A Neural Network Aero Design System for Advanced Turbo-Engines

NASA Technical Reports Server (NTRS)

Sanz, Jose M.

1999-01-01

An inverse design method calculates the blade shape that produces a prescribed input pressure distribution. By controlling this input pressure distribution the aerodynamic design objectives can easily be met. Because of the intrinsic relationship between pressure distribution and airfoil physical properties, a neural network can be trained to choose the optimal pressure distribution that would meet a set of physical requirements. The neural network technique works well not only as an interpolating device but also as an extrapolating device to achieve blade designs from a given database. Two validating test cases are discussed.

Artificial Neural Network System to Predict the Postoperative Outcome of Percutaneous Nephrolithotomy.

PubMed

Aminsharifi, Alireza; Irani, Dariush; Pooyesh, Shima; Parvin, Hamid; Dehghani, Sakineh; Yousofi, Khalilolah; Fazel, Ebrahim; Zibaie, Fatemeh

2017-05-01

To construct, train, and apply an artificial neural network (ANN) system for prediction of different outcome variables of percutaneous nephrolithotomy (PCNL). We calculated predictive accuracy, sensitivity, and precision for each outcome variable. During the study period, all adult patients who underwent PCNL at our institute were enrolled in the study. Preoperative and postoperative variables were recorded, and stone-free status was assessed perioperatively with computed tomography scans. MATLAB software was used to design and train the network in a feed forward back-propagation error adjustment scheme. Preoperative and postoperative data from 200 patients (training set) were used to analyze the effect and relative relevance of preoperative values on postoperative parameters. The validated adequately trained ANN was used to predict postoperative outcomes in the subsequent 254 adult patients (test set) whose preoperative values were serially fed into the system. To evaluate system accuracy in predicting each postoperative variable, predicted values were compared with actual outcomes. Two hundred fifty-four patients (155 [61%] males) were considered the test set. Mean stone burden was 6702.86 ± 381.6 mm 3 . Overall stone-free rate was 76.4%. Fifty-four out of 254 patients (21.3%) required ancillary procedures (shockwave lithotripsy 5.9%, transureteral lithotripsy 10.6%, and repeat PCNL 4.7%). The accuracy and sensitivity of the system in predicting different postoperative variables ranged from 81.0% to 98.2%. As a complex nonlinear mathematical model, our ANN system is an interconnected data mining tool, which prospectively analyzes and "learns" the relationships between variables. The accuracy and sensitivity of the system for predicting the stone-free rate, the need for blood transfusion, and post-PCNL ancillary procedures ranged from 81.0% to 98.2%.The stone burden and the stone morphometry were among the most significant preoperative characteristics that affected all postoperative outcome variables and they received the highest relative weight by the ANN system.

Comparison of machine learned approaches for thyroid nodule characterization from shear wave elastography images

NASA Astrophysics Data System (ADS)

Pereira, Carina; Dighe, Manjiri; Alessio, Adam M.

2018-02-01

Various Computer Aided Diagnosis (CAD) systems have been developed that characterize thyroid nodules using the features extracted from the B-mode ultrasound images and Shear Wave Elastography images (SWE). These features, however, are not perfect predictors of malignancy. In other domains, deep learning techniques such as Convolutional Neural Networks (CNNs) have outperformed conventional feature extraction based machine learning approaches. In general, fully trained CNNs require substantial volumes of data, motivating several efforts to use transfer learning with pre-trained CNNs. In this context, we sought to compare the performance of conventional feature extraction, fully trained CNNs, and transfer learning based, pre-trained CNNs for the detection of thyroid malignancy from ultrasound images. We compared these approaches applied to a data set of 964 B-mode and SWE images from 165 patients. The data were divided into 80% training/validation and 20% testing data. The highest accuracies achieved on the testing data for the conventional feature extraction, fully trained CNN, and pre-trained CNN were 0.80, 0.75, and 0.83 respectively. In this application, classification using a pre-trained network yielded the best performance, potentially due to the relatively limited sample size and sub-optimal architecture for the fully trained CNN.

Revisiting Social Network Utilization by Physicians-in-Training.

PubMed

Black, Erik W; Thompson, Lindsay A; Duff, W Patrick; Dawson, Kara; Saliba, Heidi; Black, Nicole M Paradise

2010-06-01

To measure and compare the frequency and content of online social networking among 2 cohorts of medical students and residents (2007 and 2009). Using the online social networking application Facebook, we evaluated social networking profiles for 2 cohorts of medical students (n  =  528) and residents (n  =  712) at the University of Florida in Gainesville. Objective measures included existence of a profile, whether it was made private, and whether any personally identifiable information was included. Subjective outcomes included photographic content, affiliated social groups, and personal information not generally disclosed in a doctor-patient encounter. We compared our results to our previously published and reported data from 2007. Social networking continues to be common amongst physicians-in-training, with 39.8% of residents and 69.5% of medical students maintaining Facebook accounts. Residents' participation significantly increased (P < .01) when compared to the 2007 data. Individuals in the 2009 cohort had significantly more "friends" (P < .01), belonged to more "groups" (P < .01), and were more likely to limit public access to their profiles through the use of privacy settings (P < .01) than the individuals in the 2007 cohort. Online social networking application use by physicians-in-training remains common. While most now limit access to their profiles, personal profiles that still allow public access exhibited a few instances of unprofessional behavior. Concerns remain related to the discovery of content in violation of patient privacy and the expansive and impersonal networks of online "friends" who may view profiles.

Development of Energy Efficient Clustering Protocol in Wireless Sensor Network Using Neuro-Fuzzy Approach.

PubMed

Julie, E Golden; Selvi, S Tamil

2016-01-01

Wireless sensor networks (WSNs) consist of sensor nodes with limited processing capability and limited nonrechargeable battery power. Energy consumption in WSN is a significant issue in networks for improving network lifetime. It is essential to develop an energy aware clustering protocol in WSN to reduce energy consumption for increasing network lifetime. In this paper, a neuro-fuzzy energy aware clustering scheme (NFEACS) is proposed to form optimum and energy aware clusters. NFEACS consists of two parts: fuzzy subsystem and neural network system that achieved energy efficiency in forming clusters and cluster heads in WSN. NFEACS used neural network that provides effective training set related to energy and received signal strength of all nodes to estimate the expected energy for tentative cluster heads. Sensor nodes with higher energy are trained with center location of base station to select energy aware cluster heads. Fuzzy rule is used in fuzzy logic part that inputs to form clusters. NFEACS is designed for WSN handling mobility of node. The proposed scheme NFEACS is compared with related clustering schemes, cluster-head election mechanism using fuzzy logic, and energy aware fuzzy unequal clustering. The experiment results show that NFEACS performs better than the other related schemes.

Development of Energy Efficient Clustering Protocol in Wireless Sensor Network Using Neuro-Fuzzy Approach

PubMed Central

Julie, E. Golden; Selvi, S. Tamil

2016-01-01

Wireless sensor networks (WSNs) consist of sensor nodes with limited processing capability and limited nonrechargeable battery power. Energy consumption in WSN is a significant issue in networks for improving network lifetime. It is essential to develop an energy aware clustering protocol in WSN to reduce energy consumption for increasing network lifetime. In this paper, a neuro-fuzzy energy aware clustering scheme (NFEACS) is proposed to form optimum and energy aware clusters. NFEACS consists of two parts: fuzzy subsystem and neural network system that achieved energy efficiency in forming clusters and cluster heads in WSN. NFEACS used neural network that provides effective training set related to energy and received signal strength of all nodes to estimate the expected energy for tentative cluster heads. Sensor nodes with higher energy are trained with center location of base station to select energy aware cluster heads. Fuzzy rule is used in fuzzy logic part that inputs to form clusters. NFEACS is designed for WSN handling mobility of node. The proposed scheme NFEACS is compared with related clustering schemes, cluster-head election mechanism using fuzzy logic, and energy aware fuzzy unequal clustering. The experiment results show that NFEACS performs better than the other related schemes. PMID:26881269

Multilingual Data Selection for Low Resource Speech Recognition

DTIC Science & Technology

2016-09-12

Figure 1: Identification of language clusters using scores from an LID system training languages used in the Base and OP1 evaluation periods of the Babel...the posterior scores over frames. For a set of languages that are used to train the lan- guage identification (LID) network, pairs of languages that...which are combined during test time to produce 10 dimensional language 3854 Figure 3: Identification of language clusters using scores from individually

The Role of the Global SOF Network in a Resource Constrained Environment

DTIC Science & Technology

2013-11-01

backed by actions. Step Two: Matching Skill Sets with Individuals: Who Needs to Know What? This step sounds simple enough in theory . You need to...Training Finally, step five is the recognition that, like physical training and condition- ing, cognitive skills need to be continuously refreshed. These...and maintaining the data needed , and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other

DeepMoon: Convolutional neural network trainer to identify moon craters

NASA Astrophysics Data System (ADS)

Silburt, Ari; Zhu, Chenchong; Ali-Dib, Mohamad; Menou, Kristen; Jackson, Alan

2018-05-01

DeepMoon trains a convolutional neural net using data derived from a global digital elevation map (DEM) and catalog of craters to recognize craters on the Moon. The TensorFlow-based pipeline code is divided into three parts. The first generates a set images of the Moon randomly cropped from the DEM, with corresponding crater positions and radii. The second trains a convnet using this data, and the third validates the convnet's predictions.

Deep Logic Networks: Inserting and Extracting Knowledge From Deep Belief Networks.

PubMed

Tran, Son N; d'Avila Garcez, Artur S

2018-02-01

Developments in deep learning have seen the use of layerwise unsupervised learning combined with supervised learning for fine-tuning. With this layerwise approach, a deep network can be seen as a more modular system that lends itself well to learning representations. In this paper, we investigate whether such modularity can be useful to the insertion of background knowledge into deep networks, whether it can improve learning performance when it is available, and to the extraction of knowledge from trained deep networks, and whether it can offer a better understanding of the representations learned by such networks. To this end, we use a simple symbolic language-a set of logical rules that we call confidence rules-and show that it is suitable for the representation of quantitative reasoning in deep networks. We show by knowledge extraction that confidence rules can offer a low-cost representation for layerwise networks (or restricted Boltzmann machines). We also show that layerwise extraction can produce an improvement in the accuracy of deep belief networks. Furthermore, the proposed symbolic characterization of deep networks provides a novel method for the insertion of prior knowledge and training of deep networks. With the use of this method, a deep neural-symbolic system is proposed and evaluated, with the experimental results indicating that modularity through the use of confidence rules and knowledge insertion can be beneficial to network performance.

Using beta binomials to estimate classification uncertainty for ensemble models.

PubMed

Clark, Robert D; Liang, Wenkel; Lee, Adam C; Lawless, Michael S; Fraczkiewicz, Robert; Waldman, Marvin

2014-01-01

Quantitative structure-activity (QSAR) models have enormous potential for reducing drug discovery and development costs as well as the need for animal testing. Great strides have been made in estimating their overall reliability, but to fully realize that potential, researchers and regulators need to know how confident they can be in individual predictions. Submodels in an ensemble model which have been trained on different subsets of a shared training pool represent multiple samples of the model space, and the degree of agreement among them contains information on the reliability of ensemble predictions. For artificial neural network ensembles (ANNEs) using two different methods for determining ensemble classification - one using vote tallies and the other averaging individual network outputs - we have found that the distribution of predictions across positive vote tallies can be reasonably well-modeled as a beta binomial distribution, as can the distribution of errors. Together, these two distributions can be used to estimate the probability that a given predictive classification will be in error. Large data sets comprised of logP, Ames mutagenicity, and CYP2D6 inhibition data are used to illustrate and validate the method. The distributions of predictions and errors for the training pool accurately predicted the distribution of predictions and errors for large external validation sets, even when the number of positive and negative examples in the training pool were not balanced. Moreover, the likelihood of a given compound being prospectively misclassified as a function of the degree of consensus between networks in the ensemble could in most cases be estimated accurately from the fitted beta binomial distributions for the training pool. Confidence in an individual predictive classification by an ensemble model can be accurately assessed by examining the distributions of predictions and errors as a function of the degree of agreement among the constituent submodels. Further, ensemble uncertainty estimation can often be improved by adjusting the voting or classification threshold based on the parameters of the error distribution. Finally, the profiles for models whose predictive uncertainty estimates are not reliable provide clues to that effect without the need for comparison to an external test set.

Computer-aided detection of bladder wall thickening in CT urography (CTU)

NASA Astrophysics Data System (ADS)

Cha, Kenny H.; Hadjiiski, Lubomir M.; Chan, Heang-Ping; Caoili, Elaine M.; Cohan, Richard H.; Weizer, Alon Z.; Gordon, Marshall N.; Samala, Ravi K.

2018-02-01

We are developing a computer-aided detection system for bladder cancer in CT urography (CTU). Bladder wall thickening is a manifestation of bladder cancer and its detection is more challenging than the detection of bladder masses. We first segmented the inner and outer bladder walls using our method that combined deep-learning convolutional neural network with level sets. The non-contrast-enhanced region was separated from the contrast-enhanced region with a maximum-intensity-projection-based method. The non-contrast region was smoothed and gray level threshold was applied to the contrast and non-contrast regions separately to extract the bladder wall and potential lesions. The bladder wall was transformed into a straightened thickness profile, which was analyzed to identify regions of wall thickening candidates. Volume-based features of the wall thickening candidates were analyzed with linear discriminant analysis (LDA) to differentiate bladder wall thickenings from false positives. A data set of 112 patients, 87 with wall thickening and 25 with normal bladders, was collected retrospectively with IRB approval, and split into independent training and test sets. Of the 57 training cases, 44 had bladder wall thickening and 13 were normal. Of the 55 test cases, 43 had wall thickening and 12 were normal. The LDA classifier was trained with the training set and evaluated with the test set. FROC analysis showed that the system achieved sensitivities of 93.2% and 88.4% for the training and test sets, respectively, at 0.5 FPs/case.

A novel constructive-optimizer neural network for the traveling salesman problem.

PubMed

Saadatmand-Tarzjan, Mahdi; Khademi, Morteza; Akbarzadeh-T, Mohammad-R; Moghaddam, Hamid Abrishami

2007-08-01

In this paper, a novel constructive-optimizer neural network (CONN) is proposed for the traveling salesman problem (TSP). CONN uses a feedback structure similar to Hopfield-type neural networks and a competitive training algorithm similar to the Kohonen-type self-organizing maps (K-SOMs). Consequently, CONN is composed of a constructive part, which grows the tour and an optimizer part to optimize it. In the training algorithm, an initial tour is created first and introduced to CONN. Then, it is trained in the constructive phase for adding a number of cities to the tour. Next, the training algorithm switches to the optimizer phase for optimizing the current tour by displacing the tour cities. After convergence in this phase, the training algorithm switches to the constructive phase anew and is continued until all cities are added to the tour. Furthermore, we investigate a relationship between the number of TSP cities and the number of cities to be added in each constructive phase. CONN was tested on nine sets of benchmark TSPs from TSPLIB to demonstrate its performance and efficiency. It performed better than several typical Neural networks (NNs), including KNIES_TSP_Local, KNIES_TSP_Global, Budinich's SOM, Co-Adaptive Net, and multivalued Hopfield network as wall as computationally comparable variants of the simulated annealing algorithm, in terms of both CPU time and accuracy. Furthermore, CONN converged considerably faster than expanding SOM and evolved integrated SOM and generated shorter tours compared to KNIES_DECOMPOSE. Although CONN is not yet comparable in terms of accuracy with some sophisticated computationally intensive algorithms, it converges significantly faster than they do. Generally speaking, CONN provides the best compromise between CPU time and accuracy among currently reported NNs for TSP.

Whole-Genome DNA Methylation Status Associated with Clinical PTSD Measures of OIF/OEF Veterans (Open Access)

DTIC Science & Technology

2017-07-11

Significantly enriched networks with similar functional purposes were grouped together, resulting in four network clusters (Figure 1): nervous system...relatively conservative estimate for the mean difference (that is, top 1%), 76 people per group should give 95% power to detect an individual probe...CpGIs curated from the training set were enriched for four functional clusters: PTSD-associated somatic complications, PTSD-relevant endocrine signaling

Linear induction accelerator and pulse forming networks therefor

DOEpatents

Buttram, Malcolm T.; Ginn, Jerry W.

1989-01-01

A linear induction accelerator includes a plurality of adder cavities arranged in a series and provided in a structure which is evacuated so that a vacuum inductance is provided between each adder cavity and the structure. An energy storage system for the adder cavities includes a pulsed current source and a respective plurality of bipolar converting networks connected thereto. The bipolar high-voltage, high-repetition-rate square pulse train sets and resets the cavities.

Process Control Strategies for Dual-Phase Steel Manufacturing Using ANN and ANFIS

NASA Astrophysics Data System (ADS)

Vafaeenezhad, H.; Ghanei, S.; Seyedein, S. H.; Beygi, H.; Mazinani, M.

2014-11-01

In this research, a comprehensive soft computational approach is presented for the analysis of the influencing parameters on manufacturing of dual-phase steels. A set of experimental data have been gathered to obtain the initial database used for the training and testing of both artificial neural networks (ANN) and adaptive neuro-fuzzy inference system (ANFIS). The parameters used in the strategy were intercritical annealing temperature, carbon content, and holding time which gives off martensite percentage as an output. A fraction of the data set was chosen to train both ANN and ANFIS, and the rest was put into practice to authenticate the act of the trained networks while seeing unseen data. To compare the obtained results, coefficient of determination and root mean squared error indexes were chosen. Using artificial intelligence methods, it is not necessary to consider and establish a preliminary mathematical model and formulate its affecting parameters on its definition. In conclusion, the martensite percentages corresponding to the manufacturing parameters can be determined prior to a production using these controlling algorithms. Although the results acquired from both ANN and ANFIS are very encouraging, the proposed ANFIS has enhanced performance over the ANN and takes better effect on cost-reduction profit.

Pneumothorax detection in chest radiographs using convolutional neural networks

NASA Astrophysics Data System (ADS)

Blumenfeld, Aviel; Konen, Eli; Greenspan, Hayit

2018-02-01

This study presents a computer assisted diagnosis system for the detection of pneumothorax (PTX) in chest radiographs based on a convolutional neural network (CNN) for pixel classification. Using a pixel classification approach allows utilization of the texture information in the local environment of each pixel while training a CNN model on millions of training patches extracted from a relatively small dataset. The proposed system uses a pre-processing step of lung field segmentation to overcome the large variability in the input images coming from a variety of imaging sources and protocols. Using a CNN classification, suspected pixel candidates are extracted within each lung segment. A postprocessing step follows to remove non-physiological suspected regions and noisy connected components. The overall percentage of suspected PTX area was used as a robust global decision for the presence of PTX in each lung. The system was trained on a set of 117 chest x-ray images with ground truth segmentations of the PTX regions. The system was tested on a set of 86 images and reached diagnosis accuracy of AUC=0.95. Overall preliminary results are promising and indicate the growing ability of CAD based systems to detect findings in medical imaging on a clinical level accuracy.

Using transfer learning to detect galaxy mergers

NASA Astrophysics Data System (ADS)

Ackermann, Sandro; Schawinksi, Kevin; Zhang, Ce; Weigel, Anna K.; Turp, M. Dennis

2018-05-01

We investigate the use of deep convolutional neural networks (deep CNNs) for automatic visual detection of galaxy mergers. Moreover, we investigate the use of transfer learning in conjunction with CNNs, by retraining networks first trained on pictures of everyday objects. We test the hypothesis that transfer learning is useful for improving classification performance for small training sets. This would make transfer learning useful for finding rare objects in astronomical imaging datasets. We find that these deep learning methods perform significantly better than current state-of-the-art merger detection methods based on nonparametric systems like CAS and GM20. Our method is end-to-end and robust to image noise and distortions; it can be applied directly without image preprocessing. We also find that transfer learning can act as a regulariser in some cases, leading to better overall classification accuracy (p = 0.02). Transfer learning on our full training set leads to a lowered error rate from 0.0381 down to 0.0321, a relative improvement of 15%. Finally, we perform a basic sanity-check by creating a merger sample with our method, and comparing with an already existing, manually created merger catalogue in terms of colour-mass distribution and stellar mass function.

Solar Energy Meteorological Research and Training Site: Region 5. Annual report, 30 September 1977-29 September 1978

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

Rao, C.R.N.; Hewson, E.W.

The primary facility which is to be a benchmark site for the acquisition of research quality solar radiation and solar energy related meteorological data has been set up and will be fully operational in the near future. The training program has been established with the introduction of two, two-quarter courses on solar radiation and meteorological measurements and on atmospheric radiative processes. Also, as part of the training program, a week-long workshop on solar energy measurement and instrumentation was conducted during the summer of '78 and a series of seminars on solar energy related topics, catering to both professionals and non-professionals,more » was arranged during the 1977-78 academic year. A meeting of solar radiation scientists from the five states of the region was held in Corvallis (August '78) to explore the feasibility of setting up a regional network of stations to acquire research quality solar radiation and meteorological data. Useful global irradiance measurements have been made at the five sites, making up the general quality network in Oregon, over the greater part of the year.« less

Efficient Online Learning Algorithms Based on LSTM Neural Networks.

PubMed

Ergen, Tolga; Kozat, Suleyman Serdar

2017-09-13

We investigate online nonlinear regression and introduce novel regression structures based on the long short term memory (LSTM) networks. For the introduced structures, we also provide highly efficient and effective online training methods. To train these novel LSTM-based structures, we put the underlying architecture in a state space form and introduce highly efficient and effective particle filtering (PF)-based updates. We also provide stochastic gradient descent and extended Kalman filter-based updates. Our PF-based training method guarantees convergence to the optimal parameter estimation in the mean square error sense provided that we have a sufficient number of particles and satisfy certain technical conditions. More importantly, we achieve this performance with a computational complexity in the order of the first-order gradient-based methods by controlling the number of particles. Since our approach is generic, we also introduce a gated recurrent unit (GRU)-based approach by directly replacing the LSTM architecture with the GRU architecture, where we demonstrate the superiority of our LSTM-based approach in the sequential prediction task via different real life data sets. In addition, the experimental results illustrate significant performance improvements achieved by the introduced algorithms with respect to the conventional methods over several different benchmark real life data sets.

Learning relevant features of data with multi-scale tensor networks

NASA Astrophysics Data System (ADS)

Miles Stoudenmire, E.

2018-07-01

Inspired by coarse-graining approaches used in physics, we show how similar algorithms can be adapted for data. The resulting algorithms are based on layered tree tensor networks and scale linearly with both the dimension of the input and the training set size. Computing most of the layers with an unsupervised algorithm, then optimizing just the top layer for supervised classification of the MNIST and fashion MNIST data sets gives very good results. We also discuss mixing a prior guess for supervised weights together with an unsupervised representation of the data, yielding a smaller number of features nevertheless able to give good performance.

Artificial Neural Networks applied to estimate permeability, porosity and intrinsic attenuation using seismic attributes and well-log data

NASA Astrophysics Data System (ADS)

Iturrarán-Viveros, Ursula; Parra, Jorge O.

2014-08-01

Permeability and porosity are two fundamental reservoir properties which relate to the amount of fluid contained in a reservoir and its ability to flow. The intrinsic attenuation is another important parameter since it is related to porosity, permeability, oil and gas saturation and these parameters significantly affect the seismic signature of a reservoir. We apply Artificial Neural Network (ANN) models to predict permeability (k) and porosity (ϕ) for a carbonate aquifer in southeastern Florida and to predict intrinsic attenuation (1/Q) for a sand-shale oil reservoir in northeast Texas. In this study, the Gamma test (a revolutionary estimator of the noise in a data set) has been used as a mathematically non-parametric nonlinear smooth modeling tool to choose the best input combination of seismic attributes to estimate k and ϕ, and the best combination of well-logs to estimate 1/Q. This saves time during the construction and training of ANN models and also sets a lower bound for the mean squared error to prevent over-training. The Neural Network method successfully delineates a highly permeable zone that corresponds to a high water production in the aquifer. The Gamma test found nonlinear relations that were not visible to linear regression allowing us to generalize the ANN estimations of k, ϕ and 1/Q for their respective sets of patterns that were not used during the learning phase.

Supervised neural network classification of pre-sliced cooked pork ham images using quaternionic singular values.

PubMed

Valous, Nektarios A; Mendoza, Fernando; Sun, Da-Wen; Allen, Paul

2010-03-01

The quaternionic singular value decomposition is a technique to decompose a quaternion matrix (representation of a colour image) into quaternion singular vector and singular value component matrices exposing useful properties. The objective of this study was to use a small portion of uncorrelated singular values, as robust features for the classification of sliced pork ham images, using a supervised artificial neural network classifier. Images were acquired from four qualities of sliced cooked pork ham typically consumed in Ireland (90 slices per quality), having similar appearances. Mahalanobis distances and Pearson product moment correlations were used for feature selection. Six highly discriminating features were used as input to train the neural network. An adaptive feedforward multilayer perceptron classifier was employed to obtain a suitable mapping from the input dataset. The overall correct classification performance for the training, validation and test set were 90.3%, 94.4%, and 86.1%, respectively. The results confirm that the classification performance was satisfactory. Extracting the most informative features led to the recognition of a set of different but visually quite similar textural patterns based on quaternionic singular values. Copyright 2009 Elsevier Ltd. All rights reserved.

Kannada character recognition system using neural network

NASA Astrophysics Data System (ADS)

Kumar, Suresh D. S.; Kamalapuram, Srinivasa K.; Kumar, Ajay B. R.

2013-03-01

Handwriting recognition has been one of the active and challenging research areas in the field of pattern recognition. It has numerous applications which include, reading aid for blind, bank cheques and conversion of any hand written document into structural text form. As there is no sufficient number of works on Indian language character recognition especially Kannada script among 15 major scripts in India. In this paper an attempt is made to recognize handwritten Kannada characters using Feed Forward neural networks. A handwritten Kannada character is resized into 20x30 Pixel. The resized character is used for training the neural network. Once the training process is completed the same character is given as input to the neural network with different set of neurons in hidden layer and their recognition accuracy rate for different Kannada characters has been calculated and compared. The results show that the proposed system yields good recognition accuracy rates comparable to that of other handwritten character recognition systems.

Classification of images acquired with colposcopy using artificial neural networks.

PubMed

Simões, Priscyla W; Izumi, Narjara B; Casagrande, Ramon S; Venson, Ramon; Veronezi, Carlos D; Moretti, Gustavo P; da Rocha, Edroaldo L; Cechinel, Cristian; Ceretta, Luciane B; Comunello, Eros; Martins, Paulo J; Casagrande, Rogério A; Snoeyer, Maria L; Manenti, Sandra A

2014-01-01

To explore the advantages of using artificial neural networks (ANNs) to recognize patterns in colposcopy to classify images in colposcopy. Transversal, descriptive, and analytical study of a quantitative approach with an emphasis on diagnosis. The training test e validation set was composed of images collected from patients who underwent colposcopy. These images were provided by a gynecology clinic located in the city of Criciúma (Brazil). The image database (n = 170) was divided; 48 images were used for the training process, 58 images were used for the tests, and 64 images were used for the validation. A hybrid neural network based on Kohonen self-organizing maps and multilayer perceptron (MLP) networks was used. After 126 cycles, the validation was performed. The best results reached an accuracy of 72.15%, a sensibility of 69.78%, and a specificity of 68%. Although the preliminary results still exhibit an average efficiency, the present approach is an innovative and promising technique that should be deeply explored in the context of the present study.

Mass detection in digital breast tomosynthesis: Deep convolutional neural network with transfer learning from mammography.

PubMed

Samala, Ravi K; Chan, Heang-Ping; Hadjiiski, Lubomir; Helvie, Mark A; Wei, Jun; Cha, Kenny

2016-12-01

Develop a computer-aided detection (CAD) system for masses in digital breast tomosynthesis (DBT) volume using a deep convolutional neural network (DCNN) with transfer learning from mammograms. A data set containing 2282 digitized film and digital mammograms and 324 DBT volumes were collected with IRB approval. The mass of interest on the images was marked by an experienced breast radiologist as reference standard. The data set was partitioned into a training set (2282 mammograms with 2461 masses and 230 DBT views with 228 masses) and an independent test set (94 DBT views with 89 masses). For DCNN training, the region of interest (ROI) containing the mass (true positive) was extracted from each image. False positive (FP) ROIs were identified at prescreening by their previously developed CAD systems. After data augmentation, a total of 45 072 mammographic ROIs and 37 450 DBT ROIs were obtained. Data normalization and reduction of non-uniformity in the ROIs across heterogeneous data was achieved using a background correction method applied to each ROI. A DCNN with four convolutional layers and three fully connected (FC) layers was first trained on the mammography data. Jittering and dropout techniques were used to reduce overfitting. After training with the mammographic ROIs, all weights in the first three convolutional layers were frozen, and only the last convolution layer and the FC layers were randomly initialized again and trained using the DBT training ROIs. The authors compared the performances of two CAD systems for mass detection in DBT: one used the DCNN-based approach and the other used their previously developed feature-based approach for FP reduction. The prescreening stage was identical in both systems, passing the same set of mass candidates to the FP reduction stage. For the feature-based CAD system, 3D clustering and active contour method was used for segmentation; morphological, gray level, and texture features were extracted and merged with a linear discriminant classifier to score the detected masses. For the DCNN-based CAD system, ROIs from five consecutive slices centered at each candidate were passed through the trained DCNN and a mass likelihood score was generated. The performances of the CAD systems were evaluated using free-response ROC curves and the performance difference was analyzed using a non-parametric method. Before transfer learning, the DCNN trained only on mammograms with an AUC of 0.99 classified DBT masses with an AUC of 0.81 in the DBT training set. After transfer learning with DBT, the AUC improved to 0.90. For breast-based CAD detection in the test set, the sensitivity for the feature-based and the DCNN-based CAD systems was 83% and 91%, respectively, at 1 FP/DBT volume. The difference between the performances for the two systems was statistically significant (p-value < 0.05). The image patterns learned from the mammograms were transferred to the mass detection on DBT slices through the DCNN. This study demonstrated that large data sets collected from mammography are useful for developing new CAD systems for DBT, alleviating the problem and effort of collecting entirely new large data sets for the new modality.

Improving CNN Performance Accuracies With Min-Max Objective.

PubMed

Shi, Weiwei; Gong, Yihong; Tao, Xiaoyu; Wang, Jinjun; Zheng, Nanning

2017-06-09

We propose a novel method for improving performance accuracies of convolutional neural network (CNN) without the need to increase the network complexity. We accomplish the goal by applying the proposed Min-Max objective to a layer below the output layer of a CNN model in the course of training. The Min-Max objective explicitly ensures that the feature maps learned by a CNN model have the minimum within-manifold distance for each object manifold and the maximum between-manifold distances among different object manifolds. The Min-Max objective is general and able to be applied to different CNNs with insignificant increases in computation cost. Moreover, an incremental minibatch training procedure is also proposed in conjunction with the Min-Max objective to enable the handling of large-scale training data. Comprehensive experimental evaluations on several benchmark data sets with both the image classification and face verification tasks reveal that employing the proposed Min-Max objective in the training process can remarkably improve performance accuracies of a CNN model in comparison with the same model trained without using this objective.

Min-max hyperellipsoidal clustering for anomaly detection in network security.

PubMed

Sarasamma, Suseela T; Zhu, Qiuming A

2006-08-01

A novel hyperellipsoidal clustering technique is presented for an intrusion-detection system in network security. Hyperellipsoidal clusters toward maximum intracluster similarity and minimum intercluster similarity are generated from training data sets. The novelty of the technique lies in the fact that the parameters needed to construct higher order data models in general multivariate Gaussian functions are incrementally derived from the data sets using accretive processes. The technique is implemented in a feedforward neural network that uses a Gaussian radial basis function as the model generator. An evaluation based on the inclusiveness and exclusiveness of samples with respect to specific criteria is applied to accretively learn the output clusters of the neural network. One significant advantage of this is its ability to detect individual anomaly types that are hard to detect with other anomaly-detection schemes. Applying this technique, several feature subsets of the tcptrace network-connection records that give above 95% detection at false-positive rates below 5% were identified.

A rapid learning and dynamic stepwise updating algorithm for flat neural networks and the application to time-series prediction.

PubMed

Chen, C P; Wan, J Z

1999-01-01

A fast learning algorithm is proposed to find an optimal weights of the flat neural networks (especially, the functional-link network). Although the flat networks are used for nonlinear function approximation, they can be formulated as linear systems. Thus, the weights of the networks can be solved easily using a linear least-square method. This formulation makes it easier to update the weights instantly for both a new added pattern and a new added enhancement node. A dynamic stepwise updating algorithm is proposed to update the weights of the system on-the-fly. The model is tested on several time-series data including an infrared laser data set, a chaotic time-series, a monthly flour price data set, and a nonlinear system identification problem. The simulation results are compared to existing models in which more complex architectures and more costly training are needed. The results indicate that the proposed model is very attractive to real-time processes.

Fire detection from hyperspectral data using neural network approach

NASA Astrophysics Data System (ADS)

Piscini, Alessandro; Amici, Stefania

2015-10-01

This study describes an application of artificial neural networks for the recognition of flaming areas using hyper- spectral remote sensed data. Satellite remote sensing is considered an effective and safe way to monitor active fires for environmental and people safeguarding. Neural networks are an effective and consolidated technique for the classification of satellite images. Moreover, once well trained, they prove to be very fast in the application stage for a rapid response. At flaming temperature, thanks to its low excitation energy (about 4.34 eV), potassium (K) ionize with a unique doublet emission features. This emission features can be detected remotely providing a detection map of active fire which allows in principle to separate flaming from smouldering areas of vegetation even in presence of smoke. For this study a normalised Advanced K Band Difference (AKBD) has been applied to airborne hyper spectral sensor covering a range of 400-970 nm with resolution 2.9 nm. A back propagation neural network was used for the recognition of active fires affecting the hyperspectral image. The network was trained using all channels of sensor as inputs, and the corresponding AKBD indexes as target output. In order to evaluate its generalization capabilities, the neural network was validated on two independent data sets of hyperspectral images, not used during neural network training phase. The validation results for the independent data-sets had an overall accuracy round 100% for both image and a few commission errors (0.1%), therefore demonstrating the feasibility of estimating the presence of active fires using a neural network approach. Although the validation of the neural network classifier had a few commission errors, the producer accuracies were lower due to the presence of omission errors. Image analysis revealed that those false negatives lie in "smoky" portion fire fronts, and due to the low intensity of the signal. The proposed method can be considered effective both in terms of classification accuracy and generalization capability. In particular our approach proved to be robust in the rejection of false positives, often corresponding to noisy or smoke pixels, whose presence in hyperspectral images can often undermine the performance of traditional classification algorithms. In order to improve neural network performance, future activities will include also the exploiting of hyperspectral images in the shortwave infrared region of the electromagnetic spectrum, covering wavelengths from 1400 to 2500 nm, which include significant emitted radiance from fire.

Research training of Developmental-Behavioral Pediatrics fellows: a survey of fellowship directors by Developmental-Behavioral Pediatrics Research Network.

PubMed

Wiley, Susan; Schonfeld, David J; Fredstrom, Bridget; Huffman, Lynne

2013-01-01

To describe research training in Developmental-Behavioral Pediatrics (DBP) Fellowship Programs. Thirty-five US-accredited DBP fellowships were contacted through the Developmental-Behavioral Pediatrics Research Network to complete an online survey on scholarly work and research training. With an 83% response rate, responding programs represented 110 (87 filled) fellowship positions. External funding for fellowship positions was minimal (11 positions fully funded, 13 funded above 50% of cost). Structured research training included didactic lectures, web-based training, university courses, direct mentoring, journal clubs, and required reading. Of the 159 fellows described, spanning a 5-year training period, the majority chose projects relying on their own data collection (57%) rather than joining an existing research study and focused on clinical research (86%). Among 96 fellows with completed scholarly work, 29% were observational/epidemiological studies, 22% secondary analyses of large data sets, 16% community-based research, and 15% survey design. A limited number of fellows pursued basic science, meta-analysis/critical appraisal of the literature, or analysis of public policy. Barriers to successful fellow research are as follows: lack of time and money, challenges in balancing clinical demands and protected faculty research time, limited faculty research opportunities, time or expertise, and a lack of infrastructure for fellow research mentoring. The scholarly work of fellows in DBP fellowship programs has primarily focused on clinical research using observational/epidemiological research and secondary analysis of large data set. Barriers largely in faculty time and expertise for research mentoring and inadequate funding in programs that have high clinical demands and little resources for research efforts were noted.

Modeling of cortical signals using echo state networks

NASA Astrophysics Data System (ADS)

Zhou, Hanying; Wang, Yongji; Huang, Jiangshuai

2009-10-01

Diverse modeling frameworks have been utilized with the ultimate goal of translating brain cortical signals into prediction of visible behavior. The inputs to these models are usually multidimensional neural recordings collected from relevant regions of a monkey's brain while the outputs are the associated behavior which is typically the 2-D or 3-D hand position of a primate. Here our task is to set up a proper model in order to figure out the move trajectories by input the neural signals which are simultaneously collected in the experiment. In this paper, we propose to use Echo State Networks (ESN) to map the neural firing activities into hand positions. ESN is a newly developed recurrent neural network(RNN) model. Besides its dynamic property and short term memory just as other recurrent neural networks have, it has a special echo state property which endows it with the ability to model nonlinear dynamic systems powerfully. What distinguished it from transitional recurrent neural networks most significantly is its special learning method. In this paper we train this net with a refined version of its typical training method and get a better model.

Application of Deep Learning in GLOBELAND30-2010 Product Refinement

NASA Astrophysics Data System (ADS)

Liu, T.; Chen, X.

2018-04-01

GlobeLand30, as one of the best Global Land Cover (GLC) product at 30-m resolution, has been widely used in many research fields. Due to the significant spectral confusion among different land cover types and limited textual information of Landsat data, the overall accuracy of GlobeLand30 is about 80 %. Although such accuracy is much higher than most other global land cover products, it cannot satisfy various applications. There is still a great need of an effective method to improve the quality of GlobeLand30. The explosive high-resolution satellite images and remarkable performance of Deep Learning on image classification provide a new opportunity to refine GlobeLand30. However, the performance of deep leaning depends on quality and quantity of training samples as well as model training strategy. Therefore, this paper 1) proposed an automatic training sample generation method via Google earth to build a large training sample set; and 2) explore the best training strategy for land cover classification using GoogleNet (Inception V3), one of the most widely used deep learning network. The result shows that the fine-tuning from first layer of Inception V3 using rough large sample set is the best strategy. The retrained network was then applied in one selected area from Xi'an city as a case study of GlobeLand30 refinement. The experiment results indicate that the proposed approach with Deep Learning and google earth imagery is a promising solution for further improving accuracy of GlobeLand30.

Genetic algorithm based adaptive neural network ensemble and its application in predicting carbon flux

USGS Publications Warehouse

Xue, Y.; Liu, S.; Hu, Y.; Yang, J.; Chen, Q.

2007-01-01

To improve the accuracy in prediction, Genetic Algorithm based Adaptive Neural Network Ensemble (GA-ANNE) is presented. Intersections are allowed between different training sets based on the fuzzy clustering analysis, which ensures the diversity as well as the accuracy of individual Neural Networks (NNs). Moreover, to improve the accuracy of the adaptive weights of individual NNs, GA is used to optimize the cluster centers. Empirical results in predicting carbon flux of Duke Forest reveal that GA-ANNE can predict the carbon flux more accurately than Radial Basis Function Neural Network (RBFNN), Bagging NN ensemble, and ANNE. ?? 2007 IEEE.

Bayesian convolutional neural network based MRI brain extraction on nonhuman primates.

PubMed

Zhao, Gengyan; Liu, Fang; Oler, Jonathan A; Meyerand, Mary E; Kalin, Ned H; Birn, Rasmus M

2018-07-15

Brain extraction or skull stripping of magnetic resonance images (MRI) is an essential step in neuroimaging studies, the accuracy of which can severely affect subsequent image processing procedures. Current automatic brain extraction methods demonstrate good results on human brains, but are often far from satisfactory on nonhuman primates, which are a necessary part of neuroscience research. To overcome the challenges of brain extraction in nonhuman primates, we propose a fully-automated brain extraction pipeline combining deep Bayesian convolutional neural network (CNN) and fully connected three-dimensional (3D) conditional random field (CRF). The deep Bayesian CNN, Bayesian SegNet, is used as the core segmentation engine. As a probabilistic network, it is not only able to perform accurate high-resolution pixel-wise brain segmentation, but also capable of measuring the model uncertainty by Monte Carlo sampling with dropout in the testing stage. Then, fully connected 3D CRF is used to refine the probability result from Bayesian SegNet in the whole 3D context of the brain volume. The proposed method was evaluated with a manually brain-extracted dataset comprising T1w images of 100 nonhuman primates. Our method outperforms six popular publicly available brain extraction packages and three well-established deep learning based methods with a mean Dice coefficient of 0.985 and a mean average symmetric surface distance of 0.220 mm. A better performance against all the compared methods was verified by statistical tests (all p-values < 10 -4 , two-sided, Bonferroni corrected). The maximum uncertainty of the model on nonhuman primate brain extraction has a mean value of 0.116 across all the 100 subjects. The behavior of the uncertainty was also studied, which shows the uncertainty increases as the training set size decreases, the number of inconsistent labels in the training set increases, or the inconsistency between the training set and the testing set increases. Copyright © 2018 Elsevier Inc. All rights reserved.

Dog and human inflammatory bowel disease rely on overlapping yet distinct dysbiosis networks.

PubMed

Vázquez-Baeza, Yoshiki; Hyde, Embriette R; Suchodolski, Jan S; Knight, Rob

2016-10-03

Inflammatory bowel disease (IBD) is an autoimmune condition that is difficult to diagnose, and animal models of this disease have questionable human relevance 1 . Here, we show that the dysbiosis network underlying IBD in dogs differs from that in humans, with some bacteria such as Fusobacterium switching roles between the two species (as Bacteroides fragilis switches roles between humans and mice) 2 . For example, a dysbiosis index trained on humans fails when applied to dogs, but a dog-specific dysbiosis index achieves high correlations with the overall dog microbial community diversity patterns. In addition, a random forest classifier trained on dog-specific samples achieves high discriminatory power, even when using stool samples rather than the mucosal biopsies required for high discriminatory power in humans 2 . These relationships were not detected in previously published dog IBD data sets due to their limited sample size and statistical power 3 . Taken together, these results reveal the need to train host-specific dysbiosis networks and point the way towards a generalized understanding of IBD across different mammalian models.

Photoacoustic image reconstruction via deep learning

NASA Astrophysics Data System (ADS)

Antholzer, Stephan; Haltmeier, Markus; Nuster, Robert; Schwab, Johannes

2018-02-01

Applying standard algorithms to sparse data problems in photoacoustic tomography (PAT) yields low-quality images containing severe under-sampling artifacts. To some extent, these artifacts can be reduced by iterative image reconstruction algorithms which allow to include prior knowledge such as smoothness, total variation (TV) or sparsity constraints. These algorithms tend to be time consuming as the forward and adjoint problems have to be solved repeatedly. Further, iterative algorithms have additional drawbacks. For example, the reconstruction quality strongly depends on a-priori model assumptions about the objects to be recovered, which are often not strictly satisfied in practical applications. To overcome these issues, in this paper, we develop direct and efficient reconstruction algorithms based on deep learning. As opposed to iterative algorithms, we apply a convolutional neural network, whose parameters are trained before the reconstruction process based on a set of training data. For actual image reconstruction, a single evaluation of the trained network yields the desired result. Our presented numerical results (using two different network architectures) demonstrate that the proposed deep learning approach reconstructs images with a quality comparable to state of the art iterative reconstruction methods.

Predicting high-risk preterm birth using artificial neural networks.

PubMed

Catley, Christina; Frize, Monique; Walker, C Robin; Petriu, Dorina C

2006-07-01

A reengineered approach to the early prediction of preterm birth is presented as a complimentary technique to the current procedure of using costly and invasive clinical testing on high-risk maternal populations. Artificial neural networks (ANNs) are employed as a screening tool for preterm birth on a heterogeneous maternal population; risk estimations use obstetrical variables available to physicians before 23 weeks gestation. The objective was to assess if ANNs have a potential use in obstetrical outcome estimations in low-risk maternal populations. The back-propagation feedforward ANN was trained and tested on cases with eight input variables describing the patient's obstetrical history; the output variables were: 1) preterm birth; 2) high-risk preterm birth; and 3) a refined high-risk preterm birth outcome excluding all cases where resuscitation was delivered in the form of free flow oxygen. Artificial training sets were created to increase the distribution of the underrepresented class to 20%. Training on the refined high-risk preterm birth model increased the network's sensitivity to 54.8%, compared to just over 20% for the nonartificially distributed preterm birth model.

Suicide Prevention in a Treatment Setting.

ERIC Educational Resources Information Center

Litman, Robert E.

1995-01-01

The author anticipates that sophisticated interactive computer programs will be effective in improving screening and case finding of the suicidal and that they will become invaluable in improving training for primary care providers and outpatient mental health workers. Additionally, improved communication networks will help maintain continuity of…

Design of cognitive engine for cognitive radio based on the rough sets and radial basis function neural network

NASA Astrophysics Data System (ADS)

Yang, Yanchao; Jiang, Hong; Liu, Congbin; Lan, Zhongli

2013-03-01

Cognitive radio (CR) is an intelligent wireless communication system which can dynamically adjust the parameters to improve system performance depending on the environmental change and quality of service. The core technology for CR is the design of cognitive engine, which introduces reasoning and learning methods in the field of artificial intelligence, to achieve the perception, adaptation and learning capability. Considering the dynamical wireless environment and demands, this paper proposes a design of cognitive engine based on the rough sets (RS) and radial basis function neural network (RBF_NN). The method uses experienced knowledge and environment information processed by RS module to train the RBF_NN, and then the learning model is used to reconfigure communication parameters to allocate resources rationally and improve system performance. After training learning model, the performance is evaluated according to two benchmark functions. The simulation results demonstrate the effectiveness of the model and the proposed cognitive engine can effectively achieve the goal of learning and reconfiguration in cognitive radio.

Plastic Surgery Response in Natural Disasters.

PubMed

Chung, Susan; Zimmerman, Amanda; Gaviria, Andres; Dayicioglu, Deniz

2015-06-01

Disasters cause untold damage and are often unpredictable; however, with proper preparation, these events can be better managed. The initial response has the greatest impact on the overall success of the relief effort. A well-trained multidisciplinary network of providers is necessary to ensure coordinated care for the victims of these mass casualty disasters. As members of this network of providers, plastic surgeons have the ability to efficiently address injuries sustained in mass casualty disasters and are a valuable member of the relief effort. The skill set of plastic surgeons includes techniques that can address injuries sustained in large-scale emergencies, such as the management of soft-tissue injury, tissue viability, facial fractures, and extremity salvage. An approach to disaster relief, the types of disasters encountered, the management of injuries related to mass casualty disasters, the role of plastic surgeons in the relief effort, and resource management are discussed. In order to improve preparedness in future mass casualty disasters, plastic surgeons should receive training during residency regarding the utilization of plastic surgery knowledge in the disaster setting.

Neural network analysis of crosshole tomographic images: The seismic signature of gas hydrate bearing sediments in the Mackenzie Delta (NW Canada)

NASA Astrophysics Data System (ADS)

Bauer, K.; Pratt, R. G.; Haberland, C.; Weber, M.

2008-10-01

Crosshole seismic experiments were conducted to study the in-situ properties of gas hydrate bearing sediments (GHBS) in the Mackenzie Delta (NW Canada). Seismic tomography provided images of P velocity, anisotropy, and attenuation. Self-organizing maps (SOM) are powerful neural network techniques to classify and interpret multi-attribute data sets. The coincident tomographic images are translated to a set of data vectors in order to train a Kohonen layer. The total gradient of the model vectors is determined for the trained SOM and a watershed segmentation algorithm is used to visualize and map the lithological clusters with well-defined seismic signatures. Application to the Mallik data reveals four major litho-types: (1) GHBS, (2) sands, (3) shale/coal interlayering, and (4) silt. The signature of seismic P wave characteristics distinguished for the GHBS (high velocities, strong anisotropy and attenuation) is new and can be used for new exploration strategies to map and quantify gas hydrates.

Generating Focused Molecule Libraries for Drug Discovery with Recurrent Neural Networks

PubMed Central

2017-01-01

In de novo drug design, computational strategies are used to generate novel molecules with good affinity to the desired biological target. In this work, we show that recurrent neural networks can be trained as generative models for molecular structures, similar to statistical language models in natural language processing. We demonstrate that the properties of the generated molecules correlate very well with the properties of the molecules used to train the model. In order to enrich libraries with molecules active toward a given biological target, we propose to fine-tune the model with small sets of molecules, which are known to be active against that target. Against Staphylococcus aureus, the model reproduced 14% of 6051 hold-out test molecules that medicinal chemists designed, whereas against Plasmodium falciparum (Malaria), it reproduced 28% of 1240 test molecules. When coupled with a scoring function, our model can perform the complete de novo drug design cycle to generate large sets of novel molecules for drug discovery. PMID:29392184

Cancer diagnostics using neural network sorting of processed images

NASA Astrophysics Data System (ADS)

Wyman, Charles L.; Schreeder, Marshall; Grundy, Walt; Kinser, Jason M.

1996-03-01

A combination of image processing with neural network sorting was conducted to demonstrate feasibility of automated cervical smear screening. Nuclei were isolated to generate a series of data points relating to the density and size of individual nuclei. This was followed by segmentation to isolate entire cells for subsequent generation of data points to bound the size of the cytoplasm. Data points were taken on as many as ten cells per image frame and included correlation against a series of filters providing size and density readings on nuclei. Additional point data was taken on nuclei images to refine size information and on whole cells to bound the size of the cytoplasm, twenty data points per assessed cell were generated. These data point sets, designated as neural tensors, comprise the inputs for training and use of a unique neural network to sort the images and identify those indicating evidence of disease. The neural network, named the Fast Analog Associative Memory, accumulates data and establishes lookup tables for comparison against images to be assessed. Six networks were trained to differentiate normal cells from those evidencing various levels abnormality that may lead to cancer. A blind test was conducted on 77 images to evaluate system performance. The image set included 31 positives (diseased) and 46 negatives (normal). Our system correctly identified all 31 positives and 41 of the negatives with 5 false positives. We believe this technology can lead to more efficient automated screening of cervical smears.

Quantum autoencoders for efficient compression of quantum data

NASA Astrophysics Data System (ADS)

Romero, Jonathan; Olson, Jonathan P.; Aspuru-Guzik, Alan

2017-12-01

Classical autoencoders are neural networks that can learn efficient low-dimensional representations of data in higher-dimensional space. The task of an autoencoder is, given an input x, to map x to a lower dimensional point y such that x can likely be recovered from y. The structure of the underlying autoencoder network can be chosen to represent the data on a smaller dimension, effectively compressing the input. Inspired by this idea, we introduce the model of a quantum autoencoder to perform similar tasks on quantum data. The quantum autoencoder is trained to compress a particular data set of quantum states, where a classical compression algorithm cannot be employed. The parameters of the quantum autoencoder are trained using classical optimization algorithms. We show an example of a simple programmable circuit that can be trained as an efficient autoencoder. We apply our model in the context of quantum simulation to compress ground states of the Hubbard model and molecular Hamiltonians.

Training pediatric clinical pharmacology and therapeutics specialists of the future: the needs, the reality, and opportunities for international networking.

PubMed

Gazarian, Madlen

2009-01-01

In recent years there has been a rapid and marked increase in global recognition of the need for better medicines for children, with various initiatives being implemented at global and regional levels. These exciting developments are matched by recognition of the need to build greater capacity in the field of pediatric clinical pharmacology and therapeutics to help deliver on the promise of better medicines for children. A range of pediatric medicines researchers, educators, clinical therapeutics practitioners, and experts in drug evaluation, regulation, and broader medicines policy are needed on a larger scale, in both developed and developing world settings. The current and likely future training needs to meet these diverse challenges, the current realities of trying to meet such needs, and the opportunities for international networking to help meet future training needs are discussed from a global perspective.

Enhancing deep convolutional neural network scheme for breast cancer diagnosis with unlabeled data.

PubMed

Sun, Wenqing; Tseng, Tzu-Liang Bill; Zhang, Jianying; Qian, Wei

2017-04-01

In this study we developed a graph based semi-supervised learning (SSL) scheme using deep convolutional neural network (CNN) for breast cancer diagnosis. CNN usually needs a large amount of labeled data for training and fine tuning the parameters, and our proposed scheme only requires a small portion of labeled data in training set. Four modules were included in the diagnosis system: data weighing, feature selection, dividing co-training data labeling, and CNN. 3158 region of interests (ROIs) with each containing a mass extracted from 1874 pairs of mammogram images were used for this study. Among them 100 ROIs were treated as labeled data while the rest were treated as unlabeled. The area under the curve (AUC) observed in our study was 0.8818, and the accuracy of CNN is 0.8243 using the mixed labeled and unlabeled data. Copyright © 2016. Published by Elsevier Ltd.

Supervised Detection of Anomalous Light Curves in Massive Astronomical Catalogs

NASA Astrophysics Data System (ADS)

Nun, Isadora; Pichara, Karim; Protopapas, Pavlos; Kim, Dae-Won

2014-09-01

The development of synoptic sky surveys has led to a massive amount of data for which resources needed for analysis are beyond human capabilities. In order to process this information and to extract all possible knowledge, machine learning techniques become necessary. Here we present a new methodology to automatically discover unknown variable objects in large astronomical catalogs. With the aim of taking full advantage of all information we have about known objects, our method is based on a supervised algorithm. In particular, we train a random forest classifier using known variability classes of objects and obtain votes for each of the objects in the training set. We then model this voting distribution with a Bayesian network and obtain the joint voting distribution among the training objects. Consequently, an unknown object is considered as an outlier insofar it has a low joint probability. By leaving out one of the classes on the training set, we perform a validity test and show that when the random forest classifier attempts to classify unknown light curves (the class left out), it votes with an unusual distribution among the classes. This rare voting is detected by the Bayesian network and expressed as a low joint probability. Our method is suitable for exploring massive data sets given that the training process is performed offline. We tested our algorithm on 20 million light curves from the MACHO catalog and generated a list of anomalous candidates. After analysis, we divided the candidates into two main classes of outliers: artifacts and intrinsic outliers. Artifacts were principally due to air mass variation, seasonal variation, bad calibration, or instrumental errors and were consequently removed from our outlier list and added to the training set. After retraining, we selected about 4000 objects, which we passed to a post-analysis stage by performing a cross-match with all publicly available catalogs. Within these candidates we identified certain known but rare objects such as eclipsing Cepheids, blue variables, cataclysmic variables, and X-ray sources. For some outliers there was no additional information. Among them we identified three unknown variability types and a few individual outliers that will be followed up in order to perform a deeper analysis.

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

Nun, Isadora; Pichara, Karim; Protopapas, Pavlos

The development of synoptic sky surveys has led to a massive amount of data for which resources needed for analysis are beyond human capabilities. In order to process this information and to extract all possible knowledge, machine learning techniques become necessary. Here we present a new methodology to automatically discover unknown variable objects in large astronomical catalogs. With the aim of taking full advantage of all information we have about known objects, our method is based on a supervised algorithm. In particular, we train a random forest classifier using known variability classes of objects and obtain votes for each ofmore » the objects in the training set. We then model this voting distribution with a Bayesian network and obtain the joint voting distribution among the training objects. Consequently, an unknown object is considered as an outlier insofar it has a low joint probability. By leaving out one of the classes on the training set, we perform a validity test and show that when the random forest classifier attempts to classify unknown light curves (the class left out), it votes with an unusual distribution among the classes. This rare voting is detected by the Bayesian network and expressed as a low joint probability. Our method is suitable for exploring massive data sets given that the training process is performed offline. We tested our algorithm on 20 million light curves from the MACHO catalog and generated a list of anomalous candidates. After analysis, we divided the candidates into two main classes of outliers: artifacts and intrinsic outliers. Artifacts were principally due to air mass variation, seasonal variation, bad calibration, or instrumental errors and were consequently removed from our outlier list and added to the training set. After retraining, we selected about 4000 objects, which we passed to a post-analysis stage by performing a cross-match with all publicly available catalogs. Within these candidates we identified certain known but rare objects such as eclipsing Cepheids, blue variables, cataclysmic variables, and X-ray sources. For some outliers there was no additional information. Among them we identified three unknown variability types and a few individual outliers that will be followed up in order to perform a deeper analysis.« less

Protein backbone and sidechain torsion angles predicted from NMR chemical shifts using artificial neural networks

PubMed Central

Shen, Yang; Bax, Ad

2013-01-01

A new program, TALOS-N, is introduced for predicting protein backbone torsion angles from NMR chemical shifts. The program relies far more extensively on the use of trained artificial neural networks than its predecessor, TALOS+. Validation on an independent set of proteins indicates that backbone torsion angles can be predicted for a larger, ≥ 90% fraction of the residues, with an error rate smaller than ca 3.5%, using an acceptance criterion that is nearly two-fold tighter than that used previously, and a root mean square difference between predicted and crystallographically observed (φ,ψ) torsion angles of ca 12°. TALOS-N also reports sidechain χ1 rotameric states for about 50% of the residues, and a consistency with reference structures of 89%. The program includes a neural network trained to identify secondary structure from residue sequence and chemical shifts. PMID:23728592

A neural-network potential through charge equilibration for WS2: From clusters to sheets

NASA Astrophysics Data System (ADS)

Hafizi, Roohollah; Ghasemi, S. Alireza; Hashemifar, S. Javad; Akbarzadeh, Hadi

2017-12-01

In the present work, we use a machine learning method to construct a high-dimensional potential for tungsten disulfide using a charge equilibration neural-network technique. A training set of stoichiometric WS2 clusters is prepared in the framework of density functional theory. After training the neural-network potential, the reliability and transferability of the potential are verified by performing a crystal structure search on bulk phases of WS2 and by plotting energy-area curves of two different monolayers. Then, we use the potential to investigate various triangular nano-clusters and nanotubes of WS2. In the case of nano-structures, we argue that 2H atomic configurations with sulfur rich edges are thermodynamically more stable than the other investigated configurations. We also studied a number of WS2 nanotubes which revealed that 1T tubes with armchair chirality exhibit lower bending stiffness.

Graph Representations of Flow and Transport in Fracture Networks using Machine Learning

NASA Astrophysics Data System (ADS)

Srinivasan, G.; Viswanathan, H. S.; Karra, S.; O'Malley, D.; Godinez, H. C.; Hagberg, A.; Osthus, D.; Mohd-Yusof, J.

2017-12-01

Flow and transport of fluids through fractured systems is governed by the properties and interactions at the micro-scale. Retaining information about the micro-structure such as fracture length, orientation, aperture and connectivity in mesh-based computational models results in solving for millions to billions of degrees of freedom and quickly renders the problem computationally intractable. Our approach depicts fracture networks graphically, by mapping fractures to nodes and intersections to edges, thereby greatly reducing computational burden. Additionally, we use machine learning techniques to build simulators on the graph representation, trained on data from the mesh-based high fidelity simulations to speed up computation by orders of magnitude. We demonstrate our methodology on ensembles of discrete fracture networks, dividing up the data into training and validation sets. Our machine learned graph-based solvers result in over 3 orders of magnitude speedup without any significant sacrifice in accuracy.

Two algorithms for neural-network design and training with application to channel equalization.

PubMed

Sweatman, C Z; Mulgrew, B; Gibson, G J

1998-01-01

We describe two algorithms for designing and training neural-network classifiers. The first, the linear programming slab algorithm (LPSA), is motivated by the problem of reconstructing digital signals corrupted by passage through a dispersive channel and by additive noise. It constructs a multilayer perceptron (MLP) to separate two disjoint sets by using linear programming methods to identify network parameters. The second, the perceptron learning slab algorithm (PLSA), avoids the computational costs of linear programming by using an error-correction approach to identify parameters. Both algorithms operate in highly constrained parameter spaces and are able to exploit symmetry in the classification problem. Using these algorithms, we develop a number of procedures for the adaptive equalization of a complex linear 4-quadrature amplitude modulation (QAM) channel, and compare their performance in a simulation study. Results are given for both stationary and time-varying channels, the latter based on the COST 207 GSM propagation model.

Transcranial direct current stimulation facilitates cognitive multi-task performance differentially depending on anode location and subtask

PubMed Central

Scheldrup, Melissa; Greenwood, Pamela M.; McKendrick, Ryan; Strohl, Jon; Bikson, Marom; Alam, Mahtab; McKinley, R. Andy; Parasuraman, Raja

2014-01-01

There is a need to facilitate acquisition of real world cognitive multi-tasks that require long periods of training (e.g., air traffic control, intelligence analysis, medicine). Non-invasive brain stimulation—specifically transcranial Direct Current Stimulation (tDCS)—has promise as a method to speed multi-task training. We hypothesized that during acquisition of the complex multi-task Space Fortress, subtasks that require focused attention on ship control would benefit from tDCS aimed at the dorsal attention network while subtasks that require redirection of attention would benefit from tDCS aimed at the right hemisphere ventral attention network. We compared effects of 30 min prefrontal and parietal stimulation to right and left hemispheres on subtask performance during the first 45 min of training. The strongest effects both overall and for ship flying (control and velocity subtasks) were seen with a right parietal (C4, reference to left shoulder) montage, shown by modeling to induce an electric field that includes nodes in both dorsal and ventral attention networks. This is consistent with the re-orienting hypothesis that the ventral attention network is activated along with the dorsal attention network if a new, task-relevant event occurs while visuospatial attention is focused (Corbetta et al., 2008). No effects were seen with anodes over sites that stimulated only dorsal (C3) or only ventral (F10) attention networks. The speed subtask (update memory for symbols) benefited from an F9 anode over left prefrontal cortex. These results argue for development of tDCS as a training aid in real world settings where multi-tasking is critical. PMID:25249958

Transcranial direct current stimulation facilitates cognitive multi-task performance differentially depending on anode location and subtask.

PubMed

Scheldrup, Melissa; Greenwood, Pamela M; McKendrick, Ryan; Strohl, Jon; Bikson, Marom; Alam, Mahtab; McKinley, R Andy; Parasuraman, Raja

2014-01-01

There is a need to facilitate acquisition of real world cognitive multi-tasks that require long periods of training (e.g., air traffic control, intelligence analysis, medicine). Non-invasive brain stimulation-specifically transcranial Direct Current Stimulation (tDCS)-has promise as a method to speed multi-task training. We hypothesized that during acquisition of the complex multi-task Space Fortress, subtasks that require focused attention on ship control would benefit from tDCS aimed at the dorsal attention network while subtasks that require redirection of attention would benefit from tDCS aimed at the right hemisphere ventral attention network. We compared effects of 30 min prefrontal and parietal stimulation to right and left hemispheres on subtask performance during the first 45 min of training. The strongest effects both overall and for ship flying (control and velocity subtasks) were seen with a right parietal (C4, reference to left shoulder) montage, shown by modeling to induce an electric field that includes nodes in both dorsal and ventral attention networks. This is consistent with the re-orienting hypothesis that the ventral attention network is activated along with the dorsal attention network if a new, task-relevant event occurs while visuospatial attention is focused (Corbetta et al., 2008). No effects were seen with anodes over sites that stimulated only dorsal (C3) or only ventral (F10) attention networks. The speed subtask (update memory for symbols) benefited from an F9 anode over left prefrontal cortex. These results argue for development of tDCS as a training aid in real world settings where multi-tasking is critical.

MO-DE-BRA-04: The CREATE Medical Physics Research Training Network: Training of New Generation Innovators

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

Seuntjens, J; Collins, L; Devic, S

Purpose: Over the past century, physicists have played a major role in transforming scientific discovery into everyday clinical applications. However, with the increasingly stringent requirements to regulate medical physics as a health profession, the role of physicists as scientists and innovators has become at serious risk of erosion. These challenges trigger the need for a new, revolutionized training program at the graduate level that respects scientific rigor, attention for medical physics-relevant developments in basic sciences, innovation and entrepreneurship. Methods: A grant proposal was funded by the Collaborative REsearch and Training Experience program (CREATE) of the Natural Sciences and Engineering Researchmore » Council (NSERC) of Canada. This enabled the creation of the Medical Physics Research Training Network (MPRTN) around two CAMPEP-accredited medical physics programs. Members of the network consist of medical device companies, government (research and regulatory) and academia. The MPRTN/CREATE program proposes a curriculum with three main themes: (1) radiation physics, (2) imaging & image processing and (3) radiation response, outcomes and modeling. Results: The MPRTN was created mid 2013 (mprtn.com) and features (1) four new basic Ph.D. courses; (2) industry participation in research projects; (3) formal job-readiness training with involvement of guest faculty from academia, government and industry. MPRTN activities since 2013 include 22 conferences; 7 workshops and 4 exchange travels. Three patents were filed or issued, nine awards/best papers were won. Fifteen journal publications were accepted/published, 102 conference abstracts. There are now 13 industry partners. Conclusion: A medical physics research training network has been set up with the goal to harness graduate student’s job-readiness for industry, government and academia in addition to the conventional clinical role. Two years after inception, significant successes have been booked, but the true challenge will be to demonstrate that with this training philosophy CREATE scholars gain access to a much broader job market. Supported by the Natural Sciences and Engineering Research Council (NSERC) Canada.« less

Empirical study on neural network based predictive techniques for automatic number plate recognition

NASA Astrophysics Data System (ADS)

Shashidhara, M. S.; Indrakumar, S. S.

2011-10-01

The objective of this study is to provide an easy, accurate and effective technology for the Bangalore city traffic control. This is based on the techniques of image processing and laser beam technology. The core concept chosen here is an image processing technology by the method of automatic number plate recognition system. First number plate is recognized if any vehicle breaks the traffic rules in the signals. The number is fetched from the database of the RTO office by the process of automatic database fetching. Next this sends the notice and penalty related information to the vehicle owner email-id and an SMS sent to vehicle owner. In this paper, we use of cameras with zooming options & laser beams to get accurate pictures further applied image processing techniques such as Edge detection to understand the vehicle, Identifying the location of the number plate, Identifying the number plate for further use, Plain plate number, Number plate with additional information, Number plates in the different fonts. Accessing the database of the vehicle registration office to identify the name and address and other information of the vehicle number. The updates to be made to the database for the recording of the violation and penalty issues. A feed forward artificial neural network is used for OCR. This procedure is particularly important for glyphs that are visually similar such as '8' and '9' and results in training sets of between 25,000 and 40,000 training samples. Over training of the neural network is prevented by Bayesian regularization. The neural network output value is set to 0.05 when the input is not desired glyph, and 0.95 for correct input.

NetMHC-3.0: accurate web accessible predictions of human, mouse and monkey MHC class I affinities for peptides of length 8-11.

PubMed

Lundegaard, Claus; Lamberth, Kasper; Harndahl, Mikkel; Buus, Søren; Lund, Ole; Nielsen, Morten

2008-07-01

NetMHC-3.0 is trained on a large number of quantitative peptide data using both affinity data from the Immune Epitope Database and Analysis Resource (IEDB) and elution data from SYFPEITHI. The method generates high-accuracy predictions of major histocompatibility complex (MHC): peptide binding. The predictions are based on artificial neural networks trained on data from 55 MHC alleles (43 Human and 12 non-human), and position-specific scoring matrices (PSSMs) for additional 67 HLA alleles. As only the MHC class I prediction server is available, predictions are possible for peptides of length 8-11 for all 122 alleles. artificial neural network predictions are given as actual IC(50) values whereas PSSM predictions are given as a log-odds likelihood scores. The output is optionally available as download for easy post-processing. The training method underlying the server is the best available, and has been used to predict possible MHC-binding peptides in a series of pathogen viral proteomes including SARS, Influenza and HIV, resulting in an average of 75-80% confirmed MHC binders. Here, the performance is further validated and benchmarked using a large set of newly published affinity data, non-redundant to the training set. The server is free of use and available at: http://www.cbs.dtu.dk/services/NetMHC.

An application of deep learning in the analysis of stellar spectra

NASA Astrophysics Data System (ADS)

Fabbro, S.; Venn, K. A.; O'Briain, T.; Bialek, S.; Kielty, C. L.; Jahandar, F.; Monty, S.

2018-04-01

Spectroscopic surveys require fast and efficient analysis methods to maximize their scientific impact. Here, we apply a deep neural network architecture to analyse both SDSS-III APOGEE DR13 and synthetic stellar spectra. When our convolutional neural network model (StarNet) is trained on APOGEE spectra, we show that the stellar parameters (temperature, gravity, and metallicity) are determined with similar precision and accuracy as the APOGEE pipeline. StarNet can also predict stellar parameters when trained on synthetic data, with excellent precision and accuracy for both APOGEE data and synthetic data, over a wide range of signal-to-noise ratios. In addition, the statistical uncertainties in the stellar parameter determinations are comparable to the differences between the APOGEE pipeline results and those determined independently from optical spectra. We compare StarNet to other data-driven methods; for example, StarNet and the Cannon 2 show similar behaviour when trained with the same data sets; however, StarNet performs poorly on small training sets like those used by the original Cannon. The influence of the spectral features on the stellar parameters is examined via partial derivatives of the StarNet model results with respect to the input spectra. While StarNet was developed using the APOGEE observed spectra and corresponding ASSET synthetic data, we suggest that this technique is applicable to other wavelength ranges and other spectral surveys.

[Study on artificial neural network combined with multispectral remote sensing imagery for forest site evaluation].

PubMed

Gong, Yin-Xi; He, Cheng; Yan, Fei; Feng, Zhong-Ke; Cao, Meng-Lei; Gao, Yuan; Miao, Jie; Zhao, Jin-Long

2013-10-01

Multispectral remote sensing data containing rich site information are not fully used by the classic site quality evaluation system, as it merely adopts artificial ground survey data. In order to establish a more effective site quality evaluation system, a neural network model which combined remote sensing spectra factors with site factors and site index relations was established and used to study the sublot site quality evaluation in the Wangyedian Forest Farm in Inner Mongolia Province, Chifeng City. Based on the improved back propagation artificial neural network (BPANN), this model combined multispectral remote sensing data with sublot survey data, and took larch as example, Through training data set sensitivity analysis weak or irrelevant factor was excluded, the size of neural network was simplified, and the efficiency of network training was improved. This optimal site index prediction model had an accuracy up to 95.36%, which was 9.83% higher than that of the neural network model based on classic sublot survey data, and this shows that using multi-spectral remote sensing and small class survey data to determine the status of larch index prediction model has the highest predictive accuracy. The results fully indicate the effectiveness and superiority of this method.

Training Excitatory-Inhibitory Recurrent Neural Networks for Cognitive Tasks: A Simple and Flexible Framework

PubMed Central

Wang, Xiao-Jing

2016-01-01

The ability to simultaneously record from large numbers of neurons in behaving animals has ushered in a new era for the study of the neural circuit mechanisms underlying cognitive functions. One promising approach to uncovering the dynamical and computational principles governing population responses is to analyze model recurrent neural networks (RNNs) that have been optimized to perform the same tasks as behaving animals. Because the optimization of network parameters specifies the desired output but not the manner in which to achieve this output, “trained” networks serve as a source of mechanistic hypotheses and a testing ground for data analyses that link neural computation to behavior. Complete access to the activity and connectivity of the circuit, and the ability to manipulate them arbitrarily, make trained networks a convenient proxy for biological circuits and a valuable platform for theoretical investigation. However, existing RNNs lack basic biological features such as the distinction between excitatory and inhibitory units (Dale’s principle), which are essential if RNNs are to provide insights into the operation of biological circuits. Moreover, trained networks can achieve the same behavioral performance but differ substantially in their structure and dynamics, highlighting the need for a simple and flexible framework for the exploratory training of RNNs. Here, we describe a framework for gradient descent-based training of excitatory-inhibitory RNNs that can incorporate a variety of biological knowledge. We provide an implementation based on the machine learning library Theano, whose automatic differentiation capabilities facilitate modifications and extensions. We validate this framework by applying it to well-known experimental paradigms such as perceptual decision-making, context-dependent integration, multisensory integration, parametric working memory, and motor sequence generation. Our results demonstrate the wide range of neural activity patterns and behavior that can be modeled, and suggest a unified setting in which diverse cognitive computations and mechanisms can be studied. PMID:26928718

Finding strong lenses in CFHTLS using convolutional neural networks

NASA Astrophysics Data System (ADS)

Jacobs, C.; Glazebrook, K.; Collett, T.; More, A.; McCarthy, C.

2017-10-01

We train and apply convolutional neural networks, a machine learning technique developed to learn from and classify image data, to Canada-France-Hawaii Telescope Legacy Survey (CFHTLS) imaging for the identification of potential strong lensing systems. An ensemble of four convolutional neural networks was trained on images of simulated galaxy-galaxy lenses. The training sets consisted of a total of 62 406 simulated lenses and 64 673 non-lens negative examples generated with two different methodologies. An ensemble of trained networks was applied to all of the 171 deg2 of the CFHTLS wide field image data, identifying 18 861 candidates including 63 known and 139 other potential lens candidates. A second search of 1.4 million early-type galaxies selected from the survey catalogue as potential deflectors, identified 2465 candidates including 117 previously known lens candidates, 29 confirmed lenses/high-quality lens candidates, 266 novel probable or potential lenses and 2097 candidates we classify as false positives. For the catalogue-based search we estimate a completeness of 21-28 per cent with respect to detectable lenses and a purity of 15 per cent, with a false-positive rate of 1 in 671 images tested. We predict a human astronomer reviewing candidates produced by the system would identify 20 probable lenses and 100 possible lenses per hour in a sample selected by the robot. Convolutional neural networks are therefore a promising tool for use in the search for lenses in current and forthcoming surveys such as the Dark Energy Survey and the Large Synoptic Survey Telescope.

Application of complex discrete wavelet transform in classification of Doppler signals using complex-valued artificial neural network.

PubMed

Ceylan, Murat; Ceylan, Rahime; Ozbay, Yüksel; Kara, Sadik

2008-09-01

In biomedical signal classification, due to the huge amount of data, to compress the biomedical waveform data is vital. This paper presents two different structures formed using feature extraction algorithms to decrease size of feature set in training and test data. The proposed structures, named as wavelet transform-complex-valued artificial neural network (WT-CVANN) and complex wavelet transform-complex-valued artificial neural network (CWT-CVANN), use real and complex discrete wavelet transform for feature extraction. The aim of using wavelet transform is to compress data and to reduce training time of network without decreasing accuracy rate. In this study, the presented structures were applied to the problem of classification in carotid arterial Doppler ultrasound signals. Carotid arterial Doppler ultrasound signals were acquired from left carotid arteries of 38 patients and 40 healthy volunteers. The patient group included 22 males and 16 females with an established diagnosis of the early phase of atherosclerosis through coronary or aortofemoropopliteal (lower extremity) angiographies (mean age, 59 years; range, 48-72 years). Healthy volunteers were young non-smokers who seem to not bear any risk of atherosclerosis, including 28 males and 12 females (mean age, 23 years; range, 19-27 years). Sensitivity, specificity and average detection rate were calculated for comparison, after training and test phases of all structures finished. These parameters have demonstrated that training times of CVANN and real-valued artificial neural network (RVANN) were reduced using feature extraction algorithms without decreasing accuracy rate in accordance to our aim.

Artificial Neural Networks as an Architectural Design Tool-Generating New Detail Forms Based On the Roman Corinthian Order Capital

NASA Astrophysics Data System (ADS)

Radziszewski, Kacper

2017-10-01

The following paper presents the results of the research in the field of the machine learning, investigating the scope of application of the artificial neural networks algorithms as a tool in architectural design. The computational experiment was held using the backward propagation of errors method of training the artificial neural network, which was trained based on the geometry of the details of the Roman Corinthian order capital. During the experiment, as an input training data set, five local geometry parameters combined has given the best results: Theta, Pi, Rho in spherical coordinate system based on the capital volume centroid, followed by Z value of the Cartesian coordinate system and a distance from vertical planes created based on the capital symmetry. Additionally during the experiment, artificial neural network hidden layers optimal count and structure was found, giving results of the error below 0.2% for the mentioned before input parameters. Once successfully trained artificial network, was able to mimic the details composition on any other geometry type given. Despite of calculating the transformed geometry locally and separately for each of the thousands of surface points, system could create visually attractive and diverse, complex patterns. Designed tool, based on the supervised learning method of machine learning, gives possibility of generating new architectural forms- free of the designer’s imagination bounds. Implementing the infinitely broad computational methods of machine learning, or Artificial Intelligence in general, not only could accelerate and simplify the design process, but give an opportunity to explore never seen before, unpredictable forms or everyday architectural practice solutions.

ANI-1: an extensible neural network potential with DFT accuracy at force field computational cost† †Electronic supplementary information (ESI) available. See DOI: 10.1039/c6sc05720a Click here for additional data file.

PubMed Central

Smith, J. S.

2017-01-01

Deep learning is revolutionizing many areas of science and technology, especially image, text, and speech recognition. In this paper, we demonstrate how a deep neural network (NN) trained on quantum mechanical (QM) DFT calculations can learn an accurate and transferable potential for organic molecules. We introduce ANAKIN-ME (Accurate NeurAl networK engINe for Molecular Energies) or ANI for short. ANI is a new method designed with the intent of developing transferable neural network potentials that utilize a highly-modified version of the Behler and Parrinello symmetry functions to build single-atom atomic environment vectors (AEV) as a molecular representation. AEVs provide the ability to train neural networks to data that spans both configurational and conformational space, a feat not previously accomplished on this scale. We utilized ANI to build a potential called ANI-1, which was trained on a subset of the GDB databases with up to 8 heavy atoms in order to predict total energies for organic molecules containing four atom types: H, C, N, and O. To obtain an accelerated but physically relevant sampling of molecular potential surfaces, we also proposed a Normal Mode Sampling (NMS) method for generating molecular conformations. Through a series of case studies, we show that ANI-1 is chemically accurate compared to reference DFT calculations on much larger molecular systems (up to 54 atoms) than those included in the training data set. PMID:28507695

Multi-level deep supervised networks for retinal vessel segmentation.

PubMed

Mo, Juan; Zhang, Lei

2017-12-01

Changes in the appearance of retinal blood vessels are an important indicator for various ophthalmologic and cardiovascular diseases, including diabetes, hypertension, arteriosclerosis, and choroidal neovascularization. Vessel segmentation from retinal images is very challenging because of low blood vessel contrast, intricate vessel topology, and the presence of pathologies such as microaneurysms and hemorrhages. To overcome these challenges, we propose a neural network-based method for vessel segmentation. A deep supervised fully convolutional network is developed by leveraging multi-level hierarchical features of the deep networks. To improve the discriminative capability of features in lower layers of the deep network and guide the gradient back propagation to overcome gradient vanishing, deep supervision with auxiliary classifiers is incorporated in some intermediate layers of the network. Moreover, the transferred knowledge learned from other domains is used to alleviate the issue of insufficient medical training data. The proposed approach does not rely on hand-crafted features and needs no problem-specific preprocessing or postprocessing, which reduces the impact of subjective factors. We evaluate the proposed method on three publicly available databases, the DRIVE, STARE, and CHASE_DB1 databases. Extensive experiments demonstrate that our approach achieves better or comparable performance to state-of-the-art methods with a much faster processing speed, making it suitable for real-world clinical applications. The results of cross-training experiments demonstrate its robustness with respect to the training set. The proposed approach segments retinal vessels accurately with a much faster processing speed and can be easily applied to other biomedical segmentation tasks.

Deep Recurrent Neural Networks for Supernovae Classification

NASA Astrophysics Data System (ADS)

Charnock, Tom; Moss, Adam

2017-03-01

We apply deep recurrent neural networks, which are capable of learning complex sequential information, to classify supernovae (code available at https://github.com/adammoss/supernovae). The observational time and filter fluxes are used as inputs to the network, but since the inputs are agnostic, additional data such as host galaxy information can also be included. Using the Supernovae Photometric Classification Challenge (SPCC) data, we find that deep networks are capable of learning about light curves, however the performance of the network is highly sensitive to the amount of training data. For a training size of 50% of the representational SPCC data set (around 104 supernovae) we obtain a type-Ia versus non-type-Ia classification accuracy of 94.7%, an area under the Receiver Operating Characteristic curve AUC of 0.986 and an SPCC figure-of-merit F 1 = 0.64. When using only the data for the early-epoch challenge defined by the SPCC, we achieve a classification accuracy of 93.1%, AUC of 0.977, and F 1 = 0.58, results almost as good as with the whole light curve. By employing bidirectional neural networks, we can acquire impressive classification results between supernovae types I, II and III at an accuracy of 90.4% and AUC of 0.974. We also apply a pre-trained model to obtain classification probabilities as a function of time and show that it can give early indications of supernovae type. Our method is competitive with existing algorithms and has applications for future large-scale photometric surveys.

Computer-aided detection of bladder masses in CT urography (CTU)

NASA Astrophysics Data System (ADS)

Cha, Kenny H.; Hadjiiski, Lubomir M.; Chan, Heang-Ping; Caoili, Elaine M.; Cohan, Richard H.; Weizer, Alon; Samala, Ravi K.

2017-03-01

We are developing a computer-aided detection system for bladder cancer in CT urography (CTU). We have previously developed methods for detection of bladder masses within the contrast-enhanced and the non-contrastenhanced regions of the bladder individually. In this study, we investigated methods for detection of bladder masses within the entire bladder. The bladder was segmented using our method that combined deep-learning convolutional neural network with level sets. The non-contrast-enhanced region was separated from the contrast-enhanced region with a maximum-intensity-projection-based method. The non-contrast region was smoothed and gray level threshold was applied to the contrast and non-contrast regions separately to extract the bladder wall and potential masses. The bladder wall was transformed into a straightened thickness profile, which was analyzed to identify lesion candidates in a prescreening step. The candidates were mapped back to the 3D CT volume and segmented using our auto-initialized cascaded level set (AI-CALS) segmentation method. Twenty-seven morphological features were extracted for each candidate. A data set of 57 patients with 71 biopsy-proven bladder lesions was used, which was split into independent training and test sets: 42 training cases with 52 lesions, and 15 test cases with 19 lesions. Using the training set, feature selection was performed and a linear discriminant (LDA) classifier was designed to merge the selected features for classification of bladder lesions and false positives. The trained classifier was evaluated with the test set. FROC analysis showed that the system achieved a sensitivity of 86.5% at 3.3 FPs/case for the training set, and 84.2% at 3.7 FPs/case for the test set.

Application of an Artificial Neural Network to the Prediction of OH Radical Reaction Rate Constants for Evaluating Global Warming Potential.

PubMed

Allison, Thomas C

2016-03-03

Rate constants for reactions of chemical compounds with hydroxyl radical are a key quantity used in evaluating the global warming potential of a substance. Experimental determination of these rate constants is essential, but it can also be difficult and time-consuming to produce. High-level quantum chemistry predictions of the rate constant can suffer from the same issues. Therefore, it is valuable to devise estimation schemes that can give reasonable results on a variety of chemical compounds. In this article, the construction and training of an artificial neural network (ANN) for the prediction of rate constants at 298 K for reactions of hydroxyl radical with a diverse set of molecules is described. Input to the ANN consists of counts of the chemical bonds and bends present in the target molecule. The ANN is trained using 792 (•)OH reaction rate constants taken from the NIST Chemical Kinetics Database. The mean unsigned percent error (MUPE) for the training set is 12%, and the MUPE of the testing set is 51%. It is shown that the present methodology yields rate constants of reasonable accuracy for a diverse set of inputs. The results are compared to high-quality literature values and to another estimation scheme. This ANN methodology is expected to be of use in a wide range of applications for which (•)OH reaction rate constants are required. The model uses only information that can be gathered from a 2D representation of the molecule, making the present approach particularly appealing, especially for screening applications.

National Crime Information Center (NCIC) Training Videos.

ERIC Educational Resources Information Center

Federal Bureau of Investigation, Washington, DC. National Crime Information Center.

The Federal Bureau of Investigation's National Crime Information Center (NCIC) maintains a set of computerized files of documented criminal justice information reported by a network of over 60,000 participating national, regional, state, and local agencies. The files, dealing with wanted persons, missing persons, unidentified persons, and stolen…

Application of physicochemical properties and process parameters in the development of a neural network model for prediction of tablet characteristics.

PubMed

Sovány, Tamás; Papós, Kitti; Kása, Péter; Ilič, Ilija; Srčič, Stane; Pintye-Hódi, Klára

2013-06-01

The importance of in silico modeling in the pharmaceutical industry is continuously increasing. The aim of the present study was the development of a neural network model for prediction of the postcompressional properties of scored tablets based on the application of existing data sets from our previous studies. Some important process parameters and physicochemical characteristics of the powder mixtures were used as training factors to achieve the best applicability in a wide range of possible compositions. The results demonstrated that, after some pre-processing of the factors, an appropriate prediction performance could be achieved. However, because of the poor extrapolation capacity, broadening of the training data range appears necessary.

Forecasting the daily electricity consumption in the Moscow region using artificial neural networks

NASA Astrophysics Data System (ADS)

Ivanov, V. V.; Kryanev, A. V.; Osetrov, E. S.

2017-07-01

In [1] we demonstrated the possibility in principle for short-term forecasting of daily volumes of passenger traffic in the Moscow metro with the help of artificial neural networks. During training and predicting, a set of the factors that affect the daily passenger traffic in the subway is passed to the input of the neural network. One of these factors is the daily power consumption in the Moscow region. Therefore, to predict the volume of the passenger traffic in the subway, we must first to solve the problem of forecasting the daily energy consumption in the Moscow region.

Signal recognition efficiencies of artificial neural-network pulse-shape discrimination in HPGe -decay searches

NASA Astrophysics Data System (ADS)

Caldwell, A.; Cossavella, F.; Majorovits, B.; Palioselitis, D.; Volynets, O.

2015-07-01

A pulse-shape discrimination method based on artificial neural networks was applied to pulses simulated for different background, signal and signal-like interactions inside a germanium detector. The simulated pulses were used to investigate variations of efficiencies as a function of used training set. It is verified that neural networks are well-suited to identify background pulses in true-coaxial high-purity germanium detectors. The systematic uncertainty on the signal recognition efficiency derived using signal-like evaluation samples from calibration measurements is estimated to be 5 %. This uncertainty is due to differences between signal and calibration samples.

A fuzzy neural network for intelligent data processing

NASA Astrophysics Data System (ADS)

Xie, Wei; Chu, Feng; Wang, Lipo; Lim, Eng Thiam

2005-03-01

In this paper, we describe an incrementally generated fuzzy neural network (FNN) for intelligent data processing. This FNN combines the features of initial fuzzy model self-generation, fast input selection, partition validation, parameter optimization and rule-base simplification. A small FNN is created from scratch -- there is no need to specify the initial network architecture, initial membership functions, or initial weights. Fuzzy IF-THEN rules are constantly combined and pruned to minimize the size of the network while maintaining accuracy; irrelevant inputs are detected and deleted, and membership functions and network weights are trained with a gradient descent algorithm, i.e., error backpropagation. Experimental studies on synthesized data sets demonstrate that the proposed Fuzzy Neural Network is able to achieve accuracy comparable to or higher than both a feedforward crisp neural network, i.e., NeuroRule, and a decision tree, i.e., C4.5, with more compact rule bases for most of the data sets used in our experiments. The FNN has achieved outstanding results for cancer classification based on microarray data. The excellent classification result for Small Round Blue Cell Tumors (SRBCTs) data set is shown. Compared with other published methods, we have used a much fewer number of genes for perfect classification, which will help researchers directly focus their attention on some specific genes and may lead to discovery of deep reasons of the development of cancers and discovery of drugs.

Strengthening the American Academy of Pediatric Dentistry's Public Policy Advocate Network: Identifying Advocacy Efforts and Recommendations.

PubMed

Curtis, Benjamin D; Orynich, C Ashley; Casamassimo, Paul S; Seale, N Sue; Reggiardo, Paul; Wright, Robin; Litch, C Scott

2017-09-15

The purposes of this study were to collect information on involvement, training, and barriers to participation in advocacy efforts for Public Policy Advocates (PPAs) of the American Academy of Pediatric Dentistry (AAPD) and make recommendations to the AAPD. Preliminary data were collected from the PPAs during structured AAPD program meetings, conference calls, and individual interviews. Based on these data, a survey was created, piloted, and sent electronically to all PPAs. Data were analyzed and collated by frequencies. Responses from 38 PPAs (100 percent) revealed they were involved with state legislatures and state chapters of the AAPD and American Dental Association. Eighty-two percent of the PPAs requested additional public policy training and clearer communication channels within the network. PPAs are funding their own advocacy efforts, and the time and resources spent away from patient care is a financial barrier. The Public Policy Advocate network holds a broad policy skill set and voluntarily commits time and resource to advocate for the support of the pediatric dental patient at state and federal government levels. The American Academy of Pediatric Dentistry can strengthen the PPA's self-directed leadership role at state and federal levels through formalized training, restructuring of the network, and increased resources.

Beyond Retinal Layers: A Deep Voting Model for Automated Geographic Atrophy Segmentation in SD-OCT Images

PubMed Central

Ji, Zexuan; Chen, Qiang; Niu, Sijie; Leng, Theodore; Rubin, Daniel L.

2018-01-01

Purpose To automatically and accurately segment geographic atrophy (GA) in spectral-domain optical coherence tomography (SD-OCT) images by constructing a voting system with deep neural networks without the use of retinal layer segmentation. Methods An automatic GA segmentation method for SD-OCT images based on the deep network was constructed. The structure of the deep network was composed of five layers, including one input layer, three hidden layers, and one output layer. During the training phase, the labeled A-scans with 1024 features were directly fed into the network as the input layer to obtain the deep representations. Then a soft-max classifier was trained to determine the label of each individual pixel. Finally, a voting decision strategy was used to refine the segmentation results among 10 trained models. Results Two image data sets with GA were used to evaluate the model. For the first dataset, our algorithm obtained a mean overlap ratio (OR) 86.94% ± 8.75%, absolute area difference (AAD) 11.49% ± 11.50%, and correlation coefficients (CC) 0.9857; for the second dataset, the mean OR, AAD, and CC of the proposed method were 81.66% ± 10.93%, 8.30% ± 9.09%, and 0.9952, respectively. The proposed algorithm was capable of improving over 5% and 10% segmentation accuracy, respectively, when compared with several state-of-the-art algorithms on two data sets. Conclusions Without retinal layer segmentation, the proposed algorithm could produce higher segmentation accuracy and was more stable when compared with state-of-the-art methods that relied on retinal layer segmentation results. Our model may provide reliable GA segmentations from SD-OCT images and be useful in the clinical diagnosis of advanced nonexudative AMD. Translational Relevance Based on the deep neural networks, this study presents an accurate GA segmentation method for SD-OCT images without using any retinal layer segmentation results, and may contribute to improved understanding of advanced nonexudative AMD. PMID:29302382

Beyond Retinal Layers: A Deep Voting Model for Automated Geographic Atrophy Segmentation in SD-OCT Images.

PubMed

Ji, Zexuan; Chen, Qiang; Niu, Sijie; Leng, Theodore; Rubin, Daniel L

2018-01-01

To automatically and accurately segment geographic atrophy (GA) in spectral-domain optical coherence tomography (SD-OCT) images by constructing a voting system with deep neural networks without the use of retinal layer segmentation. An automatic GA segmentation method for SD-OCT images based on the deep network was constructed. The structure of the deep network was composed of five layers, including one input layer, three hidden layers, and one output layer. During the training phase, the labeled A-scans with 1024 features were directly fed into the network as the input layer to obtain the deep representations. Then a soft-max classifier was trained to determine the label of each individual pixel. Finally, a voting decision strategy was used to refine the segmentation results among 10 trained models. Two image data sets with GA were used to evaluate the model. For the first dataset, our algorithm obtained a mean overlap ratio (OR) 86.94% ± 8.75%, absolute area difference (AAD) 11.49% ± 11.50%, and correlation coefficients (CC) 0.9857; for the second dataset, the mean OR, AAD, and CC of the proposed method were 81.66% ± 10.93%, 8.30% ± 9.09%, and 0.9952, respectively. The proposed algorithm was capable of improving over 5% and 10% segmentation accuracy, respectively, when compared with several state-of-the-art algorithms on two data sets. Without retinal layer segmentation, the proposed algorithm could produce higher segmentation accuracy and was more stable when compared with state-of-the-art methods that relied on retinal layer segmentation results. Our model may provide reliable GA segmentations from SD-OCT images and be useful in the clinical diagnosis of advanced nonexudative AMD. Based on the deep neural networks, this study presents an accurate GA segmentation method for SD-OCT images without using any retinal layer segmentation results, and may contribute to improved understanding of advanced nonexudative AMD.

A Comprehensive Training Data Set for the Development of Satellite-Based Volcanic Ash Detection Algorithms

NASA Astrophysics Data System (ADS)

Schmidl, Marius

2017-04-01

We present a comprehensive training data set covering a large range of atmospheric conditions, including disperse volcanic ash and desert dust layers. These data sets contain all information required for the development of volcanic ash detection algorithms based on artificial neural networks, urgently needed since volcanic ash in the airspace is a major concern of aviation safety authorities. Selected parts of the data are used to train the volcanic ash detection algorithm VADUGS. They contain atmospheric and surface-related quantities as well as the corresponding simulated satellite data for the channels in the infrared spectral range of the SEVIRI instrument on board MSG-2. To get realistic results, ECMWF, IASI-based, and GEOS-Chem data are used to calculate all parameters describing the environment, whereas the software package libRadtran is used to perform radiative transfer simulations returning the brightness temperatures for each atmospheric state. As optical properties are a prerequisite for radiative simulations accounting for aerosol layers, the development also included the computation of optical properties for a set of different aerosol types from different sources. A description of the developed software and the used methods is given, besides an overview of the resulting data sets.

Research on Daily Objects Detection Based on Deep Neural Network

NASA Astrophysics Data System (ADS)

Ding, Sheng; Zhao, Kun

2018-03-01

With the rapid development of deep learning, great breakthroughs have been made in the field of object detection. In this article, the deep learning algorithm is applied to the detection of daily objects, and some progress has been made in this direction. Compared with traditional object detection methods, the daily objects detection method based on deep learning is faster and more accurate. The main research work of this article: 1. collect a small data set of daily objects; 2. in the TensorFlow framework to build different models of object detection, and use this data set training model; 3. the training process and effect of the model are improved by fine-tuning the model parameters.

Automatic classification of DMSA scans using an artificial neural network

NASA Astrophysics Data System (ADS)

Wright, J. W.; Duguid, R.; Mckiddie, F.; Staff, R. T.

2014-04-01

DMSA imaging is carried out in nuclear medicine to assess the level of functional renal tissue in patients. This study investigated the use of an artificial neural network to perform diagnostic classification of these scans. Using the radiological report as the gold standard, the network was trained to classify DMSA scans as positive or negative for defects using a representative sample of 257 previously reported images. The trained network was then independently tested using a further 193 scans and achieved a binary classification accuracy of 95.9%. The performance of the network was compared with three qualified expert observers who were asked to grade each scan in the 193 image testing set on a six point defect scale, from ‘definitely normal’ to ‘definitely abnormal’. A receiver operating characteristic analysis comparison between a consensus operator, generated from the scores of the three expert observers, and the network revealed a statistically significant increase (α < 0.05) in performance between the network and operators. A further result from this work was that when suitably optimized, a negative predictive value of 100% for renal defects was achieved by the network, while still managing to identify 93% of the negative cases in the dataset. These results are encouraging for application of such a network as a screening tool or quality assurance assistant in clinical practice.

PDA usage and training: targeting curriculum for residents and faculty.

PubMed

Morris, Carl G; Church, Lili; Vincent, Chris; Rao, Ashwin

2007-06-01

Utilization of personal digital assistants (PDAs) in residency education is common, but information about their use and how residents are trained to use them is limited. Better understanding of resident and faculty PDA use and training is needed. We used a cross-sectional survey of 598 residents and faculty from the WWAMI (Washington, Wyoming, Alaska, Montana, and Idaho) Family Medicine Residency Network regarding PDA usage and training. Use of PDAs is common among residents (94%) and faculty (79%). Ninety-six percent of faculty and residents report stable or increasing frequency of use over time. The common barriers to PDA use relate to lack of time, knowledge, and formal education. Approximately half of PDA users (52%) have received some formal training; however, the majority of users report being self-taught. Faculty and residents prefer either small-group or one-on-one settings with hands-on, self-directed, interactive formats for PDA training. Large-group settings in lecture, written, or computer program formats were considered less helpful or desirable. PDAs have become a commonly used clinical tool. Lack of time and adequate training present a barrier to optimal application of PDAs in family medicine residency education.

Event Recognition Based on Deep Learning in Chinese Texts

PubMed Central

Zhang, Yajun; Liu, Zongtian; Zhou, Wen

2016-01-01

Event recognition is the most fundamental and critical task in event-based natural language processing systems. Existing event recognition methods based on rules and shallow neural networks have certain limitations. For example, extracting features using methods based on rules is difficult; methods based on shallow neural networks converge too quickly to a local minimum, resulting in low recognition precision. To address these problems, we propose the Chinese emergency event recognition model based on deep learning (CEERM). Firstly, we use a word segmentation system to segment sentences. According to event elements labeled in the CEC 2.0 corpus, we classify words into five categories: trigger words, participants, objects, time and location. Each word is vectorized according to the following six feature layers: part of speech, dependency grammar, length, location, distance between trigger word and core word and trigger word frequency. We obtain deep semantic features of words by training a feature vector set using a deep belief network (DBN), then analyze those features in order to identify trigger words by means of a back propagation neural network. Extensive testing shows that the CEERM achieves excellent recognition performance, with a maximum F-measure value of 85.17%. Moreover, we propose the dynamic-supervised DBN, which adds supervised fine-tuning to a restricted Boltzmann machine layer by monitoring its training performance. Test analysis reveals that the new DBN improves recognition performance and effectively controls the training time. Although the F-measure increases to 88.11%, the training time increases by only 25.35%. PMID:27501231

Neural Network and Regression Approximations in High Speed Civil Transport Aircraft Design Optimization

NASA Technical Reports Server (NTRS)

Patniak, Surya N.; Guptill, James D.; Hopkins, Dale A.; Lavelle, Thomas M.

1998-01-01

Nonlinear mathematical-programming-based design optimization can be an elegant method. However, the calculations required to generate the merit function, constraints, and their gradients, which are frequently required, can make the process computational intensive. The computational burden can be greatly reduced by using approximating analyzers derived from an original analyzer utilizing neural networks and linear regression methods. The experience gained from using both of these approximation methods in the design optimization of a high speed civil transport aircraft is the subject of this paper. The Langley Research Center's Flight Optimization System was selected for the aircraft analysis. This software was exercised to generate a set of training data with which a neural network and a regression method were trained, thereby producing the two approximating analyzers. The derived analyzers were coupled to the Lewis Research Center's CometBoards test bed to provide the optimization capability. With the combined software, both approximation methods were examined for use in aircraft design optimization, and both performed satisfactorily. The CPU time for solution of the problem, which had been measured in hours, was reduced to minutes with the neural network approximation and to seconds with the regression method. Instability encountered in the aircraft analysis software at certain design points was also eliminated. On the other hand, there were costs and difficulties associated with training the approximating analyzers. The CPU time required to generate the input-output pairs and to train the approximating analyzers was seven times that required for solution of the problem.

Event Recognition Based on Deep Learning in Chinese Texts.

PubMed

Zhang, Yajun; Liu, Zongtian; Zhou, Wen

2016-01-01

Event recognition is the most fundamental and critical task in event-based natural language processing systems. Existing event recognition methods based on rules and shallow neural networks have certain limitations. For example, extracting features using methods based on rules is difficult; methods based on shallow neural networks converge too quickly to a local minimum, resulting in low recognition precision. To address these problems, we propose the Chinese emergency event recognition model based on deep learning (CEERM). Firstly, we use a word segmentation system to segment sentences. According to event elements labeled in the CEC 2.0 corpus, we classify words into five categories: trigger words, participants, objects, time and location. Each word is vectorized according to the following six feature layers: part of speech, dependency grammar, length, location, distance between trigger word and core word and trigger word frequency. We obtain deep semantic features of words by training a feature vector set using a deep belief network (DBN), then analyze those features in order to identify trigger words by means of a back propagation neural network. Extensive testing shows that the CEERM achieves excellent recognition performance, with a maximum F-measure value of 85.17%. Moreover, we propose the dynamic-supervised DBN, which adds supervised fine-tuning to a restricted Boltzmann machine layer by monitoring its training performance. Test analysis reveals that the new DBN improves recognition performance and effectively controls the training time. Although the F-measure increases to 88.11%, the training time increases by only 25.35%.

Editor's highlight: Evaluation of a Microelectrode Array-based Assay for Neural Network Ontogeny using Training Set Chemicals

EPA Science Inventory

Thousands of compounds in the environment have not been characterized for developmental neurotoxicity (DNT) hazard. To address this issue, methods to screen compounds rapidly for DNT hazard evaluation are necessary and are being developed for key neurodevelopmental processes. In...

Extension of mixture-of-experts networks for binary classification of hierarchical data.

PubMed

Ng, Shu-Kay; McLachlan, Geoffrey J

2007-09-01

For many applied problems in the context of medically relevant artificial intelligence, the data collected exhibit a hierarchical or clustered structure. Ignoring the interdependence between hierarchical data can result in misleading classification. In this paper, we extend the mechanism for mixture-of-experts (ME) networks for binary classification of hierarchical data. Another extension is to quantify cluster-specific information on data hierarchy by random effects via the generalized linear mixed-effects model (GLMM). The extension of ME networks is implemented by allowing for correlation in the hierarchical data in both the gating and expert networks via the GLMM. The proposed model is illustrated using a real thyroid disease data set. In our study, we consider 7652 thyroid diagnosis records from 1984 to early 1987 with complete information on 20 attribute values. We obtain 10 independent random splits of the data into a training set and a test set in the proportions 85% and 15%. The test sets are used to assess the generalization performance of the proposed model, based on the percentage of misclassifications. For comparison, the results obtained from the ME network with independence assumption are also included. With the thyroid disease data, the misclassification rate on test sets for the extended ME network is 8.9%, compared to 13.9% for the ME network. In addition, based on model selection methods described in Section 2, a network with two experts is selected. These two expert networks can be considered as modeling two groups of patients with high and low incidence rates. Significant variation among the predicted cluster-specific random effects is detected in the patient group with low incidence rate. It is shown that the extended ME network outperforms the ME network for binary classification of hierarchical data. With the thyroid disease data, useful information on the relative log odds of patients with diagnosed conditions at different periods can be evaluated. This information can be taken into consideration for the assessment of treatment planning of the disease. The proposed extended ME network thus facilitates a more general approach to incorporate data hierarchy mechanism in network modeling.

The application of improved neural network in hydrocarbon reservoir prediction

NASA Astrophysics Data System (ADS)

Peng, Xiaobo

2013-03-01

This paper use BP neural network techniques to realize hydrocarbon reservoir predication easier and faster in tarim basin in oil wells. A grey - cascade neural network model is proposed and it is faster convergence speed and low error rate. The new method overcomes the shortcomings of traditional BP neural network convergence slow, easy to achieve extreme minimum value. This study had 220 sets of measured logging data to the sample data training mode. By changing the neuron number and types of the transfer function of hidden layers, the best work prediction model is analyzed. The conclusion is the model which can produce good prediction results in general, and can be used for hydrocarbon reservoir prediction.

Automated selection of computed tomography display parameters using neural networks

NASA Astrophysics Data System (ADS)

Zhang, Di; Neu, Scott; Valentino, Daniel J.

2001-07-01

A collection of artificial neural networks (ANN's) was trained to identify simple anatomical structures in a set of x-ray computed tomography (CT) images. These neural networks learned to associate a point in an image with the anatomical structure containing the point by using the image pixels located on the horizontal and vertical lines that ran through the point. The neural networks were integrated into a computer software tool whose function is to select an index into a list of CT window/level values from the location of the user's mouse cursor. Based upon the anatomical structure selected by the user, the software tool automatically adjusts the image display to optimally view the structure.

Method and system for pattern analysis using a coarse-coded neural network

NASA Technical Reports Server (NTRS)

Spirkovska, Liljana (Inventor); Reid, Max B. (Inventor)

1994-01-01

A method and system for performing pattern analysis with a neural network coarse-coding a pattern to be analyzed so as to form a plurality of sub-patterns collectively defined by data. Each of the sub-patterns comprises sets of pattern data. The neural network includes a plurality fields, each field being associated with one of the sub-patterns so as to receive the sub-pattern data therefrom. Training and testing by the neural network then proceeds in the usual way, with one modification: the transfer function thresholds the value obtained from summing the weighted products of each field over all sub-patterns associated with each pattern being analyzed by the system.

Routine programs of health care systems as an opportunity toward communication skills training for family physicians: A randomized field trial

PubMed Central

Zamani, Ahmad Reza; Motamedi, Narges; Farajzadegan, Ziba

2015-01-01

Background: To have high-quality primary health care services, an adequate doctor–patient communication is necessary. Because of time restrictions and limited budget in health system, an effective, feasible, and continuous training approach is important. The aim of this study is to assess the appropriateness of a communication skills training program simultaneously with routine programs of health care system. Materials and Methods: It was a randomized field trial in two health network settings during 2013. Twenty-eight family physicians through simple random sampling and 140 patients through convenience sampling participated as intervention and control group. The physicians in the intervention group (n = 14) attended six educational sessions, simultaneous organization meeting, with case discussion and peer education method. In both the groups, physicians completed communication skills knowledge and attitude questionnaires, and patients completed patient satisfaction of medical interview questionnaire at baseline, immediately after intervention, and four months postintervention. Physicians and health network administrators (stakeholders), completed a set of program evaluation forms. Descriptive statistics and Chi-square test, t-test, and repeated measure analysis of variance were used to analyze the data. Results: Use of routine program as a strategy of training was rated by stakeholders highly on “feasibility” (80.5%), “acceptability” (93.5%), “educational content and method appropriateness” (80.75%), and “ability to integrating in the health system programs” (approximate 60%). Significant improvements were found in physicians’ knowledge (P < 0.001), attitude (P < 0.001), and patients’ satisfaction (P = 0.002) in intervention group. Conclusions: Communication skills training program, simultaneous organization meeting was successfully implemented and well received by stakeholders, without considering extra time and manpower. Therefore it can be a valuable opportunity toward communication skills training. PMID:27462613

Mass detection in digital breast tomosynthesis: Deep convolutional neural network with transfer learning from mammography

PubMed Central

Chan, Heang-Ping; Hadjiiski, Lubomir; Helvie, Mark A.; Wei, Jun; Cha, Kenny

2016-01-01

Purpose: Develop a computer-aided detection (CAD) system for masses in digital breast tomosynthesis (DBT) volume using a deep convolutional neural network (DCNN) with transfer learning from mammograms. Methods: A data set containing 2282 digitized film and digital mammograms and 324 DBT volumes were collected with IRB approval. The mass of interest on the images was marked by an experienced breast radiologist as reference standard. The data set was partitioned into a training set (2282 mammograms with 2461 masses and 230 DBT views with 228 masses) and an independent test set (94 DBT views with 89 masses). For DCNN training, the region of interest (ROI) containing the mass (true positive) was extracted from each image. False positive (FP) ROIs were identified at prescreening by their previously developed CAD systems. After data augmentation, a total of 45 072 mammographic ROIs and 37 450 DBT ROIs were obtained. Data normalization and reduction of non-uniformity in the ROIs across heterogeneous data was achieved using a background correction method applied to each ROI. A DCNN with four convolutional layers and three fully connected (FC) layers was first trained on the mammography data. Jittering and dropout techniques were used to reduce overfitting. After training with the mammographic ROIs, all weights in the first three convolutional layers were frozen, and only the last convolution layer and the FC layers were randomly initialized again and trained using the DBT training ROIs. The authors compared the performances of two CAD systems for mass detection in DBT: one used the DCNN-based approach and the other used their previously developed feature-based approach for FP reduction. The prescreening stage was identical in both systems, passing the same set of mass candidates to the FP reduction stage. For the feature-based CAD system, 3D clustering and active contour method was used for segmentation; morphological, gray level, and texture features were extracted and merged with a linear discriminant classifier to score the detected masses. For the DCNN-based CAD system, ROIs from five consecutive slices centered at each candidate were passed through the trained DCNN and a mass likelihood score was generated. The performances of the CAD systems were evaluated using free-response ROC curves and the performance difference was analyzed using a non-parametric method. Results: Before transfer learning, the DCNN trained only on mammograms with an AUC of 0.99 classified DBT masses with an AUC of 0.81 in the DBT training set. After transfer learning with DBT, the AUC improved to 0.90. For breast-based CAD detection in the test set, the sensitivity for the feature-based and the DCNN-based CAD systems was 83% and 91%, respectively, at 1 FP/DBT volume. The difference between the performances for the two systems was statistically significant (p-value < 0.05). Conclusions: The image patterns learned from the mammograms were transferred to the mass detection on DBT slices through the DCNN. This study demonstrated that large data sets collected from mammography are useful for developing new CAD systems for DBT, alleviating the problem and effort of collecting entirely new large data sets for the new modality. PMID:27908154

Deep 3D Convolutional Encoder Networks With Shortcuts for Multiscale Feature Integration Applied to Multiple Sclerosis Lesion Segmentation.

PubMed

Brosch, Tom; Tang, Lisa Y W; Youngjin Yoo; Li, David K B; Traboulsee, Anthony; Tam, Roger

2016-05-01

We propose a novel segmentation approach based on deep 3D convolutional encoder networks with shortcut connections and apply it to the segmentation of multiple sclerosis (MS) lesions in magnetic resonance images. Our model is a neural network that consists of two interconnected pathways, a convolutional pathway, which learns increasingly more abstract and higher-level image features, and a deconvolutional pathway, which predicts the final segmentation at the voxel level. The joint training of the feature extraction and prediction pathways allows for the automatic learning of features at different scales that are optimized for accuracy for any given combination of image types and segmentation task. In addition, shortcut connections between the two pathways allow high- and low-level features to be integrated, which enables the segmentation of lesions across a wide range of sizes. We have evaluated our method on two publicly available data sets (MICCAI 2008 and ISBI 2015 challenges) with the results showing that our method performs comparably to the top-ranked state-of-the-art methods, even when only relatively small data sets are available for training. In addition, we have compared our method with five freely available and widely used MS lesion segmentation methods (EMS, LST-LPA, LST-LGA, Lesion-TOADS, and SLS) on a large data set from an MS clinical trial. The results show that our method consistently outperforms these other methods across a wide range of lesion sizes.

Hypotheses generation as supervised link discovery with automated class labeling on large-scale biomedical concept networks

PubMed Central

2012-01-01

Computational approaches to generate hypotheses from biomedical literature have been studied intensively in recent years. Nevertheless, it still remains a challenge to automatically discover novel, cross-silo biomedical hypotheses from large-scale literature repositories. In order to address this challenge, we first model a biomedical literature repository as a comprehensive network of biomedical concepts and formulate hypotheses generation as a process of link discovery on the concept network. We extract the relevant information from the biomedical literature corpus and generate a concept network and concept-author map on a cluster using Map-Reduce frame-work. We extract a set of heterogeneous features such as random walk based features, neighborhood features and common author features. The potential number of links to consider for the possibility of link discovery is large in our concept network and to address the scalability problem, the features from a concept network are extracted using a cluster with Map-Reduce framework. We further model link discovery as a classification problem carried out on a training data set automatically extracted from two network snapshots taken in two consecutive time duration. A set of heterogeneous features, which cover both topological and semantic features derived from the concept network, have been studied with respect to their impacts on the accuracy of the proposed supervised link discovery process. A case study of hypotheses generation based on the proposed method has been presented in the paper. PMID:22759614

VoxResNet: Deep voxelwise residual networks for brain segmentation from 3D MR images.

PubMed

Chen, Hao; Dou, Qi; Yu, Lequan; Qin, Jing; Heng, Pheng-Ann

2018-04-15

Segmentation of key brain tissues from 3D medical images is of great significance for brain disease diagnosis, progression assessment and monitoring of neurologic conditions. While manual segmentation is time-consuming, laborious, and subjective, automated segmentation is quite challenging due to the complicated anatomical environment of brain and the large variations of brain tissues. We propose a novel voxelwise residual network (VoxResNet) with a set of effective training schemes to cope with this challenging problem. The main merit of residual learning is that it can alleviate the degradation problem when training a deep network so that the performance gains achieved by increasing the network depth can be fully leveraged. With this technique, our VoxResNet is built with 25 layers, and hence can generate more representative features to deal with the large variations of brain tissues than its rivals using hand-crafted features or shallower networks. In order to effectively train such a deep network with limited training data for brain segmentation, we seamlessly integrate multi-modality and multi-level contextual information into our network, so that the complementary information of different modalities can be harnessed and features of different scales can be exploited. Furthermore, an auto-context version of the VoxResNet is proposed by combining the low-level image appearance features, implicit shape information, and high-level context together for further improving the segmentation performance. Extensive experiments on the well-known benchmark (i.e., MRBrainS) of brain segmentation from 3D magnetic resonance (MR) images corroborated the efficacy of the proposed VoxResNet. Our method achieved the first place in the challenge out of 37 competitors including several state-of-the-art brain segmentation methods. Our method is inherently general and can be readily applied as a powerful tool to many brain-related studies, where accurate segmentation of brain structures is critical. Copyright © 2017 Elsevier Inc. All rights reserved.

Evaluation of Deep Learning Representations of Spatial Storm Data

NASA Astrophysics Data System (ADS)

Gagne, D. J., II; Haupt, S. E.; Nychka, D. W.

2017-12-01

The spatial structure of a severe thunderstorm and its surrounding environment provide useful information about the potential for severe weather hazards, including tornadoes, hail, and high winds. Statistics computed over the area of a storm or from the pre-storm environment can provide descriptive information but fail to capture structural information. Because the storm environment is a complex, high-dimensional space, identifying methods to encode important spatial storm information in a low-dimensional form should aid analysis and prediction of storms by statistical and machine learning models. Principal component analysis (PCA), a more traditional approach, transforms high-dimensional data into a set of linearly uncorrelated, orthogonal components ordered by the amount of variance explained by each component. The burgeoning field of deep learning offers two potential approaches to this problem. Convolutional Neural Networks are a supervised learning method for transforming spatial data into a hierarchical set of feature maps that correspond with relevant combinations of spatial structures in the data. Generative Adversarial Networks (GANs) are an unsupervised deep learning model that uses two neural networks trained against each other to produce encoded representations of spatial data. These different spatial encoding methods were evaluated on the prediction of severe hail for a large set of storm patches extracted from the NCAR convection-allowing ensemble. Each storm patch contains information about storm structure and the near-storm environment. Logistic regression and random forest models were trained using the PCA and GAN encodings of the storm data and were compared against the predictions from a convolutional neural network. All methods showed skill over climatology at predicting the probability of severe hail. However, the verification scores among the methods were very similar and the predictions were highly correlated. Further evaluations are being performed to determine how the choice of input variables affects the results.

Radio Galaxy Zoo: compact and extended radio source classification with deep learning

NASA Astrophysics Data System (ADS)

Lukic, V.; Brüggen, M.; Banfield, J. K.; Wong, O. I.; Rudnick, L.; Norris, R. P.; Simmons, B.

2018-05-01

Machine learning techniques have been increasingly useful in astronomical applications over the last few years, for example in the morphological classification of galaxies. Convolutional neural networks have proven to be highly effective in classifying objects in image data. In the context of radio-interferometric imaging in astronomy, we looked for ways to identify multiple components of individual sources. To this effect, we design a convolutional neural network to differentiate between different morphology classes using sources from the Radio Galaxy Zoo (RGZ) citizen science project. In this first step, we focus on exploring the factors that affect the performance of such neural networks, such as the amount of training data, number and nature of layers, and the hyperparameters. We begin with a simple experiment in which we only differentiate between two extreme morphologies, using compact and multiple-component extended sources. We found that a three-convolutional layer architecture yielded very good results, achieving a classification accuracy of 97.4 per cent on a test data set. The same architecture was then tested on a four-class problem where we let the network classify sources into compact and three classes of extended sources, achieving a test accuracy of 93.5 per cent. The best-performing convolutional neural network set-up has been verified against RGZ Data Release 1 where a final test accuracy of 94.8 per cent was obtained, using both original and augmented images. The use of sigma clipping does not offer a significant benefit overall, except in cases with a small number of training images.

Effectiveness of feature and classifier algorithms in character recognition systems

NASA Astrophysics Data System (ADS)

Wilson, Charles L.

1993-04-01

At the first Census Optical Character Recognition Systems Conference, NIST generated accuracy data for more than character recognition systems. Most systems were tested on the recognition of isolated digits and upper and lower case alphabetic characters. The recognition experiments were performed on sample sizes of 58,000 digits, and 12,000 upper and lower case alphabetic characters. The algorithms used by the 26 conference participants included rule-based methods, image-based methods, statistical methods, and neural networks. The neural network methods included Multi-Layer Perceptron's, Learned Vector Quantitization, Neocognitrons, and cascaded neural networks. In this paper 11 different systems are compared using correlations between the answers of different systems, comparing the decrease in error rate as a function of confidence of recognition, and comparing the writer dependence of recognition. This comparison shows that methods that used different algorithms for feature extraction and recognition performed with very high levels of correlation. This is true for neural network systems, hybrid systems, and statistically based systems, and leads to the conclusion that neural networks have not yet demonstrated a clear superiority to more conventional statistical methods. Comparison of these results with the models of Vapnick (for estimation problems), MacKay (for Bayesian statistical models), Moody (for effective parameterization), and Boltzmann models (for information content) demonstrate that as the limits of training data variance are approached, all classifier systems have similar statistical properties. The limiting condition can only be approached for sufficiently rich feature sets because the accuracy limit is controlled by the available information content of the training set, which must pass through the feature extraction process prior to classification.

Classification of bifurcations regions in IVOCT images using support vector machine and artificial neural network models

NASA Astrophysics Data System (ADS)

Porto, C. D. N.; Costa Filho, C. F. F.; Macedo, M. M. G.; Gutierrez, M. A.; Costa, M. G. F.

2017-03-01

Studies in intravascular optical coherence tomography (IV-OCT) have demonstrated the importance of coronary bifurcation regions in intravascular medical imaging analysis, as plaques are more likely to accumulate in this region leading to coronary disease. A typical IV-OCT pullback acquires hundreds of frames, thus developing an automated tool to classify the OCT frames as bifurcation or non-bifurcation can be an important step to speed up OCT pullbacks analysis and assist automated methods for atherosclerotic plaque quantification. In this work, we evaluate the performance of two state-of-the-art classifiers, SVM and Neural Networks in the bifurcation classification task. The study included IV-OCT frames from 9 patients. In order to improve classification performance, we trained and tested the SVM with different parameters by means of a grid search and different stop criteria were applied to the Neural Network classifier: mean square error, early stop and regularization. Different sets of features were tested, using feature selection techniques: PCA, LDA and scalar feature selection with correlation. Training and test were performed in sets with a maximum of 1460 OCT frames. We quantified our results in terms of false positive rate, true positive rate, accuracy, specificity, precision, false alarm, f-measure and area under ROC curve. Neural networks obtained the best classification accuracy, 98.83%, overcoming the results found in literature. Our methods appear to offer a robust and reliable automated classification of OCT frames that might assist physicians indicating potential frames to analyze. Methods for improving neural networks generalization have increased the classification performance.

Automated embolic signal detection using Deep Convolutional Neural Network.

PubMed

Sombune, Praotasna; Phienphanich, Phongphan; Phuechpanpaisal, Sutanya; Muengtaweepongsa, Sombat; Ruamthanthong, Anuchit; Tantibundhit, Charturong

2017-07-01

This work investigated the potential of Deep Neural Network in detection of cerebral embolic signal (ES) from transcranial Doppler ultrasound (TCD). The resulting system is aimed to couple with TCD devices in diagnosing a risk of stroke in real-time with high accuracy. The Adaptive Gain Control (AGC) approach developed in our previous study is employed to capture suspected ESs in real-time. By using spectrograms of the same TCD signal dataset as that of our previous work as inputs and the same experimental setup, Deep Convolutional Neural Network (CNN), which can learn features while training, was investigated for its ability to bypass the traditional handcrafted feature extraction and selection process. Extracted feature vectors from the suspected ESs are later determined whether they are of an ES, artifact (AF) or normal (NR) interval. The effectiveness of the developed system was evaluated over 19 subjects going under procedures generating emboli. The CNN-based system could achieve in average of 83.0% sensitivity, 80.1% specificity, and 81.4% accuracy, with considerably much less time consumption in development. The certainly growing set of training samples and computational resources will contribute to high performance. Besides having potential use in various clinical ES monitoring settings, continuation of this promising study will benefit developments of wearable applications by leveraging learnable features to serve demographic differentials.

RENEB accident simulation exercise.

PubMed

Brzozowska, Beata; Ainsbury, Elizabeth; Baert, Annelot; Beaton-Green, Lindsay; Barrios, Leonardo; Barquinero, Joan Francesc; Bassinet, Celine; Beinke, Christina; Benedek, Anett; Beukes, Philip; Bortolin, Emanuela; Buraczewska, Iwona; Burbidge, Christopher; De Amicis, Andrea; De Angelis, Cinzia; Della Monaca, Sara; Depuydt, Julie; De Sanctis, Stefania; Dobos, Katalin; Domene, Mercedes Moreno; Domínguez, Inmaculada; Facco, Eva; Fattibene, Paola; Frenzel, Monika; Monteiro Gil, Octávia; Gonon, Géraldine; Gregoire, Eric; Gruel, Gaëtan; Hadjidekova, Valeria; Hatzi, Vasiliki I; Hristova, Rositsa; Jaworska, Alicja; Kis, Enikő; Kowalska, Maria; Kulka, Ulrike; Lista, Florigio; Lumniczky, Katalin; Martínez-López, Wilner; Meschini, Roberta; Moertl, Simone; Moquet, Jayne; Noditi, Mihaela; Oestreicher, Ursula; Orta Vázquez, Manuel Luis; Palma, Valentina; Pantelias, Gabriel; Montoro Pastor, Alegria; Patrono, Clarice; Piqueret-Stephan, Laure; Quattrini, Maria Cristina; Regalbuto, Elisa; Ricoul, Michelle; Roch-Lefevre, Sandrine; Roy, Laurence; Sabatier, Laure; Sarchiapone, Lucia; Sebastià, Natividad; Sommer, Sylwester; Sun, Mingzhu; Suto, Yumiko; Terzoudi, Georgia; Trompier, Francois; Vral, Anne; Wilkins, Ruth; Zafiropoulos, Demetre; Wieser, Albrecht; Woda, Clemens; Wojcik, Andrzej

2017-01-01

The RENEB accident exercise was carried out in order to train the RENEB participants in coordinating and managing potentially large data sets that would be generated in case of a major radiological event. Each participant was offered the possibility to activate the network by sending an alerting email about a simulated radiation emergency. The same participant had to collect, compile and report capacity, triage categorization and exposure scenario results obtained from all other participants. The exercise was performed over 27 weeks and involved the network consisting of 28 institutes: 21 RENEB members, four candidates and three non-RENEB partners. The duration of a single exercise never exceeded 10 days, while the response from the assisting laboratories never came later than within half a day. During each week of the exercise, around 4500 samples were reported by all service laboratories (SL) to be examined and 54 scenarios were coherently estimated by all laboratories (the standard deviation from the mean of all SL answers for a given scenario category and a set of data was not larger than 3 patient codes). Each participant received training in both the role of a reference laboratory (activating the network) and of a service laboratory (responding to an activation request). The procedures in the case of radiological event were successfully established and tested.

Quality control and improvement of cancer care: what is needed? 4th European Roundtable Meeting (ERTM) May 5th, 2017, Berlin, Germany.

PubMed

Ortmann, Olaf; Helbig, Ulrike; Torode, Julie; Schreck, Stefan; Karjalainen, Sakari; Bettio, Manola; Ringborg, Ulrik; Klinkhammer-Schalke, Monika; Bray, Freddy

2018-06-01

National Cancer Control Plans (NCCPs) often describe structural requirements for high quality cancer care. During the fourth European Roundtable Meeting (ERTM) participants shared learnings from their own national setting to formulate best practice in optimizing communication strategies between parties involved in clinical cancer registries, cancer centers and guideline groups. A decentralized model of data collection close to the patient and caregiver enhances timely completion and the quality of the data captured. Nevertheless, central coordination is necessary to define datasets, indicators, standard settings, education, training and quality control to maintain standards across the network. In particular, interaction of parties in cancer care network has to be established and maintained on a regular basis. After establishing the structural requirements of cancer care networks, communication between the different components and parties is required to analyze outcome data, provide regular reporting to all and develop strategies for continuous improvement of quality across the network.

Rapid differentiation of Ghana cocoa beans by FT-NIR spectroscopy coupled with multivariate classification

NASA Astrophysics Data System (ADS)

Teye, Ernest; Huang, Xingyi; Dai, Huang; Chen, Quansheng

2013-10-01

Quick, accurate and reliable technique for discrimination of cocoa beans according to geographical origin is essential for quality control and traceability management. This current study presents the application of Near Infrared Spectroscopy technique and multivariate classification for the differentiation of Ghana cocoa beans. A total of 194 cocoa bean samples from seven cocoa growing regions were used. Principal component analysis (PCA) was used to extract relevant information from the spectral data and this gave visible cluster trends. The performance of four multivariate classification methods: Linear discriminant analysis (LDA), K-nearest neighbors (KNN), Back propagation artificial neural network (BPANN) and Support vector machine (SVM) were compared. The performances of the models were optimized by cross validation. The results revealed that; SVM model was superior to all the mathematical methods with a discrimination rate of 100% in both the training and prediction set after preprocessing with Mean centering (MC). BPANN had a discrimination rate of 99.23% for the training set and 96.88% for prediction set. While LDA model had 96.15% and 90.63% for the training and prediction sets respectively. KNN model had 75.01% for the training set and 72.31% for prediction set. The non-linear classification methods used were superior to the linear ones. Generally, the results revealed that NIR Spectroscopy coupled with SVM model could be used successfully to discriminate cocoa beans according to their geographical origins for effective quality assurance.

Accelerated Training for Large Feedforward Neural Networks

NASA Technical Reports Server (NTRS)

Stepniewski, Slawomir W.; Jorgensen, Charles C.

1998-01-01

In this paper we introduce a new training algorithm, the scaled variable metric (SVM) method. Our approach attempts to increase the convergence rate of the modified variable metric method. It is also combined with the RBackprop algorithm, which computes the product of the matrix of second derivatives (Hessian) with an arbitrary vector. The RBackprop method allows us to avoid computationally expensive, direct line searches. In addition, it can be utilized in the new, 'predictive' updating technique of the inverse Hessian approximation. We have used directional slope testing to adjust the step size and found that this strategy works exceptionally well in conjunction with the Rbackprop algorithm. Some supplementary, but nevertheless important enhancements to the basic training scheme such as improved setting of a scaling factor for the variable metric update and computationally more efficient procedure for updating the inverse Hessian approximation are presented as well. We summarize by comparing the SVM method with four first- and second- order optimization algorithms including a very effective implementation of the Levenberg-Marquardt method. Our tests indicate promising computational speed gains of the new training technique, particularly for large feedforward networks, i.e., for problems where the training process may be the most laborious.

A Novel Approach for Lie Detection Based on F-Score and Extreme Learning Machine

PubMed Central

Gao, Junfeng; Wang, Zhao; Yang, Yong; Zhang, Wenjia; Tao, Chunyi; Guan, Jinan; Rao, Nini

2013-01-01

A new machine learning method referred to as F-score_ELM was proposed to classify the lying and truth-telling using the electroencephalogram (EEG) signals from 28 guilty and innocent subjects. Thirty-one features were extracted from the probe responses from these subjects. Then, a recently-developed classifier called extreme learning machine (ELM) was combined with F-score, a simple but effective feature selection method, to jointly optimize the number of the hidden nodes of ELM and the feature subset by a grid-searching training procedure. The method was compared to two classification models combining principal component analysis with back-propagation network and support vector machine classifiers. We thoroughly assessed the performance of these classification models including the training and testing time, sensitivity and specificity from the training and testing sets, as well as network size. The experimental results showed that the number of the hidden nodes can be effectively optimized by the proposed method. Also, F-score_ELM obtained the best classification accuracy and required the shortest training and testing time. PMID:23755136

High-Dimensional Function Approximation With Neural Networks for Large Volumes of Data.

PubMed

Andras, Peter

2018-02-01

Approximation of high-dimensional functions is a challenge for neural networks due to the curse of dimensionality. Often the data for which the approximated function is defined resides on a low-dimensional manifold and in principle the approximation of the function over this manifold should improve the approximation performance. It has been show that projecting the data manifold into a lower dimensional space, followed by the neural network approximation of the function over this space, provides a more precise approximation of the function than the approximation of the function with neural networks in the original data space. However, if the data volume is very large, the projection into the low-dimensional space has to be based on a limited sample of the data. Here, we investigate the nature of the approximation error of neural networks trained over the projection space. We show that such neural networks should have better approximation performance than neural networks trained on high-dimensional data even if the projection is based on a relatively sparse sample of the data manifold. We also find that it is preferable to use a uniformly distributed sparse sample of the data for the purpose of the generation of the low-dimensional projection. We illustrate these results considering the practical neural network approximation of a set of functions defined on high-dimensional data including real world data as well.

Feature extraction using convolutional neural network for classifying breast density in mammographic images

NASA Astrophysics Data System (ADS)

Thomaz, Ricardo L.; Carneiro, Pedro C.; Patrocinio, Ana C.

2017-03-01

Breast cancer is the leading cause of death for women in most countries. The high levels of mortality relate mostly to late diagnosis and to the direct proportionally relationship between breast density and breast cancer development. Therefore, the correct assessment of breast density is important to provide better screening for higher risk patients. However, in modern digital mammography the discrimination among breast densities is highly complex due to increased contrast and visual information for all densities. Thus, a computational system for classifying breast density might be a useful tool for aiding medical staff. Several machine-learning algorithms are already capable of classifying small number of classes with good accuracy. However, machinelearning algorithms main constraint relates to the set of features extracted and used for classification. Although well-known feature extraction techniques might provide a good set of features, it is a complex task to select an initial set during design of a classifier. Thus, we propose feature extraction using a Convolutional Neural Network (CNN) for classifying breast density by a usual machine-learning classifier. We used 307 mammographic images downsampled to 260x200 pixels to train a CNN and extract features from a deep layer. After training, the activation of 8 neurons from a deep fully connected layer are extracted and used as features. Then, these features are feedforward to a single hidden layer neural network that is cross-validated using 10-folds to classify among four classes of breast density. The global accuracy of this method is 98.4%, presenting only 1.6% of misclassification. However, the small set of samples and memory constraints required the reuse of data in both CNN and MLP-NN, therefore overfitting might have influenced the results even though we cross-validated the network. Thus, although we presented a promising method for extracting features and classifying breast density, a greater database is still required for evaluating the results.

Classifying galaxy spectra at 0.5 < z < 1 with self-organizing maps

NASA Astrophysics Data System (ADS)

Rahmani, S.; Teimoorinia, H.; Barmby, P.

2018-05-01

The spectrum of a galaxy contains information about its physical properties. Classifying spectra using templates helps elucidate the nature of a galaxy's energy sources. In this paper, we investigate the use of self-organizing maps in classifying galaxy spectra against templates. We trained semi-supervised self-organizing map networks using a set of templates covering the wavelength range from far ultraviolet to near infrared. The trained networks were used to classify the spectra of a sample of 142 galaxies with 0.5 < z < 1 and the results compared to classifications performed using K-means clustering, a supervised neural network, and chi-squared minimization. Spectra corresponding to quiescent galaxies were more likely to be classified similarly by all methods while starburst spectra showed more variability. Compared to classification using chi-squared minimization or the supervised neural network, the galaxies classed together by the self-organizing map had more similar spectra. The class ordering provided by the one-dimensional self-organizing maps corresponds to an ordering in physical properties, a potentially important feature for the exploration of large datasets.

PSF estimation for defocus blurred image based on quantum back-propagation neural network

NASA Astrophysics Data System (ADS)

Gao, Kun; Zhang, Yan; Shao, Xiao-guang; Liu, Ying-hui; Ni, Guoqiang

2010-11-01

Images obtained by an aberration-free system are defocused blur due to motion in depth and/or zooming. The precondition of restoring the degraded image is to estimate point spread function (PSF) of the imaging system as precisely as possible. But it is difficult to identify the analytic model of PSF precisely due to the complexity of the degradation process. Inspired by the similarity between the quantum process and imaging process in the probability and statistics fields, one reformed multilayer quantum neural network (QNN) is proposed to estimate PSF of the defocus blurred image. Different from the conventional artificial neural network (ANN), an improved quantum neuron model is used in the hidden layer instead, which introduces a 2-bit controlled NOT quantum gate to control output and adopts 2 texture and edge features as the input vectors. The supervised back-propagation learning rule is adopted to train network based on training sets from the historical images. Test results show that this method owns excellent features of high precision and strong generalization ability.

Deep Deconvolutional Neural Network for Target Segmentation of Nasopharyngeal Cancer in Planning Computed Tomography Images.

PubMed

Men, Kuo; Chen, Xinyuan; Zhang, Ye; Zhang, Tao; Dai, Jianrong; Yi, Junlin; Li, Yexiong

2017-01-01

Radiotherapy is one of the main treatment methods for nasopharyngeal carcinoma (NPC). It requires exact delineation of the nasopharynx gross tumor volume (GTVnx), the metastatic lymph node gross tumor volume (GTVnd), the clinical target volume (CTV), and organs at risk in the planning computed tomography images. However, this task is time-consuming and operator dependent. In the present study, we developed an end-to-end deep deconvolutional neural network (DDNN) for segmentation of these targets. The proposed DDNN is an end-to-end architecture enabling fast training and testing. It consists of two important components: an encoder network and a decoder network. The encoder network was used to extract the visual features of a medical image and the decoder network was used to recover the original resolution by deploying deconvolution. A total of 230 patients diagnosed with NPC stage I or stage II were included in this study. Data from 184 patients were chosen randomly as a training set to adjust the parameters of DDNN, and the remaining 46 patients were the test set to assess the performance of the model. The Dice similarity coefficient (DSC) was used to quantify the segmentation results of the GTVnx, GTVnd, and CTV. In addition, the performance of DDNN was compared with the VGG-16 model. The proposed DDNN method outperformed the VGG-16 in all the segmentation. The mean DSC values of DDNN were 80.9% for GTVnx, 62.3% for the GTVnd, and 82.6% for CTV, whereas VGG-16 obtained 72.3, 33.7, and 73.7% for the DSC values, respectively. DDNN can be used to segment the GTVnx and CTV accurately. The accuracy for the GTVnd segmentation was relatively low due to the considerable differences in its shape, volume, and location among patients. The accuracy is expected to increase with more training data and combination of MR images. In conclusion, DDNN has the potential to improve the consistency of contouring and streamline radiotherapy workflows, but careful human review and a considerable amount of editing will be required.

The application of neural networks to myoelectric signal analysis: a preliminary study.

PubMed

Kelly, M F; Parker, P A; Scott, R N

1990-03-01

Two neural network implementations are applied to myoelectric signal (MES) analysis tasks. The motivation behind this research is to explore more reliable methods of deriving control for multidegree of freedom arm prostheses. A discrete Hopfield network is used to calculate the time series parameters for a moving average MES model. It is demonstrated that the Hopfield network is capable of generating the same time series parameters as those produced by the conventional sequential least squares (SLS) algorithm. Furthermore, it can be extended to applications utilizing larger amounts of data, and possibly to higher order time series models, without significant degradation in computational efficiency. The second neural network implementation involves using a two-layer perceptron for classifying a single site MES based on two features, specifically the first time series parameter, and the signal power. Using these features, the perceptron is trained to distinguish between four separate arm functions. The two-dimensional decision boundaries used by the perceptron classifier are delineated. It is also demonstrated that the perceptron is able to rapidly compensate for variations when new data are incorporated into the training set. This adaptive quality suggests that perceptrons may provide a useful tool for future MES analysis.

Super-Resolution for “Jilin-1” Satellite Video Imagery via a Convolutional Network

PubMed Central

Wang, Zhongyuan; Wang, Lei; Ren, Yexian

2018-01-01

Super-resolution for satellite video attaches much significance to earth observation accuracy, and the special imaging and transmission conditions on the video satellite pose great challenges to this task. The existing deep convolutional neural-network-based methods require pre-processing or post-processing to be adapted to a high-resolution size or pixel format, leading to reduced performance and extra complexity. To this end, this paper proposes a five-layer end-to-end network structure without any pre-processing and post-processing, but imposes a reshape or deconvolution layer at the end of the network to retain the distribution of ground objects within the image. Meanwhile, we formulate a joint loss function by combining the output and high-dimensional features of a non-linear mapping network to precisely learn the desirable mapping relationship between low-resolution images and their high-resolution counterparts. Also, we use satellite video data itself as a training set, which favors consistency between training and testing images and promotes the method’s practicality. Experimental results on “Jilin-1” satellite video imagery show that this method demonstrates a superior performance in terms of both visual effects and measure metrics over competing methods. PMID:29652838

Forecasting solar proton event with artificial neural network

NASA Astrophysics Data System (ADS)

Gong, J.; Wang, J.; Xue, B.; Liu, S.; Zou, Z.

Solar proton event (SPE), relatively rare but popular in solar maximum, can bring hazard situation to spacecraft. As a special event, SPE always accompanies flare, which is also called proton flare. To produce such an eruptive event, large amount energy must be accumulated within the active region. So we can investigate the character of the active region and its evolving trend, together with other such as cm radio emission and soft X-ray background to evaluate the potential of SEP in chosen area. In order to summarize the omen of SPEs in the active regions behind the observed parameters, we employed AI technology. Full connecting neural network was chosen to fulfil this job. After constructing the network, we train it with 13 parameters that was able to exhibit the character of active regions and their evolution trend. More than 80 sets of event parameter were defined to teach the neural network to identify whether an active region was potential of SPE. Then we test this model with a data base consisting SPE and non-SPE cases that was not used to train the neural network. The result showed that 75% of the choice by the model was right.

Super-Resolution for "Jilin-1" Satellite Video Imagery via a Convolutional Network.

PubMed

Xiao, Aoran; Wang, Zhongyuan; Wang, Lei; Ren, Yexian

2018-04-13

Super-resolution for satellite video attaches much significance to earth observation accuracy, and the special imaging and transmission conditions on the video satellite pose great challenges to this task. The existing deep convolutional neural-network-based methods require pre-processing or post-processing to be adapted to a high-resolution size or pixel format, leading to reduced performance and extra complexity. To this end, this paper proposes a five-layer end-to-end network structure without any pre-processing and post-processing, but imposes a reshape or deconvolution layer at the end of the network to retain the distribution of ground objects within the image. Meanwhile, we formulate a joint loss function by combining the output and high-dimensional features of a non-linear mapping network to precisely learn the desirable mapping relationship between low-resolution images and their high-resolution counterparts. Also, we use satellite video data itself as a training set, which favors consistency between training and testing images and promotes the method's practicality. Experimental results on "Jilin-1" satellite video imagery show that this method demonstrates a superior performance in terms of both visual effects and measure metrics over competing methods.

Neural Network and Regression Soft Model Extended for PAX-300 Aircraft Engine

NASA Technical Reports Server (NTRS)

Patnaik, Surya N.; Hopkins, Dale A.

2002-01-01

In fiscal year 2001, the neural network and regression capabilities of NASA Glenn Research Center's COMETBOARDS design optimization testbed were extended to generate approximate models for the PAX-300 aircraft engine. The analytical model of the engine is defined through nine variables: the fan efficiency factor, the low pressure of the compressor, the high pressure of the compressor, the high pressure of the turbine, the low pressure of the turbine, the operating pressure, and three critical temperatures (T(sub 4), T(sub vane), and T(sub metal)). Numerical Propulsion System Simulation (NPSS) calculations of the specific fuel consumption (TSFC), as a function of the variables can become time consuming, and numerical instabilities can occur during these design calculations. "Soft" models can alleviate both deficiencies. These approximate models are generated from a set of high-fidelity input-output pairs obtained from the NPSS code and a design of the experiment strategy. A neural network and a regression model with 45 weight factors were trained for the input/output pairs. Then, the trained models were validated through a comparison with the original NPSS code. Comparisons of TSFC versus the operating pressure and of TSFC versus the three temperatures (T(sub 4), T(sub vane), and T(sub metal)) are depicted in the figures. The overall performance was satisfactory for both the regression and the neural network model. The regression model required fewer calculations than the neural network model, and it produced marginally superior results. Training the approximate methods is time consuming. Once trained, the approximate methods generated the solution with only a trivial computational effort, reducing the solution time from hours to less than a minute.

Deep multi-scale location-aware 3D convolutional neural networks for automated detection of lacunes of presumed vascular origin.

PubMed

Ghafoorian, Mohsen; Karssemeijer, Nico; Heskes, Tom; Bergkamp, Mayra; Wissink, Joost; Obels, Jiri; Keizer, Karlijn; de Leeuw, Frank-Erik; Ginneken, Bram van; Marchiori, Elena; Platel, Bram

2017-01-01

Lacunes of presumed vascular origin (lacunes) are associated with an increased risk of stroke, gait impairment, and dementia and are a primary imaging feature of the small vessel disease. Quantification of lacunes may be of great importance to elucidate the mechanisms behind neuro-degenerative disorders and is recommended as part of study standards for small vessel disease research. However, due to the different appearance of lacunes in various brain regions and the existence of other similar-looking structures, such as perivascular spaces, manual annotation is a difficult, elaborative and subjective task, which can potentially be greatly improved by reliable and consistent computer-aided detection (CAD) routines. In this paper, we propose an automated two-stage method using deep convolutional neural networks (CNN). We show that this method has good performance and can considerably benefit readers. We first use a fully convolutional neural network to detect initial candidates. In the second step, we employ a 3D CNN as a false positive reduction tool. As the location information is important to the analysis of candidate structures, we further equip the network with contextual information using multi-scale analysis and integration of explicit location features. We trained, validated and tested our networks on a large dataset of 1075 cases obtained from two different studies. Subsequently, we conducted an observer study with four trained observers and compared our method with them using a free-response operating characteristic analysis. Shown on a test set of 111 cases, the resulting CAD system exhibits performance similar to the trained human observers and achieves a sensitivity of 0.974 with 0.13 false positives per slice. A feasibility study also showed that a trained human observer would considerably benefit once aided by the CAD system.

Anti AIDS drug design with the help of neural networks

NASA Astrophysics Data System (ADS)

Tetko, I. V.; Tanchuk, V. Yu.; Luik, A. I.

1995-04-01

Artificial neural networks were used to analyze and predict the human immunodefiency virus type 1 reverse transcriptase inhibitors. Training and control set included 44 molecules (most of them are well-known substances such as AZT, TIBO, dde, etc.) The biological activities of molecules were taken from literature and rated for two classes: active and inactive compounds according to their values. We used topological indices as molecular parameters. Four most informative parameters (out of 46) were chosen using cluster analysis and original input parameters' estimation procedure and were used to predict activities of both control and new (synthesized in our institute) molecules. We applied pruning network algorithm and network ensembles to obtain the final classifier and avoid chance correlation. The increasing of neural network generalization of the data from the control set was observed, when using the aforementioned methods. The prognosis of new molecules revealed one molecule as possibly active. It was confirmed by further biological tests. The compound was as active as AZT and in order less toxic. The active compound is currently being evaluated in pre clinical trials as possible drug for anti-AIDS therapy.

High quality garbage: A neural network plastic sorter in hardware and software

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

Stanton, S.L.; Alam, M.K.; Hebner, G.A.

1993-09-01

In order to produce pure polymer streams from post-consumer waste plastics, a quick, accurate and relatively inexpensive method of sorting needs to be implemented. This technology has been demonstrated by using near-infrared spectroscopy reflectance data and neural network classification techniques. Backpropagation neural network routines have been developed to run real-time sortings in the lab, using a laboratory-grade spectrometer. In addition, a new reflectance spectrometer has been developed which is fast enough for commercial use. Initial training and test sets taken with the laboratory instrument show that a network is capable of learning 100% when classifying 5 groups of plastic (HDPEmore » and LDPE combined), and up to 100% when classifying 6 groups. Initial data sets from the new instrument have classified plastics into all seven groups with varying degrees of success. One of the initial networks has been implemented in hardware, for high speed computations, and thus rapid classification. Two neural accelerator systems have been evaluated, one based on the Intel 8017ONX chip, and another on the AT&T ANNA chip.« less

AcademyHealth's Delivery System Science Fellowship: Training Embedded Researchers to Design, Implement, and Evaluate New Models of Care.

PubMed

Kanani, Nisha; Hahn, Erin; Gould, Michael; Brunisholz, Kimberly; Savitz, Lucy; Holve, Erin

2017-07-01

AcademyHealth's Delivery System Science Fellowship (DSSF) provides a paid postdoctoral pragmatic learning experience to build capacity within learning healthcare systems to conduct research in applied settings. The fellowship provides hands-on training and professional leadership opportunities for researchers. Since its inception in 2012, the program has grown rapidly, with 16 health systems participating in the DSSF to date. In addition to specific projects conducted within health systems (and numerous publications associated with those initiatives), the DSSF has made several broader contributions to the field, including defining delivery system science, identifying a set of training objectives for researchers working in delivery systems, and developing a national collaborative network of care delivery organizations, operational leaders, and trainees. The DSSF is one promising approach to support higher-value care by promoting continuous learning and improvement in health systems. © 2017 Society of Hospital Medicine.

Effects of Cognitive Training on Resting-State Functional Connectivity of Default Mode, Salience, and Central Executive Networks.

PubMed

Cao, Weifang; Cao, Xinyi; Hou, Changyue; Li, Ting; Cheng, Yan; Jiang, Lijuan; Luo, Cheng; Li, Chunbo; Yao, Dezhong

2016-01-01

Neuroimaging studies have documented that aging can disrupt certain higher cognitive systems such as the default mode network (DMN), the salience network and the central executive network (CEN). The effect of cognitive training on higher cognitive systems remains unclear. This study used a 1-year longitudinal design to explore the cognitive training effect on three higher cognitive networks in healthy older adults. The community-living healthy older adults were divided into two groups: the multi-domain cognitive training group (24 sessions of cognitive training over a 3-months period) and the wait-list control group. All subjects underwent cognitive measurements and resting-state functional magnetic resonance imaging scanning at baseline and at 1 year after the training ended. We examined training-related changes in functional connectivity (FC) within and between three networks. Compared with the baseline, we observed maintained or increased FC within all three networks after training. The scans after training also showed maintained anti-correlation of FC between the DMN and CEN compared to the baseline. These findings demonstrated that cognitive training maintained or improved the functional integration within networks and the coupling between the DMN and CEN in older adults. Our findings suggested that multi-domain cognitive training can mitigate the aging-related dysfunction of higher cognitive networks.

Effects of Cognitive Training on Resting-State Functional Connectivity of Default Mode, Salience, and Central Executive Networks

PubMed Central

Cao, Weifang; Cao, Xinyi; Hou, Changyue; Li, Ting; Cheng, Yan; Jiang, Lijuan; Luo, Cheng; Li, Chunbo; Yao, Dezhong

2016-01-01

Neuroimaging studies have documented that aging can disrupt certain higher cognitive systems such as the default mode network (DMN), the salience network and the central executive network (CEN). The effect of cognitive training on higher cognitive systems remains unclear. This study used a 1-year longitudinal design to explore the cognitive training effect on three higher cognitive networks in healthy older adults. The community-living healthy older adults were divided into two groups: the multi-domain cognitive training group (24 sessions of cognitive training over a 3-months period) and the wait-list control group. All subjects underwent cognitive measurements and resting-state functional magnetic resonance imaging scanning at baseline and at 1 year after the training ended. We examined training-related changes in functional connectivity (FC) within and between three networks. Compared with the baseline, we observed maintained or increased FC within all three networks after training. The scans after training also showed maintained anti-correlation of FC between the DMN and CEN compared to the baseline. These findings demonstrated that cognitive training maintained or improved the functional integration within networks and the coupling between the DMN and CEN in older adults. Our findings suggested that multi-domain cognitive training can mitigate the aging-related dysfunction of higher cognitive networks. PMID:27148042

Remote voice training: A case study on space shuttle applications, appendix C

NASA Technical Reports Server (NTRS)

Mollakarimi, Cindy; Hamid, Tamin

1990-01-01

The Tile Automation System includes applications of automation and robotics technology to all aspects of the Shuttle tile processing and inspection system. An integrated set of rapid prototyping testbeds was developed which include speech recognition and synthesis, laser imaging systems, distributed Ada programming environments, distributed relational data base architectures, distributed computer network architectures, multi-media workbenches, and human factors considerations. Remote voice training in the Tile Automation System is discussed. The user is prompted over a headset by synthesized speech for the training sequences. The voice recognition units and the voice output units are remote from the user and are connected by Ethernet to the main computer system. A supervisory channel is used to monitor the training sequences. Discussions include the training approaches as well as the human factors problems and solutions for this system utilizing remote training techniques.

Tomography and generative training with quantum Boltzmann machines

NASA Astrophysics Data System (ADS)

Kieferová, Mária; Wiebe, Nathan

2017-12-01

The promise of quantum neural nets, which utilize quantum effects to model complex data sets, has made their development an aspirational goal for quantum machine learning and quantum computing in general. Here we provide methods of training quantum Boltzmann machines. Our work generalizes existing methods and provides additional approaches for training quantum neural networks that compare favorably to existing methods. We further demonstrate that quantum Boltzmann machines enable a form of partial quantum state tomography that further provides a generative model for the input quantum state. Classical Boltzmann machines are incapable of this. This verifies the long-conjectured connection between tomography and quantum machine learning. Finally, we prove that classical computers cannot simulate our training process in general unless BQP=BPP , provide lower bounds on the complexity of the training procedures and numerically investigate training for small nonstoquastic Hamiltonians.

Practice-Based Research Networks, Part II: A Descriptive Analysis of the Athletic Training Practice-Based Research Network in the Secondary School Setting

PubMed Central

McLeod, Tamara C. Valovich; Lam, Kenneth C.; Bay, R. Curtis; Sauers, Eric L.; Valier, Alison R. Snyder

2012-01-01

Context Analysis of health care service models requires the collection and evaluation of basic practice characterization data. Practice-based research networks (PBRNs) provide a framework for gathering data useful in characterizing clinical practice. Objective To describe preliminary secondary school setting practice data from the Athletic Training Practice-Based Research Network (AT-PBRN). Design Descriptive study. Setting Secondary school athletic training facilities within the AT-PBRN. Patients or Other Participants Clinicians (n = 22) and their patients (n = 2523) from the AT-PBRN. Main Outcome Measure(s) A Web-based survey was used to obtain data on clinical practice site and clinician characteristics. Patient and practice characteristics were obtained via deidentified electronic medical record data collected between September 1, 2009, and April 1, 2011. Descriptive data regarding the clinician and CPS practice characteristics are reported as percentages and frequencies. Descriptive analysis of patient encounters and practice characteristic data was performed, with the percentages and frequencies of the type of injuries recorded at initial evaluation, type of treatment received at initial evaluation, daily treatment, and daily sign-in procedures. Results The AT-PBRN had secondary school sites in 7 states, and most athletic trainers at those sites (78.2%) had less than 5 years of experience. The secondary school sites within the AT-PBRN documented 2523 patients treated across 3140 encounters. Patients most frequently sought care for a current injury (61.3%), followed by preventive services (24.0%), and new injuries (14.7%). The most common diagnoses were ankle sprain/strain (17.9%), hip sprain/strain (12.5%), concussion (12.0%), and knee pain (2.5%). The most frequent procedures were athletic trainer evaluation (53.9%), hot- or cold-pack application (26.0%), strapping (10.3%), and therapeutic exercise (5.7%). The median number of treatments per injury was 3 (interquartile range = 2, 4; range = 2–19). Conclusions These preliminary data describe services provided by clinicians within the AT-PBRN and demonstrate the usefulness of the PBRN model for obtaining such data. PMID:23068594

Deep Neural Networks as a Computational Model for Human Shape Sensitivity

PubMed Central

Op de Beeck, Hans P.

2016-01-01

Theories of object recognition agree that shape is of primordial importance, but there is no consensus about how shape might be represented, and so far attempts to implement a model of shape perception that would work with realistic stimuli have largely failed. Recent studies suggest that state-of-the-art convolutional ‘deep’ neural networks (DNNs) capture important aspects of human object perception. We hypothesized that these successes might be partially related to a human-like representation of object shape. Here we demonstrate that sensitivity for shape features, characteristic to human and primate vision, emerges in DNNs when trained for generic object recognition from natural photographs. We show that these models explain human shape judgments for several benchmark behavioral and neural stimulus sets on which earlier models mostly failed. In particular, although never explicitly trained for such stimuli, DNNs develop acute sensitivity to minute variations in shape and to non-accidental properties that have long been implicated to form the basis for object recognition. Even more strikingly, when tested with a challenging stimulus set in which shape and category membership are dissociated, the most complex model architectures capture human shape sensitivity as well as some aspects of the category structure that emerges from human judgments. As a whole, these results indicate that convolutional neural networks not only learn physically correct representations of object categories but also develop perceptually accurate representational spaces of shapes. An even more complete model of human object representations might be in sight by training deep architectures for multiple tasks, which is so characteristic in human development. PMID:27124699

Visual pathways from the perspective of cost functions and multi-task deep neural networks.

PubMed

Scholte, H Steven; Losch, Max M; Ramakrishnan, Kandan; de Haan, Edward H F; Bohte, Sander M

2018-01-01

Vision research has been shaped by the seminal insight that we can understand the higher-tier visual cortex from the perspective of multiple functional pathways with different goals. In this paper, we try to give a computational account of the functional organization of this system by reasoning from the perspective of multi-task deep neural networks. Machine learning has shown that tasks become easier to solve when they are decomposed into subtasks with their own cost function. We hypothesize that the visual system optimizes multiple cost functions of unrelated tasks and this causes the emergence of a ventral pathway dedicated to vision for perception, and a dorsal pathway dedicated to vision for action. To evaluate the functional organization in multi-task deep neural networks, we propose a method that measures the contribution of a unit towards each task, applying it to two networks that have been trained on either two related or two unrelated tasks, using an identical stimulus set. Results show that the network trained on the unrelated tasks shows a decreasing degree of feature representation sharing towards higher-tier layers while the network trained on related tasks uniformly shows high degree of sharing. We conjecture that the method we propose can be used to analyze the anatomical and functional organization of the visual system and beyond. We predict that the degree to which tasks are related is a good descriptor of the degree to which they share downstream cortical-units. Copyright © 2017 Elsevier Ltd. All rights reserved.

Underwater target classification using wavelet packets and neural networks.

PubMed

Azimi-Sadjadi, M R; Yao, D; Huang, Q; Dobeck, G J

2000-01-01

In this paper, a new subband-based classification scheme is developed for classifying underwater mines and mine-like targets from the acoustic backscattered signals. The system consists of a feature extractor using wavelet packets in conjunction with linear predictive coding (LPC), a feature selection scheme, and a backpropagation neural-network classifier. The data set used for this study consists of the backscattered signals from six different objects: two mine-like targets and four nontargets for several aspect angles. Simulation results on ten different noisy realizations and for signal-to-noise ratio (SNR) of 12 dB are presented. The receiver operating characteristic (ROC) curve of the classifier generated based on these results demonstrated excellent classification performance of the system. The generalization ability of the trained network was demonstrated by computing the error and classification rate statistics on a large data set. A multiaspect fusion scheme was also adopted in order to further improve the classification performance.

Imbalance aware lithography hotspot detection: a deep learning approach

NASA Astrophysics Data System (ADS)

Yang, Haoyu; Luo, Luyang; Su, Jing; Lin, Chenxi; Yu, Bei

2017-03-01

With the advancement of VLSI technology nodes, light diffraction caused lithographic hotspots have become a serious problem affecting manufacture yield. Lithography hotspot detection at the post-OPC stage is imperative to check potential circuit failures when transferring designed patterns onto silicon wafers. Although conventional lithography hotspot detection methods, such as machine learning, have gained satisfactory performance, with extreme scaling of transistor feature size and more and more complicated layout patterns, conventional methodologies may suffer from performance degradation. For example, manual or ad hoc feature extraction in a machine learning framework may lose important information when predicting potential errors in ultra-large-scale integrated circuit masks. In this paper, we present a deep convolutional neural network (CNN) targeting representative feature learning in lithography hotspot detection. We carefully analyze impact and effectiveness of different CNN hyper-parameters, through which a hotspot-detection-oriented neural network model is established. Because hotspot patterns are always minorities in VLSI mask design, the training data set is highly imbalanced. In this situation, a neural network is no longer reliable, because a trained model with high classification accuracy may still suffer from high false negative results (missing hotspots), which is fatal in hotspot detection problems. To address the imbalance problem, we further apply minority upsampling and random-mirror flipping before training the network. Experimental results show that our proposed neural network model achieves highly comparable or better performance on the ICCAD 2012 contest benchmark compared to state-of-the-art hotspot detectors based on deep or representative machine leaning.

Neural Network for Nanoscience Scanning Electron Microscope Image Recognition.

PubMed

Modarres, Mohammad Hadi; Aversa, Rossella; Cozzini, Stefano; Ciancio, Regina; Leto, Angelo; Brandino, Giuseppe Piero

2017-10-16

In this paper we applied transfer learning techniques for image recognition, automatic categorization, and labeling of nanoscience images obtained by scanning electron microscope (SEM). Roughly 20,000 SEM images were manually classified into 10 categories to form a labeled training set, which can be used as a reference set for future applications of deep learning enhanced algorithms in the nanoscience domain. The categories chosen spanned the range of 0-Dimensional (0D) objects such as particles, 1D nanowires and fibres, 2D films and coated surfaces, and 3D patterned surfaces such as pillars. The training set was used to retrain on the SEM dataset and to compare many convolutional neural network models (Inception-v3, Inception-v4, ResNet). We obtained compatible results by performing a feature extraction of the different models on the same dataset. We performed additional analysis of the classifier on a second test set to further investigate the results both on particular cases and from a statistical point of view. Our algorithm was able to successfully classify around 90% of a test dataset consisting of SEM images, while reduced accuracy was found in the case of images at the boundary between two categories or containing elements of multiple categories. In these cases, the image classification did not identify a predominant category with a high score. We used the statistical outcomes from testing to deploy a semi-automatic workflow able to classify and label images generated by the SEM. Finally, a separate training was performed to determine the volume fraction of coherently aligned nanowires in SEM images. The results were compared with what was obtained using the Local Gradient Orientation method. This example demonstrates the versatility and the potential of transfer learning to address specific tasks of interest in nanoscience applications.

Top-down, bottom-up, and around the jungle gym: a social exchange and networks approach to engaging afterschool programs in implementing evidence-based practices.

PubMed

Smith, Emilie Phillips; Wise, Eileen; Rosen, Howard; Rosen, Alison; Childs, Sharon; McManus, Margaret

2014-06-01

This paper uses concepts from social networks and social exchange theories to describe the implementation of evidence-based practices in afterschool programs. The members of the LEGACY Together Afterschool Project team have been involved in conducting collaborative research to migrate a behavioral strategy that has been documented to reduce disruptive behaviors in classroom settings to a new setting-that of afterschool programs. We adapted the Paxis Institute's version of the Good Behavior Game to afterschool settings which differ from in-school settings, including more fluid attendance, multiple age groupings, diverse activities that may take place simultaneously, and differences in staff training and experience (Barrish et al. in J Appl Behav Anal 2(2):119-124, 1969; Embry et al. in The Pax Good Behavior Game. Hazelden, Center City, 2003; Hynes et al. in J Child Serv 4(3):4-20, 2009; Kellam et al. in Drug Alcohol Depend 95:S5-S28, 2008; Tingstrom et al. in Behav Modif 30(2):225-253, 2006). This paper presents the experiences of the three adult groups involved in the implementation process who give first-person accounts of implementation: (1) university-based scientist-practitioners, (2) community partners who trained and provided technical assistance/coaching, and (3) an afterschool program administrator. We introduce here the AIMS model used to frame the implementation process conceptualized by this town-gown collaborative team. AIMS builds upon previous work in implementation science using four phases in which the three collaborators have overlapping roles: approach/engagement, implementation, monitoring, and sustainability. Within all four phases principles of Social Exchange Theory and Social Network Theory are highlighted.

Setup of a Parameterized FE Model for the Die Roll Prediction in Fine Blanking using Artificial Neural Networks

NASA Astrophysics Data System (ADS)

Stanke, J.; Trauth, D.; Feuerhack, A.; Klocke, F.

2017-09-01

Die roll is a morphological feature of fine blanked sheared edges. The die roll reduces the functional part of the sheared edge. To compensate for the die roll thicker sheet metal strips and secondary machining must be used. However, in order to avoid this, the influence of various fine blanking process parameters on the die roll has been experimentally and numerically studied, but there is still a lack of knowledge on the effects of some factors and especially factor interactions on the die roll. Recent changes in the field of artificial intelligence motivate the hybrid use of the finite element method and artificial neural networks to account for these non-considered parameters. Therefore, a set of simulations using a validated finite element model of fine blanking is firstly used to train an artificial neural network. Then the artificial neural network is trained with thousands of experimental trials. Thus, the objective of this contribution is to develop an artificial neural network that reliably predicts the die roll. Therefore, in this contribution, the setup of a fully parameterized 2D FE model is presented that will be used for batch training of an artificial neural network. The FE model enables an automatic variation of the edge radii of blank punch and die plate, the counter and blank holder force, the sheet metal thickness and part diameter, V-ring height and position, cutting velocity as well as material parameters covered by the Hensel-Spittel model for 16MnCr5 (1.7131, AISI/SAE 5115). The FE model is validated using experimental trails. The results of this contribution is a FE model suitable to perform 9.623 simulations and to pass the simulated die roll width and height automatically to an artificial neural network.

Probing the functions of long non-coding RNAs by exploiting the topology of global association and interaction network.

PubMed

Deng, Lei; Wu, Hongjie; Liu, Chuyao; Zhan, Weihua; Zhang, Jingpu

2018-06-01

Long non-coding RNAs (lncRNAs) are involved in many biological processes, such as immune response, development, differentiation and gene imprinting and are associated with diseases and cancers. But the functions of the vast majority of lncRNAs are still unknown. Predicting the biological functions of lncRNAs is one of the key challenges in the post-genomic era. In our work, We first build a global network including a lncRNA similarity network, a lncRNA-protein association network and a protein-protein interaction network according to the expressions and interactions, then extract the topological feature vectors of the global network. Using these features, we present an SVM-based machine learning approach, PLNRGO, to annotate human lncRNAs. In PLNRGO, we construct a training data set according to the proteins with GO annotations and train a binary classifier for each GO term. We assess the performance of PLNRGO on our manually annotated lncRNA benchmark and a protein-coding gene benchmark with known functional annotations. As a result, the performance of our method is significantly better than that of other state-of-the-art methods in terms of maximum F-measure and coverage. Copyright © 2018 Elsevier Ltd. All rights reserved.

Economic Development: A Community College Faces the Challenge.

ERIC Educational Resources Information Center

Crowley, Jackie; Boatright, Joyce

1988-01-01

Faced with a changing economy, the Houston Community College System (Texas) set goals and objectives to pull together its resources in response to the crisis in the Houston area. This article describes successful strategies and approaches used to network, market, and develop company partnerships and training activities to meet local needs. (Author)

DeepID-Net: Deformable Deep Convolutional Neural Networks for Object Detection.

PubMed

Ouyang, Wanli; Zeng, Xingyu; Wang, Xiaogang; Qiu, Shi; Luo, Ping; Tian, Yonglong; Li, Hongsheng; Yang, Shuo; Wang, Zhe; Li, Hongyang; Loy, Chen Change; Wang, Kun; Yan, Junjie; Tang, Xiaoou

2016-07-07

In this paper, we propose deformable deep convolutional neural networks for generic object detection. This new deep learning object detection framework has innovations in multiple aspects. In the proposed new deep architecture, a new deformation constrained pooling (def-pooling) layer models the deformation of object parts with geometric constraint and penalty. A new pre-training strategy is proposed to learn feature representations more suitable for the object detection task and with good generalization capability. By changing the net structures, training strategies, adding and removing some key components in the detection pipeline, a set of models with large diversity are obtained, which significantly improves the effectiveness of model averaging. The proposed approach improves the mean averaged precision obtained by RCNN [16], which was the state-of-the-art, from 31% to 50.3% on the ILSVRC2014 detection test set. It also outperforms the winner of ILSVRC2014, GoogLeNet, by 6.1%. Detailed component-wise analysis is also provided through extensive experimental evaluation, which provides a global view for people to understand the deep learning object detection pipeline.

Alcoholism Detection by Data Augmentation and Convolutional Neural Network with Stochastic Pooling.

PubMed

Wang, Shui-Hua; Lv, Yi-Ding; Sui, Yuxiu; Liu, Shuai; Wang, Su-Jing; Zhang, Yu-Dong

2017-11-17

Alcohol use disorder (AUD) is an important brain disease. It alters the brain structure. Recently, scholars tend to use computer vision based techniques to detect AUD. We collected 235 subjects, 114 alcoholic and 121 non-alcoholic. Among the 235 image, 100 images were used as training set, and data augmentation method was used. The rest 135 images were used as test set. Further, we chose the latest powerful technique-convolutional neural network (CNN) based on convolutional layer, rectified linear unit layer, pooling layer, fully connected layer, and softmax layer. We also compared three different pooling techniques: max pooling, average pooling, and stochastic pooling. The results showed that our method achieved a sensitivity of 96.88%, a specificity of 97.18%, and an accuracy of 97.04%. Our method was better than three state-of-the-art approaches. Besides, stochastic pooling performed better than other max pooling and average pooling. We validated CNN with five convolution layers and two fully connected layers performed the best. The GPU yielded a 149× acceleration in training and a 166× acceleration in test, compared to CPU.

Evolvable rough-block-based neural network and its biomedical application to hypoglycemia detection system.

PubMed

San, Phyo Phyo; Ling, Sai Ho; Nuryani; Nguyen, Hung

2014-08-01

This paper focuses on the hybridization technology using rough sets concepts and neural computing for decision and classification purposes. Based on the rough set properties, the lower region and boundary region are defined to partition the input signal to a consistent (predictable) part and an inconsistent (random) part. In this way, the neural network is designed to deal only with the boundary region, which mainly consists of an inconsistent part of applied input signal causing inaccurate modeling of the data set. Owing to different characteristics of neural network (NN) applications, the same structure of conventional NN might not give the optimal solution. Based on the knowledge of application in this paper, a block-based neural network (BBNN) is selected as a suitable classifier due to its ability to evolve internal structures and adaptability in dynamic environments. This architecture will systematically incorporate the characteristics of application to the structure of hybrid rough-block-based neural network (R-BBNN). A global training algorithm, hybrid particle swarm optimization with wavelet mutation is introduced for parameter optimization of proposed R-BBNN. The performance of the proposed R-BBNN algorithm was evaluated by an application to the field of medical diagnosis using real hypoglycemia episodes in patients with Type 1 diabetes mellitus. The performance of the proposed hybrid system has been compared with some of the existing neural networks. The comparison results indicated that the proposed method has improved classification performance and results in early convergence of the network.

Applying deep neural networks to HEP job classification

NASA Astrophysics Data System (ADS)

Wang, L.; Shi, J.; Yan, X.

2015-12-01

The cluster of IHEP computing center is a middle-sized computing system which provides 10 thousands CPU cores, 5 PB disk storage, and 40 GB/s IO throughput. Its 1000+ users come from a variety of HEP experiments. In such a system, job classification is an indispensable task. Although experienced administrator can classify a HEP job by its IO pattern, it is unpractical to classify millions of jobs manually. We present how to solve this problem with deep neural networks in a supervised learning way. Firstly, we built a training data set of 320K samples by an IO pattern collection agent and a semi-automatic process of sample labelling. Then we implemented and trained DNNs models with Torch. During the process of model training, several meta-parameters was tuned with cross-validations. Test results show that a 5- hidden-layer DNNs model achieves 96% precision on the classification task. By comparison, it outperforms a linear model by 8% precision.

Crystal surface analysis using matrix textural features classified by a probabilistic neural network

NASA Astrophysics Data System (ADS)

Sawyer, Curry R.; Quach, Viet; Nason, Donald; van den Berg, Lodewijk

1991-12-01

A system is under development in which surface quality of a growing bulk mercuric iodide crystal is monitored by video camera at regular intervals for early detection of growth irregularities. Mercuric iodide single crystals are employed in radiation detectors. A microcomputer system is used for image capture and processing. The digitized image is divided into multiple overlapping sub-images and features are extracted from each sub-image based on statistical measures of the gray tone distribution, according to the method of Haralick. Twenty parameters are derived from each sub-image and presented to a probabilistic neural network (PNN) for classification. This number of parameters was found to be optimal for the system. The PNN is a hierarchical, feed-forward network that can be rapidly reconfigured as additional training data become available. Training data is gathered by reviewing digital images of many crystals during their growth cycle and compiling two sets of images, those with and without irregularities.

A novel and generalized approach in the inversion of geoelectrical resistivity data using Artificial Neural Networks (ANN)

NASA Astrophysics Data System (ADS)

Raj, A. Stanley; Srinivas, Y.; Oliver, D. Hudson; Muthuraj, D.

2014-03-01

The non-linear apparent resistivity problem in the subsurface study of the earth takes into account the model parameters in terms of resistivity and thickness of individual subsurface layers using the trained synthetic data by means of Artificial Neural Networks (ANN). Here we used a single layer feed-forward neural network with fast back propagation learning algorithm. So on proper training of back propagation networks it tends to give the resistivity and thickness of the subsurface layer model of the field resistivity data with reference to the synthetic data trained in the appropriate network. During training, the weights and biases of the network are iteratively adjusted to make network performance function level more efficient. On adequate training, errors are minimized and the best result is obtained using the artificial neural networks. The network is trained with more number of VES data and this trained network is demonstrated by the field data. The accuracy of inversion depends upon the number of data trained. In this novel and specially designed algorithm, the interpretation of the vertical electrical sounding has been done successfully with the more accurate layer model.

Evolutionary and Neural Computing Based Decision Support System for Disease Diagnosis from Clinical Data Sets in Medical Practice.

PubMed

Sudha, M

2017-09-27

As a recent trend, various computational intelligence and machine learning approaches have been used for mining inferences hidden in the large clinical databases to assist the clinician in strategic decision making. In any target data the irrelevant information may be detrimental, causing confusion for the mining algorithm and degrades the prediction outcome. To address this issue, this study attempts to identify an intelligent approach to assist disease diagnostic procedure using an optimal set of attributes instead of all attributes present in the clinical data set. In this proposed Application Specific Intelligent Computing (ASIC) decision support system, a rough set based genetic algorithm is employed in pre-processing phase and a back propagation neural network is applied in training and testing phase. ASIC has two phases, the first phase handles outliers, noisy data, and missing values to obtain a qualitative target data to generate appropriate attribute reduct sets from the input data using rough computing based genetic algorithm centred on a relative fitness function measure. The succeeding phase of this system involves both training and testing of back propagation neural network classifier on the selected reducts. The model performance is evaluated with widely adopted existing classifiers. The proposed ASIC system for clinical decision support has been tested with breast cancer, fertility diagnosis and heart disease data set from the University of California at Irvine (UCI) machine learning repository. The proposed system outperformed the existing approaches attaining the accuracy rate of 95.33%, 97.61%, and 93.04% for breast cancer, fertility issue and heart disease diagnosis.

Biosignals learning and synthesis using deep neural networks.

PubMed

Belo, David; Rodrigues, João; Vaz, João R; Pezarat-Correia, Pedro; Gamboa, Hugo

2017-09-25

Modeling physiological signals is a complex task both for understanding and synthesize biomedical signals. We propose a deep neural network model that learns and synthesizes biosignals, validated by the morphological equivalence of the original ones. This research could lead the creation of novel algorithms for signal reconstruction in heavily noisy data and source detection in biomedical engineering field. The present work explores the gated recurrent units (GRU) employed in the training of respiration (RESP), electromyograms (EMG) and electrocardiograms (ECG). Each signal is pre-processed, segmented and quantized in a specific number of classes, corresponding to the amplitude of each sample and fed to the model, which is composed by an embedded matrix, three GRU blocks and a softmax function. This network is trained by adjusting its internal parameters, acquiring the representation of the abstract notion of the next value based on the previous ones. The simulated signal was generated by forecasting a random value and re-feeding itself. The resulting generated signals are similar with the morphological expression of the originals. During the learning process, after a set of iterations, the model starts to grasp the basic morphological characteristics of the signal and later their cyclic characteristics. After training, these models' prediction are closer to the signals that trained them, specially the RESP and ECG. This synthesis mechanism has shown relevant results that inspire the use to characterize signals from other physiological sources.

Cerebral vessels segmentation for light-sheet microscopy image using convolutional neural networks

NASA Astrophysics Data System (ADS)

Hu, Chaoen; Hui, Hui; Wang, Shuo; Dong, Di; Liu, Xia; Yang, Xin; Tian, Jie

2017-03-01

Cerebral vessel segmentation is an important step in image analysis for brain function and brain disease studies. To extract all the cerebrovascular patterns, including arteries and capillaries, some filter-based methods are used to segment vessels. However, the design of accurate and robust vessel segmentation algorithms is still challenging, due to the variety and complexity of images, especially in cerebral blood vessel segmentation. In this work, we addressed a problem of automatic and robust segmentation of cerebral micro-vessels structures in cerebrovascular images acquired by light-sheet microscope for mouse. To segment micro-vessels in large-scale image data, we proposed a convolutional neural networks (CNNs) architecture trained by 1.58 million pixels with manual label. Three convolutional layers and one fully connected layer were used in the CNNs model. We extracted a patch of size 32x32 pixels in each acquired brain vessel image as training data set to feed into CNNs for classification. This network was trained to output the probability that the center pixel of input patch belongs to vessel structures. To build the CNNs architecture, a series of mouse brain vascular images acquired from a commercial light sheet fluorescence microscopy (LSFM) system were used for training the model. The experimental results demonstrated that our approach is a promising method for effectively segmenting micro-vessels structures in cerebrovascular images with vessel-dense, nonuniform gray-level and long-scale contrast regions.

Continuous medical education (CME): do we need CME for paediatric environmental medicine in Europe?

PubMed

Boese-O'Reilly, Stephan

2007-10-01

The knowledge about the influence of environmental hazards on children's health is increasing enormously. European Ministers of Health and Environment, like many other stakeholders, identified the environmental hazards in Europe for the health of children as so serious, that they called for a "Children's Environment and Health Action Plan for Europe (CEHAPE)" approved in June 2004. The knowledge of paediatricians and other health care providers on children's health and environment in Europe is insufficient, due to the lack of training in environmental medicine for medical students, clinical trainees and postgraduates. Only continuous medical education in environmental medicine can help to fill this gap of knowledge and is thereby urgently needed. The World Health Organization developed a training package for health care providers for children's health and environment, containing excellent material for paediatric training events. The International Network on Children's Health, Environment and Safety (INCHES) developed additional training material for paediatricians within the Children's Health, Environment and Safety Training (CHEST) project. The German Network Children's Health and Environment offers training for paediatric doctors' assistants in primary prevention. To improve knowledge about children's health and environment at all levels in paediatric settings, greater efforts of national institutions, paediatric associations and other institutions are necessary. It is time to strengthen existing structures and to introduce, where necessary, new structures for training in environmental medicine.

Co-expression network analysis identified six hub genes in association with metastasis risk and prognosis in hepatocellular carcinoma

PubMed Central

Feng, Juerong; Zhou, Rui; Chang, Ying; Liu, Jing; Zhao, Qiu

2017-01-01

Hepatocellular carcinoma (HCC) has a high incidence and mortality worldwide, and its carcinogenesis and progression are influenced by a complex network of gene interactions. A weighted gene co-expression network was constructed to identify gene modules associated with the clinical traits in HCC (n = 214). Among the 13 modules, high correlation was only found between the red module and metastasis risk (classified by the HCC metastasis gene signature) (R2 = −0.74). Moreover, in the red module, 34 network hub genes for metastasis risk were identified, six of which (ABAT, AGXT, ALDH6A1, CYP4A11, DAO and EHHADH) were also hub nodes in the protein-protein interaction network of the module genes. Thus, a total of six hub genes were identified. In validation, all hub genes showed a negative correlation with the four-stage HCC progression (P for trend < 0.05) in the test set. Furthermore, in the training set, HCC samples with any hub gene lowly expressed demonstrated a higher recurrence rate and poorer survival rate (hazard ratios with 95% confidence intervals > 1). RNA-sequencing data of 142 HCC samples showed consistent results in the prognosis. Gene set enrichment analysis (GSEA) demonstrated that in the samples with any hub gene highly expressed, a total of 24 functional gene sets were enriched, most of which focused on amino acid metabolism and oxidation. In conclusion, co-expression network analysis identified six hub genes in association with HCC metastasis risk and prognosis, which might improve the prognosis by influencing amino acid metabolism and oxidation. PMID:28430663

Neural networks for data mining electronic text collections

NASA Astrophysics Data System (ADS)

Walker, Nicholas; Truman, Gregory

1997-04-01

The use of neural networks in information retrieval and text analysis has primarily suffered from the issues of adequate document representation, the ability to scale to very large collections, dynamism in the face of new information and the practical difficulties of basing the design on the use of supervised training sets. Perhaps the most important approach to begin solving these problems is the use of `intermediate entities' which reduce the dimensionality of document representations and the size of documents collections to manageable levels coupled with the use of unsupervised neural network paradigms. This paper describes the issues, a fully configured neural network-based text analysis system--dataHARVEST--aimed at data mining text collections which begins this process, along with the remaining difficulties and potential ways forward.

Testing of Haar-Like Feature in Region of Interest Detection for Automated Target Recognition (ATR) System

NASA Technical Reports Server (NTRS)

Zhang, Yuhan; Lu, Dr. Thomas

2010-01-01

The objectives of this project were to develop a ROI (Region of Interest) detector using Haar-like feature similar to the face detection in Intel's OpenCV library, implement it in Matlab code, and test the performance of the new ROI detector against the existing ROI detector that uses Optimal Trade-off Maximum Average Correlation Height filter (OTMACH). The ROI detector included 3 parts: 1, Automated Haar-like feature selection in finding a small set of the most relevant Haar-like features for detecting ROIs that contained a target. 2, Having the small set of Haar-like features from the last step, a neural network needed to be trained to recognize ROIs with targets by taking the Haar-like features as inputs. 3, using the trained neural network from the last step, a filtering method needed to be developed to process the neural network responses into a small set of regions of interests. This needed to be coded in Matlab. All the 3 parts needed to be coded in Matlab. The parameters in the detector needed to be trained by machine learning and tested with specific datasets. Since OpenCV library and Haar-like feature were not available in Matlab, the Haar-like feature calculation needed to be implemented in Matlab. The codes for Adaptive Boosting and max/min filters in Matlab could to be found from the Internet but needed to be integrated to serve the purpose of this project. The performance of the new detector was tested by comparing the accuracy and the speed of the new detector against the existing OTMACH detector. The speed was referred as the average speed to find the regions of interests in an image. The accuracy was measured by the number of false positives (false alarms) at the same detection rate between the two detectors.

A neural network approach for enhancing information extraction from multispectral image data

USGS Publications Warehouse

Liu, J.; Shao, G.; Zhu, H.; Liu, S.

2005-01-01

A back-propagation artificial neural network (ANN) was applied to classify multispectral remote sensing imagery data. The classification procedure included four steps: (i) noisy training that adds minor random variations to the sampling data to make the data more representative and to reduce the training sample size; (ii) iterative or multi-tier classification that reclassifies the unclassified pixels by making a subset of training samples from the original training set, which means the neural model can focus on fewer classes; (iii) spectral channel selection based on neural network weights that can distinguish the relative importance of each channel in the classification process to simplify the ANN model; and (iv) voting rules that adjust the accuracy of classification and produce outputs of different confidence levels. The Purdue Forest, located west of Purdue University, West Lafayette, Indiana, was chosen as the test site. The 1992 Landsat thematic mapper imagery was used as the input data. High-quality airborne photographs of the same Lime period were used for the ground truth. A total of 11 land use and land cover classes were defined, including water, broadleaved forest, coniferous forest, young forest, urban and road, and six types of cropland-grassland. The experiment, indicated that the back-propagation neural network application was satisfactory in distinguishing different land cover types at US Geological Survey levels II-III. The single-tier classification reached an overall accuracy of 85%. and the multi-tier classification an overall accuracy of 95%. For the whole test, region, the final output of this study reached an overall accuracy of 87%. ?? 2005 CASI.

On the improvement of neural cryptography using erroneous transmitted information with error prediction.

PubMed

Allam, Ahmed M; Abbas, Hazem M

2010-12-01

Neural cryptography deals with the problem of "key exchange" between two neural networks using the mutual learning concept. The two networks exchange their outputs (in bits) and the key between the two communicating parties is eventually represented in the final learned weights, when the two networks are said to be synchronized. Security of neural synchronization is put at risk if an attacker is capable of synchronizing with any of the two parties during the training process. Therefore, diminishing the probability of such a threat improves the reliability of exchanging the output bits through a public channel. The synchronization with feedback algorithm is one of the existing algorithms that enhances the security of neural cryptography. This paper proposes three new algorithms to enhance the mutual learning process. They mainly depend on disrupting the attacker confidence in the exchanged outputs and input patterns during training. The first algorithm is called "Do not Trust My Partner" (DTMP), which relies on one party sending erroneous output bits, with the other party being capable of predicting and correcting this error. The second algorithm is called "Synchronization with Common Secret Feedback" (SCSFB), where inputs are kept partially secret and the attacker has to train its network on input patterns that are different from the training sets used by the communicating parties. The third algorithm is a hybrid technique combining the features of the DTMP and SCSFB. The proposed approaches are shown to outperform the synchronization with feedback algorithm in the time needed for the parties to synchronize.

Building a virtual network in a community health research training program.

PubMed

Lau, F; Hayward, R

2000-01-01

To describe the experiences, lessons, and implications of building a virtual network as part of a two-year community health research training program in a Canadian province. An action research field study in which 25 health professionals from 17 health regions participated in a seven-week training course on health policy, management, economics, research methods, data analysis, and computer technology. The participants then returned to their regions to apply the knowledge in different community health research projects. Ongoing faculty consultations and support were provided as needed. Each participant was given a notebook computer with the necessary software, Internet access, and technical support for two years, to access information resources, engage in group problem solving, share ideas and knowledge, and collaborate on projects. Data collected over two years consisted of program documents, records of interviews with participants and staff, meeting notes, computer usage statistics, automated online surveys, computer conference postings, program Web site, and course feedback. The analysis consisted of detailed review and comparison of the data from different sources. NUD*IST was then used to validate earlier study findings. The ten key lessons are that role clarity, technology vision, implementation staging, protected time, just-in-time training, ongoing facilitation, work integration, participatory design, relationship building, and the demonstration of results are essential ingredients for building a successful network. This study provides a descriptive model of the processes involved in developing, in the community health setting, virtual networks that can be used as the basis for future research and as a practical guide for managers.

A novel end-to-end classifier using domain transferred deep convolutional neural networks for biomedical images.

PubMed

Pang, Shuchao; Yu, Zhezhou; Orgun, Mehmet A

2017-03-01

Highly accurate classification of biomedical images is an essential task in the clinical diagnosis of numerous medical diseases identified from those images. Traditional image classification methods combined with hand-crafted image feature descriptors and various classifiers are not able to effectively improve the accuracy rate and meet the high requirements of classification of biomedical images. The same also holds true for artificial neural network models directly trained with limited biomedical images used as training data or directly used as a black box to extract the deep features based on another distant dataset. In this study, we propose a highly reliable and accurate end-to-end classifier for all kinds of biomedical images via deep learning and transfer learning. We first apply domain transferred deep convolutional neural network for building a deep model; and then develop an overall deep learning architecture based on the raw pixels of original biomedical images using supervised training. In our model, we do not need the manual design of the feature space, seek an effective feature vector classifier or segment specific detection object and image patches, which are the main technological difficulties in the adoption of traditional image classification methods. Moreover, we do not need to be concerned with whether there are large training sets of annotated biomedical images, affordable parallel computing resources featuring GPUs or long times to wait for training a perfect deep model, which are the main problems to train deep neural networks for biomedical image classification as observed in recent works. With the utilization of a simple data augmentation method and fast convergence speed, our algorithm can achieve the best accuracy rate and outstanding classification ability for biomedical images. We have evaluated our classifier on several well-known public biomedical datasets and compared it with several state-of-the-art approaches. We propose a robust automated end-to-end classifier for biomedical images based on a domain transferred deep convolutional neural network model that shows a highly reliable and accurate performance which has been confirmed on several public biomedical image datasets. Copyright © 2017 Elsevier Ireland Ltd. All rights reserved.

Massive-training support vector regression and Gaussian process for false-positive reduction in computer-aided detection of polyps in CT colonography

PubMed Central

Xu, Jian-Wu; Suzuki, Kenji

2011-01-01

Purpose: A massive-training artificial neural network (MTANN) has been developed for the reduction of false positives (FPs) in computer-aided detection (CADe) of polyps in CT colonography (CTC). A major limitation of the MTANN is the long training time. To address this issue, the authors investigated the feasibility of two state-of-the-art regression models, namely, support vector regression (SVR) and Gaussian process regression (GPR) models, in the massive-training framework and developed massive-training SVR (MTSVR) and massive-training GPR (MTGPR) for the reduction of FPs in CADe of polyps. Methods: The authors applied SVR and GPR as volume-processing techniques in the distinction of polyps from FP detections in a CTC CADe scheme. Unlike artificial neural networks (ANNs), both SVR and GPR are memory-based methods that store a part of or the entire training data for testing. Therefore, their training is generally fast and they are able to improve the efficiency of the massive-training methodology. Rooted in a maximum margin property, SVR offers excellent generalization ability and robustness to outliers. On the other hand, GPR approaches nonlinear regression from a Bayesian perspective, which produces both the optimal estimated function and the covariance associated with the estimation. Therefore, both SVR and GPR, as the state-of-the-art nonlinear regression models, are able to offer a performance comparable or potentially superior to that of ANN, with highly efficient training. Both MTSVR and MTGPR were trained directly with voxel values from CTC images. A 3D scoring method based on a 3D Gaussian weighting function was applied to the outputs of MTSVR and MTGPR for distinction between polyps and nonpolyps. To test the performance of the proposed models, the authors compared them to the original MTANN in the distinction between actual polyps and various types of FPs in terms of training time reduction and FP reduction performance. The authors’ CTC database consisted of 240 CTC data sets obtained from 120 patients in the supine and prone positions. The training set consisted of 27 patients, 10 of which had 10 polyps. The authors selected 10 nonpolyps (i.e., FP sources) from the training set. These ten polyps and ten nonpolyps were used for training the proposed models. The testing set consisted of 93 patients, including 19 polyps in 7 patients and 86 negative patients with 474 FPs produced by an original CADe scheme. Results: With the MTSVR, the training time was reduced by a factor of 190, while a FP reduction performance [by-polyp sensitivity of 94.7% (18∕19) with 2.5 (230∕93) FPs∕patient] comparable to that of the original MTANN [the same sensitivity with 2.6 (244∕93) FPs∕patient] was achieved. The classification performance in terms of the area under the receiver-operating-characteristic curve value of the MTGPR (0.82) was statistically significantly higher than that of the original MTANN (0.77), with a two-sided p-value of 0.03. The MTGPR yielded a 94.7% (18∕19) by-polyp sensitivity at a FP rate of 2.5 (235∕93) per patient and reduced the training time by a factor of 1.3. Conclusions: Both MTSVR and MTGPR improve the efficiency of the training in the massive-training framework while maintaining a comparable performance. PMID:21626922

Detecting and preventing error propagation via competitive learning.

PubMed

Silva, Thiago Christiano; Zhao, Liang

2013-05-01

Semisupervised learning is a machine learning approach which is able to employ both labeled and unlabeled samples in the training process. It is an important mechanism for autonomous systems due to the ability of exploiting the already acquired information and for exploring the new knowledge in the learning space at the same time. In these cases, the reliability of the labels is a crucial factor, because mislabeled samples may propagate wrong labels to a portion of or even the entire data set. This paper has the objective of addressing the error propagation problem originated by these mislabeled samples by presenting a mechanism embedded in a network-based (graph-based) semisupervised learning method. Such a procedure is based on a combined random-preferential walk of particles in a network constructed from the input data set. The particles of the same class cooperate among them, while the particles of different classes compete with each other to propagate class labels to the whole network. Computer simulations conducted on synthetic and real-world data sets reveal the effectiveness of the model. Copyright © 2012 Elsevier Ltd. All rights reserved.

The NASA Regional Planetary Image Facility (RPIF) Network: A Key Resource for Accessing and Using Planetary Spatial Data

NASA Astrophysics Data System (ADS)

Hagerty, J. J.

2017-12-01

The role of the NASA Regional Planetary Image Facility (RPIF) Network is evolving as new science-ready spatial data products continue to be created and as key historical planetary data sets are digitized. Specifically, the RPIF Network is poised to serve specialized knowledge and services in a user-friendly manner that removes most barriers to locating, accessing, and exploiting planetary spatial data, thus providing a critical data access role within a spatial data infrastructure. The goal of the Network is to provide support and training to a broad audience of planetary spatial data users. In an effort to meet the planetary science community's evolving needs, we are focusing on the following objectives: Maintain and improve the delivery of historical data accumulated over the past four decades so as not to lose critical, historical information. This is being achieved by systematically digitizing fragile materials, allowing increased access and preserving them at the same time. Help users locate, access, visualize, and exploit planetary science data. Many of the facilities have begun to establish Guest User Facilities that allow researchers to use and/or be trained on GIS equipment and other specialized tools like Socet Set/GXP photogrammetry workstations for generating digital elevation maps. Improve the connection between the Network nodes while also leveraging the unique resources of each node. To achieve this goal, each facility is developing and sharing searchable databases of their collections, including robust metadata in a standards compliant way. Communicate more effectively and regularly with the planetary science community in an effort to make potential users aware of resources and services provided by the Network, while also engaging community members in discussions about community needs. Provide a regional resource for the science community, colleges, universities, museums, media, and the public to access planetary data. Introduce new strategies for visualizing planetary data and products (e.g., 3D printing and virtual reality platforms/experiences). We anticipate that in a few years virtual reality tools will be an integral part of data analysis, providing more intuitive understanding of multiple complex data sets.

Low-cost autonomous perceptron neural network inspired by quantum computation

NASA Astrophysics Data System (ADS)

Zidan, Mohammed; Abdel-Aty, Abdel-Haleem; El-Sadek, Alaa; Zanaty, E. A.; Abdel-Aty, Mahmoud

2017-11-01

Achieving low cost learning with reliable accuracy is one of the important goals to achieve intelligent machines to save time, energy and perform learning process over limited computational resources machines. In this paper, we propose an efficient algorithm for a perceptron neural network inspired by quantum computing composite from a single neuron to classify inspirable linear applications after a single training iteration O(1). The algorithm is applied over a real world data set and the results are outer performs the other state-of-the art algorithms.

Data fusion with artificial neural networks (ANN) for classification of earth surface from microwave satellite measurements

NASA Technical Reports Server (NTRS)

Lure, Y. M. Fleming; Grody, Norman C.; Chiou, Y. S. Peter; Yeh, H. Y. Michael

1993-01-01

A data fusion system with artificial neural networks (ANN) is used for fast and accurate classification of five earth surface conditions and surface changes, based on seven SSMI multichannel microwave satellite measurements. The measurements include brightness temperatures at 19, 22, 37, and 85 GHz at both H and V polarizations (only V at 22 GHz). The seven channel measurements are processed through a convolution computation such that all measurements are located at same grid. Five surface classes including non-scattering surface, precipitation over land, over ocean, snow, and desert are identified from ground-truth observations. The system processes sensory data in three consecutive phases: (1) pre-processing to extract feature vectors and enhance separability among detected classes; (2) preliminary classification of Earth surface patterns using two separate and parallely acting classifiers: back-propagation neural network and binary decision tree classifiers; and (3) data fusion of results from preliminary classifiers to obtain the optimal performance in overall classification. Both the binary decision tree classifier and the fusion processing centers are implemented by neural network architectures. The fusion system configuration is a hierarchical neural network architecture, in which each functional neural net will handle different processing phases in a pipelined fashion. There is a total of around 13,500 samples for this analysis, of which 4 percent are used as the training set and 96 percent as the testing set. After training, this classification system is able to bring up the detection accuracy to 94 percent compared with 88 percent for back-propagation artificial neural networks and 80 percent for binary decision tree classifiers. The neural network data fusion classification is currently under progress to be integrated in an image processing system at NOAA and to be implemented in a prototype of a massively parallel and dynamically reconfigurable Modular Neural Ring (MNR).

Satisfaction with web-based training in an integrated healthcare delivery network: do age, education, computer skills and attitudes matter?

PubMed Central

Atreja, Ashish; Mehta, Neil B; Jain, Anil K; Harris, CM; Ishwaran, Hemant; Avital, Michel; Fishleder, Andrew J

2008-01-01

Background Healthcare institutions spend enormous time and effort to train their workforce. Web-based training can potentially streamline this process. However the deployment of web-based training in a large-scale setting with a diverse healthcare workforce has not been evaluated. The aim of this study was to evaluate the satisfaction of healthcare professionals with web-based training and to determine the predictors of such satisfaction including age, education status and computer proficiency. Methods Observational, cross-sectional survey of healthcare professionals from six hospital systems in an integrated delivery network. We measured overall satisfaction to web-based training and response to survey items measuring Website Usability, Course Usefulness, Instructional Design Effectiveness, Computer Proficiency and Self-learning Attitude. Results A total of 17,891 healthcare professionals completed the web-based training on HIPAA Privacy Rule; and of these, 13,537 completed the survey (response rate 75.6%). Overall course satisfaction was good (median, 4; scale, 1 to 5) with more than 75% of the respondents satisfied with the training (rating 4 or 5) and 65% preferring web-based training over traditional instructor-led training (rating 4 or 5). Multivariable ordinal regression revealed 3 key predictors of satisfaction with web-based training: Instructional Design Effectiveness, Website Usability and Course Usefulness. Demographic predictors such as gender, age and education did not have an effect on satisfaction. Conclusion The study shows that web-based training when tailored to learners' background, is perceived as a satisfactory mode of learning by an interdisciplinary group of healthcare professionals, irrespective of age, education level or prior computer experience. Future studies should aim to measure the long-term outcomes of web-based training. PMID:18922178

Single-hidden-layer feed-forward quantum neural network based on Grover learning.

PubMed

Liu, Cheng-Yi; Chen, Chein; Chang, Ching-Ter; Shih, Lun-Min

2013-09-01

In this paper, a novel single-hidden-layer feed-forward quantum neural network model is proposed based on some concepts and principles in the quantum theory. By combining the quantum mechanism with the feed-forward neural network, we defined quantum hidden neurons and connected quantum weights, and used them as the fundamental information processing unit in a single-hidden-layer feed-forward neural network. The quantum neurons make a wide range of nonlinear functions serve as the activation functions in the hidden layer of the network, and the Grover searching algorithm outstands the optimal parameter setting iteratively and thus makes very efficient neural network learning possible. The quantum neuron and weights, along with a Grover searching algorithm based learning, result in a novel and efficient neural network characteristic of reduced network, high efficient training and prospect application in future. Some simulations are taken to investigate the performance of the proposed quantum network and the result show that it can achieve accurate learning. Copyright © 2013 Elsevier Ltd. All rights reserved.

Segmentation of skin lesions in chronic graft versus host disease photographs with fully convolutional networks

NASA Astrophysics Data System (ADS)

Wang, Jianing; Chen, Fuyao; Dellalana, Laura E.; Jagasia, Madan H.; Tkaczyk, Eric R.; Dawant, Benoit M.

2018-02-01

Chronic graft-versus-host disease (cGVHD) is a frequent and potentially life-threatening complication of allogeneic hematopoietic stem cell transplantation (HCT) and commonly affects the skin, resulting in distressing patient morbidity. The percentage of involved body surface area (BSA) is commonly used for diagnosing and scoring the severity of cGVHD. However, the segmentation of the involved BSA from patient whole body serial photography is challenging because (1) it is difficult to design traditional segmentation method that rely on hand crafted features as the appearance of cGVHD lesions can be drastically different from patient to patient; (2) to the best of our knowledge, currently there is no publicavailable labelled image set of cGVHD skin for training deep networks to segment the involved BSA. In this preliminary study we create a small labelled image set of skin cGVHD, and we explore the possibility to use a fully convolutional neural network (FCN) to segment the skin lesion in the images. We use a commercial stereoscopic Vectra H1 camera (Canfield Scientific) to acquire 400 3D photographs of 17 cGVHD patients aged between 22 and 72. A rotational data augmentation process is then applied, which rotates the 3D photos through 10 predefined angles, producing one 2D projection image at each position. This results in 4000 2D images that constitute our cGVHD image set. A FCN model is trained and tested using our images. We show that our method achieves encouraging results for segmenting cGVHD skin lesion in photographic images.

Broadband seismic effects from train vibrations

NASA Astrophysics Data System (ADS)

Fuchs, Florian; Bokelmann, Götz

2017-04-01

Seismologists rarely study train induced vibrations which are mainly regarded an unwanted source of noise for classical seismological applications such as earthquake monitoring. A few seismological studies try to utilize train vibrations however as active sources, e.g. for subsurface imaging, but they do not focus on the characteristics of the train signal itself. Most available studies on train induced vibrations take an engineering approach and aim at better understanding the generation and short-distance propagation of train induced vibrations, mainly for mitigation and construction purposes. They mostly rely on numerical simulations and/or short-period or accelerometer recordings obtained directly on the train track or up to few hundred meters away and almost no studies exist with seismic recordings further away from the track. In some of these previous studies sharp and equidistant peaks are present in the vibration spectrum of heavy freight trains, but they do not attempt to explain them. Here we show and analyze various train vibration signals obtained from a set of seismic broadband stations installed in the context of the temporary, large-scale regional seismic network AlpArray. The geometrical restrictions of this seismic network combined with budget and safety considerations resulted in a number of broad-band instruments deployed in the vicinity of busy railway lines. On these stations we observe very characteristic seismic signals associated with different types of trains, typically showing pronounced equidistant spectral lines over a wide frequency range. In this study we analyze the nature of such signals and discuss if they are generated by a source effect or by wave propagation effects in near-surface soil layers.

Improving Suicide Prevention in Dutch Regions by Creating Local Suicide Prevention Action Networks (SUPRANET): A Study Protocol.

PubMed

Gilissen, Renske; De Beurs, Derek; Mokkenstorm, Jan; Mérelle, Saskia; Donker, Gé; Terpstra, Sanne; Derijck, Carla; Franx, Gerdien

2017-03-28

The European Alliance against Depression (EAAD) program is to be introduced in The Netherlands from 2017 onwards. This program to combat suicide consists of interventions on four levels: (1) increasing the awareness of suicide by local media campaigns; (2) training local gatekeepers, such as teachers or police officers; (3) targeting high-risk persons in the community; and (4) training and support of professionals in primary care settings. The implementation starts in seven Dutch pilot regions. Each region is designated as a Suicide Prevention Action NETwork (SUPRANET). This paper describes the SUPRANET program components and the evaluation of its feasibility and impact. The findings will be used to facilitate the national implementation of EAAD in The Netherlands and to add new findings to the existing literature on EAAD.

Improving Suicide Prevention in Dutch Regions by Creating Local Suicide Prevention Action Networks (SUPRANET): A Study Protocol

PubMed Central

Gilissen, Renske; De Beurs, Derek; Mokkenstorm, Jan; Mérelle, Saskia; Donker, Gé; Terpstra, Sanne; Derijck, Carla; Franx, Gerdien

2017-01-01

The European Alliance against Depression (EAAD) program is to be introduced in The Netherlands from 2017 onwards. This program to combat suicide consists of interventions on four levels: (1) increasing the awareness of suicide by local media campaigns; (2) training local gatekeepers, such as teachers or police officers; (3) targeting high-risk persons in the community; and (4) training and support of professionals in primary care settings. The implementation starts in seven Dutch pilot regions. Each region is designated as a Suicide Prevention Action NETwork (SUPRANET). This paper describes the SUPRANET program components and the evaluation of its feasibility and impact. The findings will be used to facilitate the national implementation of EAAD in The Netherlands and to add new findings to the existing literature on EAAD. PMID:28350367

Resource constrained design of artificial neural networks using comparator neural network

NASA Technical Reports Server (NTRS)

Wah, Benjamin W.; Karnik, Tanay S.

1992-01-01

We present a systematic design method executed under resource constraints for automating the design of artificial neural networks using the back error propagation algorithm. Our system aims at finding the best possible configuration for solving the given application with proper tradeoff between the training time and the network complexity. The design of such a system is hampered by three related problems. First, there are infinitely many possible network configurations, each may take an exceedingly long time to train; hence, it is impossible to enumerate and train all of them to completion within fixed time, space, and resource constraints. Second, expert knowledge on predicting good network configurations is heuristic in nature and is application dependent, rendering it difficult to characterize fully in the design process. A learning procedure that refines this knowledge based on examples on training neural networks for various applications is, therefore, essential. Third, the objective of the network to be designed is ill-defined, as it is based on a subjective tradeoff between the training time and the network cost. A design process that proposes alternate configurations under different cost-performance tradeoff is important. We have developed a Design System which schedules the available time, divided into quanta, for testing alternative network configurations. Its goal is to select/generate and test alternative network configurations in each quantum, and find the best network when time is expended. Since time is limited, a dynamic schedule that determines the network configuration to be tested in each quantum is developed. The schedule is based on relative comparison of predicted training times of alternative network configurations using comparator network paradigm. The comparator network has been trained to compare training times for a large variety of traces of TSSE-versus-time collected during back-propagation learning of various applications.

Social Networking Web Sites: Teaching Appropriate Social Competence to Students with Emotional and Behavioral Disorders

ERIC Educational Resources Information Center

Morgan, Joseph J.

2010-01-01

The Internet has opened a variety of different avenues for people to interact with each other. As new digital environments are developed, new sets of social skills are needed to appropriately interact. Students with emotional and behavioral disorders often have deficits in social competence and require specialized training in specific social…

Practice-based research networks, part II: a descriptive analysis of the athletic training practice-based research network in the secondary school setting.

PubMed

Valovich McLeod, Tamara C; Lam, Kenneth C; Bay, R Curtis; Sauers, Eric L; Snyder Valier, Alison R

2012-01-01

Analysis of health care service models requires the collection and evaluation of basic practice characterization data. Practice-based research networks (PBRNs) provide a framework for gathering data useful in characterizing clinical practice. To describe preliminary secondary school setting practice data from the Athletic Training Practice-Based Research Network (AT-PBRN). Descriptive study. Secondary school athletic training facilities within the AT-PBRN. Clinicians (n = 22) and their patients (n = 2523) from the AT-PBRN. A Web-based survey was used to obtain data on clinical practice site and clinician characteristics. Patient and practice characteristics were obtained via deidentified electronic medical record data collected between September 1, 2009, and April 1, 2011. Descriptive data regarding the clinician and CPS practice characteristics are reported as percentages and frequencies. Descriptive analysis of patient encounters and practice characteristic data was performed, with the percentages and frequencies of the type of injuries recorded at initial evaluation, type of treatment received at initial evaluation, daily treatment, and daily sign-in procedures. The AT-PBRN had secondary school sites in 7 states, and most athletic trainers at those sites (78.2%) had less than 5 years of experience. The secondary school sites within the AT-PBRN documented 2523 patients treated across 3140 encounters. Patients most frequently sought care for a current injury (61.3%), followed by preventive services (24.0%), and new injuries (14.7%). The most common diagnoses were ankle sprain/strain (17.9%), hip sprain/strain (12.5%), concussion (12.0%), and knee pain (2.5%). The most frequent procedures were athletic trainer evaluation (53.9%), hot- or cold-pack application (26.0%), strapping (10.3%), and therapeutic exercise (5.7%). The median number of treatments per injury was 3 (interquartile range = 2, 4; range = 2-19). These preliminary data describe services provided by clinicians within the AT-PBRN and demonstrate the usefulness of the PBRN model for obtaining such data.

Statistical generation of training sets for measuring NO3(-), NH4(+) and major ions in natural waters using an ion selective electrode array.

PubMed

Mueller, Amy V; Hemond, Harold F

2016-05-18

Knowledge of ionic concentrations in natural waters is essential to understand watershed processes. Inorganic nitrogen, in the form of nitrate and ammonium ions, is a key nutrient as well as a participant in redox, acid-base, and photochemical processes of natural waters, leading to spatiotemporal patterns of ion concentrations at scales as small as meters or hours. Current options for measurement in situ are costly, relying primarily on instruments adapted from laboratory methods (e.g., colorimetric, UV absorption); free-standing and inexpensive ISE sensors for NO3(-) and NH4(+) could be attractive alternatives if interferences from other constituents were overcome. Multi-sensor arrays, coupled with appropriate non-linear signal processing, offer promise in this capacity but have not yet successfully achieved signal separation for NO3(-) and NH4(+)in situ at naturally occurring levels in unprocessed water samples. A novel signal processor, underpinned by an appropriate sensor array, is proposed that overcomes previous limitations by explicitly integrating basic chemical constraints (e.g., charge balance). This work further presents a rationalized process for the development of such in situ instrumentation for NO3(-) and NH4(+), including a statistical-modeling strategy for instrument design, training/calibration, and validation. Statistical analysis reveals that historical concentrations of major ionic constituents in natural waters across New England strongly covary and are multi-modal. This informs the design of a statistically appropriate training set, suggesting that the strong covariance of constituents across environmental samples can be exploited through appropriate signal processing mechanisms to further improve estimates of minor constituents. Two artificial neural network architectures, one expanded to incorporate knowledge of basic chemical constraints, were tested to process outputs of a multi-sensor array, trained using datasets of varying degrees of statistical representativeness to natural water samples. The accuracy of ANN results improves monotonically with the statistical representativeness of the training set (error decreases by ∼5×), while the expanded neural network architecture contributes a further factor of 2-3.5 decrease in error when trained with the most representative sample set. Results using the most statistically accurate set of training samples (which retain environmentally relevant ion concentrations but avoid the potential interference of humic acids) demonstrated accurate, unbiased quantification of nitrate and ammonium at natural environmental levels (±20% down to <10 μM), as well as the major ions Na(+), K(+), Ca(2+), Mg(2+), Cl(-), and SO4(2-), in unprocessed samples. These results show promise for the development of new in situ instrumentation for the support of scientific field work.

Modified-hybrid optical neural network filter for multiple object recognition within cluttered scenes

NASA Astrophysics Data System (ADS)

Kypraios, Ioannis; Young, Rupert C. D.; Chatwin, Chris R.

2009-08-01

Motivated by the non-linear interpolation and generalization abilities of the hybrid optical neural network filter between the reference and non-reference images of the true-class object we designed the modifiedhybrid optical neural network filter. We applied an optical mask to the hybrid optical neural network's filter input. The mask was built with the constant weight connections of a randomly chosen image included in the training set. The resulted design of the modified-hybrid optical neural network filter is optimized for performing best in cluttered scenes of the true-class object. Due to the shift invariance properties inherited by its correlator unit the filter can accommodate multiple objects of the same class to be detected within an input cluttered image. Additionally, the architecture of the neural network unit of the general hybrid optical neural network filter allows the recognition of multiple objects of different classes within the input cluttered image by modifying the output layer of the unit. We test the modified-hybrid optical neural network filter for multiple objects of the same and of different classes' recognition within cluttered input images and video sequences of cluttered scenes. The filter is shown to exhibit with a single pass over the input data simultaneously out-of-plane rotation, shift invariance and good clutter tolerance. It is able to successfully detect and classify correctly the true-class objects within background clutter for which there has been no previous training.

Modeling and control of magnetorheological fluid dampers using neural networks

NASA Astrophysics Data System (ADS)

Wang, D. H.; Liao, W. H.

2005-02-01

Due to the inherent nonlinear nature of magnetorheological (MR) fluid dampers, one of the challenging aspects for utilizing these devices to achieve high system performance is the development of accurate models and control algorithms that can take advantage of their unique characteristics. In this paper, the direct identification and inverse dynamic modeling for MR fluid dampers using feedforward and recurrent neural networks are studied. The trained direct identification neural network model can be used to predict the damping force of the MR fluid damper on line, on the basis of the dynamic responses across the MR fluid damper and the command voltage, and the inverse dynamic neural network model can be used to generate the command voltage according to the desired damping force through supervised learning. The architectures and the learning methods of the dynamic neural network models and inverse neural network models for MR fluid dampers are presented, and some simulation results are discussed. Finally, the trained neural network models are applied to predict and control the damping force of the MR fluid damper. Moreover, validation methods for the neural network models developed are proposed and used to evaluate their performance. Validation results with different data sets indicate that the proposed direct identification dynamic model using the recurrent neural network can be used to predict the damping force accurately and the inverse identification dynamic model using the recurrent neural network can act as a damper controller to generate the command voltage when the MR fluid damper is used in a semi-active mode.

An evaluation of the American Indian Air Quality Training Program

NASA Astrophysics Data System (ADS)

Quartaroli, Marylynn

During centuries of geographic, economic, and cultural domination, the federal government held the responsibility for the management of environmental issues on tribal lands. Today, tribes are reasserting their sovereignty in many ways, including the development of their own environmental programs. Tribal agencies desperately search for tribal members who are qualified to make decisions for the benefit of the tribes from both Western scientific and traditional cultural viewpoints. To meet this need, the American Indian Air Quality Training Program (AIAQTP) offers technical and regulatory training courses that are both scientifically up-to-date and culturally responsive to this community. This study is an evaluation of these courses. To supplement data from existing program documents and databases, I also observed five courses, sent follow-up questionnaires, and interviewed lead instructors and course participants to develop an understanding of their perceptions of the training received. Computer analysis of this quantitative and qualitative data revealed patterns and themes; an external reviewer also independently analyzed the data set. The training courses offered by AIAQTP were judged to have merit and value by the course instructors, the participants, the external evaluator, and me. Designed to be both culturally responsive and technically rigorous, these courses provided relevant and useful information and skills to the tribal environmental professionals in attendance, meeting the demands of their jobs. Although not all training needs or expectations were met, the study participants indicated their intentions to continue their education and training in air quality and other environmental media. A significant benefit of attendance at AIAQTP training courses was the development of a network of tribal professionals across the nation that acts as a support system for the implementation and continuation of changes in the professional practice for the trainees and their tribal environmental programs. As in all educational settings, there is room for improvement in course design and delivery. This study suggested that incorporating cultural considerations plays a role in effective scientific and technical training for Native students. Further, actively promoting the development of professional networks can assist in establishing a geographically dispersed community of practice supportive of personal and institutional change.

NNAlign: a platform to construct and evaluate artificial neural network models of receptor-ligand interactions.

PubMed

Nielsen, Morten; Andreatta, Massimo

2017-07-03

Peptides are extensively used to characterize functional or (linear) structural aspects of receptor-ligand interactions in biological systems, e.g. SH2, SH3, PDZ peptide-recognition domains, the MHC membrane receptors and enzymes such as kinases and phosphatases. NNAlign is a method for the identification of such linear motifs in biological sequences. The algorithm aligns the amino acid or nucleotide sequences provided as training set, and generates a model of the sequence motif detected in the data. The webserver allows setting up cross-validation experiments to estimate the performance of the model, as well as evaluations on independent data. Many features of the training sequences can be encoded as input, and the network architecture is highly customizable. The results returned by the server include a graphical representation of the motif identified by the method, performance values and a downloadable model that can be applied to scan protein sequences for occurrence of the motif. While its performance for the characterization of peptide-MHC interactions is widely documented, we extended NNAlign to be applicable to other receptor-ligand systems as well. Version 2.0 supports alignments with insertions and deletions, encoding of receptor pseudo-sequences, and custom alphabets for the training sequences. The server is available at http://www.cbs.dtu.dk/services/NNAlign-2.0. © The Author(s) 2017. Published by Oxford University Press on behalf of Nucleic Acids Research.

Reinforced Adversarial Neural Computer for de Novo Molecular Design.

PubMed

Putin, Evgeny; Asadulaev, Arip; Ivanenkov, Yan; Aladinskiy, Vladimir; Sanchez-Lengeling, Benjamin; Aspuru-Guzik, Alán; Zhavoronkov, Alex

2018-06-12

In silico modeling is a crucial milestone in modern drug design and development. Although computer-aided approaches in this field are well-studied, the application of deep learning methods in this research area is at the beginning. In this work, we present an original deep neural network (DNN) architecture named RANC (Reinforced Adversarial Neural Computer) for the de novo design of novel small-molecule organic structures based on the generative adversarial network (GAN) paradigm and reinforcement learning (RL). As a generator RANC uses a differentiable neural computer (DNC), a category of neural networks, with increased generation capabilities due to the addition of an explicit memory bank, which can mitigate common problems found in adversarial settings. The comparative results have shown that RANC trained on the SMILES string representation of the molecules outperforms its first DNN-based counterpart ORGANIC by several metrics relevant to drug discovery: the number of unique structures, passing medicinal chemistry filters (MCFs), Muegge criteria, and high QED scores. RANC is able to generate structures that match the distributions of the key chemical features/descriptors (e.g., MW, logP, TPSA) and lengths of the SMILES strings in the training data set. Therefore, RANC can be reasonably regarded as a promising starting point to develop novel molecules with activity against different biological targets or pathways. In addition, this approach allows scientists to save time and covers a broad chemical space populated with novel and diverse compounds.

Detection of high-grade atypia nuclei in breast cancer imaging

NASA Astrophysics Data System (ADS)

Noël, Henri; Roux, Ludovic; Lu, Shijian; Boudier, Thomas

2015-03-01

Along with mitotic count, nuclear pleomorphism or nuclear atypia is an important criterion for the grading of breast cancer in histopathology. Though some works have been done in mitosis detection (ICPR 2012,1 MICCAI 2013,2 and ICPR 2014), not much work has been dedicated to automated nuclear atypia grading, especially the most difficult task of detection of grade 3 nuclei. We propose the use of Convolutional Neural Networks for the automated detection of cell nuclei, using images from the three grades of breast cancer for training. The images were obtained from ICPR contests. Additional manual annotation was performed to classify pixels into five classes: stroma, nuclei, lymphocytes, mitosis and fat. At total of 3,000 thumbnail images of 101 × 101 pixels were used for training. By dividing this training set in an 80/20 ratio we could obtain good training results (around 90%). We tested our CNN on images of the three grades which were not in the training set. High grades nuclei were correctly classified. We then thresholded the classification map and performed basic analysis to keep only rounded objects. Our results show that mostly all atypical nuclei were correctly detected.

Fault zone identification in the eastern part of the Persian Gulf based on combined seismic attributes

NASA Astrophysics Data System (ADS)

Mirkamali, M. S.; Keshavarz FK, N.; Bakhtiari, M. R.

2013-02-01

Faults, as main pathways for fluids, play a critical role in creating regions of high porosity and permeability, in cutting cap rock and in the migration of hydrocarbons into the reservoir. Therefore, accurate identification of fault zones is very important in maximizing production from petroleum traps. Image processing and modern visualization techniques are provided for better mapping of objects of interest. In this study, the application of fault mapping in the identification of fault zones within the Mishan and Aghajari formations above the Guri base unconformity surface in the eastern part of Persian Gulf is investigated. Seismic single- and multi-trace attribute analyses are employed separately to determine faults in a vertical section, but different kinds of geological objects cannot be identified using individual attributes only. A mapping model is utilized to improve the identification of the faults, giving more accurate results. This method is based on combinations of all individual relevant attributes using a neural network system to create combined attributes, which gives an optimal view of the object of interest. Firstly, a set of relevant attributes were separately calculated on the vertical section. Then, at interpreted positions, some example training locations were manually selected in each fault and non-fault class by an interpreter. A neural network was trained on combinations of the attributes extracted at the example training locations to generate an optimized fault cube. Finally, the results of the fault and nonfault probability cube were estimated, which the neural network applied to the entire data set. The fault probability cube was obtained with higher mapping accuracy and greater contrast, and with fewer disturbances in comparison with individual attributes. The computed results of this study can support better understanding of the data, providing fault zone mapping with reliable results.

Fusion of shallow and deep features for classification of high-resolution remote sensing images

NASA Astrophysics Data System (ADS)

Gao, Lang; Tian, Tian; Sun, Xiao; Li, Hang

2018-02-01

Effective spectral and spatial pixel description plays a significant role for the classification of high resolution remote sensing images. Current approaches of pixel-based feature extraction are of two main kinds: one includes the widelyused principal component analysis (PCA) and gray level co-occurrence matrix (GLCM) as the representative of the shallow spectral and shape features, and the other refers to the deep learning-based methods which employ deep neural networks and have made great promotion on classification accuracy. However, the former traditional features are insufficient to depict complex distribution of high resolution images, while the deep features demand plenty of samples to train the network otherwise over fitting easily occurs if only limited samples are involved in the training. In view of the above, we propose a GLCM-based convolution neural network (CNN) approach to extract features and implement classification for high resolution remote sensing images. The employment of GLCM is able to represent the original images and eliminate redundant information and undesired noises. Meanwhile, taking shallow features as the input of deep network will contribute to a better guidance and interpretability. In consideration of the amount of samples, some strategies such as L2 regularization and dropout methods are used to prevent over-fitting. The fine-tuning strategy is also used in our study to reduce training time and further enhance the generalization performance of the network. Experiments with popular data sets such as PaviaU data validate that our proposed method leads to a performance improvement compared to individual involved approaches.

Force Field for Water Based on Neural Network.

PubMed

Wang, Hao; Yang, Weitao

2018-05-18

We developed a novel neural network based force field for water based on training with high level ab initio theory. The force field was built based on electrostatically embedded many-body expansion method truncated at binary interactions. Many-body expansion method is a common strategy to partition the total Hamiltonian of large systems into a hierarchy of few-body terms. Neural networks were trained to represent electrostatically embedded one-body and two-body interactions, which require as input only one and two water molecule calculations at the level of ab initio electronic structure method CCSD/aug-cc-pVDZ embedded in the molecular mechanics water environment, making it efficient as a general force field construction approach. Structural and dynamic properties of liquid water calculated with our force field show good agreement with experimental results. We constructed two sets of neural network based force fields: non-polarizable and polarizable force fields. Simulation results show that the non-polarizable force field using fixed TIP3P charges has already behaved well, since polarization effects and many-body effects are implicitly included due to the electrostatic embedding scheme. Our results demonstrate that the electrostatically embedded many-body expansion combined with neural network provides a promising and systematic way to build the next generation force fields at high accuracy and low computational costs, especially for large systems.

Correcting wave predictions with artificial neural networks

NASA Astrophysics Data System (ADS)

Makarynskyy, O.; Makarynska, D.

2003-04-01

The predictions of wind waves with different lead times are necessary in a large scope of coastal and open ocean activities. Numerical wave models, which usually provide this information, are based on deterministic equations that do not entirely account for the complexity and uncertainty of the wave generation and dissipation processes. An attempt to improve wave parameters short-term forecasts based on artificial neural networks is reported. In recent years, artificial neural networks have been used in a number of coastal engineering applications due to their ability to approximate the nonlinear mathematical behavior without a priori knowledge of interrelations among the elements within a system. The common multilayer feed-forward networks, with a nonlinear transfer functions in the hidden layers, were developed and employed to forecast the wave characteristics over one hour intervals starting from one up to 24 hours, and to correct these predictions. Three non-overlapping data sets of wave characteristics, both from a buoy, moored roughly 60 miles west of the Aran Islands, west coast of Ireland, were used to train and validate the neural nets involved. The networks were trained with error back propagation algorithm. Time series plots and scatterplots of the wave characteristics as well as tables with statistics show an improvement of the results achieved due to the correction procedure employed.

Design of microstrip patch antennas using knowledge insertion through retraining

NASA Astrophysics Data System (ADS)

Divakar, T. V. S.; Sudhakar, A.

2018-04-01

The traditional way of analyzing/designing neural network is to collect experimental data and train neural network. Then, the trained neural network acts as global approximate function. The network is then used to calculate parameters for unknown configurations. The main drawback of this method is one does not have enough experimental data, cost of prototypes being a major factor [1-4]. Therefore, in this method the author collected training data from available approximate formulas with in full design range and trained the network with it. After successful training, the network is retrained with available measured results. This simple way inserts experimental knowledge into the network [5]. This method is tested for rectangular microstrip antenna and circular microstrip antenna.