3D Imaging of Microbial Biofilms: Integration of Synchrotron Imaging and an Interactive Visualization Interface

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

Thomas, Mathew; Marshall, Matthew J.; Miller, Erin A.

2014-08-26

Understanding the interactions of structured communities known as “biofilms” and other complex matrixes is possible through the X-ray micro tomography imaging of the biofilms. Feature detection and image processing for this type of data focuses on efficiently identifying and segmenting biofilms and bacteria in the datasets. The datasets are very large and often require manual interventions due to low contrast between objects and high noise levels. Thus new software is required for the effectual interpretation and analysis of the data. This work specifies the evolution and application of the ability to analyze and visualize high resolution X-ray micro tomography datasets.

A large dataset of synthetic SEM images of powder materials and their ground truth 3D structures.

PubMed

DeCost, Brian L; Holm, Elizabeth A

2016-12-01

This data article presents a data set comprised of 2048 synthetic scanning electron microscope (SEM) images of powder materials and descriptions of the corresponding 3D structures that they represent. These images were created using open source rendering software, and the generating scripts are included with the data set. Eight particle size distributions are represented with 256 independent images from each. The particle size distributions are relatively similar to each other, so that the dataset offers a useful benchmark to assess the fidelity of image analysis techniques. The characteristics of the PSDs and the resulting images are described and analyzed in more detail in the research article "Characterizing powder materials using keypoint-based computer vision methods" (B.L. DeCost, E.A. Holm, 2016) [1]. These data are freely available in a Mendeley Data archive "A large dataset of synthetic SEM images of powder materials and their ground truth 3D structures" (B.L. DeCost, E.A. Holm, 2016) located at http://dx.doi.org/10.17632/tj4syyj9mr.1[2] for any academic, educational, or research purposes.

Image Harvest: an open-source platform for high-throughput plant image processing and analysis.

PubMed

Knecht, Avi C; Campbell, Malachy T; Caprez, Adam; Swanson, David R; Walia, Harkamal

2016-05-01

High-throughput plant phenotyping is an effective approach to bridge the genotype-to-phenotype gap in crops. Phenomics experiments typically result in large-scale image datasets, which are not amenable for processing on desktop computers, thus creating a bottleneck in the image-analysis pipeline. Here, we present an open-source, flexible image-analysis framework, called Image Harvest (IH), for processing images originating from high-throughput plant phenotyping platforms. Image Harvest is developed to perform parallel processing on computing grids and provides an integrated feature for metadata extraction from large-scale file organization. Moreover, the integration of IH with the Open Science Grid provides academic researchers with the computational resources required for processing large image datasets at no cost. Image Harvest also offers functionalities to extract digital traits from images to interpret plant architecture-related characteristics. To demonstrate the applications of these digital traits, a rice (Oryza sativa) diversity panel was phenotyped and genome-wide association mapping was performed using digital traits that are used to describe different plant ideotypes. Three major quantitative trait loci were identified on rice chromosomes 4 and 6, which co-localize with quantitative trait loci known to regulate agronomically important traits in rice. Image Harvest is an open-source software for high-throughput image processing that requires a minimal learning curve for plant biologists to analyzephenomics datasets. © The Author 2016. Published by Oxford University Press on behalf of the Society for Experimental Biology.

CIFAR10-DVS: An Event-Stream Dataset for Object Classification

PubMed Central

Li, Hongmin; Liu, Hanchao; Ji, Xiangyang; Li, Guoqi; Shi, Luping

2017-01-01

Neuromorphic vision research requires high-quality and appropriately challenging event-stream datasets to support continuous improvement of algorithms and methods. However, creating event-stream datasets is a time-consuming task, which needs to be recorded using the neuromorphic cameras. Currently, there are limited event-stream datasets available. In this work, by utilizing the popular computer vision dataset CIFAR-10, we converted 10,000 frame-based images into 10,000 event streams using a dynamic vision sensor (DVS), providing an event-stream dataset of intermediate difficulty in 10 different classes, named as “CIFAR10-DVS.” The conversion of event-stream dataset was implemented by a repeated closed-loop smooth (RCLS) movement of frame-based images. Unlike the conversion of frame-based images by moving the camera, the image movement is more realistic in respect of its practical applications. The repeated closed-loop image movement generates rich local intensity changes in continuous time which are quantized by each pixel of the DVS camera to generate events. Furthermore, a performance benchmark in event-driven object classification is provided based on state-of-the-art classification algorithms. This work provides a large event-stream dataset and an initial benchmark for comparison, which may boost algorithm developments in even-driven pattern recognition and object classification. PMID:28611582

CIFAR10-DVS: An Event-Stream Dataset for Object Classification.

PubMed

Li, Hongmin; Liu, Hanchao; Ji, Xiangyang; Li, Guoqi; Shi, Luping

2017-01-01

Neuromorphic vision research requires high-quality and appropriately challenging event-stream datasets to support continuous improvement of algorithms and methods. However, creating event-stream datasets is a time-consuming task, which needs to be recorded using the neuromorphic cameras. Currently, there are limited event-stream datasets available. In this work, by utilizing the popular computer vision dataset CIFAR-10, we converted 10,000 frame-based images into 10,000 event streams using a dynamic vision sensor (DVS), providing an event-stream dataset of intermediate difficulty in 10 different classes, named as "CIFAR10-DVS." The conversion of event-stream dataset was implemented by a repeated closed-loop smooth (RCLS) movement of frame-based images. Unlike the conversion of frame-based images by moving the camera, the image movement is more realistic in respect of its practical applications. The repeated closed-loop image movement generates rich local intensity changes in continuous time which are quantized by each pixel of the DVS camera to generate events. Furthermore, a performance benchmark in event-driven object classification is provided based on state-of-the-art classification algorithms. This work provides a large event-stream dataset and an initial benchmark for comparison, which may boost algorithm developments in even-driven pattern recognition and object classification.

High-throughput Analysis of Large Microscopy Image Datasets on CPU-GPU Cluster Platforms

PubMed Central

Teodoro, George; Pan, Tony; Kurc, Tahsin M.; Kong, Jun; Cooper, Lee A. D.; Podhorszki, Norbert; Klasky, Scott; Saltz, Joel H.

2014-01-01

Analysis of large pathology image datasets offers significant opportunities for the investigation of disease morphology, but the resource requirements of analysis pipelines limit the scale of such studies. Motivated by a brain cancer study, we propose and evaluate a parallel image analysis application pipeline for high throughput computation of large datasets of high resolution pathology tissue images on distributed CPU-GPU platforms. To achieve efficient execution on these hybrid systems, we have built runtime support that allows us to express the cancer image analysis application as a hierarchical data processing pipeline. The application is implemented as a coarse-grain pipeline of stages, where each stage may be further partitioned into another pipeline of fine-grain operations. The fine-grain operations are efficiently managed and scheduled for computation on CPUs and GPUs using performance aware scheduling techniques along with several optimizations, including architecture aware process placement, data locality conscious task assignment, data prefetching, and asynchronous data copy. These optimizations are employed to maximize the utilization of the aggregate computing power of CPUs and GPUs and minimize data copy overheads. Our experimental evaluation shows that the cooperative use of CPUs and GPUs achieves significant improvements on top of GPU-only versions (up to 1.6×) and that the execution of the application as a set of fine-grain operations provides more opportunities for runtime optimizations and attains better performance than coarser-grain, monolithic implementations used in other works. An implementation of the cancer image analysis pipeline using the runtime support was able to process an image dataset consisting of 36,848 4Kx4K-pixel image tiles (about 1.8TB uncompressed) in less than 4 minutes (150 tiles/second) on 100 nodes of a state-of-the-art hybrid cluster system. PMID:25419546

Distributed memory parallel Markov random fields using graph partitioning

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

Heinemann, C.; Perciano, T.; Ushizima, D.

Markov random fields (MRF) based algorithms have attracted a large amount of interest in image analysis due to their ability to exploit contextual information about data. Image data generated by experimental facilities, though, continues to grow larger and more complex, making it more difficult to analyze in a reasonable amount of time. Applying image processing algorithms to large datasets requires alternative approaches to circumvent performance problems. Aiming to provide scientists with a new tool to recover valuable information from such datasets, we developed a general purpose distributed memory parallel MRF-based image analysis framework (MPI-PMRF). MPI-PMRF overcomes performance and memory limitationsmore » by distributing data and computations across processors. The proposed approach was successfully tested with synthetic and experimental datasets. Additionally, the performance of the MPI-PMRF framework is analyzed through a detailed scalability study. We show that a performance increase is obtained while maintaining an accuracy of the segmentation results higher than 98%. The contributions of this paper are: (a) development of a distributed memory MRF framework; (b) measurement of the performance increase of the proposed approach; (c) verification of segmentation accuracy in both synthetic and experimental, real-world datasets« less

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

Wendelberger, Laura Jean

In large datasets, it is time consuming or even impossible to pick out interesting images. Our proposed solution is to find statistics to quantify the information in each image and use those to identify and pick out images of interest.

High performance computing environment for multidimensional image analysis

PubMed Central

Rao, A Ravishankar; Cecchi, Guillermo A; Magnasco, Marcelo

2007-01-01

Background The processing of images acquired through microscopy is a challenging task due to the large size of datasets (several gigabytes) and the fast turnaround time required. If the throughput of the image processing stage is significantly increased, it can have a major impact in microscopy applications. Results We present a high performance computing (HPC) solution to this problem. This involves decomposing the spatial 3D image into segments that are assigned to unique processors, and matched to the 3D torus architecture of the IBM Blue Gene/L machine. Communication between segments is restricted to the nearest neighbors. When running on a 2 Ghz Intel CPU, the task of 3D median filtering on a typical 256 megabyte dataset takes two and a half hours, whereas by using 1024 nodes of Blue Gene, this task can be performed in 18.8 seconds, a 478× speedup. Conclusion Our parallel solution dramatically improves the performance of image processing, feature extraction and 3D reconstruction tasks. This increased throughput permits biologists to conduct unprecedented large scale experiments with massive datasets. PMID:17634099

High performance computing environment for multidimensional image analysis.

PubMed

Rao, A Ravishankar; Cecchi, Guillermo A; Magnasco, Marcelo

2007-07-10

The processing of images acquired through microscopy is a challenging task due to the large size of datasets (several gigabytes) and the fast turnaround time required. If the throughput of the image processing stage is significantly increased, it can have a major impact in microscopy applications. We present a high performance computing (HPC) solution to this problem. This involves decomposing the spatial 3D image into segments that are assigned to unique processors, and matched to the 3D torus architecture of the IBM Blue Gene/L machine. Communication between segments is restricted to the nearest neighbors. When running on a 2 Ghz Intel CPU, the task of 3D median filtering on a typical 256 megabyte dataset takes two and a half hours, whereas by using 1024 nodes of Blue Gene, this task can be performed in 18.8 seconds, a 478x speedup. Our parallel solution dramatically improves the performance of image processing, feature extraction and 3D reconstruction tasks. This increased throughput permits biologists to conduct unprecedented large scale experiments with massive datasets.

Image-based query-by-example for big databases of galaxy images

NASA Astrophysics Data System (ADS)

Shamir, Lior; Kuminski, Evan

2017-01-01

Very large astronomical databases containing millions or even billions of galaxy images have been becoming increasingly important tools in astronomy research. However, in many cases the very large size makes it more difficult to analyze these data manually, reinforcing the need for computer algorithms that can automate the data analysis process. An example of such task is the identification of galaxies of a certain morphology of interest. For instance, if a rare galaxy is identified it is reasonable to expect that more galaxies of similar morphology exist in the database, but it is virtually impossible to manually search these databases to identify such galaxies. Here we describe computer vision and pattern recognition methodology that receives a galaxy image as an input, and searches automatically a large dataset of galaxies to return a list of galaxies that are visually similar to the query galaxy. The returned list is not necessarily complete or clean, but it provides a substantial reduction of the original database into a smaller dataset, in which the frequency of objects visually similar to the query galaxy is much higher. Experimental results show that the algorithm can identify rare galaxies such as ring galaxies among datasets of 10,000 astronomical objects.

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.

Using Functional or Structural Magnetic Resonance Images and Personal Characteristic Data to Identify ADHD and Autism

PubMed Central

Ghiassian, Sina; Greiner, Russell; Jin, Ping; Brown, Matthew R. G.

2016-01-01

A clinical tool that can diagnose psychiatric illness using functional or structural magnetic resonance (MR) brain images has the potential to greatly assist physicians and improve treatment efficacy. Working toward the goal of automated diagnosis, we propose an approach for automated classification of ADHD and autism based on histogram of oriented gradients (HOG) features extracted from MR brain images, as well as personal characteristic data features. We describe a learning algorithm that can produce effective classifiers for ADHD and autism when run on two large public datasets. The algorithm is able to distinguish ADHD from control with hold-out accuracy of 69.6% (over baseline 55.0%) using personal characteristics and structural brain scan features when trained on the ADHD-200 dataset (769 participants in training set, 171 in test set). It is able to distinguish autism from control with hold-out accuracy of 65.0% (over baseline 51.6%) using functional images with personal characteristic data when trained on the Autism Brain Imaging Data Exchange (ABIDE) dataset (889 participants in training set, 222 in test set). These results outperform all previously presented methods on both datasets. To our knowledge, this is the first demonstration of a single automated learning process that can produce classifiers for distinguishing patients vs. controls from brain imaging data with above-chance accuracy on large datasets for two different psychiatric illnesses (ADHD and autism). Working toward clinical applications requires robustness against real-world conditions, including the substantial variability that often exists among data collected at different institutions. It is therefore important that our algorithm was successful with the large ADHD-200 and ABIDE datasets, which include data from hundreds of participants collected at multiple institutions. While the resulting classifiers are not yet clinically relevant, this work shows that there is a signal in the (f)MRI data that a learning algorithm is able to find. We anticipate this will lead to yet more accurate classifiers, over these and other psychiatric disorders, working toward the goal of a clinical tool for high accuracy differential diagnosis. PMID:28030565

Universal Batch Steganalysis

DTIC Science & Technology

2014-06-30

steganalysis) in large-scale datasets such as might be obtained by monitoring a corporate network or social network. Identifying guilty actors...guilty’ user (of steganalysis) in large-scale datasets such as might be obtained by monitoring a corporate network or social network. Identifying guilty...floating point operations (1 TFLOPs) for a 1 megapixel image. We designed a new implementation using Compute Unified Device Architecture (CUDA) on NVIDIA

Uvf - Unified Volume Format: A General System for Efficient Handling of Large Volumetric Datasets.

PubMed

Krüger, Jens; Potter, Kristin; Macleod, Rob S; Johnson, Christopher

2008-01-01

With the continual increase in computing power, volumetric datasets with sizes ranging from only a few megabytes to petascale are generated thousands of times per day. Such data may come from an ordinary source such as simple everyday medical imaging procedures, while larger datasets may be generated from cluster-based scientific simulations or measurements of large scale experiments. In computer science an incredible amount of work worldwide is put into the efficient visualization of these datasets. As researchers in the field of scientific visualization, we often have to face the task of handling very large data from various sources. This data usually comes in many different data formats. In medical imaging, the DICOM standard is well established, however, most research labs use their own data formats to store and process data. To simplify the task of reading the many different formats used with all of the different visualization programs, we present a system for the efficient handling of many types of large scientific datasets (see Figure 1 for just a few examples). While primarily targeted at structured volumetric data, UVF can store just about any type of structured and unstructured data. The system is composed of a file format specification with a reference implementation of a reader. It is not only a common, easy to implement format but also allows for efficient rendering of most datasets without the need to convert the data in memory.

Combining deep residual neural network features with supervised machine learning algorithms to classify diverse food image datasets.

PubMed

McAllister, Patrick; Zheng, Huiru; Bond, Raymond; Moorhead, Anne

2018-04-01

Obesity is increasing worldwide and can cause many chronic conditions such as type-2 diabetes, heart disease, sleep apnea, and some cancers. Monitoring dietary intake through food logging is a key method to maintain a healthy lifestyle to prevent and manage obesity. Computer vision methods have been applied to food logging to automate image classification for monitoring dietary intake. In this work we applied pretrained ResNet-152 and GoogleNet convolutional neural networks (CNNs), initially trained using ImageNet Large Scale Visual Recognition Challenge (ILSVRC) dataset with MatConvNet package, to extract features from food image datasets; Food 5K, Food-11, RawFooT-DB, and Food-101. Deep features were extracted from CNNs and used to train machine learning classifiers including artificial neural network (ANN), support vector machine (SVM), Random Forest, and Naive Bayes. Results show that using ResNet-152 deep features with SVM with RBF kernel can accurately detect food items with 99.4% accuracy using Food-5K validation food image dataset and 98.8% with Food-5K evaluation dataset using ANN, SVM-RBF, and Random Forest classifiers. Trained with ResNet-152 features, ANN can achieve 91.34%, 99.28% when applied to Food-11 and RawFooT-DB food image datasets respectively and SVM with RBF kernel can achieve 64.98% with Food-101 image dataset. From this research it is clear that using deep CNN features can be used efficiently for diverse food item image classification. The work presented in this research shows that pretrained ResNet-152 features provide sufficient generalisation power when applied to a range of food image classification tasks. Copyright © 2018 Elsevier Ltd. All rights reserved.

Object recognition using deep convolutional neural networks with complete transfer and partial frozen layers

NASA Astrophysics Data System (ADS)

Kruithof, Maarten C.; Bouma, Henri; Fischer, Noëlle M.; Schutte, Klamer

2016-10-01

Object recognition is important to understand the content of video and allow flexible querying in a large number of cameras, especially for security applications. Recent benchmarks show that deep convolutional neural networks are excellent approaches for object recognition. This paper describes an approach of domain transfer, where features learned from a large annotated dataset are transferred to a target domain where less annotated examples are available as is typical for the security and defense domain. Many of these networks trained on natural images appear to learn features similar to Gabor filters and color blobs in the first layer. These first-layer features appear to be generic for many datasets and tasks while the last layer is specific. In this paper, we study the effect of copying all layers and fine-tuning a variable number. We performed an experiment with a Caffe-based network on 1000 ImageNet classes that are randomly divided in two equal subgroups for the transfer from one to the other. We copy all layers and vary the number of layers that is fine-tuned and the size of the target dataset. We performed additional experiments with the Keras platform on CIFAR-10 dataset to validate general applicability. We show with both platforms and both datasets that the accuracy on the target dataset improves when more target data is used. When the target dataset is large, it is beneficial to freeze only a few layers. For a large target dataset, the network without transfer learning performs better than the transfer network, especially if many layers are frozen. When the target dataset is small, it is beneficial to transfer (and freeze) many layers. For a small target dataset, the transfer network boosts generalization and it performs much better than the network without transfer learning. Learning time can be reduced by freezing many layers in a network.

Automated Analysis of Fluorescence Microscopy Images to Identify Protein-Protein Interactions

DOE PAGES

Venkatraman, S.; Doktycz, M. J.; Qi, H.; ...

2006-01-01

The identification of protein interactions is important for elucidating biological networks. One obstacle in comprehensive interaction studies is the analyses of large datasets, particularly those containing images. Development of an automated system to analyze an image-based protein interaction dataset is needed. Such an analysis system is described here, to automatically extract features from fluorescence microscopy images obtained from a bacterial protein interaction assay. These features are used to relay quantitative values that aid in the automated scoring of positive interactions. Experimental observations indicate that identifying at least 50% positive cells in an image is sufficient to detect a protein interaction.more » Based on this criterion, the automated system presents 100% accuracy in detecting positive interactions for a dataset of 16 images. Algorithms were implemented using MATLAB and the software developed is available on request from the authors.« less

iSBatch: a batch-processing platform for data analysis and exploration of live-cell single-molecule microscopy images and other hierarchical datasets.

PubMed

Caldas, Victor E A; Punter, Christiaan M; Ghodke, Harshad; Robinson, Andrew; van Oijen, Antoine M

2015-10-01

Recent technical advances have made it possible to visualize single molecules inside live cells. Microscopes with single-molecule sensitivity enable the imaging of low-abundance proteins, allowing for a quantitative characterization of molecular properties. Such data sets contain information on a wide spectrum of important molecular properties, with different aspects highlighted in different imaging strategies. The time-lapsed acquisition of images provides information on protein dynamics over long time scales, giving insight into expression dynamics and localization properties. Rapid burst imaging reveals properties of individual molecules in real-time, informing on their diffusion characteristics, binding dynamics and stoichiometries within complexes. This richness of information, however, adds significant complexity to analysis protocols. In general, large datasets of images must be collected and processed in order to produce statistically robust results and identify rare events. More importantly, as live-cell single-molecule measurements remain on the cutting edge of imaging, few protocols for analysis have been established and thus analysis strategies often need to be explored for each individual scenario. Existing analysis packages are geared towards either single-cell imaging data or in vitro single-molecule data and typically operate with highly specific algorithms developed for particular situations. Our tool, iSBatch, instead allows users to exploit the inherent flexibility of the popular open-source package ImageJ, providing a hierarchical framework in which existing plugins or custom macros may be executed over entire datasets or portions thereof. This strategy affords users freedom to explore new analysis protocols within large imaging datasets, while maintaining hierarchical relationships between experiments, samples, fields of view, cells, and individual molecules.

Robust Statistical Fusion of Image Labels

PubMed Central

Landman, Bennett A.; Asman, Andrew J.; Scoggins, Andrew G.; Bogovic, John A.; Xing, Fangxu; Prince, Jerry L.

2011-01-01

Image labeling and parcellation (i.e. assigning structure to a collection of voxels) are critical tasks for the assessment of volumetric and morphometric features in medical imaging data. The process of image labeling is inherently error prone as images are corrupted by noise and artifacts. Even expert interpretations are subject to subjectivity and the precision of the individual raters. Hence, all labels must be considered imperfect with some degree of inherent variability. One may seek multiple independent assessments to both reduce this variability and quantify the degree of uncertainty. Existing techniques have exploited maximum a posteriori statistics to combine data from multiple raters and simultaneously estimate rater reliabilities. Although quite successful, wide-scale application has been hampered by unstable estimation with practical datasets, for example, with label sets with small or thin objects to be labeled or with partial or limited datasets. As well, these approaches have required each rater to generate a complete dataset, which is often impossible given both human foibles and the typical turnover rate of raters in a research or clinical environment. Herein, we propose a robust approach to improve estimation performance with small anatomical structures, allow for missing data, account for repeated label sets, and utilize training/catch trial data. With this approach, numerous raters can label small, overlapping portions of a large dataset, and rater heterogeneity can be robustly controlled while simultaneously estimating a single, reliable label set and characterizing uncertainty. The proposed approach enables many individuals to collaborate in the construction of large datasets for labeling tasks (e.g., human parallel processing) and reduces the otherwise detrimental impact of rater unavailability. PMID:22010145

Automatic Diabetic Macular Edema Detection in Fundus Images Using Publicly Available Datasets

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

Giancardo, Luca; Meriaudeau, Fabrice; Karnowski, Thomas Paul

2011-01-01

Diabetic macular edema (DME) is a common vision threatening complication of diabetic retinopathy. In a large scale screening environment DME can be assessed by detecting exudates (a type of bright lesions) in fundus images. In this work, we introduce a new methodology for diagnosis of DME using a novel set of features based on colour, wavelet decomposition and automatic lesion segmentation. These features are employed to train a classifier able to automatically diagnose DME. We present a new publicly available dataset with ground-truth data containing 169 patients from various ethnic groups and levels of DME. This and other two publiclymore » available datasets are employed to evaluate our algorithm. We are able to achieve diagnosis performance comparable to retina experts on the MESSIDOR (an independently labelled dataset with 1200 images) with cross-dataset testing. Our algorithm is robust to segmentation uncertainties, does not need ground truth at lesion level, and is very fast, generating a diagnosis on an average of 4.4 seconds per image on an 2.6 GHz platform with an unoptimised Matlab implementation.« less

Intellicount: High-Throughput Quantification of Fluorescent Synaptic Protein Puncta by Machine Learning

PubMed Central

Fantuzzo, J. A.; Mirabella, V. R.; Zahn, J. D.

2017-01-01

Abstract Synapse formation analyses can be performed by imaging and quantifying fluorescent signals of synaptic markers. Traditionally, these analyses are done using simple or multiple thresholding and segmentation approaches or by labor-intensive manual analysis by a human observer. Here, we describe Intellicount, a high-throughput, fully-automated synapse quantification program which applies a novel machine learning (ML)-based image processing algorithm to systematically improve region of interest (ROI) identification over simple thresholding techniques. Through processing large datasets from both human and mouse neurons, we demonstrate that this approach allows image processing to proceed independently of carefully set thresholds, thus reducing the need for human intervention. As a result, this method can efficiently and accurately process large image datasets with minimal interaction by the experimenter, making it less prone to bias and less liable to human error. Furthermore, Intellicount is integrated into an intuitive graphical user interface (GUI) that provides a set of valuable features, including automated and multifunctional figure generation, routine statistical analyses, and the ability to run full datasets through nested folders, greatly expediting the data analysis process. PMID:29218324

Image processing for optical mapping.

PubMed

Ravindran, Prabu; Gupta, Aditya

2015-01-01

Optical Mapping is an established single-molecule, whole-genome analysis system, which has been used to gain a comprehensive understanding of genomic structure and to study structural variation of complex genomes. A critical component of Optical Mapping system is the image processing module, which extracts single molecule restriction maps from image datasets of immobilized, restriction digested and fluorescently stained large DNA molecules. In this review, we describe robust and efficient image processing techniques to process these massive datasets and extract accurate restriction maps in the presence of noise, ambiguity and confounding artifacts. We also highlight a few applications of the Optical Mapping system.

Large scale validation of the M5L lung CAD on heterogeneous CT datasets.

PubMed

Torres, E Lopez; Fiorina, E; Pennazio, F; Peroni, C; Saletta, M; Camarlinghi, N; Fantacci, M E; Cerello, P

2015-04-01

M5L, a fully automated computer-aided detection (CAD) system for the detection and segmentation of lung nodules in thoracic computed tomography (CT), is presented and validated on several image datasets. M5L is the combination of two independent subsystems, based on the Channeler Ant Model as a segmentation tool [lung channeler ant model (lungCAM)] and on the voxel-based neural approach. The lungCAM was upgraded with a scan equalization module and a new procedure to recover the nodules connected to other lung structures; its classification module, which makes use of a feed-forward neural network, is based of a small number of features (13), so as to minimize the risk of lacking generalization, which could be possible given the large difference between the size of the training and testing datasets, which contain 94 and 1019 CTs, respectively. The lungCAM (standalone) and M5L (combined) performance was extensively tested on 1043 CT scans from three independent datasets, including a detailed analysis of the full Lung Image Database Consortium/Image Database Resource Initiative database, which is not yet found in literature. The lungCAM and M5L performance is consistent across the databases, with a sensitivity of about 70% and 80%, respectively, at eight false positive findings per scan, despite the variable annotation criteria and acquisition and reconstruction conditions. A reduced sensitivity is found for subtle nodules and ground glass opacities (GGO) structures. A comparison with other CAD systems is also presented. The M5L performance on a large and heterogeneous dataset is stable and satisfactory, although the development of a dedicated module for GGOs detection could further improve it, as well as an iterative optimization of the training procedure. The main aim of the present study was accomplished: M5L results do not deteriorate when increasing the dataset size, making it a candidate for supporting radiologists on large scale screenings and clinical programs.

OMERO and Bio-Formats 5: flexible access to large bioimaging datasets at scale

NASA Astrophysics Data System (ADS)

Moore, Josh; Linkert, Melissa; Blackburn, Colin; Carroll, Mark; Ferguson, Richard K.; Flynn, Helen; Gillen, Kenneth; Leigh, Roger; Li, Simon; Lindner, Dominik; Moore, William J.; Patterson, Andrew J.; Pindelski, Blazej; Ramalingam, Balaji; Rozbicki, Emil; Tarkowska, Aleksandra; Walczysko, Petr; Allan, Chris; Burel, Jean-Marie; Swedlow, Jason

2015-03-01

The Open Microscopy Environment (OME) has built and released Bio-Formats, a Java-based proprietary file format conversion tool and OMERO, an enterprise data management platform under open source licenses. In this report, we describe new versions of Bio-Formats and OMERO that are specifically designed to support large, multi-gigabyte or terabyte scale datasets that are routinely collected across most domains of biological and biomedical research. Bio- Formats reads image data directly from native proprietary formats, bypassing the need for conversion into a standard format. It implements the concept of a file set, a container that defines the contents of multi-dimensional data comprised of many files. OMERO uses Bio-Formats to read files natively, and provides a flexible access mechanism that supports several different storage and access strategies. These new capabilities of OMERO and Bio-Formats make them especially useful for use in imaging applications like digital pathology, high content screening and light sheet microscopy that create routinely large datasets that must be managed and analyzed.

Automatic labeling of MR brain images through extensible learning and atlas forests.

PubMed

Xu, Lijun; Liu, Hong; Song, Enmin; Yan, Meng; Jin, Renchao; Hung, Chih-Cheng

2017-12-01

Multiatlas-based method is extensively used in MR brain images segmentation because of its simplicity and robustness. This method provides excellent accuracy although it is time consuming and limited in terms of obtaining information about new atlases. In this study, an automatic labeling of MR brain images through extensible learning and atlas forest is presented to address these limitations. We propose an extensible learning model which allows the multiatlas-based framework capable of managing the datasets with numerous atlases or dynamic atlas datasets and simultaneously ensure the accuracy of automatic labeling. Two new strategies are used to reduce the time and space complexity and improve the efficiency of the automatic labeling of brain MR images. First, atlases are encoded to atlas forests through random forest technology to reduce the time consumed for cross-registration between atlases and target image, and a scatter spatial vector is designed to eliminate errors caused by inaccurate registration. Second, an atlas selection method based on the extensible learning model is used to select atlases for target image without traversing the entire dataset and then obtain the accurate labeling. The labeling results of the proposed method were evaluated in three public datasets, namely, IBSR, LONI LPBA40, and ADNI. With the proposed method, the dice coefficient metric values on the three datasets were 84.17 ± 4.61%, 83.25 ± 4.29%, and 81.88 ± 4.53% which were 5% higher than those of the conventional method, respectively. The efficiency of the extensible learning model was evaluated by state-of-the-art methods for labeling of MR brain images. Experimental results showed that the proposed method could achieve accurate labeling for MR brain images without traversing the entire datasets. In the proposed multiatlas-based method, extensible learning and atlas forests were applied to control the automatic labeling of brain anatomies on large atlas datasets or dynamic atlas datasets and obtain accurate results. © 2017 American Association of Physicists in Medicine.

Exudate-based diabetic macular edema detection in fundus images using publicly available datasets

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

Giancardo, Luca; Meriaudeau, Fabrice; Karnowski, Thomas Paul

2011-01-01

Diabetic macular edema (DME) is a common vision threatening complication of diabetic retinopathy. In a large scale screening environment DME can be assessed by detecting exudates (a type of bright lesions) in fundus images. In this work, we introduce a new methodology for diagnosis of DME using a novel set of features based on colour, wavelet decomposition and automatic lesion segmentation. These features are employed to train a classifier able to automatically diagnose DME through the presence of exudation. We present a new publicly available dataset with ground-truth data containing 169 patients from various ethnic groups and levels of DME.more » This and other two publicly available datasets are employed to evaluate our algorithm. We are able to achieve diagnosis performance comparable to retina experts on the MESSIDOR (an independently labelled dataset with 1200 images) with cross-dataset testing (e.g., the classifier was trained on an independent dataset and tested on MESSIDOR). Our algorithm obtained an AUC between 0.88 and 0.94 depending on the dataset/features used. Additionally, it does not need ground truth at lesion level to reject false positives and is computationally efficient, as it generates a diagnosis on an average of 4.4 s (9.3 s, considering the optic nerve localization) per image on an 2.6 GHz platform with an unoptimized Matlab implementation.« less

I'll take that to go: Big data bags and minimal identifiers for exchange of large, complex datasets

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

Chard, Kyle; D'Arcy, Mike; Heavner, Benjamin D.

Big data workflows often require the assembly and exchange of complex, multi-element datasets. For example, in biomedical applications, the input to an analytic pipeline can be a dataset consisting thousands of images and genome sequences assembled from diverse repositories, requiring a description of the contents of the dataset in a concise and unambiguous form. Typical approaches to creating datasets for big data workflows assume that all data reside in a single location, requiring costly data marshaling and permitting errors of omission and commission because dataset members are not explicitly specified. We address these issues by proposing simple methods and toolsmore » for assembling, sharing, and analyzing large and complex datasets that scientists can easily integrate into their daily workflows. These tools combine a simple and robust method for describing data collections (BDBags), data descriptions (Research Objects), and simple persistent identifiers (Minids) to create a powerful ecosystem of tools and services for big data analysis and sharing. We present these tools and use biomedical case studies to illustrate their use for the rapid assembly, sharing, and analysis of large datasets.« less

A spline-based regression parameter set for creating customized DARTEL MRI brain templates from infancy to old age.

PubMed

Wilke, Marko

2018-02-01

This dataset contains the regression parameters derived by analyzing segmented brain MRI images (gray matter and white matter) from a large population of healthy subjects, using a multivariate adaptive regression splines approach. A total of 1919 MRI datasets ranging in age from 1-75 years from four publicly available datasets (NIH, C-MIND, fCONN, and IXI) were segmented using the CAT12 segmentation framework, writing out gray matter and white matter images normalized using an affine-only spatial normalization approach. These images were then subjected to a six-step DARTEL procedure, employing an iterative non-linear registration approach and yielding increasingly crisp intermediate images. The resulting six datasets per tissue class were then analyzed using multivariate adaptive regression splines, using the CerebroMatic toolbox. This approach allows for flexibly modelling smoothly varying trajectories while taking into account demographic (age, gender) as well as technical (field strength, data quality) predictors. The resulting regression parameters described here can be used to generate matched DARTEL or SHOOT templates for a given population under study, from infancy to old age. The dataset and the algorithm used to generate it are publicly available at https://irc.cchmc.org/software/cerebromatic.php.

Coherent Image Layout using an Adaptive Visual Vocabulary

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

Dillard, Scott E.; Henry, Michael J.; Bohn, Shawn J.

When querying a huge image database containing millions of images, the result of the query may still contain many thousands of images that need to be presented to the user. We consider the problem of arranging such a large set of images into a visually coherent layout, one that places similar images next to each other. Image similarity is determined using a bag-of-features model, and the layout is constructed from a hierarchical clustering of the image set by mapping an in-order traversal of the hierarchy tree into a space-filling curve. This layout method provides strong locality guarantees so we aremore » able to quantitatively evaluate performance using standard image retrieval benchmarks. Performance of the bag-of-features method is best when the vocabulary is learned on the image set being clustered. Because learning a large, discriminative vocabulary is a computationally demanding task, we present a novel method for efficiently adapting a generic visual vocabulary to a particular dataset. We evaluate our clustering and vocabulary adaptation methods on a variety of image datasets and show that adapting a generic vocabulary to a particular set of images improves performance on both hierarchical clustering and image retrieval tasks.« less

Data augmentation-assisted deep learning of hand-drawn partially colored sketches for visual search

PubMed Central

Muhammad, Khan; Baik, Sung Wook

2017-01-01

In recent years, image databases are growing at exponential rates, making their management, indexing, and retrieval, very challenging. Typical image retrieval systems rely on sample images as queries. However, in the absence of sample query images, hand-drawn sketches are also used. The recent adoption of touch screen input devices makes it very convenient to quickly draw shaded sketches of objects to be used for querying image databases. This paper presents a mechanism to provide access to visual information based on users’ hand-drawn partially colored sketches using touch screen devices. A key challenge for sketch-based image retrieval systems is to cope with the inherent ambiguity in sketches due to the lack of colors, textures, shading, and drawing imperfections. To cope with these issues, we propose to fine-tune a deep convolutional neural network (CNN) using augmented dataset to extract features from partially colored hand-drawn sketches for query specification in a sketch-based image retrieval framework. The large augmented dataset contains natural images, edge maps, hand-drawn sketches, de-colorized, and de-texturized images which allow CNN to effectively model visual contents presented to it in a variety of forms. The deep features extracted from CNN allow retrieval of images using both sketches and full color images as queries. We also evaluated the role of partial coloring or shading in sketches to improve the retrieval performance. The proposed method is tested on two large datasets for sketch recognition and sketch-based image retrieval and achieved better classification and retrieval performance than many existing methods. PMID:28859140

Classification of foods by transferring knowledge from ImageNet dataset

NASA Astrophysics Data System (ADS)

Heravi, Elnaz J.; Aghdam, Hamed H.; Puig, Domenec

2017-03-01

Automatic classification of foods is a way to control food intake and tackle with obesity. However, it is a challenging problem since foods are highly deformable and complex objects. Results on ImageNet dataset have revealed that Convolutional Neural Network has a great expressive power to model natural objects. Nonetheless, it is not trivial to train a ConvNet from scratch for classification of foods. This is due to the fact that ConvNets require large datasets and to our knowledge there is not a large public dataset of food for this purpose. Alternative solution is to transfer knowledge from trained ConvNets to the domain of foods. In this work, we study how transferable are state-of-art ConvNets to the task of food classification. We also propose a method for transferring knowledge from a bigger ConvNet to a smaller ConvNet by keeping its accuracy similar to the bigger ConvNet. Our experiments on UECFood256 datasets show that Googlenet, VGG and residual networks produce comparable results if we start transferring knowledge from appropriate layer. In addition, we show that our method is able to effectively transfer knowledge to the smaller ConvNet using unlabeled samples.

Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning

PubMed Central

Hoo-Chang, Shin; Roth, Holger R.; Gao, Mingchen; Lu, Le; Xu, Ziyue; Nogues, Isabella; Yao, Jianhua; Mollura, Daniel

2016-01-01

Remarkable progress has been made in image recognition, primarily due to the availability of large-scale annotated datasets (i.e. ImageNet) and the revival of deep convolutional neural networks (CNN). CNNs enable learning data-driven, highly representative, layered hierarchical image features from sufficient training data. However, obtaining datasets as comprehensively annotated as ImageNet in the medical imaging domain remains a challenge. There are currently three major techniques that successfully employ CNNs to medical image classification: training the CNN from scratch, using off-the-shelf pre-trained CNN features, and conducting unsupervised CNN pre-training with supervised fine-tuning. Another effective method is transfer learning, i.e., fine-tuning CNN models (supervised) pre-trained from natural image dataset to medical image tasks (although domain transfer between two medical image datasets is also possible). In this paper, we exploit three important, but previously understudied factors of employing deep convolutional neural networks to computer-aided detection problems. We first explore and evaluate different CNN architectures. The studied models contain 5 thousand to 160 million parameters, and vary in numbers of layers. We then evaluate the influence of dataset scale and spatial image context on performance. Finally, we examine when and why transfer learning from pre-trained ImageNet (via fine-tuning) can be useful. We study two specific computeraided detection (CADe) problems, namely thoraco-abdominal lymph node (LN) detection and interstitial lung disease (ILD) classification. We achieve the state-of-the-art performance on the mediastinal LN detection, with 85% sensitivity at 3 false positive per patient, and report the first five-fold cross-validation classification results on predicting axial CT slices with ILD categories. Our extensive empirical evaluation, CNN model analysis and valuable insights can be extended to the design of high performance CAD systems for other medical imaging tasks. PMID:26886976

Predicting Classifier Performance with Limited Training Data: Applications to Computer-Aided Diagnosis in Breast and Prostate Cancer

PubMed Central

Basavanhally, Ajay; Viswanath, Satish; Madabhushi, Anant

2015-01-01

Clinical trials increasingly employ medical imaging data in conjunction with supervised classifiers, where the latter require large amounts of training data to accurately model the system. Yet, a classifier selected at the start of the trial based on smaller and more accessible datasets may yield inaccurate and unstable classification performance. In this paper, we aim to address two common concerns in classifier selection for clinical trials: (1) predicting expected classifier performance for large datasets based on error rates calculated from smaller datasets and (2) the selection of appropriate classifiers based on expected performance for larger datasets. We present a framework for comparative evaluation of classifiers using only limited amounts of training data by using random repeated sampling (RRS) in conjunction with a cross-validation sampling strategy. Extrapolated error rates are subsequently validated via comparison with leave-one-out cross-validation performed on a larger dataset. The ability to predict error rates as dataset size increases is demonstrated on both synthetic data as well as three different computational imaging tasks: detecting cancerous image regions in prostate histopathology, differentiating high and low grade cancer in breast histopathology, and detecting cancerous metavoxels in prostate magnetic resonance spectroscopy. For each task, the relationships between 3 distinct classifiers (k-nearest neighbor, naive Bayes, Support Vector Machine) are explored. Further quantitative evaluation in terms of interquartile range (IQR) suggests that our approach consistently yields error rates with lower variability (mean IQRs of 0.0070, 0.0127, and 0.0140) than a traditional RRS approach (mean IQRs of 0.0297, 0.0779, and 0.305) that does not employ cross-validation sampling for all three datasets. PMID:25993029

Combining semi-automated image analysis techniques with machine learning algorithms to accelerate large-scale genetic studies.

PubMed

Atkinson, Jonathan A; Lobet, Guillaume; Noll, Manuel; Meyer, Patrick E; Griffiths, Marcus; Wells, Darren M

2017-10-01

Genetic analyses of plant root systems require large datasets of extracted architectural traits. To quantify such traits from images of root systems, researchers often have to choose between automated tools (that are prone to error and extract only a limited number of architectural traits) or semi-automated ones (that are highly time consuming). We trained a Random Forest algorithm to infer architectural traits from automatically extracted image descriptors. The training was performed on a subset of the dataset, then applied to its entirety. This strategy allowed us to (i) decrease the image analysis time by 73% and (ii) extract meaningful architectural traits based on image descriptors. We also show that these traits are sufficient to identify the quantitative trait loci that had previously been discovered using a semi-automated method. We have shown that combining semi-automated image analysis with machine learning algorithms has the power to increase the throughput of large-scale root studies. We expect that such an approach will enable the quantification of more complex root systems for genetic studies. We also believe that our approach could be extended to other areas of plant phenotyping. © The Authors 2017. Published by Oxford University Press.

Combining semi-automated image analysis techniques with machine learning algorithms to accelerate large-scale genetic studies

PubMed Central

Atkinson, Jonathan A.; Lobet, Guillaume; Noll, Manuel; Meyer, Patrick E.; Griffiths, Marcus

2017-01-01

Abstract Genetic analyses of plant root systems require large datasets of extracted architectural traits. To quantify such traits from images of root systems, researchers often have to choose between automated tools (that are prone to error and extract only a limited number of architectural traits) or semi-automated ones (that are highly time consuming). We trained a Random Forest algorithm to infer architectural traits from automatically extracted image descriptors. The training was performed on a subset of the dataset, then applied to its entirety. This strategy allowed us to (i) decrease the image analysis time by 73% and (ii) extract meaningful architectural traits based on image descriptors. We also show that these traits are sufficient to identify the quantitative trait loci that had previously been discovered using a semi-automated method. We have shown that combining semi-automated image analysis with machine learning algorithms has the power to increase the throughput of large-scale root studies. We expect that such an approach will enable the quantification of more complex root systems for genetic studies. We also believe that our approach could be extended to other areas of plant phenotyping. PMID:29020748

Histopathological Breast Cancer Image Classification by Deep Neural Network Techniques Guided by Local Clustering.

PubMed

Nahid, Abdullah-Al; Mehrabi, Mohamad Ali; Kong, Yinan

2018-01-01

Breast Cancer is a serious threat and one of the largest causes of death of women throughout the world. The identification of cancer largely depends on digital biomedical photography analysis such as histopathological images by doctors and physicians. Analyzing histopathological images is a nontrivial task, and decisions from investigation of these kinds of images always require specialised knowledge. However, Computer Aided Diagnosis (CAD) techniques can help the doctor make more reliable decisions. The state-of-the-art Deep Neural Network (DNN) has been recently introduced for biomedical image analysis. Normally each image contains structural and statistical information. This paper classifies a set of biomedical breast cancer images (BreakHis dataset) using novel DNN techniques guided by structural and statistical information derived from the images. Specifically a Convolutional Neural Network (CNN), a Long-Short-Term-Memory (LSTM), and a combination of CNN and LSTM are proposed for breast cancer image classification. Softmax and Support Vector Machine (SVM) layers have been used for the decision-making stage after extracting features utilising the proposed novel DNN models. In this experiment the best Accuracy value of 91.00% is achieved on the 200x dataset, the best Precision value 96.00% is achieved on the 40x dataset, and the best F -Measure value is achieved on both the 40x and 100x datasets.

PitScan: Computer-Assisted Feature Detection

NASA Astrophysics Data System (ADS)

Wagner, R. V.; Robinson, M. S.

2018-04-01

We developed PitScan to assist in searching the very large LROC image dataset for pits — unusual <200m wide vertical-walled holes in the Moon's surface. PitScan reduces analysts' workload by pre-filtering images to identify possible pits.

A New Approach to Create Image Control Networks in ISIS

NASA Astrophysics Data System (ADS)

Becker, K. J.; Berry, K. L.; Mapel, J. A.; Walldren, J. C.

2017-06-01

A new approach was used to create a feature-based control point network that required the development of new tools in the Integrated Software for Imagers and Spectrometers (ISIS3) system to process very large datasets.

Systematic Processing of Clementine Data for Scientific Analyses

NASA Technical Reports Server (NTRS)

Mcewen, A. S.

1993-01-01

If fully successful, the Clementine mission will return about 3,000,000 lunar images and more than 5000 images of Geographos. Effective scientific analyses of such large datasets require systematic processing efforts. Concepts for two such efforts are described: glogal multispectral imaging of the moon; and videos of Geographos.

Deep learning and face recognition: the state of the art

NASA Astrophysics Data System (ADS)

Balaban, Stephen

2015-05-01

Deep Neural Networks (DNNs) have established themselves as a dominant technique in machine learning. DNNs have been top performers on a wide variety of tasks including image classification, speech recognition, and face recognition.1-3 Convolutional neural networks (CNNs) have been used in nearly all of the top performing methods on the Labeled Faces in the Wild (LFW) dataset.3-6 In this talk and accompanying paper, I attempt to provide a review and summary of the deep learning techniques used in the state-of-the-art. In addition, I highlight the need for both larger and more challenging public datasets to benchmark these systems. Despite the ability of DNNs and autoencoders to perform unsupervised feature learning, modern facial recognition pipelines still require domain specific engineering in the form of re-alignment. For example, in Facebook's recent DeepFace paper, a 3D "frontalization" step lies at the beginning of the pipeline. This step creates a 3D face model for the incoming image and then uses a series of affine transformations of the fiducial points to "frontalize" the image. This step enables the DeepFace system to use a neural network architecture with locally connected layers without weight sharing as opposed to standard convolutional layers.6 Deep learning techniques combined with large datasets have allowed research groups to surpass human level performance on the LFW dataset.3, 5 The high accuracy (99.63% for FaceNet at the time of publishing) and utilization of outside data (hundreds of millions of images in the case of Google's FaceNet) suggest that current face verification benchmarks such as LFW may not be challenging enough, nor provide enough data, for current techniques.3, 5 There exist a variety of organizations with mobile photo sharing applications that would be capable of releasing a very large scale and highly diverse dataset of facial images captured on mobile devices. Such an "ImageNet for Face Recognition" would likely receive a warm welcome from researchers and practitioners alike.

A large-scale solar dynamics observatory image dataset for computer vision applications.

PubMed

Kucuk, Ahmet; Banda, Juan M; Angryk, Rafal A

2017-01-01

The National Aeronautics Space Agency (NASA) Solar Dynamics Observatory (SDO) mission has given us unprecedented insight into the Sun's activity. By capturing approximately 70,000 images a day, this mission has created one of the richest and biggest repositories of solar image data available to mankind. With such massive amounts of information, researchers have been able to produce great advances in detecting solar events. In this resource, we compile SDO solar data into a single repository in order to provide the computer vision community with a standardized and curated large-scale dataset of several hundred thousand solar events found on high resolution solar images. This publicly available resource, along with the generation source code, will accelerate computer vision research on NASA's solar image data by reducing the amount of time spent performing data acquisition and curation from the multiple sources we have compiled. By improving the quality of the data with thorough curation, we anticipate a wider adoption and interest from the computer vision to the solar physics community.

Hierarchical storage of large volume of multidector CT data using distributed servers

NASA Astrophysics Data System (ADS)

Ratib, Osman; Rosset, Antoine; Heuberger, Joris; Bandon, David

2006-03-01

Multidector scanners and hybrid multimodality scanners have the ability to generate large number of high-resolution images resulting in very large data sets. In most cases, these datasets are generated for the sole purpose of generating secondary processed images and 3D rendered images as well as oblique and curved multiplanar reformatted images. It is therefore not essential to archive the original images after they have been processed. We have developed an architecture of distributed archive servers for temporary storage of large image datasets for 3D rendering and image processing without the need for long term storage in PACS archive. With the relatively low cost of storage devices it is possible to configure these servers to hold several months or even years of data, long enough for allowing subsequent re-processing if required by specific clinical situations. We tested the latest generation of RAID servers provided by Apple computers with a capacity of 5 TBytes. We implemented a peer-to-peer data access software based on our Open-Source image management software called OsiriX, allowing remote workstations to directly access DICOM image files located on the server through a new technology called "bonjour". This architecture offers a seamless integration of multiple servers and workstations without the need for central database or complex workflow management tools. It allows efficient access to image data from multiple workstation for image analysis and visualization without the need for image data transfer. It provides a convenient alternative to centralized PACS architecture while avoiding complex and time-consuming data transfer and storage.

Supervised learning based multimodal MRI brain tumour segmentation using texture features from supervoxels.

PubMed

Soltaninejad, Mohammadreza; Yang, Guang; Lambrou, Tryphon; Allinson, Nigel; Jones, Timothy L; Barrick, Thomas R; Howe, Franklyn A; Ye, Xujiong

2018-04-01

Accurate segmentation of brain tumour in magnetic resonance images (MRI) is a difficult task due to various tumour types. Using information and features from multimodal MRI including structural MRI and isotropic (p) and anisotropic (q) components derived from the diffusion tensor imaging (DTI) may result in a more accurate analysis of brain images. We propose a novel 3D supervoxel based learning method for segmentation of tumour in multimodal MRI brain images (conventional MRI and DTI). Supervoxels are generated using the information across the multimodal MRI dataset. For each supervoxel, a variety of features including histograms of texton descriptor, calculated using a set of Gabor filters with different sizes and orientations, and first order intensity statistical features are extracted. Those features are fed into a random forests (RF) classifier to classify each supervoxel into tumour core, oedema or healthy brain tissue. The method is evaluated on two datasets: 1) Our clinical dataset: 11 multimodal images of patients and 2) BRATS 2013 clinical dataset: 30 multimodal images. For our clinical dataset, the average detection sensitivity of tumour (including tumour core and oedema) using multimodal MRI is 86% with balanced error rate (BER) 7%; while the Dice score for automatic tumour segmentation against ground truth is 0.84. The corresponding results of the BRATS 2013 dataset are 96%, 2% and 0.89, respectively. The method demonstrates promising results in the segmentation of brain tumour. Adding features from multimodal MRI images can largely increase the segmentation accuracy. The method provides a close match to expert delineation across all tumour grades, leading to a faster and more reproducible method of brain tumour detection and delineation to aid patient management. Copyright © 2018 Elsevier B.V. All rights reserved.

Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning.

PubMed

Shin, Hoo-Chang; Roth, Holger R; Gao, Mingchen; Lu, Le; Xu, Ziyue; Nogues, Isabella; Yao, Jianhua; Mollura, Daniel; Summers, Ronald M

2016-05-01

Remarkable progress has been made in image recognition, primarily due to the availability of large-scale annotated datasets and deep convolutional neural networks (CNNs). CNNs enable learning data-driven, highly representative, hierarchical image features from sufficient training data. However, obtaining datasets as comprehensively annotated as ImageNet in the medical imaging domain remains a challenge. There are currently three major techniques that successfully employ CNNs to medical image classification: training the CNN from scratch, using off-the-shelf pre-trained CNN features, and conducting unsupervised CNN pre-training with supervised fine-tuning. Another effective method is transfer learning, i.e., fine-tuning CNN models pre-trained from natural image dataset to medical image tasks. In this paper, we exploit three important, but previously understudied factors of employing deep convolutional neural networks to computer-aided detection problems. We first explore and evaluate different CNN architectures. The studied models contain 5 thousand to 160 million parameters, and vary in numbers of layers. We then evaluate the influence of dataset scale and spatial image context on performance. Finally, we examine when and why transfer learning from pre-trained ImageNet (via fine-tuning) can be useful. We study two specific computer-aided detection (CADe) problems, namely thoraco-abdominal lymph node (LN) detection and interstitial lung disease (ILD) classification. We achieve the state-of-the-art performance on the mediastinal LN detection, and report the first five-fold cross-validation classification results on predicting axial CT slices with ILD categories. Our extensive empirical evaluation, CNN model analysis and valuable insights can be extended to the design of high performance CAD systems for other medical imaging tasks.

TeraStitcher - A tool for fast automatic 3D-stitching of teravoxel-sized microscopy images

PubMed Central

2012-01-01

Background Further advances in modern microscopy are leading to teravoxel-sized tiled 3D images at high resolution, thus increasing the dimension of the stitching problem of at least two orders of magnitude. The existing software solutions do not seem adequate to address the additional requirements arising from these datasets, such as the minimization of memory usage and the need to process just a small portion of data. Results We propose a free and fully automated 3D Stitching tool designed to match the special requirements coming out of teravoxel-sized tiled microscopy images that is able to stitch them in a reasonable time even on workstations with limited resources. The tool was tested on teravoxel-sized whole mouse brain images with micrometer resolution and it was also compared with the state-of-the-art stitching tools on megavoxel-sized publicy available datasets. This comparison confirmed that the solutions we adopted are suited for stitching very large images and also perform well on datasets with different characteristics. Indeed, some of the algorithms embedded in other stitching tools could be easily integrated in our framework if they turned out to be more effective on other classes of images. To this purpose, we designed a software architecture which separates the strategies that use efficiently memory resources from the algorithms which may depend on the characteristics of the acquired images. Conclusions TeraStitcher is a free tool that enables the stitching of Teravoxel-sized tiled microscopy images even on workstations with relatively limited resources of memory (<8 GB) and processing power. It exploits the knowledge of approximate tile positions and uses ad-hoc strategies and algorithms designed for such very large datasets. The produced images can be saved into a multiresolution representation to be efficiently retrieved and processed. We provide TeraStitcher both as standalone application and as plugin of the free software Vaa3D. PMID:23181553

MMX-I: data-processing software for multimodal X-ray imaging and tomography.

PubMed

Bergamaschi, Antoine; Medjoubi, Kadda; Messaoudi, Cédric; Marco, Sergio; Somogyi, Andrea

2016-05-01

A new multi-platform freeware has been developed for the processing and reconstruction of scanning multi-technique X-ray imaging and tomography datasets. The software platform aims to treat different scanning imaging techniques: X-ray fluorescence, phase, absorption and dark field and any of their combinations, thus providing an easy-to-use data processing tool for the X-ray imaging user community. A dedicated data input stream copes with the input and management of large datasets (several hundred GB) collected during a typical multi-technique fast scan at the Nanoscopium beamline and even on a standard PC. To the authors' knowledge, this is the first software tool that aims at treating all of the modalities of scanning multi-technique imaging and tomography experiments.

Benchmark of Machine Learning Methods for Classification of a SENTINEL-2 Image

NASA Astrophysics Data System (ADS)

Pirotti, F.; Sunar, F.; Piragnolo, M.

2016-06-01

Thanks to mainly ESA and USGS, a large bulk of free images of the Earth is readily available nowadays. One of the main goals of remote sensing is to label images according to a set of semantic categories, i.e. image classification. This is a very challenging issue since land cover of a specific class may present a large spatial and spectral variability and objects may appear at different scales and orientations. In this study, we report the results of benchmarking 9 machine learning algorithms tested for accuracy and speed in training and classification of land-cover classes in a Sentinel-2 dataset. The following machine learning methods (MLM) have been tested: linear discriminant analysis, k-nearest neighbour, random forests, support vector machines, multi layered perceptron, multi layered perceptron ensemble, ctree, boosting, logarithmic regression. The validation is carried out using a control dataset which consists of an independent classification in 11 land-cover classes of an area about 60 km2, obtained by manual visual interpretation of high resolution images (20 cm ground sampling distance) by experts. In this study five out of the eleven classes are used since the others have too few samples (pixels) for testing and validating subsets. The classes used are the following: (i) urban (ii) sowable areas (iii) water (iv) tree plantations (v) grasslands. Validation is carried out using three different approaches: (i) using pixels from the training dataset (train), (ii) using pixels from the training dataset and applying cross-validation with the k-fold method (kfold) and (iii) using all pixels from the control dataset. Five accuracy indices are calculated for the comparison between the values predicted with each model and control values over three sets of data: the training dataset (train), the whole control dataset (full) and with k-fold cross-validation (kfold) with ten folds. Results from validation of predictions of the whole dataset (full) show the random forests method with the highest values; kappa index ranging from 0.55 to 0.42 respectively with the most and least number pixels for training. The two neural networks (multi layered perceptron and its ensemble) and the support vector machines - with default radial basis function kernel - methods follow closely with comparable performance.

Quantifying Uncertainties in Land-Surface Microwave Emissivity Retrievals

NASA Technical Reports Server (NTRS)

Tian, Yudong; Peters-Lidard, Christa D.; Harrison, Kenneth W.; Prigent, Catherine; Norouzi, Hamidreza; Aires, Filipe; Boukabara, Sid-Ahmed; Furuzawa, Fumie A.; Masunaga, Hirohiko

2013-01-01

Uncertainties in the retrievals of microwaveland-surface emissivities are quantified over two types of land surfaces: desert and tropical rainforest. Retrievals from satellite-based microwave imagers, including the Special Sensor Microwave Imager, the Tropical Rainfall Measuring Mission Microwave Imager, and the Advanced Microwave Scanning Radiometer for Earth Observing System, are studied. Our results show that there are considerable differences between the retrievals from different sensors and from different groups over these two land-surface types. In addition, the mean emissivity values show different spectral behavior across the frequencies. With the true emissivity assumed largely constant over both of the two sites throughout the study period, the differences are largely attributed to the systematic and random errors inthe retrievals. Generally, these retrievals tend to agree better at lower frequencies than at higher ones, with systematic differences ranging 1%-4% (3-12 K) over desert and 1%-7% (3-20 K) over rainforest. The random errors within each retrieval dataset are in the range of 0.5%-2% (2-6 K). In particular, at 85.5/89.0 GHz, there are very large differences between the different retrieval datasets, and within each retrieval dataset itself. Further investigation reveals that these differences are most likely caused by rain/cloud contamination, which can lead to random errors up to 10-17 K under the most severe conditions.

SOURCE EXPLORER: Towards Web Browser Based Tools for Astronomical Source Visualization and Analysis

NASA Astrophysics Data System (ADS)

Young, M. D.; Hayashi, S.; Gopu, A.

2014-05-01

As a new generation of large format, high-resolution imagers come online (ODI, DECAM, LSST, etc.) we are faced with the daunting prospect of astronomical images containing upwards of hundreds of thousands of identifiable sources. Visualizing and interacting with such large datasets using traditional astronomical tools appears to be unfeasible, and a new approach is required. We present here a method for the display and analysis of arbitrarily large source datasets using dynamically scaling levels of detail, enabling scientists to rapidly move from large-scale spatial overviews down to the level of individual sources and everything in-between. Based on the recognized standards of HTML5+JavaScript, we enable observers and archival users to interact with their images and sources from any modern computer without having to install specialized software. We demonstrate the ability to produce large-scale source lists from the images themselves, as well as overlaying data from publicly available source ( 2MASS, GALEX, SDSS, etc.) or user provided source lists. A high-availability cluster of computational nodes allows us to produce these source maps on demand and customized based on user input. User-generated source lists and maps are persistent across sessions and are available for further plotting, analysis, refinement, and culling.

High-performance computing in image registration

NASA Astrophysics Data System (ADS)

Zanin, Michele; Remondino, Fabio; Dalla Mura, Mauro

2012-10-01

Thanks to the recent technological advances, a large variety of image data is at our disposal with variable geometric, radiometric and temporal resolution. In many applications the processing of such images needs high performance computing techniques in order to deliver timely responses e.g. for rapid decisions or real-time actions. Thus, parallel or distributed computing methods, Digital Signal Processor (DSP) architectures, Graphical Processing Unit (GPU) programming and Field-Programmable Gate Array (FPGA) devices have become essential tools for the challenging issue of processing large amount of geo-data. The article focuses on the processing and registration of large datasets of terrestrial and aerial images for 3D reconstruction, diagnostic purposes and monitoring of the environment. For the image alignment procedure, sets of corresponding feature points need to be automatically extracted in order to successively compute the geometric transformation that aligns the data. The feature extraction and matching are ones of the most computationally demanding operations in the processing chain thus, a great degree of automation and speed is mandatory. The details of the implemented operations (named LARES) exploiting parallel architectures and GPU are thus presented. The innovative aspects of the implementation are (i) the effectiveness on a large variety of unorganized and complex datasets, (ii) capability to work with high-resolution images and (iii) the speed of the computations. Examples and comparisons with standard CPU processing are also reported and commented.

Analysis Of The IJCNN 2011 UTL Challenge

DTIC Science & Technology

2012-01-13

large datasets from various application domains: handwriting recognition, image recognition, video processing, text processing, and ecology. The goal...validation and final evaluation sets consist of 4096 examples each. Dataset Domain Features Sparsity Devel. Transf. AVICENNA Handwriting 120 0% 150205...documents [3]. Transfer learning methods could accelerate the application of handwriting recognizers to historical manuscript by reducing the need for

A photogrammetric technique for generation of an accurate multispectral optical flow dataset

NASA Astrophysics Data System (ADS)

Kniaz, V. V.

2017-06-01

A presence of an accurate dataset is the key requirement for a successful development of an optical flow estimation algorithm. A large number of freely available optical flow datasets were developed in recent years and gave rise for many powerful algorithms. However most of the datasets include only images captured in the visible spectrum. This paper is focused on the creation of a multispectral optical flow dataset with an accurate ground truth. The generation of an accurate ground truth optical flow is a rather complex problem, as no device for error-free optical flow measurement was developed to date. Existing methods for ground truth optical flow estimation are based on hidden textures, 3D modelling or laser scanning. Such techniques are either work only with a synthetic optical flow or provide a sparse ground truth optical flow. In this paper a new photogrammetric method for generation of an accurate ground truth optical flow is proposed. The method combines the benefits of the accuracy and density of a synthetic optical flow datasets with the flexibility of laser scanning based techniques. A multispectral dataset including various image sequences was generated using the developed method. The dataset is freely available on the accompanying web site.

A dataset of images and morphological profiles of 30 000 small-molecule treatments using the Cell Painting assay

PubMed Central

Bray, Mark-Anthony; Gustafsdottir, Sigrun M; Rohban, Mohammad H; Singh, Shantanu; Ljosa, Vebjorn; Sokolnicki, Katherine L; Bittker, Joshua A; Bodycombe, Nicole E; Dančík, Vlado; Hasaka, Thomas P; Hon, Cindy S; Kemp, Melissa M; Li, Kejie; Walpita, Deepika; Wawer, Mathias J; Golub, Todd R; Schreiber, Stuart L; Clemons, Paul A; Shamji, Alykhan F

2017-01-01

Abstract Background Large-scale image sets acquired by automated microscopy of perturbed samples enable a detailed comparison of cell states induced by each perturbation, such as a small molecule from a diverse library. Highly multiplexed measurements of cellular morphology can be extracted from each image and subsequently mined for a number of applications. Findings This microscopy dataset includes 919 265 five-channel fields of view, representing 30 616 tested compounds, available at “The Cell Image Library” (CIL) repository. It also includes data files containing morphological features derived from each cell in each image, both at the single-cell level and population-averaged (i.e., per-well) level; the image analysis workflows that generated the morphological features are also provided. Quality-control metrics are provided as metadata, indicating fields of view that are out-of-focus or containing highly fluorescent material or debris. Lastly, chemical annotations are supplied for the compound treatments applied. Conclusions Because computational algorithms and methods for handling single-cell morphological measurements are not yet routine, the dataset serves as a useful resource for the wider scientific community applying morphological (image-based) profiling. The dataset can be mined for many purposes, including small-molecule library enrichment and chemical mechanism-of-action studies, such as target identification. Integration with genetically perturbed datasets could enable identification of small-molecule mimetics of particular disease- or gene-related phenotypes that could be useful as probes or potential starting points for development of future therapeutics. PMID:28327978

Cardiovascular imaging environment: will the future be cloud-based?

PubMed

Kawel-Boehm, Nadine; Bluemke, David A

2017-07-01

In cardiovascular CT and MR imaging large datasets have to be stored, post-processed, analyzed and distributed. Beside basic assessment of volume and function in cardiac magnetic resonance imaging e.g., more sophisticated quantitative analysis is requested requiring specific software. Several institutions cannot afford various types of software and provide expertise to perform sophisticated analysis. Areas covered: Various cloud services exist related to data storage and analysis specifically for cardiovascular CT and MR imaging. Instead of on-site data storage, cloud providers offer flexible storage services on a pay-per-use basis. To avoid purchase and maintenance of specialized software for cardiovascular image analysis, e.g. to assess myocardial iron overload, MR 4D flow and fractional flow reserve, evaluation can be performed with cloud based software by the consumer or complete analysis is performed by the cloud provider. However, challenges to widespread implementation of cloud services include regulatory issues regarding patient privacy and data security. Expert commentary: If patient privacy and data security is guaranteed cloud imaging is a valuable option to cope with storage of large image datasets and offer sophisticated cardiovascular image analysis for institutions of all sizes.

A simple rapid process for semi-automated brain extraction from magnetic resonance images of the whole mouse head.

PubMed

Delora, Adam; Gonzales, Aaron; Medina, Christopher S; Mitchell, Adam; Mohed, Abdul Faheem; Jacobs, Russell E; Bearer, Elaine L

2016-01-15

Magnetic resonance imaging (MRI) is a well-developed technique in neuroscience. Limitations in applying MRI to rodent models of neuropsychiatric disorders include the large number of animals required to achieve statistical significance, and the paucity of automation tools for the critical early step in processing, brain extraction, which prepares brain images for alignment and voxel-wise statistics. This novel timesaving automation of template-based brain extraction ("skull-stripping") is capable of quickly and reliably extracting the brain from large numbers of whole head images in a single step. The method is simple to install and requires minimal user interaction. This method is equally applicable to different types of MR images. Results were evaluated with Dice and Jacquard similarity indices and compared in 3D surface projections with other stripping approaches. Statistical comparisons demonstrate that individual variation of brain volumes are preserved. A downloadable software package not otherwise available for extraction of brains from whole head images is included here. This software tool increases speed, can be used with an atlas or a template from within the dataset, and produces masks that need little further refinement. Our new automation can be applied to any MR dataset, since the starting point is a template mask generated specifically for that dataset. The method reliably and rapidly extracts brain images from whole head images, rendering them useable for subsequent analytical processing. This software tool will accelerate the exploitation of mouse models for the investigation of human brain disorders by MRI. Copyright © 2015 Elsevier B.V. All rights reserved.

GPU accelerated fuzzy connected image segmentation by using CUDA.

PubMed

Zhuge, Ying; Cao, Yong; Miller, Robert W

2009-01-01

Image segmentation techniques using fuzzy connectedness principles have shown their effectiveness in segmenting a variety of objects in several large applications in recent years. However, one problem of these algorithms has been their excessive computational requirements when processing large image datasets. Nowadays commodity graphics hardware provides high parallel computing power. In this paper, we present a parallel fuzzy connected image segmentation algorithm on Nvidia's Compute Unified Device Architecture (CUDA) platform for segmenting large medical image data sets. Our experiments based on three data sets with small, medium, and large data size demonstrate the efficiency of the parallel algorithm, which achieves a speed-up factor of 7.2x, 7.3x, and 14.4x, correspondingly, for the three data sets over the sequential implementation of fuzzy connected image segmentation algorithm on CPU.

The first MICCAI challenge on PET tumor segmentation.

PubMed

Hatt, Mathieu; Laurent, Baptiste; Ouahabi, Anouar; Fayad, Hadi; Tan, Shan; Li, Laquan; Lu, Wei; Jaouen, Vincent; Tauber, Clovis; Czakon, Jakub; Drapejkowski, Filip; Dyrka, Witold; Camarasu-Pop, Sorina; Cervenansky, Frédéric; Girard, Pascal; Glatard, Tristan; Kain, Michael; Yao, Yao; Barillot, Christian; Kirov, Assen; Visvikis, Dimitris

2018-02-01

Automatic functional volume segmentation in PET images is a challenge that has been addressed using a large array of methods. A major limitation for the field has been the lack of a benchmark dataset that would allow direct comparison of the results in the various publications. In the present work, we describe a comparison of recent methods on a large dataset following recommendations by the American Association of Physicists in Medicine (AAPM) task group (TG) 211, which was carried out within a MICCAI (Medical Image Computing and Computer Assisted Intervention) challenge. Organization and funding was provided by France Life Imaging (FLI). A dataset of 176 images combining simulated, phantom and clinical images was assembled. A website allowed the participants to register and download training data (n = 19). Challengers then submitted encapsulated pipelines on an online platform that autonomously ran the algorithms on the testing data (n = 157) and evaluated the results. The methods were ranked according to the arithmetic mean of sensitivity and positive predictive value. Sixteen teams registered but only four provided manuscripts and pipeline(s) for a total of 10 methods. In addition, results using two thresholds and the Fuzzy Locally Adaptive Bayesian (FLAB) were generated. All competing methods except one performed with median accuracy above 0.8. The method with the highest score was the convolutional neural network-based segmentation, which significantly outperformed 9 out of 12 of the other methods, but not the improved K-Means, Gaussian Model Mixture and Fuzzy C-Means methods. The most rigorous comparative study of PET segmentation algorithms to date was carried out using a dataset that is the largest used in such studies so far. The hierarchy amongst the methods in terms of accuracy did not depend strongly on the subset of datasets or the metrics (or combination of metrics). All the methods submitted by the challengers except one demonstrated good performance with median accuracy scores above 0.8. Copyright © 2017 Elsevier B.V. All rights reserved.

MMX-I: data-processing software for multimodal X-ray imaging and tomography

PubMed Central

Bergamaschi, Antoine; Medjoubi, Kadda; Messaoudi, Cédric; Marco, Sergio; Somogyi, Andrea

2016-01-01

A new multi-platform freeware has been developed for the processing and reconstruction of scanning multi-technique X-ray imaging and tomography datasets. The software platform aims to treat different scanning imaging techniques: X-ray fluorescence, phase, absorption and dark field and any of their combinations, thus providing an easy-to-use data processing tool for the X-ray imaging user community. A dedicated data input stream copes with the input and management of large datasets (several hundred GB) collected during a typical multi-technique fast scan at the Nanoscopium beamline and even on a standard PC. To the authors’ knowledge, this is the first software tool that aims at treating all of the modalities of scanning multi-technique imaging and tomography experiments. PMID:27140159

Improved opponent color local binary patterns: an effective local image descriptor for color texture classification

NASA Astrophysics Data System (ADS)

Bianconi, Francesco; Bello-Cerezo, Raquel; Napoletano, Paolo

2018-01-01

Texture classification plays a major role in many computer vision applications. Local binary patterns (LBP) encoding schemes have largely been proven to be very effective for this task. Improved LBP (ILBP) are conceptually simple, easy to implement, and highly effective LBP variants based on a point-to-average thresholding scheme instead of a point-to-point one. We propose the use of this encoding scheme for extracting intra- and interchannel features for color texture classification. We experimentally evaluated the resulting improved opponent color LBP alone and in concatenation with the ILBP of the local color contrast map on a set of image classification tasks over 9 datasets of generic color textures and 11 datasets of biomedical textures. The proposed approach outperformed other grayscale and color LBP variants in nearly all the datasets considered and proved competitive even against image features from last generation convolutional neural networks, particularly for the classification of biomedical images.

An Automatic Segmentation Method Combining an Active Contour Model and a Classification Technique for Detecting Polycomb-group Proteinsin High-Throughput Microscopy Images.

PubMed

Gregoretti, Francesco; Cesarini, Elisa; Lanzuolo, Chiara; Oliva, Gennaro; Antonelli, Laura

2016-01-01

The large amount of data generated in biological experiments that rely on advanced microscopy can be handled only with automated image analysis. Most analyses require a reliable cell image segmentation eventually capable of detecting subcellular structures.We present an automatic segmentation method to detect Polycomb group (PcG) proteins areas isolated from nuclei regions in high-resolution fluorescent cell image stacks. It combines two segmentation algorithms that use an active contour model and a classification technique serving as a tool to better understand the subcellular three-dimensional distribution of PcG proteins in live cell image sequences. We obtained accurate results throughout several cell image datasets, coming from different cell types and corresponding to different fluorescent labels, without requiring elaborate adjustments to each dataset.

Accessing Multi-Dimensional Images and Data Cubes in the Virtual Observatory

NASA Astrophysics Data System (ADS)

Tody, Douglas; Plante, R. L.; Berriman, G. B.; Cresitello-Dittmar, M.; Good, J.; Graham, M.; Greene, G.; Hanisch, R. J.; Jenness, T.; Lazio, J.; Norris, P.; Pevunova, O.; Rots, A. H.

2014-01-01

Telescopes across the spectrum are routinely producing multi-dimensional images and datasets, such as Doppler velocity cubes, polarization datasets, and time-resolved “movies.” Examples of current telescopes producing such multi-dimensional images include the JVLA, ALMA, and the IFU instruments on large optical and near-infrared wavelength telescopes. In the near future, both the LSST and JWST will also produce such multi-dimensional images routinely. High-energy instruments such as Chandra produce event datasets that are also a form of multi-dimensional data, in effect being a very sparse multi-dimensional image. Ensuring that the data sets produced by these telescopes can be both discovered and accessed by the community is essential and is part of the mission of the Virtual Observatory (VO). The Virtual Astronomical Observatory (VAO, http://www.usvao.org/), in conjunction with its international partners in the International Virtual Observatory Alliance (IVOA), has developed a protocol and an initial demonstration service designed for the publication, discovery, and access of arbitrarily large multi-dimensional images. The protocol describing multi-dimensional images is the Simple Image Access Protocol, version 2, which provides the minimal set of metadata required to characterize a multi-dimensional image for its discovery and access. A companion Image Data Model formally defines the semantics and structure of multi-dimensional images independently of how they are serialized, while providing capabilities such as support for sparse data that are essential to deal effectively with large cubes. A prototype data access service has been deployed and tested, using a suite of multi-dimensional images from a variety of telescopes. The prototype has demonstrated the capability to discover and remotely access multi-dimensional data via standard VO protocols. The prototype informs the specification of a protocol that will be submitted to the IVOA for approval, with an operational data cube service to be delivered in mid-2014. An associated user-installable VO data service framework will provide the capabilities required to publish VO-compatible multi-dimensional images or data cubes.

Self-organizing maps: a versatile tool for the automatic analysis of untargeted imaging datasets.

PubMed

Franceschi, Pietro; Wehrens, Ron

2014-04-01

MS-based imaging approaches allow for location-specific identification of chemical components in biological samples, opening up possibilities of much more detailed understanding of biological processes and mechanisms. Data analysis, however, is challenging, mainly because of the sheer size of such datasets. This article presents a novel approach based on self-organizing maps, extending previous work in order to be able to handle the large number of variables present in high-resolution mass spectra. The key idea is to generate prototype images, representing spatial distributions of ions, rather than prototypical mass spectra. This allows for a two-stage approach, first generating typical spatial distributions and associated m/z bins, and later analyzing the interesting bins in more detail using accurate masses. The possibilities and advantages of the new approach are illustrated on an in-house dataset of apple slices. © 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

Medical imaging informatics based solutions for human performance analytics

NASA Astrophysics Data System (ADS)

Verma, Sneha; McNitt-Gray, Jill; Liu, Brent J.

2018-03-01

For human performance analysis, extensive experimental trials are often conducted to identify the underlying cause or long-term consequences of certain pathologies and to improve motor functions by examining the movement patterns of affected individuals. Data collected for human performance analysis includes high-speed video, surveys, spreadsheets, force data recordings from instrumented surfaces etc. These datasets are recorded from various standalone sources and therefore captured in different folder structures as well as in varying formats depending on the hardware configurations. Therefore, data integration and synchronization present a huge challenge while handling these multimedia datasets specifically for large datasets. Another challenge faced by researchers is querying large quantity of unstructured data and to design feedbacks/reporting tools for users who need to use datasets at various levels. In the past, database server storage solutions have been introduced to securely store these datasets. However, to automate the process of uploading raw files, various file manipulation steps are required. In the current workflow, this file manipulation and structuring is done manually and is not feasible for large amounts of data. However, by attaching metadata files and data dictionaries with these raw datasets, they can provide information and structure needed for automated server upload. We introduce one such system for metadata creation for unstructured multimedia data based on the DICOM data model design. We will discuss design and implementation of this system and evaluate this system with data set collected for movement analysis study. The broader aim of this paper is to present a solutions space achievable based on medical imaging informatics design and methods for improvement in workflow for human performance analysis in a biomechanics research lab.

Genetic and Diagnostic Biomarker Development in ASD Toddlers Using Resting State Functional MRI

DTIC Science & Technology

2015-09-01

for public release; distribution unlimited Autism spectrum disorder (ASD); biomarker; early brain development; intrinsic functional brain networks...three large neuroimaging/neurobehavioral datasets to identify brain-imaging based biomarkers for Autism Spectrum Disorders (ASD). At Yale, we focus...neurobehavioral!datasets!in!order!to!identify! brainFimaging!based!biomarkers!for! Autism ! Spectrum ! Disorders !(ASD),!including!1)!BrainMap,! developed!and

Retinal fundus images for glaucoma analysis: the RIGA dataset

NASA Astrophysics Data System (ADS)

Almazroa, Ahmed; Alodhayb, Sami; Osman, Essameldin; Ramadan, Eslam; Hummadi, Mohammed; Dlaim, Mohammed; Alkatee, Muhannad; Raahemifar, Kaamran; Lakshminarayanan, Vasudevan

2018-03-01

Glaucoma neuropathy is a major cause of irreversible blindness worldwide. Current models of chronic care will not be able to close the gap of growing prevalence of glaucoma and challenges for access to healthcare services. Teleophthalmology is being developed to close this gap. In order to develop automated techniques for glaucoma detection which can be used in tele-ophthalmology we have developed a large retinal fundus dataset. A de-identified dataset of retinal fundus images for glaucoma analysis (RIGA) was derived from three sources for a total of 750 images. The optic cup and disc boundaries for each image was marked and annotated manually by six experienced ophthalmologists and included the cup to disc (CDR) estimates. Six parameters were extracted and assessed (the disc area and centroid, cup area and centroid, horizontal and vertical cup to disc ratios) among the ophthalmologists. The inter-observer annotations were compared by calculating the standard deviation (SD) for every image between the six ophthalmologists in order to determine if the outliers amongst the six and was used to filter the corresponding images. The data set will be made available to the research community in order to crowd source other analysis from other research groups in order to develop, validate and implement analysis algorithms appropriate for tele-glaucoma assessment. The RIGA dataset can be freely accessed online through University of Michigan, Deep Blue website (doi:10.7302/Z23R0R29).

Large-Scale Image Analytics Using Deep Learning

NASA Astrophysics Data System (ADS)

Ganguly, S.; Nemani, R. R.; Basu, S.; Mukhopadhyay, S.; Michaelis, A.; Votava, P.

2014-12-01

High resolution land cover classification maps are needed to increase the accuracy of current Land ecosystem and climate model outputs. Limited studies are in place that demonstrates the state-of-the-art in deriving very high resolution (VHR) land cover products. In addition, most methods heavily rely on commercial softwares that are difficult to scale given the region of study (e.g. continents to globe). Complexities in present approaches relate to (a) scalability of the algorithm, (b) large image data processing (compute and memory intensive), (c) computational cost, (d) massively parallel architecture, and (e) machine learning automation. In addition, VHR satellite datasets are of the order of terabytes and features extracted from these datasets are of the order of petabytes. In our present study, we have acquired the National Agricultural Imaging Program (NAIP) dataset for the Continental United States at a spatial resolution of 1-m. This data comes as image tiles (a total of quarter million image scenes with ~60 million pixels) and has a total size of ~100 terabytes for a single acquisition. Features extracted from the entire dataset would amount to ~8-10 petabytes. In our proposed approach, we have implemented a novel semi-automated machine learning algorithm rooted on the principles of "deep learning" to delineate the percentage of tree cover. In order to perform image analytics in such a granular system, it is mandatory to devise an intelligent archiving and query system for image retrieval, file structuring, metadata processing and filtering of all available image scenes. Using the Open NASA Earth Exchange (NEX) initiative, which is a partnership with Amazon Web Services (AWS), we have developed an end-to-end architecture for designing the database and the deep belief network (following the distbelief computing model) to solve a grand challenge of scaling this process across quarter million NAIP tiles that cover the entire Continental United States. The AWS core components that we use to solve this problem are DynamoDB along with S3 for database query and storage, ElastiCache shared memory architecture for image segmentation, Elastic Map Reduce (EMR) for image feature extraction, and the memory optimized Elastic Cloud Compute (EC2) for the learning algorithm.

Digital tissue and what it may reveal about the brain.

PubMed

Morgan, Josh L; Lichtman, Jeff W

2017-10-30

Imaging as a means of scientific data storage has evolved rapidly over the past century from hand drawings, to photography, to digital images. Only recently can sufficiently large datasets be acquired, stored, and processed such that tissue digitization can actually reveal more than direct observation of tissue. One field where this transformation is occurring is connectomics: the mapping of neural connections in large volumes of digitized brain tissue.

Multiple Auto-Adapting Color Balancing for Large Number of Images

NASA Astrophysics Data System (ADS)

Zhou, X.

2015-04-01

This paper presents a powerful technology of color balance between images. It does not only work for small number of images but also work for unlimited large number of images. Multiple adaptive methods are used. To obtain color seamless mosaic dataset, local color is adjusted adaptively towards the target color. Local statistics of the source images are computed based on the so-called adaptive dodging window. The adaptive target colors are statistically computed according to multiple target models. The gamma function is derived from the adaptive target and the adaptive source local stats. It is applied to the source images to obtain the color balanced output images. Five target color surface models are proposed. They are color point (or single color), color grid, 1st, 2nd and 3rd 2D polynomials. Least Square Fitting is used to obtain the polynomial target color surfaces. Target color surfaces are automatically computed based on all source images or based on an external target image. Some special objects such as water and snow are filtered by percentage cut or a given mask. Excellent results are achieved. The performance is extremely fast to support on-the-fly color balancing for large number of images (possible of hundreds of thousands images). Detailed algorithm and formulae are described. Rich examples including big mosaic datasets (e.g., contains 36,006 images) are given. Excellent results and performance are presented. The results show that this technology can be successfully used in various imagery to obtain color seamless mosaic. This algorithm has been successfully using in ESRI ArcGis.

Algorithm sensitivity analysis and parameter tuning for tissue image segmentation pipelines

PubMed Central

Kurç, Tahsin M.; Taveira, Luís F. R.; Melo, Alba C. M. A.; Gao, Yi; Kong, Jun; Saltz, Joel H.

2017-01-01

Abstract Motivation: Sensitivity analysis and parameter tuning are important processes in large-scale image analysis. They are very costly because the image analysis workflows are required to be executed several times to systematically correlate output variations with parameter changes or to tune parameters. An integrated solution with minimum user interaction that uses effective methodologies and high performance computing is required to scale these studies to large imaging datasets and expensive analysis workflows. Results: The experiments with two segmentation workflows show that the proposed approach can (i) quickly identify and prune parameters that are non-influential; (ii) search a small fraction (about 100 points) of the parameter search space with billions to trillions of points and improve the quality of segmentation results (Dice and Jaccard metrics) by as much as 1.42× compared to the results from the default parameters; (iii) attain good scalability on a high performance cluster with several effective optimizations. Conclusions: Our work demonstrates the feasibility of performing sensitivity analyses, parameter studies and auto-tuning with large datasets. The proposed framework can enable the quantification of error estimations and output variations in image segmentation pipelines. Availability and Implementation: Source code: https://github.com/SBU-BMI/region-templates/. Contact: teodoro@unb.br Supplementary information: Supplementary data are available at Bioinformatics online. PMID:28062445

Algorithm sensitivity analysis and parameter tuning for tissue image segmentation pipelines.

PubMed

Teodoro, George; Kurç, Tahsin M; Taveira, Luís F R; Melo, Alba C M A; Gao, Yi; Kong, Jun; Saltz, Joel H

2017-04-01

Sensitivity analysis and parameter tuning are important processes in large-scale image analysis. They are very costly because the image analysis workflows are required to be executed several times to systematically correlate output variations with parameter changes or to tune parameters. An integrated solution with minimum user interaction that uses effective methodologies and high performance computing is required to scale these studies to large imaging datasets and expensive analysis workflows. The experiments with two segmentation workflows show that the proposed approach can (i) quickly identify and prune parameters that are non-influential; (ii) search a small fraction (about 100 points) of the parameter search space with billions to trillions of points and improve the quality of segmentation results (Dice and Jaccard metrics) by as much as 1.42× compared to the results from the default parameters; (iii) attain good scalability on a high performance cluster with several effective optimizations. Our work demonstrates the feasibility of performing sensitivity analyses, parameter studies and auto-tuning with large datasets. The proposed framework can enable the quantification of error estimations and output variations in image segmentation pipelines. Source code: https://github.com/SBU-BMI/region-templates/ . teodoro@unb.br. Supplementary data are available at Bioinformatics online. © The Author 2016. Published by Oxford University Press.

A short feature vector for image matching: The Log-Polar Magnitude feature descriptor

PubMed Central

Hast, Anders; Wählby, Carolina; Sintorn, Ida-Maria

2017-01-01

The choice of an optimal feature detector-descriptor combination for image matching often depends on the application and the image type. In this paper, we propose the Log-Polar Magnitude feature descriptor—a rotation, scale, and illumination invariant descriptor that achieves comparable performance to SIFT on a large variety of image registration problems but with much shorter feature vectors. The descriptor is based on the Log-Polar Transform followed by a Fourier Transform and selection of the magnitude spectrum components. Selecting different frequency components allows optimizing for image patterns specific for a particular application. In addition, by relying only on coordinates of the found features and (optionally) feature sizes our descriptor is completely detector independent. We propose 48- or 56-long feature vectors that potentially can be shortened even further depending on the application. Shorter feature vectors result in better memory usage and faster matching. This combined with the fact that the descriptor does not require a time-consuming feature orientation estimation (the rotation invariance is achieved solely by using the magnitude spectrum of the Log-Polar Transform) makes it particularly attractive to applications with limited hardware capacity. Evaluation is performed on the standard Oxford dataset and two different microscopy datasets; one with fluorescence and one with transmission electron microscopy images. Our method performs better than SURF and comparable to SIFT on the Oxford dataset, and better than SIFT on both microscopy datasets indicating that it is particularly useful in applications with microscopy images. PMID:29190737

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

Lopez Torres, E., E-mail: Ernesto.Lopez.Torres@cern.ch, E-mail: cerello@to.infn.it; Fiorina, E.; Pennazio, F.

Purpose: M5L, a fully automated computer-aided detection (CAD) system for the detection and segmentation of lung nodules in thoracic computed tomography (CT), is presented and validated on several image datasets. Methods: M5L is the combination of two independent subsystems, based on the Channeler Ant Model as a segmentation tool [lung channeler ant model (lungCAM)] and on the voxel-based neural approach. The lungCAM was upgraded with a scan equalization module and a new procedure to recover the nodules connected to other lung structures; its classification module, which makes use of a feed-forward neural network, is based of a small number ofmore » features (13), so as to minimize the risk of lacking generalization, which could be possible given the large difference between the size of the training and testing datasets, which contain 94 and 1019 CTs, respectively. The lungCAM (standalone) and M5L (combined) performance was extensively tested on 1043 CT scans from three independent datasets, including a detailed analysis of the full Lung Image Database Consortium/Image Database Resource Initiative database, which is not yet found in literature. Results: The lungCAM and M5L performance is consistent across the databases, with a sensitivity of about 70% and 80%, respectively, at eight false positive findings per scan, despite the variable annotation criteria and acquisition and reconstruction conditions. A reduced sensitivity is found for subtle nodules and ground glass opacities (GGO) structures. A comparison with other CAD systems is also presented. Conclusions: The M5L performance on a large and heterogeneous dataset is stable and satisfactory, although the development of a dedicated module for GGOs detection could further improve it, as well as an iterative optimization of the training procedure. The main aim of the present study was accomplished: M5L results do not deteriorate when increasing the dataset size, making it a candidate for supporting radiologists on large scale screenings and clinical programs.« less

Mesoscale brain explorer, a flexible python-based image analysis and visualization tool.

PubMed

Haupt, Dirk; Vanni, Matthieu P; Bolanos, Federico; Mitelut, Catalin; LeDue, Jeffrey M; Murphy, Tim H

2017-07-01

Imaging of mesoscale brain activity is used to map interactions between brain regions. This work has benefited from the pioneering studies of Grinvald et al., who employed optical methods to image brain function by exploiting the properties of intrinsic optical signals and small molecule voltage-sensitive dyes. Mesoscale interareal brain imaging techniques have been advanced by cell targeted and selective recombinant indicators of neuronal activity. Spontaneous resting state activity is often collected during mesoscale imaging to provide the basis for mapping of connectivity relationships using correlation. However, the information content of mesoscale datasets is vast and is only superficially presented in manuscripts given the need to constrain measurements to a fixed set of frequencies, regions of interest, and other parameters. We describe a new open source tool written in python, termed mesoscale brain explorer (MBE), which provides an interface to process and explore these large datasets. The platform supports automated image processing pipelines with the ability to assess multiple trials and combine data from different animals. The tool provides functions for temporal filtering, averaging, and visualization of functional connectivity relations using time-dependent correlation. Here, we describe the tool and show applications, where previously published datasets were reanalyzed using MBE.

Steerable Principal Components for Space-Frequency Localized Images*

PubMed Central

Landa, Boris; Shkolnisky, Yoel

2017-01-01

As modern scientific image datasets typically consist of a large number of images of high resolution, devising methods for their accurate and efficient processing is a central research task. In this paper, we consider the problem of obtaining the steerable principal components of a dataset, a procedure termed “steerable PCA” (steerable principal component analysis). The output of the procedure is the set of orthonormal basis functions which best approximate the images in the dataset and all of their planar rotations. To derive such basis functions, we first expand the images in an appropriate basis, for which the steerable PCA reduces to the eigen-decomposition of a block-diagonal matrix. If we assume that the images are well localized in space and frequency, then such an appropriate basis is the prolate spheroidal wave functions (PSWFs). We derive a fast method for computing the PSWFs expansion coefficients from the images' equally spaced samples, via a specialized quadrature integration scheme, and show that the number of required quadrature nodes is similar to the number of pixels in each image. We then establish that our PSWF-based steerable PCA is both faster and more accurate then existing methods, and more importantly, provides us with rigorous error bounds on the entire procedure. PMID:29081879

An open, multi-vendor, multi-field-strength brain MR dataset and analysis of publicly available skull stripping methods agreement.

PubMed

Souza, Roberto; Lucena, Oeslle; Garrafa, Julia; Gobbi, David; Saluzzi, Marina; Appenzeller, Simone; Rittner, Letícia; Frayne, Richard; Lotufo, Roberto

2018-04-15

This paper presents an open, multi-vendor, multi-field strength magnetic resonance (MR) T1-weighted volumetric brain imaging dataset, named Calgary-Campinas-359 (CC-359). The dataset is composed of images of older healthy adults (29-80 years) acquired on scanners from three vendors (Siemens, Philips and General Electric) at both 1.5 T and 3 T. CC-359 is comprised of 359 datasets, approximately 60 subjects per vendor and magnetic field strength. The dataset is approximately age and gender balanced, subject to the constraints of the available images. It provides consensus brain extraction masks for all volumes generated using supervised classification. Manual segmentation results for twelve randomly selected subjects performed by an expert are also provided. The CC-359 dataset allows investigation of 1) the influences of both vendor and magnetic field strength on quantitative analysis of brain MR; 2) parameter optimization for automatic segmentation methods; and potentially 3) machine learning classifiers with big data, specifically those based on deep learning methods, as these approaches require a large amount of data. To illustrate the utility of this dataset, we compared to the results of a supervised classifier, the results of eight publicly available skull stripping methods and one publicly available consensus algorithm. A linear mixed effects model analysis indicated that vendor (p-value<0.001) and magnetic field strength (p-value<0.001) have statistically significant impacts on skull stripping results. Copyright © 2017 Elsevier Inc. All rights reserved.

Multi-scale pixel-based image fusion using multivariate empirical mode decomposition.

PubMed

Rehman, Naveed ur; Ehsan, Shoaib; Abdullah, Syed Muhammad Umer; Akhtar, Muhammad Jehanzaib; Mandic, Danilo P; McDonald-Maier, Klaus D

2015-05-08

A novel scheme to perform the fusion of multiple images using the multivariate empirical mode decomposition (MEMD) algorithm is proposed. Standard multi-scale fusion techniques make a priori assumptions regarding input data, whereas standard univariate empirical mode decomposition (EMD)-based fusion techniques suffer from inherent mode mixing and mode misalignment issues, characterized respectively by either a single intrinsic mode function (IMF) containing multiple scales or the same indexed IMFs corresponding to multiple input images carrying different frequency information. We show that MEMD overcomes these problems by being fully data adaptive and by aligning common frequency scales from multiple channels, thus enabling their comparison at a pixel level and subsequent fusion at multiple data scales. We then demonstrate the potential of the proposed scheme on a large dataset of real-world multi-exposure and multi-focus images and compare the results against those obtained from standard fusion algorithms, including the principal component analysis (PCA), discrete wavelet transform (DWT) and non-subsampled contourlet transform (NCT). A variety of image fusion quality measures are employed for the objective evaluation of the proposed method. We also report the results of a hypothesis testing approach on our large image dataset to identify statistically-significant performance differences.

Multi-Scale Pixel-Based Image Fusion Using Multivariate Empirical Mode Decomposition

PubMed Central

Rehman, Naveed ur; Ehsan, Shoaib; Abdullah, Syed Muhammad Umer; Akhtar, Muhammad Jehanzaib; Mandic, Danilo P.; McDonald-Maier, Klaus D.

2015-01-01

A novel scheme to perform the fusion of multiple images using the multivariate empirical mode decomposition (MEMD) algorithm is proposed. Standard multi-scale fusion techniques make a priori assumptions regarding input data, whereas standard univariate empirical mode decomposition (EMD)-based fusion techniques suffer from inherent mode mixing and mode misalignment issues, characterized respectively by either a single intrinsic mode function (IMF) containing multiple scales or the same indexed IMFs corresponding to multiple input images carrying different frequency information. We show that MEMD overcomes these problems by being fully data adaptive and by aligning common frequency scales from multiple channels, thus enabling their comparison at a pixel level and subsequent fusion at multiple data scales. We then demonstrate the potential of the proposed scheme on a large dataset of real-world multi-exposure and multi-focus images and compare the results against those obtained from standard fusion algorithms, including the principal component analysis (PCA), discrete wavelet transform (DWT) and non-subsampled contourlet transform (NCT). A variety of image fusion quality measures are employed for the objective evaluation of the proposed method. We also report the results of a hypothesis testing approach on our large image dataset to identify statistically-significant performance differences. PMID:26007714

Virtual unfolding of light sheet fluorescence microscopy dataset for quantitative analysis of the mouse intestine

NASA Astrophysics Data System (ADS)

Candeo, Alessia; Sana, Ilenia; Ferrari, Eleonora; Maiuri, Luigi; D'Andrea, Cosimo; Valentini, Gianluca; Bassi, Andrea

2016-05-01

Light sheet fluorescence microscopy has proven to be a powerful tool to image fixed and chemically cleared samples, providing in depth and high resolution reconstructions of intact mouse organs. We applied light sheet microscopy to image the mouse intestine. We found that large portions of the sample can be readily visualized, assessing the organ status and highlighting the presence of regions with impaired morphology. Yet, three-dimensional (3-D) sectioning of the intestine leads to a large dataset that produces unnecessary storage and processing overload. We developed a routine that extracts the relevant information from a large image stack and provides quantitative analysis of the intestine morphology. This result was achieved by a three step procedure consisting of: (1) virtually unfold the 3-D reconstruction of the intestine; (2) observe it layer-by-layer; and (3) identify distinct villi and statistically analyze multiple samples belonging to different intestinal regions. Even if the procedure has been developed for the murine intestine, most of the underlying concepts have a general applicability.

Classification of Large-Scale Remote Sensing Images for Automatic Identification of Health Hazards: Smoke Detection Using an Autologistic Regression Classifier.

PubMed

Wolters, Mark A; Dean, C B

2017-01-01

Remote sensing images from Earth-orbiting satellites are a potentially rich data source for monitoring and cataloguing atmospheric health hazards that cover large geographic regions. A method is proposed for classifying such images into hazard and nonhazard regions using the autologistic regression model, which may be viewed as a spatial extension of logistic regression. The method includes a novel and simple approach to parameter estimation that makes it well suited to handling the large and high-dimensional datasets arising from satellite-borne instruments. The methodology is demonstrated on both simulated images and a real application to the identification of forest fire smoke.

Path Searching Based Crease Detection for Large Scale Scanned Document Images

NASA Astrophysics Data System (ADS)

Zhang, Jifu; Li, Yi; Li, Shutao; Sun, Bin; Sun, Jun

2017-12-01

Since the large size documents are usually folded for preservation, creases will occur in the scanned images. In this paper, a crease detection method is proposed to locate the crease pixels for further processing. According to the imaging process of contactless scanners, the shading on both sides of the crease usually varies a lot. Based on this observation, a convex hull based algorithm is adopted to extract the shading information of the scanned image. Then, the possible crease path can be achieved by applying the vertical filter and morphological operations on the shading image. Finally, the accurate crease is detected via Dijkstra path searching. Experimental results on the dataset of real scanned newspapers demonstrate that the proposed method can obtain accurate locations of the creases in the large size document images.

Immersive Interaction, Manipulation and Analysis of Large 3D Datasets for Planetary and Earth Sciences

NASA Astrophysics Data System (ADS)

Pariser, O.; Calef, F.; Manning, E. M.; Ardulov, V.

2017-12-01

We will present implementation and study of several use-cases of utilizing Virtual Reality (VR) for immersive display, interaction and analysis of large and complex 3D datasets. These datasets have been acquired by the instruments across several Earth, Planetary and Solar Space Robotics Missions. First, we will describe the architecture of the common application framework that was developed to input data, interface with VR display devices and program input controllers in various computing environments. Tethered and portable VR technologies will be contrasted and advantages of each highlighted. We'll proceed to presenting experimental immersive analytics visual constructs that enable augmentation of 3D datasets with 2D ones such as images and statistical and abstract data. We will conclude by presenting comparative analysis with traditional visualization applications and share the feedback provided by our users: scientists and engineers.

A handheld computer-aided diagnosis system and simulated analysis

NASA Astrophysics Data System (ADS)

Su, Mingjian; Zhang, Xuejun; Liu, Brent; Su, Kening; Louie, Ryan

2016-03-01

This paper describes a Computer Aided Diagnosis (CAD) system based on cellphone and distributed cluster. One of the bottlenecks in building a CAD system for clinical practice is the storage and process of mass pathology samples freely among different devices, and normal pattern matching algorithm on large scale image set is very time consuming. Distributed computation on cluster has demonstrated the ability to relieve this bottleneck. We develop a system enabling the user to compare the mass image to a dataset with feature table by sending datasets to Generic Data Handler Module in Hadoop, where the pattern recognition is undertaken for the detection of skin diseases. A single and combination retrieval algorithm to data pipeline base on Map Reduce framework is used in our system in order to make optimal choice between recognition accuracy and system cost. The profile of lesion area is drawn by doctors manually on the screen, and then uploads this pattern to the server. In our evaluation experiment, an accuracy of 75% diagnosis hit rate is obtained by testing 100 patients with skin illness. Our system has the potential help in building a novel medical image dataset by collecting large amounts of gold standard during medical diagnosis. Once the project is online, the participants are free to join and eventually an abundant sample dataset will soon be gathered enough for learning. These results demonstrate our technology is very promising and expected to be used in clinical practice.

Implementing a Parallel Image Edge Detection Algorithm Based on the Otsu-Canny Operator on the Hadoop Platform.

PubMed

Cao, Jianfang; Chen, Lichao; Wang, Min; Tian, Yun

2018-01-01

The Canny operator is widely used to detect edges in images. However, as the size of the image dataset increases, the edge detection performance of the Canny operator decreases and its runtime becomes excessive. To improve the runtime and edge detection performance of the Canny operator, in this paper, we propose a parallel design and implementation for an Otsu-optimized Canny operator using a MapReduce parallel programming model that runs on the Hadoop platform. The Otsu algorithm is used to optimize the Canny operator's dual threshold and improve the edge detection performance, while the MapReduce parallel programming model facilitates parallel processing for the Canny operator to solve the processing speed and communication cost problems that occur when the Canny edge detection algorithm is applied to big data. For the experiments, we constructed datasets of different scales from the Pascal VOC2012 image database. The proposed parallel Otsu-Canny edge detection algorithm performs better than other traditional edge detection algorithms. The parallel approach reduced the running time by approximately 67.2% on a Hadoop cluster architecture consisting of 5 nodes with a dataset of 60,000 images. Overall, our approach system speeds up the system by approximately 3.4 times when processing large-scale datasets, which demonstrates the obvious superiority of our method. The proposed algorithm in this study demonstrates both better edge detection performance and improved time performance.

Extension of research data repository system to support direct compute access to biomedical datasets: enhancing Dataverse to support large datasets

PubMed Central

McKinney, Bill; Meyer, Peter A.; Crosas, Mercè; Sliz, Piotr

2016-01-01

Access to experimental X-ray diffraction image data is important for validation and reproduction of macromolecular models and indispensable for the development of structural biology processing methods. In response to the evolving needs of the structural biology community, we recently established a diffraction data publication system, the Structural Biology Data Grid (SBDG, data.sbgrid.org), to preserve primary experimental datasets supporting scientific publications. All datasets published through the SBDG are freely available to the research community under a public domain dedication license, with metadata compliant with the DataCite Schema (schema.datacite.org). A proof-of-concept study demonstrated community interest and utility. Publication of large datasets is a challenge shared by several fields, and the SBDG has begun collaborating with the Institute for Quantitative Social Science at Harvard University to extend the Dataverse (dataverse.org) open-source data repository system to structural biology datasets. Several extensions are necessary to support the size and metadata requirements for structural biology datasets. In this paper, we describe one such extension—functionality supporting preservation of filesystem structure within Dataverse—which is essential for both in-place computation and supporting non-http data transfers. PMID:27862010

XML-based data model and architecture for a knowledge-based grid-enabled problem-solving environment for high-throughput biological imaging.

PubMed

Ahmed, Wamiq M; Lenz, Dominik; Liu, Jia; Paul Robinson, J; Ghafoor, Arif

2008-03-01

High-throughput biological imaging uses automated imaging devices to collect a large number of microscopic images for analysis of biological systems and validation of scientific hypotheses. Efficient manipulation of these datasets for knowledge discovery requires high-performance computational resources, efficient storage, and automated tools for extracting and sharing such knowledge among different research sites. Newly emerging grid technologies provide powerful means for exploiting the full potential of these imaging techniques. Efficient utilization of grid resources requires the development of knowledge-based tools and services that combine domain knowledge with analysis algorithms. In this paper, we first investigate how grid infrastructure can facilitate high-throughput biological imaging research, and present an architecture for providing knowledge-based grid services for this field. We identify two levels of knowledge-based services. The first level provides tools for extracting spatiotemporal knowledge from image sets and the second level provides high-level knowledge management and reasoning services. We then present cellular imaging markup language, an extensible markup language-based language for modeling of biological images and representation of spatiotemporal knowledge. This scheme can be used for spatiotemporal event composition, matching, and automated knowledge extraction and representation for large biological imaging datasets. We demonstrate the expressive power of this formalism by means of different examples and extensive experimental results.

Updates to FuncLab, a Matlab based GUI for handling receiver functions

NASA Astrophysics Data System (ADS)

Porritt, Robert W.; Miller, Meghan S.

2018-02-01

Receiver functions are a versatile tool commonly used in seismic imaging. Depending on how they are processed, they can be used to image discontinuity structure within the crust or mantle or they can be inverted for seismic velocity either directly or jointly with complementary datasets. However, modern studies generally require large datasets which can be challenging to handle; therefore, FuncLab was originally written as an interactive Matlab GUI to assist in handling these large datasets. This software uses a project database to allow interactive trace editing, data visualization, H-κ stacking for crustal thickness and Vp/Vs ratio, and common conversion point stacking while minimizing computational costs. Since its initial release, significant advances have been made in the implementation of web services and changes in the underlying Matlab platform have necessitated a significant revision to the software. Here, we present revisions to the software, including new features such as data downloading via irisFetch.m, receiver function calculations via processRFmatlab, on-the-fly cross-section tools, interface picking, and more. In the descriptions of the tools, we present its application to a test dataset in Michigan, Wisconsin, and neighboring areas following the passage of USArray Transportable Array. The software is made available online at https://robporritt.wordpress.com/software.

TDat: An Efficient Platform for Processing Petabyte-Scale Whole-Brain Volumetric Images.

PubMed

Li, Yuxin; Gong, Hui; Yang, Xiaoquan; Yuan, Jing; Jiang, Tao; Li, Xiangning; Sun, Qingtao; Zhu, Dan; Wang, Zhenyu; Luo, Qingming; Li, Anan

2017-01-01

Three-dimensional imaging of whole mammalian brains at single-neuron resolution has generated terabyte (TB)- and even petabyte (PB)-sized datasets. Due to their size, processing these massive image datasets can be hindered by the computer hardware and software typically found in biological laboratories. To fill this gap, we have developed an efficient platform named TDat, which adopts a novel data reformatting strategy by reading cuboid data and employing parallel computing. In data reformatting, TDat is more efficient than any other software. In data accessing, we adopted parallelization to fully explore the capability for data transmission in computers. We applied TDat in large-volume data rigid registration and neuron tracing in whole-brain data with single-neuron resolution, which has never been demonstrated in other studies. We also showed its compatibility with various computing platforms, image processing software and imaging systems.

The coming paradigm shift: A transition from manual to automated microscopy.

PubMed

Farahani, Navid; Monteith, Corey E

2016-01-01

The field of pathology has used light microscopy (LM) extensively since the mid-19(th) century for examination of histological tissue preparations. This technology has remained the foremost tool in use by pathologists even as other fields have undergone a great change in recent years through new technologies. However, as new microscopy techniques are perfected and made available, this reliance on the standard LM will likely begin to change. Advanced imaging involving both diffraction-limited and subdiffraction techniques are bringing nondestructive, high-resolution, molecular-level imaging to pathology. Some of these technologies can produce three-dimensional (3D) datasets from sampled tissues. In addition, block-face/tissue-sectioning techniques are already providing automated, large-scale 3D datasets of whole specimens. These datasets allow pathologists to see an entire sample with all of its spatial information intact, and furthermore allow image analysis such as detection, segmentation, and classification, which are impossible in standard LM. It is likely that these technologies herald a major paradigm shift in the field of pathology.

Detection of neuron membranes in electron microscopy images using a serial neural network architecture.

PubMed

Jurrus, Elizabeth; Paiva, Antonio R C; Watanabe, Shigeki; Anderson, James R; Jones, Bryan W; Whitaker, Ross T; Jorgensen, Erik M; Marc, Robert E; Tasdizen, Tolga

2010-12-01

Study of nervous systems via the connectome, the map of connectivities of all neurons in that system, is a challenging problem in neuroscience. Towards this goal, neurobiologists are acquiring large electron microscopy datasets. However, the shear volume of these datasets renders manual analysis infeasible. Hence, automated image analysis methods are required for reconstructing the connectome from these very large image collections. Segmentation of neurons in these images, an essential step of the reconstruction pipeline, is challenging because of noise, anisotropic shapes and brightness, and the presence of confounding structures. The method described in this paper uses a series of artificial neural networks (ANNs) in a framework combined with a feature vector that is composed of image intensities sampled over a stencil neighborhood. Several ANNs are applied in series allowing each ANN to use the classification context provided by the previous network to improve detection accuracy. We develop the method of serial ANNs and show that the learned context does improve detection over traditional ANNs. We also demonstrate advantages over previous membrane detection methods. The results are a significant step towards an automated system for the reconstruction of the connectome. Copyright 2010 Elsevier B.V. All rights reserved.

Segmentation of mosaicism in cervicographic images using support vector machines

NASA Astrophysics Data System (ADS)

Xue, Zhiyun; Long, L. Rodney; Antani, Sameer; Jeronimo, Jose; Thoma, George R.

2009-02-01

The National Library of Medicine (NLM), in collaboration with the National Cancer Institute (NCI), is creating a large digital repository of cervicographic images for the study of uterine cervix cancer prevention. One of the research goals is to automatically detect diagnostic bio-markers in these images. Reliable bio-marker segmentation in large biomedical image collections is a challenging task due to the large variation in image appearance. Methods described in this paper focus on segmenting mosaicism, which is an important vascular feature used to visually assess the degree of cervical intraepithelial neoplasia. The proposed approach uses support vector machines (SVM) trained on a ground truth dataset annotated by medical experts (which circumvents the need for vascular structure extraction). We have evaluated the performance of the proposed algorithm and experimentally demonstrated its feasibility.

Pathological brain detection based on wavelet entropy and Hu moment invariants.

PubMed

Zhang, Yudong; Wang, Shuihua; Sun, Ping; Phillips, Preetha

2015-01-01

With the aim of developing an accurate pathological brain detection system, we proposed a novel automatic computer-aided diagnosis (CAD) to detect pathological brains from normal brains obtained by magnetic resonance imaging (MRI) scanning. The problem still remained a challenge for technicians and clinicians, since MR imaging generated an exceptionally large information dataset. A new two-step approach was proposed in this study. We used wavelet entropy (WE) and Hu moment invariants (HMI) for feature extraction, and the generalized eigenvalue proximal support vector machine (GEPSVM) for classification. To further enhance classification accuracy, the popular radial basis function (RBF) kernel was employed. The 10 runs of k-fold stratified cross validation result showed that the proposed "WE + HMI + GEPSVM + RBF" method was superior to existing methods w.r.t. classification accuracy. It obtained the average classification accuracies of 100%, 100%, and 99.45% over Dataset-66, Dataset-160, and Dataset-255, respectively. The proposed method is effective and can be applied to realistic use.

IMAGE EXPLORER: Astronomical Image Analysis on an HTML5-based Web Application

NASA Astrophysics Data System (ADS)

Gopu, A.; Hayashi, S.; Young, M. D.

2014-05-01

Large datasets produced by recent astronomical imagers cause the traditional paradigm for basic visual analysis - typically downloading one's entire image dataset and using desktop clients like DS9, Aladin, etc. - to not scale, despite advances in desktop computing power and storage. This paper describes Image Explorer, a web framework that offers several of the basic visualization and analysis functionality commonly provided by tools like DS9, on any HTML5 capable web browser on various platforms. It uses a combination of the modern HTML5 canvas, JavaScript, and several layers of lossless PNG tiles producted from the FITS image data. Astronomers are able to rapidly and simultaneously open up several images on their web-browser, adjust the intensity min/max cutoff or its scaling function, and zoom level, apply color-maps, view position and FITS header information, execute typically used data reduction codes on the corresponding FITS data using the FRIAA framework, and overlay tiles for source catalog objects, etc.

Extraction of Built-Up Areas Using Convolutional Neural Networks and Transfer Learning from SENTINEL-2 Satellite Images

NASA Astrophysics Data System (ADS)

Bramhe, V. S.; Ghosh, S. K.; Garg, P. K.

2018-04-01

With rapid globalization, the extent of built-up areas is continuously increasing. Extraction of features for classifying built-up areas that are more robust and abstract is a leading research topic from past many years. Although, various studies have been carried out where spatial information along with spectral features has been utilized to enhance the accuracy of classification. Still, these feature extraction techniques require a large number of user-specific parameters and generally application specific. On the other hand, recently introduced Deep Learning (DL) techniques requires less number of parameters to represent more abstract aspects of the data without any manual effort. Since, it is difficult to acquire high-resolution datasets for applications that require large scale monitoring of areas. Therefore, in this study Sentinel-2 image has been used for built-up areas extraction. In this work, pre-trained Convolutional Neural Networks (ConvNets) i.e. Inception v3 and VGGNet are employed for transfer learning. Since these networks are trained on generic images of ImageNet dataset which are having very different characteristics from satellite images. Therefore, weights of networks are fine-tuned using data derived from Sentinel-2 images. To compare the accuracies with existing shallow networks, two state of art classifiers i.e. Gaussian Support Vector Machine (SVM) and Back-Propagation Neural Network (BP-NN) are also implemented. Both SVM and BP-NN gives 84.31 % and 82.86 % overall accuracies respectively. Inception-v3 and VGGNet gives 89.43 % of overall accuracy using fine-tuned VGGNet and 92.10 % when using Inception-v3. The results indicate high accuracy of proposed fine-tuned ConvNets on a 4-channel Sentinel-2 dataset for built-up area extraction.

A deep learning method for classifying mammographic breast density categories.

PubMed

Mohamed, Aly A; Berg, Wendie A; Peng, Hong; Luo, Yahong; Jankowitz, Rachel C; Wu, Shandong

2018-01-01

Mammographic breast density is an established risk marker for breast cancer and is visually assessed by radiologists in routine mammogram image reading, using four qualitative Breast Imaging and Reporting Data System (BI-RADS) breast density categories. It is particularly difficult for radiologists to consistently distinguish the two most common and most variably assigned BI-RADS categories, i.e., "scattered density" and "heterogeneously dense". The aim of this work was to investigate a deep learning-based breast density classifier to consistently distinguish these two categories, aiming at providing a potential computerized tool to assist radiologists in assigning a BI-RADS category in current clinical workflow. In this study, we constructed a convolutional neural network (CNN)-based model coupled with a large (i.e., 22,000 images) digital mammogram imaging dataset to evaluate the classification performance between the two aforementioned breast density categories. All images were collected from a cohort of 1,427 women who underwent standard digital mammography screening from 2005 to 2016 at our institution. The truths of the density categories were based on standard clinical assessment made by board-certified breast imaging radiologists. Effects of direct training from scratch solely using digital mammogram images and transfer learning of a pretrained model on a large nonmedical imaging dataset were evaluated for the specific task of breast density classification. In order to measure the classification performance, the CNN classifier was also tested on a refined version of the mammogram image dataset by removing some potentially inaccurately labeled images. Receiver operating characteristic (ROC) curves and the area under the curve (AUC) were used to measure the accuracy of the classifier. The AUC was 0.9421 when the CNN-model was trained from scratch on our own mammogram images, and the accuracy increased gradually along with an increased size of training samples. Using the pretrained model followed by a fine-tuning process with as few as 500 mammogram images led to an AUC of 0.9265. After removing the potentially inaccurately labeled images, AUC was increased to 0.9882 and 0.9857 for without and with the pretrained model, respectively, both significantly higher (P < 0.001) than when using the full imaging dataset. Our study demonstrated high classification accuracies between two difficult to distinguish breast density categories that are routinely assessed by radiologists. We anticipate that our approach will help enhance current clinical assessment of breast density and better support consistent density notification to patients in breast cancer screening. © 2017 American Association of Physicists in Medicine.

Human3.6M: Large Scale Datasets and Predictive Methods for 3D Human Sensing in Natural Environments.

PubMed

Ionescu, Catalin; Papava, Dragos; Olaru, Vlad; Sminchisescu, Cristian

2014-07-01

We introduce a new dataset, Human3.6M, of 3.6 Million accurate 3D Human poses, acquired by recording the performance of 5 female and 6 male subjects, under 4 different viewpoints, for training realistic human sensing systems and for evaluating the next generation of human pose estimation models and algorithms. Besides increasing the size of the datasets in the current state-of-the-art by several orders of magnitude, we also aim to complement such datasets with a diverse set of motions and poses encountered as part of typical human activities (taking photos, talking on the phone, posing, greeting, eating, etc.), with additional synchronized image, human motion capture, and time of flight (depth) data, and with accurate 3D body scans of all the subject actors involved. We also provide controlled mixed reality evaluation scenarios where 3D human models are animated using motion capture and inserted using correct 3D geometry, in complex real environments, viewed with moving cameras, and under occlusion. Finally, we provide a set of large-scale statistical models and detailed evaluation baselines for the dataset illustrating its diversity and the scope for improvement by future work in the research community. Our experiments show that our best large-scale model can leverage our full training set to obtain a 20% improvement in performance compared to a training set of the scale of the largest existing public dataset for this problem. Yet the potential for improvement by leveraging higher capacity, more complex models with our large dataset, is substantially vaster and should stimulate future research. The dataset together with code for the associated large-scale learning models, features, visualization tools, as well as the evaluation server, is available online at http://vision.imar.ro/human3.6m.

Implementing a Parallel Image Edge Detection Algorithm Based on the Otsu-Canny Operator on the Hadoop Platform

PubMed Central

Wang, Min; Tian, Yun

2018-01-01

The Canny operator is widely used to detect edges in images. However, as the size of the image dataset increases, the edge detection performance of the Canny operator decreases and its runtime becomes excessive. To improve the runtime and edge detection performance of the Canny operator, in this paper, we propose a parallel design and implementation for an Otsu-optimized Canny operator using a MapReduce parallel programming model that runs on the Hadoop platform. The Otsu algorithm is used to optimize the Canny operator's dual threshold and improve the edge detection performance, while the MapReduce parallel programming model facilitates parallel processing for the Canny operator to solve the processing speed and communication cost problems that occur when the Canny edge detection algorithm is applied to big data. For the experiments, we constructed datasets of different scales from the Pascal VOC2012 image database. The proposed parallel Otsu-Canny edge detection algorithm performs better than other traditional edge detection algorithms. The parallel approach reduced the running time by approximately 67.2% on a Hadoop cluster architecture consisting of 5 nodes with a dataset of 60,000 images. Overall, our approach system speeds up the system by approximately 3.4 times when processing large-scale datasets, which demonstrates the obvious superiority of our method. The proposed algorithm in this study demonstrates both better edge detection performance and improved time performance. PMID:29861711

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.

Extension of research data repository system to support direct compute access to biomedical datasets: enhancing Dataverse to support large datasets.

PubMed

McKinney, Bill; Meyer, Peter A; Crosas, Mercè; Sliz, Piotr

2017-01-01

Access to experimental X-ray diffraction image data is important for validation and reproduction of macromolecular models and indispensable for the development of structural biology processing methods. In response to the evolving needs of the structural biology community, we recently established a diffraction data publication system, the Structural Biology Data Grid (SBDG, data.sbgrid.org), to preserve primary experimental datasets supporting scientific publications. All datasets published through the SBDG are freely available to the research community under a public domain dedication license, with metadata compliant with the DataCite Schema (schema.datacite.org). A proof-of-concept study demonstrated community interest and utility. Publication of large datasets is a challenge shared by several fields, and the SBDG has begun collaborating with the Institute for Quantitative Social Science at Harvard University to extend the Dataverse (dataverse.org) open-source data repository system to structural biology datasets. Several extensions are necessary to support the size and metadata requirements for structural biology datasets. In this paper, we describe one such extension-functionality supporting preservation of file system structure within Dataverse-which is essential for both in-place computation and supporting non-HTTP data transfers. © 2016 New York Academy of Sciences.

HCP: A Flexible CNN Framework for Multi-label Image Classification.

PubMed

Wei, Yunchao; Xia, Wei; Lin, Min; Huang, Junshi; Ni, Bingbing; Dong, Jian; Zhao, Yao; Yan, Shuicheng

2015-10-26

Convolutional Neural Network (CNN) has demonstrated promising performance in single-label image classification tasks. However, how CNN best copes with multi-label images still remains an open problem, mainly due to the complex underlying object layouts and insufficient multi-label training images. In this work, we propose a flexible deep CNN infrastructure, called Hypotheses-CNN-Pooling (HCP), where an arbitrary number of object segment hypotheses are taken as the inputs, then a shared CNN is connected with each hypothesis, and finally the CNN output results from different hypotheses are aggregated with max pooling to produce the ultimate multi-label predictions. Some unique characteristics of this flexible deep CNN infrastructure include: 1) no ground-truth bounding box information is required for training; 2) the whole HCP infrastructure is robust to possibly noisy and/or redundant hypotheses; 3) the shared CNN is flexible and can be well pre-trained with a large-scale single-label image dataset, e.g., ImageNet; and 4) it may naturally output multi-label prediction results. Experimental results on Pascal VOC 2007 and VOC 2012 multi-label image datasets well demonstrate the superiority of the proposed HCP infrastructure over other state-of-the-arts. In particular, the mAP reaches 90.5% by HCP only and 93.2% after the fusion with our complementary result in [44] based on hand-crafted features on the VOC 2012 dataset.

Parallel task processing of very large datasets

NASA Astrophysics Data System (ADS)

Romig, Phillip Richardson, III

This research concerns the use of distributed computer technologies for the analysis and management of very large datasets. Improvements in sensor technology, an emphasis on global change research, and greater access to data warehouses all are increase the number of non-traditional users of remotely sensed data. We present a framework for distributed solutions to the challenges of datasets which exceed the online storage capacity of individual workstations. This framework, called parallel task processing (PTP), incorporates both the task- and data-level parallelism exemplified by many image processing operations. An implementation based on the principles of PTP, called Tricky, is also presented. Additionally, we describe the challenges and practical issues in modeling the performance of parallel task processing with large datasets. We present a mechanism for estimating the running time of each unit of work within a system and an algorithm that uses these estimates to simulate the execution environment and produce estimated runtimes. Finally, we describe and discuss experimental results which validate the design. Specifically, the system (a) is able to perform computation on datasets which exceed the capacity of any one disk, (b) provides reduction of overall computation time as a result of the task distribution even with the additional cost of data transfer and management, and (c) in the simulation mode accurately predicts the performance of the real execution environment.

Computer-based analysis of microvascular alterations in a mouse model for Alzheimer's disease

NASA Astrophysics Data System (ADS)

Heinzer, Stefan; Müller, Ralph; Stampanoni, Marco; Abela, Rafael; Meyer, Eric P.; Ulmann-Schuler, Alexandra; Krucker, Thomas

2007-03-01

Vascular factors associated with Alzheimer's disease (AD) have recently gained increased attention. To investigate changes in vascular, particularly microvascular architecture, we developed a hierarchical imaging framework to obtain large-volume, high-resolution 3D images from brains of transgenic mice modeling AD. In this paper, we present imaging and data analysis methods which allow compiling unique characteristics from several hundred gigabytes of image data. Image acquisition is based on desktop micro-computed tomography (µCT) and local synchrotron-radiation µCT (SRµCT) scanning with a nominal voxel size of 16 µm and 1.4 µm, respectively. Two visualization approaches were implemented: stacks of Z-buffer projections for fast data browsing, and progressive-mesh based surface rendering for detailed 3D visualization of the large datasets. In a first step, image data was assessed visually via a Java client connected to a central database. Identified characteristics of interest were subsequently quantified using global morphometry software. To obtain even deeper insight into microvascular alterations, tree analysis software was developed providing local morphometric parameters such as number of vessel segments or vessel tortuosity. In the context of ever increasing image resolution and large datasets, computer-aided analysis has proven both powerful and indispensable. The hierarchical approach maintains the context of local phenomena, while proper visualization and morphometry provide the basis for detailed analysis of the pathology related to structure. Beyond analysis of microvascular changes in AD this framework will have significant impact considering that vascular changes are involved in other neurodegenerative diseases as well as in cancer, cardiovascular disease, asthma, and arthritis.

Effective 2D-3D medical image registration using Support Vector Machine.

PubMed

Qi, Wenyuan; Gu, Lixu; Zhao, Qiang

2008-01-01

Registration of pre-operative 3D volume dataset and intra-operative 2D images gradually becomes an important technique to assist radiologists in diagnosing complicated diseases easily and quickly. In this paper, we proposed a novel 2D/3D registration framework based on Support Vector Machine (SVM) to compensate the disadvantages of generating large number of DRR images in the stage of intra-operation. Estimated similarity metric distribution could be built up from the relationship between parameters of transform and prior sparse target metric values by means of SVR method. Based on which, global optimal parameters of transform are finally searched out by an optimizer in order to guide 3D volume dataset to match intra-operative 2D image. Experiments reveal that our proposed registration method improved performance compared to conventional registration method and also provided a precise registration result efficiently.

GeoPAT: A toolbox for pattern-based information retrieval from large geospatial databases

NASA Astrophysics Data System (ADS)

Jasiewicz, Jarosław; Netzel, Paweł; Stepinski, Tomasz

2015-07-01

Geospatial Pattern Analysis Toolbox (GeoPAT) is a collection of GRASS GIS modules for carrying out pattern-based geospatial analysis of images and other spatial datasets. The need for pattern-based analysis arises when images/rasters contain rich spatial information either because of their very high resolution or their very large spatial extent. Elementary units of pattern-based analysis are scenes - patches of surface consisting of a complex arrangement of individual pixels (patterns). GeoPAT modules implement popular GIS algorithms, such as query, overlay, and segmentation, to operate on the grid of scenes. To achieve these capabilities GeoPAT includes a library of scene signatures - compact numerical descriptors of patterns, and a library of distance functions - providing numerical means of assessing dissimilarity between scenes. Ancillary GeoPAT modules use these functions to construct a grid of scenes or to assign signatures to individual scenes having regular or irregular geometries. Thus GeoPAT combines knowledge retrieval from patterns with mapping tasks within a single integrated GIS environment. GeoPAT is designed to identify and analyze complex, highly generalized classes in spatial datasets. Examples include distinguishing between different styles of urban settlements using VHR images, delineating different landscape types in land cover maps, and mapping physiographic units from DEM. The concept of pattern-based spatial analysis is explained and the roles of all modules and functions are described. A case study example pertaining to delineation of landscape types in a subregion of NLCD is given. Performance evaluation is included to highlight GeoPAT's applicability to very large datasets. The GeoPAT toolbox is available for download from

The Function Biomedical Informatics Research Network Data Repository

PubMed Central

Keator, David B.; van Erp, Theo G.M.; Turner, Jessica A.; Glover, Gary H.; Mueller, Bryon A.; Liu, Thomas T.; Voyvodic, James T.; Rasmussen, Jerod; Calhoun, Vince D.; Lee, Hyo Jong; Toga, Arthur W.; McEwen, Sarah; Ford, Judith M.; Mathalon, Daniel H.; Diaz, Michele; O’Leary, Daniel S.; Bockholt, H. Jeremy; Gadde, Syam; Preda, Adrian; Wible, Cynthia G.; Stern, Hal S.; Belger, Aysenil; McCarthy, Gregory; Ozyurt, Burak; Potkin, Steven G.

2015-01-01

The Function Biomedical Informatics Research Network (FBIRN) developed methods and tools for conducting multi-scanner functional magnetic resonance imaging (fMRI) studies. Method and tool development were based on two major goals: 1) to assess the major sources of variation in fMRI studies conducted across scanners, including instrumentation, acquisition protocols, challenge tasks, and analysis methods, and 2) to provide a distributed network infrastructure and an associated federated database to host and query large, multi-site, fMRI and clinical datasets. In the process of achieving these goals the FBIRN test bed generated several multi-scanner brain imaging data sets to be shared with the wider scientific community via the BIRN Data Repository (BDR). The FBIRN Phase 1 dataset consists of a traveling subject study of 5 healthy subjects, each scanned on 10 different 1.5 to 4 Tesla scanners. The FBIRN Phase 2 and Phase 3 datasets consist of subjects with schizophrenia or schizoaffective disorder along with healthy comparison subjects scanned at multiple sites. In this paper, we provide concise descriptions of FBIRN’s multi-scanner brain imaging data sets and details about the BIRN Data Repository instance of the Human Imaging Database (HID) used to publicly share the data. PMID:26364863

Identification of autism spectrum disorder using deep learning and the ABIDE dataset.

PubMed

Heinsfeld, Anibal Sólon; Franco, Alexandre Rosa; Craddock, R Cameron; Buchweitz, Augusto; Meneguzzi, Felipe

2018-01-01

The goal of the present study was to apply deep learning algorithms to identify autism spectrum disorder (ASD) patients from large brain imaging dataset, based solely on the patients brain activation patterns. We investigated ASD patients brain imaging data from a world-wide multi-site database known as ABIDE (Autism Brain Imaging Data Exchange). ASD is a brain-based disorder characterized by social deficits and repetitive behaviors. According to recent Centers for Disease Control data, ASD affects one in 68 children in the United States. We investigated patterns of functional connectivity that objectively identify ASD participants from functional brain imaging data, and attempted to unveil the neural patterns that emerged from the classification. The results improved the state-of-the-art by achieving 70% accuracy in identification of ASD versus control patients in the dataset. The patterns that emerged from the classification show an anticorrelation of brain function between anterior and posterior areas of the brain; the anticorrelation corroborates current empirical evidence of anterior-posterior disruption in brain connectivity in ASD. We present the results and identify the areas of the brain that contributed most to differentiating ASD from typically developing controls as per our deep learning model.

The Optimization of Trained and Untrained Image Classification Algorithms for Use on Large Spatial Datasets

NASA Technical Reports Server (NTRS)

Kocurek, Michael J.

2005-01-01

The HARVIST project seeks to automatically provide an accurate, interactive interface to predict crop yield over the entire United States. In order to accomplish this goal, large images must be quickly and automatically classified by crop type. Current trained and untrained classification algorithms, while accurate, are highly inefficient when operating on large datasets. This project sought to develop new variants of two standard trained and untrained classification algorithms that are optimized to take advantage of the spatial nature of image data. The first algorithm, harvist-cluster, utilizes divide-and-conquer techniques to precluster an image in the hopes of increasing overall clustering speed. The second algorithm, harvistSVM, utilizes support vector machines (SVMs), a type of trained classifier. It seeks to increase classification speed by applying a "meta-SVM" to a quick (but inaccurate) SVM to approximate a slower, yet more accurate, SVM. Speedups were achieved by tuning the algorithm to quickly identify when the quick SVM was incorrect, and then reclassifying low-confidence pixels as necessary. Comparing the classification speeds of both algorithms to known baselines showed a slight speedup for large values of k (the number of clusters) for harvist-cluster, and a significant speedup for harvistSVM. Future work aims to automate the parameter tuning process required for harvistSVM, and further improve classification accuracy and speed. Additionally, this research will move documents created in Canvas into ArcGIS. The launch of the Mars Reconnaissance Orbiter (MRO) will provide a wealth of image data such as global maps of Martian weather and high resolution global images of Mars. The ability to store this new data in a georeferenced format will support future Mars missions by providing data for landing site selection and the search for water on Mars.

Parallel and Efficient Sensitivity Analysis of Microscopy Image Segmentation Workflows in Hybrid Systems

PubMed Central

Barreiros, Willian; Teodoro, George; Kurc, Tahsin; Kong, Jun; Melo, Alba C. M. A.; Saltz, Joel

2017-01-01

We investigate efficient sensitivity analysis (SA) of algorithms that segment and classify image features in a large dataset of high-resolution images. Algorithm SA is the process of evaluating variations of methods and parameter values to quantify differences in the output. A SA can be very compute demanding because it requires re-processing the input dataset several times with different parameters to assess variations in output. In this work, we introduce strategies to efficiently speed up SA via runtime optimizations targeting distributed hybrid systems and reuse of computations from runs with different parameters. We evaluate our approach using a cancer image analysis workflow on a hybrid cluster with 256 nodes, each with an Intel Phi and a dual socket CPU. The SA attained a parallel efficiency of over 90% on 256 nodes. The cooperative execution using the CPUs and the Phi available in each node with smart task assignment strategies resulted in an additional speedup of about 2×. Finally, multi-level computation reuse lead to an additional speedup of up to 2.46× on the parallel version. The level of performance attained with the proposed optimizations will allow the use of SA in large-scale studies. PMID:29081725

Atlas-guided cluster analysis of large tractography datasets.

PubMed

Ros, Christian; Güllmar, Daniel; Stenzel, Martin; Mentzel, Hans-Joachim; Reichenbach, Jürgen Rainer

2013-01-01

Diffusion Tensor Imaging (DTI) and fiber tractography are important tools to map the cerebral white matter microstructure in vivo and to model the underlying axonal pathways in the brain with three-dimensional fiber tracts. As the fast and consistent extraction of anatomically correct fiber bundles for multiple datasets is still challenging, we present a novel atlas-guided clustering framework for exploratory data analysis of large tractography datasets. The framework uses an hierarchical cluster analysis approach that exploits the inherent redundancy in large datasets to time-efficiently group fiber tracts. Structural information of a white matter atlas can be incorporated into the clustering to achieve an anatomically correct and reproducible grouping of fiber tracts. This approach facilitates not only the identification of the bundles corresponding to the classes of the atlas; it also enables the extraction of bundles that are not present in the atlas. The new technique was applied to cluster datasets of 46 healthy subjects. Prospects of automatic and anatomically correct as well as reproducible clustering are explored. Reconstructed clusters were well separated and showed good correspondence to anatomical bundles. Using the atlas-guided cluster approach, we observed consistent results across subjects with high reproducibility. In order to investigate the outlier elimination performance of the clustering algorithm, scenarios with varying amounts of noise were simulated and clustered with three different outlier elimination strategies. By exploiting the multithreading capabilities of modern multiprocessor systems in combination with novel algorithms, our toolkit clusters large datasets in a couple of minutes. Experiments were conducted to investigate the achievable speedup and to demonstrate the high performance of the clustering framework in a multiprocessing environment.

Secure access control and large scale robust representation for online multimedia event detection.

PubMed

Liu, Changyu; Lu, Bin; Li, Huiling

2014-01-01

We developed an online multimedia event detection (MED) system. However, there are a secure access control issue and a large scale robust representation issue when we want to integrate traditional event detection algorithms into the online environment. For the first issue, we proposed a tree proxy-based and service-oriented access control (TPSAC) model based on the traditional role based access control model. Verification experiments were conducted on the CloudSim simulation platform, and the results showed that the TPSAC model is suitable for the access control of dynamic online environments. For the second issue, inspired by the object-bank scene descriptor, we proposed a 1000-object-bank (1000OBK) event descriptor. Feature vectors of the 1000OBK were extracted from response pyramids of 1000 generic object detectors which were trained on standard annotated image datasets, such as the ImageNet dataset. A spatial bag of words tiling approach was then adopted to encode these feature vectors for bridging the gap between the objects and events. Furthermore, we performed experiments in the context of event classification on the challenging TRECVID MED 2012 dataset, and the results showed that the robust 1000OBK event descriptor outperforms the state-of-the-art approaches.

A high-throughput system for high-quality tomographic reconstruction of large datasets at Diamond Light Source

PubMed Central

Atwood, Robert C.; Bodey, Andrew J.; Price, Stephen W. T.; Basham, Mark; Drakopoulos, Michael

2015-01-01

Tomographic datasets collected at synchrotrons are becoming very large and complex, and, therefore, need to be managed efficiently. Raw images may have high pixel counts, and each pixel can be multidimensional and associated with additional data such as those derived from spectroscopy. In time-resolved studies, hundreds of tomographic datasets can be collected in sequence, yielding terabytes of data. Users of tomographic beamlines are drawn from various scientific disciplines, and many are keen to use tomographic reconstruction software that does not require a deep understanding of reconstruction principles. We have developed Savu, a reconstruction pipeline that enables users to rapidly reconstruct data to consistently create high-quality results. Savu is designed to work in an ‘orthogonal’ fashion, meaning that data can be converted between projection and sinogram space throughout the processing workflow as required. The Savu pipeline is modular and allows processing strategies to be optimized for users' purposes. In addition to the reconstruction algorithms themselves, it can include modules for identification of experimental problems, artefact correction, general image processing and data quality assessment. Savu is open source, open licensed and ‘facility-independent’: it can run on standard cluster infrastructure at any institution. PMID:25939626

Multiresolution persistent homology for excessively large biomolecular datasets

NASA Astrophysics Data System (ADS)

Xia, Kelin; Zhao, Zhixiong; Wei, Guo-Wei

2015-10-01

Although persistent homology has emerged as a promising tool for the topological simplification of complex data, it is computationally intractable for large datasets. We introduce multiresolution persistent homology to handle excessively large datasets. We match the resolution with the scale of interest so as to represent large scale datasets with appropriate resolution. We utilize flexibility-rigidity index to access the topological connectivity of the data set and define a rigidity density for the filtration analysis. By appropriately tuning the resolution of the rigidity density, we are able to focus the topological lens on the scale of interest. The proposed multiresolution topological analysis is validated by a hexagonal fractal image which has three distinct scales. We further demonstrate the proposed method for extracting topological fingerprints from DNA molecules. In particular, the topological persistence of a virus capsid with 273 780 atoms is successfully analyzed which would otherwise be inaccessible to the normal point cloud method and unreliable by using coarse-grained multiscale persistent homology. The proposed method has also been successfully applied to the protein domain classification, which is the first time that persistent homology is used for practical protein domain analysis, to our knowledge. The proposed multiresolution topological method has potential applications in arbitrary data sets, such as social networks, biological networks, and graphs.

Scalable isosurface visualization of massive datasets on commodity off-the-shelf clusters

PubMed Central

Bajaj, Chandrajit

2009-01-01

Tomographic imaging and computer simulations are increasingly yielding massive datasets. Interactive and exploratory visualizations have rapidly become indispensable tools to study large volumetric imaging and simulation data. Our scalable isosurface visualization framework on commodity off-the-shelf clusters is an end-to-end parallel and progressive platform, from initial data access to the final display. Interactive browsing of extracted isosurfaces is made possible by using parallel isosurface extraction, and rendering in conjunction with a new specialized piece of image compositing hardware called Metabuffer. In this paper, we focus on the back end scalability by introducing a fully parallel and out-of-core isosurface extraction algorithm. It achieves scalability by using both parallel and out-of-core processing and parallel disks. It statically partitions the volume data to parallel disks with a balanced workload spectrum, and builds I/O-optimal external interval trees to minimize the number of I/O operations of loading large data from disk. We also describe an isosurface compression scheme that is efficient for progress extraction, transmission and storage of isosurfaces. PMID:19756231

DoctorEye: A clinically driven multifunctional platform, for accurate processing of tumors in medical images.

PubMed

Skounakis, Emmanouil; Farmaki, Christina; Sakkalis, Vangelis; Roniotis, Alexandros; Banitsas, Konstantinos; Graf, Norbert; Marias, Konstantinos

2010-01-01

This paper presents a novel, open access interactive platform for 3D medical image analysis, simulation and visualization, focusing in oncology images. The platform was developed through constant interaction and feedback from expert clinicians integrating a thorough analysis of their requirements while having an ultimate goal of assisting in accurately delineating tumors. It allows clinicians not only to work with a large number of 3D tomographic datasets but also to efficiently annotate multiple regions of interest in the same session. Manual and semi-automatic segmentation techniques combined with integrated correction tools assist in the quick and refined delineation of tumors while different users can add different components related to oncology such as tumor growth and simulation algorithms for improving therapy planning. The platform has been tested by different users and over large number of heterogeneous tomographic datasets to ensure stability, usability, extensibility and robustness with promising results. the platform, a manual and tutorial videos are available at: http://biomodeling.ics.forth.gr. it is free to use under the GNU General Public License.

Deep ensemble learning of virtual endoluminal views for polyp detection in CT colonography

NASA Astrophysics Data System (ADS)

Umehara, Kensuke; Näppi, Janne J.; Hironaka, Toru; Regge, Daniele; Ishida, Takayuki; Yoshida, Hiroyuki

2017-03-01

Robust training of a deep convolutional neural network (DCNN) requires a very large number of annotated datasets that are currently not available in CT colonography (CTC). We previously demonstrated that deep transfer learning provides an effective approach for robust application of a DCNN in CTC. However, at high detection accuracy, the differentiation of small polyps from non-polyps was still challenging. In this study, we developed and evaluated a deep ensemble learning (DEL) scheme for reviewing of virtual endoluminal images to improve the performance of computer-aided detection (CADe) of polyps in CTC. Nine different types of image renderings were generated from virtual endoluminal images of polyp candidates detected by a conventional CADe system. Eleven DCNNs that represented three types of publically available pre-trained DCNN models were re-trained by transfer learning to identify polyps from the virtual endoluminal images. A DEL scheme that determines the final detected polyps by a review of the nine types of VE images was developed by combining the DCNNs using a random forest classifier as a meta-classifier. For evaluation, we sampled 154 CTC cases from a large CTC screening trial and divided the cases randomly into a training dataset and a test dataset. At 3.9 falsepositive (FP) detections per patient on average, the detection sensitivities of the conventional CADe system, the highestperforming single DCNN, and the DEL scheme were 81.3%, 90.7%, and 93.5%, respectively, for polyps ≥6 mm in size. For small polyps, the DEL scheme reduced the number of false positives by up to 83% over that of using a single DCNN alone. These preliminary results indicate that the DEL scheme provides an effective approach for improving the polyp detection performance of CADe in CTC, especially for small polyps.

Collection of sequential imaging events for research in breast cancer screening

NASA Astrophysics Data System (ADS)

Patel, M. N.; Young, K.; Halling-Brown, M. D.

2016-03-01

Due to the huge amount of research involving medical images, there is a widely accepted need for comprehensive collections of medical images to be made available for research. This demand led to the design and implementation of a flexible image repository, which retrospectively collects images and data from multiple sites throughout the UK. The OPTIMAM Medical Image Database (OMI-DB) was created to provide a centralized, fully annotated dataset for research. The database contains both processed and unprocessed images, associated data, annotations and expert-determined ground truths. Collection has been ongoing for over three years, providing the opportunity to collect sequential imaging events. Extensive alterations to the identification, collection, processing and storage arms of the system have been undertaken to support the introduction of sequential events, including interval cancers. These updates to the collection systems allow the acquisition of many more images, but more importantly, allow one to build on the existing high-dimensional data stored in the OMI-DB. A research dataset of this scale, which includes original normal and subsequent malignant cases along with expert derived and clinical annotations, is currently unique. These data provide a powerful resource for future research and has initiated new research projects, amongst which, is the quantification of normal cases by applying a large number of quantitative imaging features, with a priori knowledge that eventually these cases develop a malignancy. This paper describes, extensions to the OMI-DB collection systems and tools and discusses the prospective applications of having such a rich dataset for future research applications.

Medical Image Data and Datasets in the Era of Machine Learning-Whitepaper from the 2016 C-MIMI Meeting Dataset Session.

PubMed

Kohli, Marc D; Summers, Ronald M; Geis, J Raymond

2017-08-01

At the first annual Conference on Machine Intelligence in Medical Imaging (C-MIMI), held in September 2016, a conference session on medical image data and datasets for machine learning identified multiple issues. The common theme from attendees was that everyone participating in medical image evaluation with machine learning is data starved. There is an urgent need to find better ways to collect, annotate, and reuse medical imaging data. Unique domain issues with medical image datasets require further study, development, and dissemination of best practices and standards, and a coordinated effort among medical imaging domain experts, medical imaging informaticists, government and industry data scientists, and interested commercial, academic, and government entities. High-level attributes of reusable medical image datasets suitable to train, test, validate, verify, and regulate ML products should be better described. NIH and other government agencies should promote and, where applicable, enforce, access to medical image datasets. We should improve communication among medical imaging domain experts, medical imaging informaticists, academic clinical and basic science researchers, government and industry data scientists, and interested commercial entities.

An open access thyroid ultrasound image database

NASA Astrophysics Data System (ADS)

Pedraza, Lina; Vargas, Carlos; Narváez, Fabián.; Durán, Oscar; Muñoz, Emma; Romero, Eduardo

2015-01-01

Computer aided diagnosis systems (CAD) have been developed to assist radiologists in the detection and diagnosis of abnormalities and a large number of pattern recognition techniques have been proposed to obtain a second opinion. Most of these strategies have been evaluated using different datasets making their performance incomparable. In this work, an open access database of thyroid ultrasound images is presented. The dataset consists of a set of B-mode Ultrasound images, including a complete annotation and diagnostic description of suspicious thyroid lesions by expert radiologists. Several types of lesions as thyroiditis, cystic nodules, adenomas and thyroid cancers were included while an accurate lesion delineation is provided in XML format. The diagnostic description of malignant lesions was confirmed by biopsy. The proposed new database is expected to be a resource for the community to assess different CAD systems.

Random forest regression for magnetic resonance image synthesis.

PubMed

Jog, Amod; Carass, Aaron; Roy, Snehashis; Pham, Dzung L; Prince, Jerry L

2017-01-01

By choosing different pulse sequences and their parameters, magnetic resonance imaging (MRI) can generate a large variety of tissue contrasts. This very flexibility, however, can yield inconsistencies with MRI acquisitions across datasets or scanning sessions that can in turn cause inconsistent automated image analysis. Although image synthesis of MR images has been shown to be helpful in addressing this problem, an inability to synthesize both T 2 -weighted brain images that include the skull and FLuid Attenuated Inversion Recovery (FLAIR) images has been reported. The method described herein, called REPLICA, addresses these limitations. REPLICA is a supervised random forest image synthesis approach that learns a nonlinear regression to predict intensities of alternate tissue contrasts given specific input tissue contrasts. Experimental results include direct image comparisons between synthetic and real images, results from image analysis tasks on both synthetic and real images, and comparison against other state-of-the-art image synthesis methods. REPLICA is computationally fast, and is shown to be comparable to other methods on tasks they are able to perform. Additionally REPLICA has the capability to synthesize both T 2 -weighted images of the full head and FLAIR images, and perform intensity standardization between different imaging datasets. Copyright © 2016 Elsevier B.V. All rights reserved.

LIPS database with LIPService: a microscopic image database of intracellular structures in Arabidopsis guard cells.

PubMed

Higaki, Takumi; Kutsuna, Natsumaro; Hasezawa, Seiichiro

2013-05-16

Intracellular configuration is an important feature of cell status. Recent advances in microscopic imaging techniques allow us to easily obtain a large number of microscopic images of intracellular structures. In this circumstance, automated microscopic image recognition techniques are of extreme importance to future phenomics/visible screening approaches. However, there was no benchmark microscopic image dataset for intracellular organelles in a specified plant cell type. We previously established the Live Images of Plant Stomata (LIPS) database, a publicly available collection of optical-section images of various intracellular structures of plant guard cells, as a model system of environmental signal perception and transduction. Here we report recent updates to the LIPS database and the establishment of a database table, LIPService. We updated the LIPS dataset and established a new interface named LIPService to promote efficient inspection of intracellular structure configurations. Cell nuclei, microtubules, actin microfilaments, mitochondria, chloroplasts, endoplasmic reticulum, peroxisomes, endosomes, Golgi bodies, and vacuoles can be filtered using probe names or morphometric parameters such as stomatal aperture. In addition to the serial optical sectional images of the original LIPS database, new volume-rendering data for easy web browsing of three-dimensional intracellular structures have been released to allow easy inspection of their configurations or relationships with cell status/morphology. We also demonstrated the utility of the new LIPS image database for automated organelle recognition of images from another plant cell image database with image clustering analyses. The updated LIPS database provides a benchmark image dataset for representative intracellular structures in Arabidopsis guard cells. The newly released LIPService allows users to inspect the relationship between organellar three-dimensional configurations and morphometrical parameters.

A data colocation grid framework for big data medical image processing: backend design

NASA Astrophysics Data System (ADS)

Bao, Shunxing; Huo, Yuankai; Parvathaneni, Prasanna; Plassard, Andrew J.; Bermudez, Camilo; Yao, Yuang; Lyu, Ilwoo; Gokhale, Aniruddha; Landman, Bennett A.

2018-03-01

When processing large medical imaging studies, adopting high performance grid computing resources rapidly becomes important. We recently presented a "medical image processing-as-a-service" grid framework that offers promise in utilizing the Apache Hadoop ecosystem and HBase for data colocation by moving computation close to medical image storage. However, the framework has not yet proven to be easy to use in a heterogeneous hardware environment. Furthermore, the system has not yet validated when considering variety of multi-level analysis in medical imaging. Our target design criteria are (1) improving the framework's performance in a heterogeneous cluster, (2) performing population based summary statistics on large datasets, and (3) introducing a table design scheme for rapid NoSQL query. In this paper, we present a heuristic backend interface application program interface (API) design for Hadoop and HBase for Medical Image Processing (HadoopBase-MIP). The API includes: Upload, Retrieve, Remove, Load balancer (for heterogeneous cluster) and MapReduce templates. A dataset summary statistic model is discussed and implemented by MapReduce paradigm. We introduce a HBase table scheme for fast data query to better utilize the MapReduce model. Briefly, 5153 T1 images were retrieved from a university secure, shared web database and used to empirically access an in-house grid with 224 heterogeneous CPU cores. Three empirical experiments results are presented and discussed: (1) load balancer wall-time improvement of 1.5-fold compared with a framework with built-in data allocation strategy, (2) a summary statistic model is empirically verified on grid framework and is compared with the cluster when deployed with a standard Sun Grid Engine (SGE), which reduces 8-fold of wall clock time and 14-fold of resource time, and (3) the proposed HBase table scheme improves MapReduce computation with 7 fold reduction of wall time compare with a naïve scheme when datasets are relative small. The source code and interfaces have been made publicly available.

A Data Colocation Grid Framework for Big Data Medical Image Processing: Backend Design.

PubMed

Bao, Shunxing; Huo, Yuankai; Parvathaneni, Prasanna; Plassard, Andrew J; Bermudez, Camilo; Yao, Yuang; Lyu, Ilwoo; Gokhale, Aniruddha; Landman, Bennett A

2018-03-01

When processing large medical imaging studies, adopting high performance grid computing resources rapidly becomes important. We recently presented a "medical image processing-as-a-service" grid framework that offers promise in utilizing the Apache Hadoop ecosystem and HBase for data colocation by moving computation close to medical image storage. However, the framework has not yet proven to be easy to use in a heterogeneous hardware environment. Furthermore, the system has not yet validated when considering variety of multi-level analysis in medical imaging. Our target design criteria are (1) improving the framework's performance in a heterogeneous cluster, (2) performing population based summary statistics on large datasets, and (3) introducing a table design scheme for rapid NoSQL query. In this paper, we present a heuristic backend interface application program interface (API) design for Hadoop & HBase for Medical Image Processing (HadoopBase-MIP). The API includes: Upload, Retrieve, Remove, Load balancer (for heterogeneous cluster) and MapReduce templates. A dataset summary statistic model is discussed and implemented by MapReduce paradigm. We introduce a HBase table scheme for fast data query to better utilize the MapReduce model. Briefly, 5153 T1 images were retrieved from a university secure, shared web database and used to empirically access an in-house grid with 224 heterogeneous CPU cores. Three empirical experiments results are presented and discussed: (1) load balancer wall-time improvement of 1.5-fold compared with a framework with built-in data allocation strategy, (2) a summary statistic model is empirically verified on grid framework and is compared with the cluster when deployed with a standard Sun Grid Engine (SGE), which reduces 8-fold of wall clock time and 14-fold of resource time, and (3) the proposed HBase table scheme improves MapReduce computation with 7 fold reduction of wall time compare with a naïve scheme when datasets are relative small. The source code and interfaces have been made publicly available.

A Data Colocation Grid Framework for Big Data Medical Image Processing: Backend Design

PubMed Central

Huo, Yuankai; Parvathaneni, Prasanna; Plassard, Andrew J.; Bermudez, Camilo; Yao, Yuang; Lyu, Ilwoo; Gokhale, Aniruddha; Landman, Bennett A.

2018-01-01

When processing large medical imaging studies, adopting high performance grid computing resources rapidly becomes important. We recently presented a "medical image processing-as-a-service" grid framework that offers promise in utilizing the Apache Hadoop ecosystem and HBase for data colocation by moving computation close to medical image storage. However, the framework has not yet proven to be easy to use in a heterogeneous hardware environment. Furthermore, the system has not yet validated when considering variety of multi-level analysis in medical imaging. Our target design criteria are (1) improving the framework’s performance in a heterogeneous cluster, (2) performing population based summary statistics on large datasets, and (3) introducing a table design scheme for rapid NoSQL query. In this paper, we present a heuristic backend interface application program interface (API) design for Hadoop & HBase for Medical Image Processing (HadoopBase-MIP). The API includes: Upload, Retrieve, Remove, Load balancer (for heterogeneous cluster) and MapReduce templates. A dataset summary statistic model is discussed and implemented by MapReduce paradigm. We introduce a HBase table scheme for fast data query to better utilize the MapReduce model. Briefly, 5153 T1 images were retrieved from a university secure, shared web database and used to empirically access an in-house grid with 224 heterogeneous CPU cores. Three empirical experiments results are presented and discussed: (1) load balancer wall-time improvement of 1.5-fold compared with a framework with built-in data allocation strategy, (2) a summary statistic model is empirically verified on grid framework and is compared with the cluster when deployed with a standard Sun Grid Engine (SGE), which reduces 8-fold of wall clock time and 14-fold of resource time, and (3) the proposed HBase table scheme improves MapReduce computation with 7 fold reduction of wall time compare with a naïve scheme when datasets are relative small. The source code and interfaces have been made publicly available. PMID:29887668

Automatic training and reliability estimation for 3D ASM applied to cardiac MRI segmentation

NASA Astrophysics Data System (ADS)

Tobon-Gomez, Catalina; Sukno, Federico M.; Butakoff, Constantine; Huguet, Marina; Frangi, Alejandro F.

2012-07-01

Training active shape models requires collecting manual ground-truth meshes in a large image database. While shape information can be reused across multiple imaging modalities, intensity information needs to be imaging modality and protocol specific. In this context, this study has two main purposes: (1) to test the potential of using intensity models learned from MRI simulated datasets and (2) to test the potential of including a measure of reliability during the matching process to increase robustness. We used a population of 400 virtual subjects (XCAT phantom), and two clinical populations of 40 and 45 subjects. Virtual subjects were used to generate simulated datasets (MRISIM simulator). Intensity models were trained both on simulated and real datasets. The trained models were used to segment the left ventricle (LV) and right ventricle (RV) from real datasets. Segmentations were also obtained with and without reliability information. Performance was evaluated with point-to-surface and volume errors. Simulated intensity models obtained average accuracy comparable to inter-observer variability for LV segmentation. The inclusion of reliability information reduced volume errors in hypertrophic patients (EF errors from 17 ± 57% to 10 ± 18% LV MASS errors from -27 ± 22 g to -14 ± 25 g), and in heart failure patients (EF errors from -8 ± 42% to -5 ± 14%). The RV model of the simulated images needs further improvement to better resemble image intensities around the myocardial edges. Both for real and simulated models, reliability information increased segmentation robustness without penalizing accuracy.

Automatic training and reliability estimation for 3D ASM applied to cardiac MRI segmentation.

PubMed

Tobon-Gomez, Catalina; Sukno, Federico M; Butakoff, Constantine; Huguet, Marina; Frangi, Alejandro F

2012-07-07

Training active shape models requires collecting manual ground-truth meshes in a large image database. While shape information can be reused across multiple imaging modalities, intensity information needs to be imaging modality and protocol specific. In this context, this study has two main purposes: (1) to test the potential of using intensity models learned from MRI simulated datasets and (2) to test the potential of including a measure of reliability during the matching process to increase robustness. We used a population of 400 virtual subjects (XCAT phantom), and two clinical populations of 40 and 45 subjects. Virtual subjects were used to generate simulated datasets (MRISIM simulator). Intensity models were trained both on simulated and real datasets. The trained models were used to segment the left ventricle (LV) and right ventricle (RV) from real datasets. Segmentations were also obtained with and without reliability information. Performance was evaluated with point-to-surface and volume errors. Simulated intensity models obtained average accuracy comparable to inter-observer variability for LV segmentation. The inclusion of reliability information reduced volume errors in hypertrophic patients (EF errors from 17 ± 57% to 10 ± 18%; LV MASS errors from -27 ± 22 g to -14 ± 25 g), and in heart failure patients (EF errors from -8 ± 42% to -5 ± 14%). The RV model of the simulated images needs further improvement to better resemble image intensities around the myocardial edges. Both for real and simulated models, reliability information increased segmentation robustness without penalizing accuracy.

A unified framework of image latent feature learning on Sina microblog

NASA Astrophysics Data System (ADS)

Wei, Jinjin; Jin, Zhigang; Zhou, Yuan; Zhang, Rui

2015-10-01

Large-scale user-contributed images with texts are rapidly increasing on the social media websites, such as Sina microblog. However, the noise and incomplete correspondence between the images and the texts give rise to the difficulty in precise image retrieval and ranking. In this paper, a hypergraph-based learning framework is proposed for image ranking, which simultaneously utilizes visual feature, textual content and social link information to estimate the relevance between images. Representing each image as a vertex in the hypergraph, complex relationship between images can be reflected exactly. Then updating the weight of hyperedges throughout the hypergraph learning process, the effect of different edges can be adaptively modulated in the constructed hypergraph. Furthermore, the popularity degree of the image is employed to re-rank the retrieval results. Comparative experiments on a large-scale Sina microblog data-set demonstrate the effectiveness of the proposed approach.

The PREP pipeline: standardized preprocessing for large-scale EEG analysis.

PubMed

Bigdely-Shamlo, Nima; Mullen, Tim; Kothe, Christian; Su, Kyung-Min; Robbins, Kay A

2015-01-01

The technology to collect brain imaging and physiological measures has become portable and ubiquitous, opening the possibility of large-scale analysis of real-world human imaging. By its nature, such data is large and complex, making automated processing essential. This paper shows how lack of attention to the very early stages of an EEG preprocessing pipeline can reduce the signal-to-noise ratio and introduce unwanted artifacts into the data, particularly for computations done in single precision. We demonstrate that ordinary average referencing improves the signal-to-noise ratio, but that noisy channels can contaminate the results. We also show that identification of noisy channels depends on the reference and examine the complex interaction of filtering, noisy channel identification, and referencing. We introduce a multi-stage robust referencing scheme to deal with the noisy channel-reference interaction. We propose a standardized early-stage EEG processing pipeline (PREP) and discuss the application of the pipeline to more than 600 EEG datasets. The pipeline includes an automatically generated report for each dataset processed. Users can download the PREP pipeline as a freely available MATLAB library from http://eegstudy.org/prepcode.

Automatic detection of blurred images in UAV image sets

NASA Astrophysics Data System (ADS)

Sieberth, Till; Wackrow, Rene; Chandler, Jim H.

2016-12-01

Unmanned aerial vehicles (UAV) have become an interesting and active research topic for photogrammetry. Current research is based on images acquired by an UAV, which have a high ground resolution and good spectral and radiometrical resolution, due to the low flight altitudes combined with a high resolution camera. UAV image flights are also cost effective and have become attractive for many applications including, change detection in small scale areas. One of the main problems preventing full automation of data processing of UAV imagery is the degradation effect of blur caused by camera movement during image acquisition. This can be caused by the normal flight movement of the UAV as well as strong winds, turbulence or sudden operator inputs. This blur disturbs the visual analysis and interpretation of the data, causes errors and can degrade the accuracy in automatic photogrammetric processing algorithms. The detection and removal of these images is currently achieved manually, which is both time consuming and prone to error, particularly for large image-sets. To increase the quality of data processing an automated process is necessary, which must be both reliable and quick. This paper describes the development of an automatic filtering process, which is based upon the quantification of blur in an image. Images with known blur are processed digitally to determine a quantifiable measure of image blur. The algorithm is required to process UAV images fast and reliably to relieve the operator from detecting blurred images manually. The newly developed method makes it possible to detect blur caused by linear camera displacement and is based on human detection of blur. Humans detect blurred images best by comparing it to other images in order to establish whether an image is blurred or not. The developed algorithm simulates this procedure by creating an image for comparison using image processing. Creating internally a comparable image makes the method independent of additional images. However, the calculated blur value named SIEDS (saturation image edge difference standard-deviation) on its own does not provide an absolute number to judge if an image is blurred or not. To achieve a reliable judgement of image sharpness the SIEDS value has to be compared to other SIEDS values from the same dataset. The speed and reliability of the method was tested using a range of different UAV datasets. Two datasets will be presented in this paper to demonstrate the effectiveness of the algorithm. The algorithm proves to be fast and the returned values are optically correct, making the algorithm applicable for UAV datasets. Additionally, a close range dataset was processed to determine whether the method is also useful for close range applications. The results show that the method is also reliable for close range images, which significantly extends the field of application for the algorithm.

A method for normalizing pathology images to improve feature extraction for quantitative pathology.

PubMed

Tam, Allison; Barker, Jocelyn; Rubin, Daniel

2016-01-01

With the advent of digital slide scanning technologies and the potential proliferation of large repositories of digital pathology images, many research studies can leverage these data for biomedical discovery and to develop clinical applications. However, quantitative analysis of digital pathology images is impeded by batch effects generated by varied staining protocols and staining conditions of pathological slides. To overcome this problem, this paper proposes a novel, fully automated stain normalization method to reduce batch effects and thus aid research in digital pathology applications. Their method, intensity centering and histogram equalization (ICHE), normalizes a diverse set of pathology images by first scaling the centroids of the intensity histograms to a common point and then applying a modified version of contrast-limited adaptive histogram equalization. Normalization was performed on two datasets of digitized hematoxylin and eosin (H&E) slides of different tissue slices from the same lung tumor, and one immunohistochemistry dataset of digitized slides created by restaining one of the H&E datasets. The ICHE method was evaluated based on image intensity values, quantitative features, and the effect on downstream applications, such as a computer aided diagnosis. For comparison, three methods from the literature were reimplemented and evaluated using the same criteria. The authors found that ICHE not only improved performance compared with un-normalized images, but in most cases showed improvement compared with previous methods for correcting batch effects in the literature. ICHE may be a useful preprocessing step a digital pathology image processing pipeline.

Single-Image Super Resolution for Multispectral Remote Sensing Data Using Convolutional Neural Networks

NASA Astrophysics Data System (ADS)

Liebel, L.; Körner, M.

2016-06-01

In optical remote sensing, spatial resolution of images is crucial for numerous applications. Space-borne systems are most likely to be affected by a lack of spatial resolution, due to their natural disadvantage of a large distance between the sensor and the sensed object. Thus, methods for single-image super resolution are desirable to exceed the limits of the sensor. Apart from assisting visual inspection of datasets, post-processing operations—e.g., segmentation or feature extraction—can benefit from detailed and distinguishable structures. In this paper, we show that recently introduced state-of-the-art approaches for single-image super resolution of conventional photographs, making use of deep learning techniques, such as convolutional neural networks (CNN), can successfully be applied to remote sensing data. With a huge amount of training data available, end-to-end learning is reasonably easy to apply and can achieve results unattainable using conventional handcrafted algorithms. We trained our CNN on a specifically designed, domain-specific dataset, in order to take into account the special characteristics of multispectral remote sensing data. This dataset consists of publicly available SENTINEL-2 images featuring 13 spectral bands, a ground resolution of up to 10m, and a high radiometric resolution and thus satisfying our requirements in terms of quality and quantity. In experiments, we obtained results superior compared to competing approaches trained on generic image sets, which failed to reasonably scale satellite images with a high radiometric resolution, as well as conventional interpolation methods.

Imaging mass spectrometry statistical analysis.

PubMed

Jones, Emrys A; Deininger, Sören-Oliver; Hogendoorn, Pancras C W; Deelder, André M; McDonnell, Liam A

2012-08-30

Imaging mass spectrometry is increasingly used to identify new candidate biomarkers. This clinical application of imaging mass spectrometry is highly multidisciplinary: expertise in mass spectrometry is necessary to acquire high quality data, histology is required to accurately label the origin of each pixel's mass spectrum, disease biology is necessary to understand the potential meaning of the imaging mass spectrometry results, and statistics to assess the confidence of any findings. Imaging mass spectrometry data analysis is further complicated because of the unique nature of the data (within the mass spectrometry field); several of the assumptions implicit in the analysis of LC-MS/profiling datasets are not applicable to imaging. The very large size of imaging datasets and the reporting of many data analysis routines, combined with inadequate training and accessible reviews, have exacerbated this problem. In this paper we provide an accessible review of the nature of imaging data and the different strategies by which the data may be analyzed. Particular attention is paid to the assumptions of the data analysis routines to ensure that the reader is apprised of their correct usage in imaging mass spectrometry research. Copyright © 2012 Elsevier B.V. All rights reserved.

TLEM 2.0 - a comprehensive musculoskeletal geometry dataset for subject-specific modeling of lower extremity.

PubMed

Carbone, V; Fluit, R; Pellikaan, P; van der Krogt, M M; Janssen, D; Damsgaard, M; Vigneron, L; Feilkas, T; Koopman, H F J M; Verdonschot, N

2015-03-18

When analyzing complex biomechanical problems such as predicting the effects of orthopedic surgery, subject-specific musculoskeletal models are essential to achieve reliable predictions. The aim of this paper is to present the Twente Lower Extremity Model 2.0, a new comprehensive dataset of the musculoskeletal geometry of the lower extremity, which is based on medical imaging data and dissection performed on the right lower extremity of a fresh male cadaver. Bone, muscle and subcutaneous fat (including skin) volumes were segmented from computed tomography and magnetic resonance images scans. Inertial parameters were estimated from the image-based segmented volumes. A complete cadaver dissection was performed, in which bony landmarks, attachments sites and lines-of-action of 55 muscle actuators and 12 ligaments, bony wrapping surfaces, and joint geometry were measured. The obtained musculoskeletal geometry dataset was finally implemented in the AnyBody Modeling System (AnyBody Technology A/S, Aalborg, Denmark), resulting in a model consisting of 12 segments, 11 joints and 21 degrees of freedom, and including 166 muscle-tendon elements for each leg. The new TLEM 2.0 dataset was purposely built to be easily combined with novel image-based scaling techniques, such as bone surface morphing, muscle volume registration and muscle-tendon path identification, in order to obtain subject-specific musculoskeletal models in a quick and accurate way. The complete dataset, including CT and MRI scans and segmented volume and surfaces, is made available at http://www.utwente.nl/ctw/bw/research/projects/TLEMsafe for the biomechanical community, in order to accelerate the development and adoption of subject-specific models on large scale. TLEM 2.0 is freely shared for non-commercial use only, under acceptance of the TLEMsafe Research License Agreement. Copyright © 2014 Elsevier Ltd. All rights reserved.

SAR image classification based on CNN in real and simulation datasets

NASA Astrophysics Data System (ADS)

Peng, Lijiang; Liu, Ming; Liu, Xiaohua; Dong, Liquan; Hui, Mei; Zhao, Yuejin

2018-04-01

Convolution neural network (CNN) has made great success in image classification tasks. Even in the field of synthetic aperture radar automatic target recognition (SAR-ATR), state-of-art results has been obtained by learning deep representation of features on the MSTAR benchmark. However, the raw data of MSTAR have shortcomings in training a SAR-ATR model because of high similarity in background among the SAR images of each kind. This indicates that the CNN would learn the hierarchies of features of backgrounds as well as the targets. To validate the influence of the background, some other SAR images datasets have been made which contains the simulation SAR images of 10 manufactured targets such as tank and fighter aircraft, and the backgrounds of simulation SAR images are sampled from the whole original MSTAR data. The simulation datasets contain the dataset that the backgrounds of each kind images correspond to the one kind of backgrounds of MSTAR targets or clutters and the dataset that each image shares the random background of whole MSTAR targets or clutters. In addition, mixed datasets of MSTAR and simulation datasets had been made to use in the experiments. The CNN architecture proposed in this paper are trained on all datasets mentioned above. The experimental results shows that the architecture can get high performances on all datasets even the backgrounds of the images are miscellaneous, which indicates the architecture can learn a good representation of the targets even though the drastic changes on background.

Unsupervised feature learning for autonomous rock image classification

NASA Astrophysics Data System (ADS)

Shu, Lei; McIsaac, Kenneth; Osinski, Gordon R.; Francis, Raymond

2017-09-01

Autonomous rock image classification can enhance the capability of robots for geological detection and enlarge the scientific returns, both in investigation on Earth and planetary surface exploration on Mars. Since rock textural images are usually inhomogeneous and manually hand-crafting features is not always reliable, we propose an unsupervised feature learning method to autonomously learn the feature representation for rock images. In our tests, rock image classification using the learned features shows that the learned features can outperform manually selected features. Self-taught learning is also proposed to learn the feature representation from a large database of unlabelled rock images of mixed class. The learned features can then be used repeatedly for classification of any subclass. This takes advantage of the large dataset of unlabelled rock images and learns a general feature representation for many kinds of rocks. We show experimental results supporting the feasibility of self-taught learning on rock images.

Crowdsourcing-based evaluation of privacy in HDR images

NASA Astrophysics Data System (ADS)

Korshunov, Pavel; Nemoto, Hiromi; Skodras, Athanassios; Ebrahimi, Touradj

2014-05-01

The ability of High Dynamic Range imaging (HDRi) to capture details in high-contrast environments, making both dark and bright regions clearly visible, has a strong implication on privacy. However, the extent to which HDRi affects privacy when it is used instead of typical Standard Dynamic Range imaging (SDRi) is not yet clear. In this paper, we investigate the effect of HDRi on privacy via crowdsourcing evaluation using the Microworkers platform. Due to the lack of HDRi standard privacy evaluation dataset, we have created such dataset containing people of varying gender, race, and age, shot indoor and outdoor and under large range of lighting conditions. We evaluate the tone-mapped versions of these images, obtained by several representative tone-mapping algorithms, using subjective privacy evaluation methodology. Evaluation was performed using crowdsourcing-based framework, because it is a popular and effective alternative to traditional lab-based assessment. The results of the experiments demonstrate a significant loss of privacy when even tone-mapped versions of HDR images are used compared to typical SDR images shot with a standard exposure.

Atlas-Guided Cluster Analysis of Large Tractography Datasets

PubMed Central

Ros, Christian; Güllmar, Daniel; Stenzel, Martin; Mentzel, Hans-Joachim; Reichenbach, Jürgen Rainer

2013-01-01

Diffusion Tensor Imaging (DTI) and fiber tractography are important tools to map the cerebral white matter microstructure in vivo and to model the underlying axonal pathways in the brain with three-dimensional fiber tracts. As the fast and consistent extraction of anatomically correct fiber bundles for multiple datasets is still challenging, we present a novel atlas-guided clustering framework for exploratory data analysis of large tractography datasets. The framework uses an hierarchical cluster analysis approach that exploits the inherent redundancy in large datasets to time-efficiently group fiber tracts. Structural information of a white matter atlas can be incorporated into the clustering to achieve an anatomically correct and reproducible grouping of fiber tracts. This approach facilitates not only the identification of the bundles corresponding to the classes of the atlas; it also enables the extraction of bundles that are not present in the atlas. The new technique was applied to cluster datasets of 46 healthy subjects. Prospects of automatic and anatomically correct as well as reproducible clustering are explored. Reconstructed clusters were well separated and showed good correspondence to anatomical bundles. Using the atlas-guided cluster approach, we observed consistent results across subjects with high reproducibility. In order to investigate the outlier elimination performance of the clustering algorithm, scenarios with varying amounts of noise were simulated and clustered with three different outlier elimination strategies. By exploiting the multithreading capabilities of modern multiprocessor systems in combination with novel algorithms, our toolkit clusters large datasets in a couple of minutes. Experiments were conducted to investigate the achievable speedup and to demonstrate the high performance of the clustering framework in a multiprocessing environment. PMID:24386292

Web-based platform for collaborative medical imaging research

NASA Astrophysics Data System (ADS)

Rittner, Leticia; Bento, Mariana P.; Costa, André L.; Souza, Roberto M.; Machado, Rubens C.; Lotufo, Roberto A.

2015-03-01

Medical imaging research depends basically on the availability of large image collections, image processing and analysis algorithms, hardware and a multidisciplinary research team. It has to be reproducible, free of errors, fast, accessible through a large variety of devices spread around research centers and conducted simultaneously by a multidisciplinary team. Therefore, we propose a collaborative research environment, named Adessowiki, where tools and datasets are integrated and readily available in the Internet through a web browser. Moreover, processing history and all intermediate results are stored and displayed in automatic generated web pages for each object in the research project or clinical study. It requires no installation or configuration from the client side and offers centralized tools and specialized hardware resources, since processing takes place in the cloud.

Mars Global Geologic Mapping: About Half Way Done

NASA Technical Reports Server (NTRS)

Tanaka, K. L.; Dohm, J. M.; Irwin, R.; Kolb, E. J.; Skinner, J. A., Jr.; Hare, T. M.

2009-01-01

We are in the third year of a five-year effort to map the geology of Mars using mainly Mars Global Surveyor, Mars Express, and Mars Odyssey imaging and altimetry datasets. Previously, we have reported on details of project management, mapping datasets (local and regional), initial and anticipated mapping approaches, and tactics of map unit delineation and description [1-2]. For example, we have seen how the multiple types and huge quantity of image data as well as more accurate and detailed altimetry data now available allow for broader and deeper geologic perspectives, based largely on improved landform perception, characterization, and analysis. Here, we describe mapping and unit delineation results thus far, a new unit identified in the northern plains, and remaining steps to complete the map.

On the Multi-Modal Object Tracking and Image Fusion Using Unsupervised Deep Learning Methodologies

NASA Astrophysics Data System (ADS)

LaHaye, N.; Ott, J.; Garay, M. J.; El-Askary, H. M.; Linstead, E.

2017-12-01

The number of different modalities of remote-sensors has been on the rise, resulting in large datasets with different complexity levels. Such complex datasets can provide valuable information separately, yet there is a bigger value in having a comprehensive view of them combined. As such, hidden information can be deduced through applying data mining techniques on the fused data. The curse of dimensionality of such fused data, due to the potentially vast dimension space, hinders our ability to have deep understanding of them. This is because each dataset requires a user to have instrument-specific and dataset-specific knowledge for optimum and meaningful usage. Once a user decides to use multiple datasets together, deeper understanding of translating and combining these datasets in a correct and effective manner is needed. Although there exists data centric techniques, generic automated methodologies that can potentially solve this problem completely don't exist. Here we are developing a system that aims to gain a detailed understanding of different data modalities. Such system will provide an analysis environment that gives the user useful feedback and can aid in research tasks. In our current work, we show the initial outputs our system implementation that leverages unsupervised deep learning techniques so not to burden the user with the task of labeling input data, while still allowing for a detailed machine understanding of the data. Our goal is to be able to track objects, like cloud systems or aerosols, across different image-like data-modalities. The proposed system is flexible, scalable and robust to understand complex likenesses within multi-modal data in a similar spatio-temporal range, and also to be able to co-register and fuse these images when needed.

A Fully-Automated Subcortical and Ventricular Shape Generation Pipeline Preserving Smoothness and Anatomical Topology

PubMed Central

Tang, Xiaoying; Luo, Yuan; Chen, Zhibin; Huang, Nianwei; Johnson, Hans J.; Paulsen, Jane S.; Miller, Michael I.

2018-01-01

In this paper, we present a fully-automated subcortical and ventricular shape generation pipeline that acts on structural magnetic resonance images (MRIs) of the human brain. Principally, the proposed pipeline consists of three steps: (1) automated structure segmentation using the diffeomorphic multi-atlas likelihood-fusion algorithm; (2) study-specific shape template creation based on the Delaunay triangulation; (3) deformation-based shape filtering using the large deformation diffeomorphic metric mapping for surfaces. The proposed pipeline is shown to provide high accuracy, sufficient smoothness, and accurate anatomical topology. Two datasets focused upon Huntington's disease (HD) were used for evaluating the performance of the proposed pipeline. The first of these contains a total of 16 MRI scans, each with a gold standard available, on which the proposed pipeline's outputs were observed to be highly accurate and smooth when compared with the gold standard. Visual examinations and outlier analyses on the second dataset, which contains a total of 1,445 MRI scans, revealed 100% success rates for the putamen, the thalamus, the globus pallidus, the amygdala, and the lateral ventricle in both hemispheres and rates no smaller than 97% for the bilateral hippocampus and caudate. Another independent dataset, consisting of 15 atlas images and 20 testing images, was also used to quantitatively evaluate the proposed pipeline, with high accuracy having been obtained. In short, the proposed pipeline is herein demonstrated to be effective, both quantitatively and qualitatively, using a large collection of MRI scans. PMID:29867332

A Fully-Automated Subcortical and Ventricular Shape Generation Pipeline Preserving Smoothness and Anatomical Topology.

PubMed

Tang, Xiaoying; Luo, Yuan; Chen, Zhibin; Huang, Nianwei; Johnson, Hans J; Paulsen, Jane S; Miller, Michael I

2018-01-01

In this paper, we present a fully-automated subcortical and ventricular shape generation pipeline that acts on structural magnetic resonance images (MRIs) of the human brain. Principally, the proposed pipeline consists of three steps: (1) automated structure segmentation using the diffeomorphic multi-atlas likelihood-fusion algorithm; (2) study-specific shape template creation based on the Delaunay triangulation; (3) deformation-based shape filtering using the large deformation diffeomorphic metric mapping for surfaces. The proposed pipeline is shown to provide high accuracy, sufficient smoothness, and accurate anatomical topology. Two datasets focused upon Huntington's disease (HD) were used for evaluating the performance of the proposed pipeline. The first of these contains a total of 16 MRI scans, each with a gold standard available, on which the proposed pipeline's outputs were observed to be highly accurate and smooth when compared with the gold standard. Visual examinations and outlier analyses on the second dataset, which contains a total of 1,445 MRI scans, revealed 100% success rates for the putamen, the thalamus, the globus pallidus, the amygdala, and the lateral ventricle in both hemispheres and rates no smaller than 97% for the bilateral hippocampus and caudate. Another independent dataset, consisting of 15 atlas images and 20 testing images, was also used to quantitatively evaluate the proposed pipeline, with high accuracy having been obtained. In short, the proposed pipeline is herein demonstrated to be effective, both quantitatively and qualitatively, using a large collection of MRI scans.

Intervertebral disc detection in X-ray images using faster R-CNN.

PubMed

Ruhan Sa; Owens, William; Wiegand, Raymond; Studin, Mark; Capoferri, Donald; Barooha, Kenneth; Greaux, Alexander; Rattray, Robert; Hutton, Adam; Cintineo, John; Chaudhary, Vipin

2017-07-01

Automatic identification of specific osseous landmarks on the spinal radiograph can be used to automate calculations for correcting ligament instability and injury, which affect 75% of patients injured in motor vehicle accidents. In this work, we propose to use deep learning based object detection method as the first step towards identifying landmark points in lateral lumbar X-ray images. The significant breakthrough of deep learning technology has made it a prevailing choice for perception based applications, however, the lack of large annotated training dataset has brought challenges to utilizing the technology in medical image processing field. In this work, we propose to fine tune a deep network, Faster-RCNN, a state-of-the-art deep detection network in natural image domain, using small annotated clinical datasets. In the experiment we show that, by using only 81 lateral lumbar X-Ray training images, one can achieve much better performance compared to traditional sliding window detection method on hand crafted features. Furthermore, we fine-tuned the network using 974 training images and tested on 108 images, which achieved average precision of 0.905 with average computation time of 3 second per image, which greatly outperformed traditional methods in terms of accuracy and efficiency.

Semi-supervised manifold learning with affinity regularization for Alzheimer's disease identification using positron emission tomography imaging.

PubMed

Lu, Shen; Xia, Yong; Cai, Tom Weidong; Feng, David Dagan

2015-01-01

Dementia, Alzheimer's disease (AD) in particular is a global problem and big threat to the aging population. An image based computer-aided dementia diagnosis method is needed to providing doctors help during medical image examination. Many machine learning based dementia classification methods using medical imaging have been proposed and most of them achieve accurate results. However, most of these methods make use of supervised learning requiring fully labeled image dataset, which usually is not practical in real clinical environment. Using large amount of unlabeled images can improve the dementia classification performance. In this study we propose a new semi-supervised dementia classification method based on random manifold learning with affinity regularization. Three groups of spatial features are extracted from positron emission tomography (PET) images to construct an unsupervised random forest which is then used to regularize the manifold learning objective function. The proposed method, stat-of-the-art Laplacian support vector machine (LapSVM) and supervised SVM are applied to classify AD and normal controls (NC). The experiment results show that learning with unlabeled images indeed improves the classification performance. And our method outperforms LapSVM on the same dataset.

Anomaly detection for medical images based on a one-class classification

NASA Astrophysics Data System (ADS)

Wei, Qi; Ren, Yinhao; Hou, Rui; Shi, Bibo; Lo, Joseph Y.; Carin, Lawrence

2018-02-01

Detecting an anomaly such as a malignant tumor or a nodule from medical images including mammogram, CT or PET images is still an ongoing research problem drawing a lot of attention with applications in medical diagnosis. A conventional way to address this is to learn a discriminative model using training datasets of negative and positive samples. The learned model can be used to classify a testing sample into a positive or negative class. However, in medical applications, the high unbalance between negative and positive samples poses a difficulty for learning algorithms, as they will be biased towards the majority group, i.e., the negative one. To address this imbalanced data issue as well as leverage the huge amount of negative samples, i.e., normal medical images, we propose to learn an unsupervised model to characterize the negative class. To make the learned model more flexible and extendable for medical images of different scales, we have designed an autoencoder based on a deep neural network to characterize the negative patches decomposed from large medical images. A testing image is decomposed into patches and then fed into the learned autoencoder to reconstruct these patches themselves. The reconstruction error of one patch is used to classify this patch into a binary class, i.e., a positive or a negative one, leading to a one-class classifier. The positive patches highlight the suspicious areas containing anomalies in a large medical image. The proposed method has been tested on InBreast dataset and achieves an AUC of 0.84. The main contribution of our work can be summarized as follows. 1) The proposed one-class learning requires only data from one class, i.e., the negative data; 2) The patch-based learning makes the proposed method scalable to images of different sizes and helps avoid the large scale problem for medical images; 3) The training of the proposed deep convolutional neural network (DCNN) based auto-encoder is fast and stable.

Learning visual balance from large-scale datasets of aesthetically highly rated images

NASA Astrophysics Data System (ADS)

Jahanian, Ali; Vishwanathan, S. V. N.; Allebach, Jan P.

2015-03-01

The concept of visual balance is innate for humans, and influences how we perceive visual aesthetics and cognize harmony. Although visual balance is a vital principle of design and taught in schools of designs, it is barely quantified. On the other hand, with emergence of automantic/semi-automatic visual designs for self-publishing, learning visual balance and computationally modeling it, may escalate aesthetics of such designs. In this paper, we present how questing for understanding visual balance inspired us to revisit one of the well-known theories in visual arts, the so called theory of "visual rightness", elucidated by Arnheim. We define Arnheim's hypothesis as a design mining problem with the goal of learning visual balance from work of professionals. We collected a dataset of 120K images that are aesthetically highly rated, from a professional photography website. We then computed factors that contribute to visual balance based on the notion of visual saliency. We fitted a mixture of Gaussians to the saliency maps of the images, and obtained the hotspots of the images. Our inferred Gaussians align with Arnheim's hotspots, and confirm his theory. Moreover, the results support the viability of the center of mass, symmetry, as well as the Rule of Thirds in our dataset.

Wearable Device-Based Gait Recognition Using Angle Embedded Gait Dynamic Images and a Convolutional Neural Network.

PubMed

Zhao, Yongjia; Zhou, Suiping

2017-02-28

The widespread installation of inertial sensors in smartphones and other wearable devices provides a valuable opportunity to identify people by analyzing their gait patterns, for either cooperative or non-cooperative circumstances. However, it is still a challenging task to reliably extract discriminative features for gait recognition with noisy and complex data sequences collected from casually worn wearable devices like smartphones. To cope with this problem, we propose a novel image-based gait recognition approach using the Convolutional Neural Network (CNN) without the need to manually extract discriminative features. The CNN's input image, which is encoded straightforwardly from the inertial sensor data sequences, is called Angle Embedded Gait Dynamic Image (AE-GDI). AE-GDI is a new two-dimensional representation of gait dynamics, which is invariant to rotation and translation. The performance of the proposed approach in gait authentication and gait labeling is evaluated using two datasets: (1) the McGill University dataset, which is collected under realistic conditions; and (2) the Osaka University dataset with the largest number of subjects. Experimental results show that the proposed approach achieves competitive recognition accuracy over existing approaches and provides an effective parametric solution for identification among a large number of subjects by gait patterns.

Wearable Device-Based Gait Recognition Using Angle Embedded Gait Dynamic Images and a Convolutional Neural Network

PubMed Central

Zhao, Yongjia; Zhou, Suiping

2017-01-01

The widespread installation of inertial sensors in smartphones and other wearable devices provides a valuable opportunity to identify people by analyzing their gait patterns, for either cooperative or non-cooperative circumstances. However, it is still a challenging task to reliably extract discriminative features for gait recognition with noisy and complex data sequences collected from casually worn wearable devices like smartphones. To cope with this problem, we propose a novel image-based gait recognition approach using the Convolutional Neural Network (CNN) without the need to manually extract discriminative features. The CNN’s input image, which is encoded straightforwardly from the inertial sensor data sequences, is called Angle Embedded Gait Dynamic Image (AE-GDI). AE-GDI is a new two-dimensional representation of gait dynamics, which is invariant to rotation and translation. The performance of the proposed approach in gait authentication and gait labeling is evaluated using two datasets: (1) the McGill University dataset, which is collected under realistic conditions; and (2) the Osaka University dataset with the largest number of subjects. Experimental results show that the proposed approach achieves competitive recognition accuracy over existing approaches and provides an effective parametric solution for identification among a large number of subjects by gait patterns. PMID:28264503

Detection and Monitoring of Oil Spills Using Moderate/High-Resolution Remote Sensing Images.

PubMed

Li, Ying; Cui, Can; Liu, Zexi; Liu, Bingxin; Xu, Jin; Zhu, Xueyuan; Hou, Yongchao

2017-07-01

Current marine oil spill detection and monitoring methods using high-resolution remote sensing imagery are quite limited. This study presented a new bottom-up and top-down visual saliency model. We used Landsat 8, GF-1, MAMS, HJ-1 oil spill imagery as dataset. A simplified, graph-based visual saliency model was used to extract bottom-up saliency. It could identify the regions with high visual saliency object in the ocean. A spectral similarity match model was used to obtain top-down saliency. It could distinguish oil regions and exclude the other salient interference by spectrums. The regions of interest containing oil spills were integrated using these complementary saliency detection steps. Then, the genetic neural network was used to complete the image classification. These steps increased the speed of analysis. For the test dataset, the average running time of the entire process to detect regions of interest was 204.56 s. During image segmentation, the oil spill was extracted using a genetic neural network. The classification results showed that the method had a low false-alarm rate (high accuracy of 91.42%) and was able to increase the speed of the detection process (fast runtime of 19.88 s). The test image dataset was composed of different types of features over large areas in complicated imaging conditions. The proposed model was proved to be robust in complex sea conditions.

3D reconstruction software comparison for short sequences

NASA Astrophysics Data System (ADS)

Strupczewski, Adam; Czupryński, BłaŻej

2014-11-01

Large scale multiview reconstruction is recently a very popular area of research. There are many open source tools that can be downloaded and run on a personal computer. However, there are few, if any, comparisons between all the available software in terms of accuracy on small datasets that a single user can create. The typical datasets for testing of the software are archeological sites or cities, comprising thousands of images. This paper presents a comparison of currently available open source multiview reconstruction software for small datasets. It also compares the open source solutions with a simple structure from motion pipeline developed by the authors from scratch with the use of OpenCV and Eigen libraries.

An integration of minimum local feature representation methods to recognize large variation of foods

NASA Astrophysics Data System (ADS)

Razali, Mohd Norhisham bin; Manshor, Noridayu; Halin, Alfian Abdul; Mustapha, Norwati; Yaakob, Razali

2017-10-01

Local invariant features have shown to be successful in describing object appearances for image classification tasks. Such features are robust towards occlusion and clutter and are also invariant against scale and orientation changes. This makes them suitable for classification tasks with little inter-class similarity and large intra-class difference. In this paper, we propose an integrated representation of the Speeded-Up Robust Feature (SURF) and Scale Invariant Feature Transform (SIFT) descriptors, using late fusion strategy. The proposed representation is used for food recognition from a dataset of food images with complex appearance variations. The Bag of Features (BOF) approach is employed to enhance the discriminative ability of the local features. Firstly, the individual local features are extracted to construct two kinds of visual vocabularies, representing SURF and SIFT. The visual vocabularies are then concatenated and fed into a Linear Support Vector Machine (SVM) to classify the respective food categories. Experimental results demonstrate impressive overall recognition at 82.38% classification accuracy based on the challenging UEC-Food100 dataset.

A Virtual Reality Visualization Tool for Neuron Tracing

PubMed Central

Usher, Will; Klacansky, Pavol; Federer, Frederick; Bremer, Peer-Timo; Knoll, Aaron; Angelucci, Alessandra; Pascucci, Valerio

2017-01-01

Tracing neurons in large-scale microscopy data is crucial to establishing a wiring diagram of the brain, which is needed to understand how neural circuits in the brain process information and generate behavior. Automatic techniques often fail for large and complex datasets, and connectomics researchers may spend weeks or months manually tracing neurons using 2D image stacks. We present a design study of a new virtual reality (VR) system, developed in collaboration with trained neuroanatomists, to trace neurons in microscope scans of the visual cortex of primates. We hypothesize that using consumer-grade VR technology to interact with neurons directly in 3D will help neuroscientists better resolve complex cases and enable them to trace neurons faster and with less physical and mental strain. We discuss both the design process and technical challenges in developing an interactive system to navigate and manipulate terabyte-sized image volumes in VR. Using a number of different datasets, we demonstrate that, compared to widely used commercial software, consumer-grade VR presents a promising alternative for scientists. PMID:28866520

A Virtual Reality Visualization Tool for Neuron Tracing.

PubMed

Usher, Will; Klacansky, Pavol; Federer, Frederick; Bremer, Peer-Timo; Knoll, Aaron; Yarch, Jeff; Angelucci, Alessandra; Pascucci, Valerio

2018-01-01

Tracing neurons in large-scale microscopy data is crucial to establishing a wiring diagram of the brain, which is needed to understand how neural circuits in the brain process information and generate behavior. Automatic techniques often fail for large and complex datasets, and connectomics researchers may spend weeks or months manually tracing neurons using 2D image stacks. We present a design study of a new virtual reality (VR) system, developed in collaboration with trained neuroanatomists, to trace neurons in microscope scans of the visual cortex of primates. We hypothesize that using consumer-grade VR technology to interact with neurons directly in 3D will help neuroscientists better resolve complex cases and enable them to trace neurons faster and with less physical and mental strain. We discuss both the design process and technical challenges in developing an interactive system to navigate and manipulate terabyte-sized image volumes in VR. Using a number of different datasets, we demonstrate that, compared to widely used commercial software, consumer-grade VR presents a promising alternative for scientists.

Deep image mining for diabetic retinopathy screening.

PubMed

Quellec, Gwenolé; Charrière, Katia; Boudi, Yassine; Cochener, Béatrice; Lamard, Mathieu

2017-07-01

Deep learning is quickly becoming the leading methodology for medical image analysis. Given a large medical archive, where each image is associated with a diagnosis, efficient pathology detectors or classifiers can be trained with virtually no expert knowledge about the target pathologies. However, deep learning algorithms, including the popular ConvNets, are black boxes: little is known about the local patterns analyzed by ConvNets to make a decision at the image level. A solution is proposed in this paper to create heatmaps showing which pixels in images play a role in the image-level predictions. In other words, a ConvNet trained for image-level classification can be used to detect lesions as well. A generalization of the backpropagation method is proposed in order to train ConvNets that produce high-quality heatmaps. The proposed solution is applied to diabetic retinopathy (DR) screening in a dataset of almost 90,000 fundus photographs from the 2015 Kaggle Diabetic Retinopathy competition and a private dataset of almost 110,000 photographs (e-ophtha). For the task of detecting referable DR, very good detection performance was achieved: A z =0.954 in Kaggle's dataset and A z =0.949 in e-ophtha. Performance was also evaluated at the image level and at the lesion level in the DiaretDB1 dataset, where four types of lesions are manually segmented: microaneurysms, hemorrhages, exudates and cotton-wool spots. For the task of detecting images containing these four lesion types, the proposed detector, which was trained to detect referable DR, outperforms recent algorithms trained to detect those lesions specifically, with pixel-level supervision. At the lesion level, the proposed detector outperforms heatmap generation algorithms for ConvNets. This detector is part of the Messidor® system for mobile eye pathology screening. Because it does not rely on expert knowledge or manual segmentation for detecting relevant patterns, the proposed solution is a promising image mining tool, which has the potential to discover new biomarkers in images. Copyright © 2017 Elsevier B.V. All rights reserved.

Cell nuclei and cytoplasm joint segmentation using the sliding band filter.

PubMed

Quelhas, Pedro; Marcuzzo, Monica; Mendonça, Ana Maria; Campilho, Aurélio

2010-08-01

Microscopy cell image analysis is a fundamental tool for biological research. In particular, multivariate fluorescence microscopy is used to observe different aspects of cells in cultures. It is still common practice to perform analysis tasks by visual inspection of individual cells which is time consuming, exhausting and prone to induce subjective bias. This makes automatic cell image analysis essential for large scale, objective studies of cell cultures. Traditionally the task of automatic cell analysis is approached through the use of image segmentation methods for extraction of cells' locations and shapes. Image segmentation, although fundamental, is neither an easy task in computer vision nor is it robust to image quality changes. This makes image segmentation for cell detection semi-automated requiring frequent tuning of parameters. We introduce a new approach for cell detection and shape estimation in multivariate images based on the sliding band filter (SBF). This filter's design makes it adequate to detect overall convex shapes and as such it performs well for cell detection. Furthermore, the parameters involved are intuitive as they are directly related to the expected cell size. Using the SBF filter we detect cells' nucleus and cytoplasm location and shapes. Based on the assumption that each cell has the same approximate shape center in both nuclei and cytoplasm fluorescence channels, we guide cytoplasm shape estimation by the nuclear detections improving performance and reducing errors. Then we validate cell detection by gathering evidence from nuclei and cytoplasm channels. Additionally, we include overlap correction and shape regularization steps which further improve the estimated cell shapes. The approach is evaluated using two datasets with different types of data: a 20 images benchmark set of simulated cell culture images, containing 1000 simulated cells; a 16 images Drosophila melanogaster Kc167 dataset containing 1255 cells, stained for DNA and actin. Both image datasets present a difficult problem due to the high variability of cell shapes and frequent cluster overlap between cells. On the Drosophila dataset our approach achieved a precision/recall of 95%/69% and 82%/90% for nuclei and cytoplasm detection respectively and an overall accuracy of 76%.

SchizConnect: Mediating Neuroimaging Databases on Schizophrenia and Related Disorders for Large-Scale Integration

PubMed Central

Wang, Lei; Alpert, Kathryn I.; Calhoun, Vince D.; Cobia, Derin J.; Keator, David B.; King, Margaret D.; Kogan, Alexandr; Landis, Drew; Tallis, Marcelo; Turner, Matthew D.; Potkin, Steven G.; Turner, Jessica A.; Ambite, Jose Luis

2015-01-01

SchizConnect (www.schizconnect.org) is built to address the issues of multiple data repositories in schizophrenia neuroimaging studies. It includes a level of mediation—translating across data sources—so that the user can place one query, e.g. for diffusion images from male individuals with schizophrenia, and find out from across participating data sources how many datasets there are, as well as downloading the imaging and related data. The current version handles the Data Usage Agreements across different studies, as well as interpreting database-specific terminologies into a common framework. New data repositories can also be mediated to bring immediate access to existing datasets. Compared with centralized, upload data sharing models, SchizConnect is a unique, virtual database with a focus on schizophrenia and related disorders that can mediate live data as information are being updated at each data source. It is our hope that SchizConnect can facilitate testing new hypotheses through aggregated datasets, promoting discovery related to the mechanisms underlying schizophrenic dysfunction. PMID:26142271

Secure Access Control and Large Scale Robust Representation for Online Multimedia Event Detection

PubMed Central

Liu, Changyu; Li, Huiling

2014-01-01

We developed an online multimedia event detection (MED) system. However, there are a secure access control issue and a large scale robust representation issue when we want to integrate traditional event detection algorithms into the online environment. For the first issue, we proposed a tree proxy-based and service-oriented access control (TPSAC) model based on the traditional role based access control model. Verification experiments were conducted on the CloudSim simulation platform, and the results showed that the TPSAC model is suitable for the access control of dynamic online environments. For the second issue, inspired by the object-bank scene descriptor, we proposed a 1000-object-bank (1000OBK) event descriptor. Feature vectors of the 1000OBK were extracted from response pyramids of 1000 generic object detectors which were trained on standard annotated image datasets, such as the ImageNet dataset. A spatial bag of words tiling approach was then adopted to encode these feature vectors for bridging the gap between the objects and events. Furthermore, we performed experiments in the context of event classification on the challenging TRECVID MED 2012 dataset, and the results showed that the robust 1000OBK event descriptor outperforms the state-of-the-art approaches. PMID:25147840

Training Scalable Restricted Boltzmann Machines Using a Quantum Annealer

NASA Astrophysics Data System (ADS)

Kumar, V.; Bass, G.; Dulny, J., III

2016-12-01

Machine learning and the optimization involved therein is of critical importance for commercial and military applications. Due to the computational complexity of many-variable optimization, the conventional approach is to employ meta-heuristic techniques to find suboptimal solutions. Quantum Annealing (QA) hardware offers a completely novel approach with the potential to obtain significantly better solutions with large speed-ups compared to traditional computing. In this presentation, we describe our development of new machine learning algorithms tailored for QA hardware. We are training restricted Boltzmann machines (RBMs) using QA hardware on large, high-dimensional commercial datasets. Traditional optimization heuristics such as contrastive divergence and other closely related techniques are slow to converge, especially on large datasets. Recent studies have indicated that QA hardware when used as a sampler provides better training performance compared to conventional approaches. Most of these studies have been limited to moderately-sized datasets due to the hardware restrictions imposed by exisitng QA devices, which make it difficult to solve real-world problems at scale. In this work we develop novel strategies to circumvent this issue. We discuss scale-up techniques such as enhanced embedding and partitioned RBMs which allow large commercial datasets to be learned using QA hardware. We present our initial results obtained by training an RBM as an autoencoder on an image dataset. The results obtained so far indicate that the convergence rates can be improved significantly by increasing RBM network connectivity. These ideas can be readily applied to generalized Boltzmann machines and we are currently investigating this in an ongoing project.

PANDA: a pipeline toolbox for analyzing brain diffusion images.

PubMed

Cui, Zaixu; Zhong, Suyu; Xu, Pengfei; He, Yong; Gong, Gaolang

2013-01-01

Diffusion magnetic resonance imaging (dMRI) is widely used in both scientific research and clinical practice in in-vivo studies of the human brain. While a number of post-processing packages have been developed, fully automated processing of dMRI datasets remains challenging. Here, we developed a MATLAB toolbox named "Pipeline for Analyzing braiN Diffusion imAges" (PANDA) for fully automated processing of brain diffusion images. The processing modules of a few established packages, including FMRIB Software Library (FSL), Pipeline System for Octave and Matlab (PSOM), Diffusion Toolkit and MRIcron, were employed in PANDA. Using any number of raw dMRI datasets from different subjects, in either DICOM or NIfTI format, PANDA can automatically perform a series of steps to process DICOM/NIfTI to diffusion metrics [e.g., fractional anisotropy (FA) and mean diffusivity (MD)] that are ready for statistical analysis at the voxel-level, the atlas-level and the Tract-Based Spatial Statistics (TBSS)-level and can finish the construction of anatomical brain networks for all subjects. In particular, PANDA can process different subjects in parallel, using multiple cores either in a single computer or in a distributed computing environment, thus greatly reducing the time cost when dealing with a large number of datasets. In addition, PANDA has a friendly graphical user interface (GUI), allowing the user to be interactive and to adjust the input/output settings, as well as the processing parameters. As an open-source package, PANDA is freely available at http://www.nitrc.org/projects/panda/. This novel toolbox is expected to substantially simplify the image processing of dMRI datasets and facilitate human structural connectome studies.

Plant species classification using flower images—A comparative study of local feature representations

PubMed Central

Seeland, Marco; Rzanny, Michael; Alaqraa, Nedal; Wäldchen, Jana; Mäder, Patrick

2017-01-01

Steady improvements of image description methods induced a growing interest in image-based plant species classification, a task vital to the study of biodiversity and ecological sensitivity. Various techniques have been proposed for general object classification over the past years and several of them have already been studied for plant species classification. However, results of these studies are selective in the evaluated steps of a classification pipeline, in the utilized datasets for evaluation, and in the compared baseline methods. No study is available that evaluates the main competing methods for building an image representation on the same datasets allowing for generalized findings regarding flower-based plant species classification. The aim of this paper is to comparatively evaluate methods, method combinations, and their parameters towards classification accuracy. The investigated methods span from detection, extraction, fusion, pooling, to encoding of local features for quantifying shape and color information of flower images. We selected the flower image datasets Oxford Flower 17 and Oxford Flower 102 as well as our own Jena Flower 30 dataset for our experiments. Findings show large differences among the various studied techniques and that their wisely chosen orchestration allows for high accuracies in species classification. We further found that true local feature detectors in combination with advanced encoding methods yield higher classification results at lower computational costs compared to commonly used dense sampling and spatial pooling methods. Color was found to be an indispensable feature for high classification results, especially while preserving spatial correspondence to gray-level features. In result, our study provides a comprehensive overview of competing techniques and the implications of their main parameters for flower-based plant species classification. PMID:28234999

Digital Rocks Portal: a sustainable platform for imaged dataset sharing, translation and automated analysis

NASA Astrophysics Data System (ADS)

Prodanovic, M.; Esteva, M.; Hanlon, M.; Nanda, G.; Agarwal, P.

2015-12-01

Recent advances in imaging have provided a wealth of 3D datasets that reveal pore space microstructure (nm to cm length scale) and allow investigation of nonlinear flow and mechanical phenomena from first principles using numerical approaches. This framework has popularly been called "digital rock physics". Researchers, however, have trouble storing and sharing the datasets both due to their size and the lack of standardized image types and associated metadata for volumetric datasets. This impedes scientific cross-validation of the numerical approaches that characterize large scale porous media properties, as well as development of multiscale approaches required for correct upscaling. A single research group typically specializes in an imaging modality and/or related modeling on a single length scale, and lack of data-sharing infrastructure makes it difficult to integrate different length scales. We developed a sustainable, open and easy-to-use repository called the Digital Rocks Portal, that (1) organizes images and related experimental measurements of different porous materials, (2) improves access to them for a wider community of geosciences or engineering researchers not necessarily trained in computer science or data analysis. Once widely accepter, the repository will jumpstart productivity and enable scientific inquiry and engineering decisions founded on a data-driven basis. This is the first repository of its kind. We show initial results on incorporating essential software tools and pipelines that make it easier for researchers to store and reuse data, and for educators to quickly visualize and illustrate concepts to a wide audience. For data sustainability and continuous access, the portal is implemented within the reliable, 24/7 maintained High Performance Computing Infrastructure supported by the Texas Advanced Computing Center (TACC) at the University of Texas at Austin. Long-term storage is provided through the University of Texas System Research Cyber-infrastructure initiative.

Using kittens to unlock photo-sharing website datasets for environmental applications

NASA Astrophysics Data System (ADS)

Gascoin, Simon

2016-04-01

Mining photo-sharing websites is a promising approach to complement in situ and satellite observations of the environment, however a challenge is to deal with the large degree of noise inherent to online social datasets. Here I explored the value of the Flickr image hosting website database to monitor the snow cover in the Pyrenees. Using the Flickr application programming interface (API) I queried all the public images metadata tagged at least with one of the following words: "snow", "neige", "nieve", "neu" (snow in French, Spanish and Catalan languages). The search was limited to the geo-tagged pictures taken in the Pyrenees area. However, the number of public pictures available in the Flickr database for a given time interval depends on several factors, including the Flickr website popularity and the development of digital photography. Thus, I also searched for all Flickr images tagged with "chat", "gat" or "gato" (cat in French, Spanish and Catalan languages). The tag "cat" was not considered in order to exclude the results from North America where Flickr got popular earlier than in Europe. The number of "cat" images per month was used to fit a model of the number of images uploaded in Flickr with time. This model was used to remove this trend in the numbers of snow-tagged photographs. The resulting time series was compared to a time series of the snow cover area derived from the MODIS satellite over the same region. Both datasets are well correlated; in particular they exhibit the same seasonal evolution, although the inter-annual variabilities are less similar. I will also discuss which other factors may explain the main discrepancies in order to further decrease the noise in the Flickr dataset.

The Livermore Brain: Massive Deep Learning Networks Enabled by High Performance Computing

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

Chen, Barry Y.

The proliferation of inexpensive sensor technologies like the ubiquitous digital image sensors has resulted in the collection and sharing of vast amounts of unsorted and unexploited raw data. Companies and governments who are able to collect and make sense of large datasets to help them make better decisions more rapidly will have a competitive advantage in the information era. Machine Learning technologies play a critical role for automating the data understanding process; however, to be maximally effective, useful intermediate representations of the data are required. These representations or “features” are transformations of the raw data into a form where patternsmore » are more easily recognized. Recent breakthroughs in Deep Learning have made it possible to learn these features from large amounts of labeled data. The focus of this project is to develop and extend Deep Learning algorithms for learning features from vast amounts of unlabeled data and to develop the HPC neural network training platform to support the training of massive network models. This LDRD project succeeded in developing new unsupervised feature learning algorithms for images and video and created a scalable neural network training toolkit for HPC. Additionally, this LDRD helped create the world’s largest freely-available image and video dataset supporting open multimedia research and used this dataset for training our deep neural networks. This research helped LLNL capture several work-for-others (WFO) projects, attract new talent, and establish collaborations with leading academic and commercial partners. Finally, this project demonstrated the successful training of the largest unsupervised image neural network using HPC resources and helped establish LLNL leadership at the intersection of Machine Learning and HPC research.« less

A PDS Archive for Observations of Mercury's Na Exosphere

NASA Astrophysics Data System (ADS)

Backes, C.; Cassidy, T.; Merkel, A. W.; Killen, R. M.; Potter, A. E.

2016-12-01

We present a data product consisting of ground-based observations of Mercury's sodium exosphere. We have amassed a sizeable dataset of several thousand spectral observations of Mercury's exosphere from the McMath-Pierce solar telescope. Over the last year, a data reduction pipeline has been developed and refined to process and reconstruct these spectral images into low resolution images of sodium D2 emission. This dataset, which extends over two decades, will provide an unprecedented opportunity to analyze the dynamics of Mercury's mid to high-latitude exospheric emissions, which have long been attributed to solar wind ion bombardment. This large archive of observations will be of great use to the Mercury science community in studying the effects of space weather on Mercury's tenuous exosphere. When completely processed, images in this dataset will show the observed spatial distribution of Na D2 in the Mercurian exosphere, have measurements of this sodium emission per pixel in units of kilorayleighs, and be available through NASA's Planetary Data System. The overall goal of the presentation will be to provide the Planetary Science community with a clear picture of what information and data this archival product will make available.

Quantitative Evaluation of Automated Skull-Stripping Methods Applied to Contemporary and Legacy Images: Effects of Diagnosis, Bias Correction, and Slice Location

PubMed Central

Fennema-Notestine, Christine; Ozyurt, I. Burak; Clark, Camellia P.; Morris, Shaunna; Bischoff-Grethe, Amanda; Bondi, Mark W.; Jernigan, Terry L.; Fischl, Bruce; Segonne, Florent; Shattuck, David W.; Leahy, Richard M.; Rex, David E.; Toga, Arthur W.; Zou, Kelly H.; BIRN, Morphometry; Brown, Gregory G.

2008-01-01

Performance of automated methods to isolate brain from nonbrain tissues in magnetic resonance (MR) structural images may be influenced by MR signal inhomogeneities, type of MR image set, regional anatomy, and age and diagnosis of subjects studied. The present study compared the performance of four methods: Brain Extraction Tool (BET; Smith [2002]: Hum Brain Mapp 17:143–155); 3dIntracranial (Ward [1999] Milwaukee: Biophysics Research Institute, Medical College of Wisconsin; in AFNI); a Hybrid Watershed algorithm (HWA, Segonne et al. [2004] Neuroimage 22:1060–1075; in FreeSurfer); and Brain Surface Extractor (BSE, Sandor and Leahy [1997] IEEE Trans Med Imag 16:41–54; Shattuck et al. [2001] Neuroimage 13:856 – 876) to manually stripped images. The methods were applied to uncorrected and bias-corrected datasets; Legacy and Contemporary T1-weighted image sets; and four diagnostic groups (depressed, Alzheimer’s, young and elderly control). To provide a criterion for outcome assessment, two experts manually stripped six sagittal sections for each dataset in locations where brain and nonbrain tissue are difficult to distinguish. Methods were compared on Jaccard similarity coefficients, Hausdorff distances, and an Expectation-Maximization algorithm. Methods tended to perform better on contemporary datasets; bias correction did not significantly improve method performance. Mesial sections were most difficult for all methods. Although AD image sets were most difficult to strip, HWA and BSE were more robust across diagnostic groups compared with 3dIntracranial and BET. With respect to specificity, BSE tended to perform best across all groups, whereas HWA was more sensitive than other methods. The results of this study may direct users towards a method appropriate to their T1-weighted datasets and improve the efficiency of processing for large, multisite neuroimaging studies. PMID:15986433

Development of a novel constellation based landmark detection algorithm

NASA Astrophysics Data System (ADS)

Ghayoor, Ali; Vaidya, Jatin G.; Johnson, Hans J.

2013-03-01

Anatomical landmarks such as the anterior commissure (AC) and posterior commissure (PC) are commonly used by researchers for co-registration of images. In this paper, we present a novel, automated approach for landmark detection that combines morphometric constraining and statistical shape models to provide accurate estimation of landmark points. This method is made robust to large rotations in initial head orientation by extracting extra information of the eye centers using a radial Hough transform and exploiting the centroid of head mass (CM) using a novel estimation approach. To evaluate the effectiveness of this method, the algorithm is trained on a set of 20 images with manually selected landmarks, and a test dataset is used to compare the automatically detected against the manually detected landmark locations of the AC, PC, midbrain-pons junction (MPJ), and fourth ventricle notch (VN4). The results show that the proposed method is accurate as the average error between the automatically and manually labeled landmark points is less than 1 mm. Also, the algorithm is highly robust as it was successfully run on a large dataset that included different kinds of images with various orientation, spacing, and origin.

A method for medulloblastoma tumor differentiation based on convolutional neural networks and transfer learning

NASA Astrophysics Data System (ADS)

Cruz-Roa, Angel; Arévalo, John; Judkins, Alexander; Madabhushi, Anant; González, Fabio

2015-12-01

Convolutional neural networks (CNN) have been very successful at addressing different computer vision tasks thanks to their ability to learn image representations directly from large amounts of labeled data. Features learned from a dataset can be used to represent images from a different dataset via an approach called transfer learning. In this paper we apply transfer learning to the challenging task of medulloblastoma tumor differentiation. We compare two different CNN models which were previously trained in two different domains (natural and histopathology images). The first CNN is a state-of-the-art approach in computer vision, a large and deep CNN with 16-layers, Visual Geometry Group (VGG) CNN. The second (IBCa-CNN) is a 2-layer CNN trained for invasive breast cancer tumor classification. Both CNNs are used as visual feature extractors of histopathology image regions of anaplastic and non-anaplastic medulloblastoma tumor from digitized whole-slide images. The features from the two models are used, separately, to train a softmax classifier to discriminate between anaplastic and non-anaplastic medulloblastoma image regions. Experimental results show that the transfer learning approach produce competitive results in comparison with the state of the art approaches for IBCa detection. Results also show that features extracted from the IBCa-CNN have better performance in comparison with features extracted from the VGG-CNN. The former obtains 89.8% while the latter obtains 76.6% in terms of average accuracy.

High End Visualization of Geophysical Datasets Using Immersive Technology: The SIO Visualization Center.

NASA Astrophysics Data System (ADS)

Newman, R. L.

2002-12-01

How many images can you display at one time with Power Point without getting "postage stamps"? Do you have fantastic datasets that you cannot view because your computer is too slow/small? Do you assume a few 2-D images of a 3-D picture are sufficient? High-end visualization centers can minimize and often eliminate these problems. The new visualization center [http://siovizcenter.ucsd.edu] at Scripps Institution of Oceanography [SIO] immerses users into a virtual world by projecting 3-D images onto a Panoram GVR-120E wall-sized floor-to-ceiling curved screen [7' x 23'] that has 3.2 mega-pixels of resolution. The Infinite Reality graphics subsystem is driven by a single-pipe SGI Onyx 3400 with a system bandwidth of 44 Gbps. The Onyx is powered by 16 MIPS R12K processors and 16 GB of addressable memory. The system is also equipped with transmitters and LCD shutter glasses which permit stereographic 3-D viewing of high-resolution images. This center is ideal for groups of up to 60 people who can simultaneously view these large-format images. A wide range of hardware and software is available, giving the users a totally immersive working environment in which to display, analyze, and discuss large datasets. The system enables simultaneous display of video and audio streams from sources such as SGI megadesktop and stereo megadesktop, S-VHS video, DVD video, and video from a Macintosh or PC. For instance, one-third of the screen might be displaying S-VHS video from a remotely-operated-vehicle [ROV], while the remaining portion of the screen might be used for an interactive 3-D flight over the same parcel of seafloor. The video and audio combinations using this system are numerous, allowing users to combine and explore data and images in innovative ways, greatly enhancing scientists' ability to visualize, understand and collaborate on complex datasets. In the not-distant future, with the rapid growth in networking speeds in the US, it will be possible for Earth Sciences Departments to collaborate effectively while limiting the amount of physical travel required. This includes porting visualization content to the popular, low-cost Geowall visualization systems, and providing web-based access to databanks filled with stock geoscience visualizations.

A tool for the estimation of the distribution of landslide area in R

NASA Astrophysics Data System (ADS)

Rossi, M.; Cardinali, M.; Fiorucci, F.; Marchesini, I.; Mondini, A. C.; Santangelo, M.; Ghosh, S.; Riguer, D. E. L.; Lahousse, T.; Chang, K. T.; Guzzetti, F.

2012-04-01

We have developed a tool in R (the free software environment for statistical computing, http://www.r-project.org/) to estimate the probability density and the frequency density of landslide area. The tool implements parametric and non-parametric approaches to the estimation of the probability density and the frequency density of landslide area, including: (i) Histogram Density Estimation (HDE), (ii) Kernel Density Estimation (KDE), and (iii) Maximum Likelihood Estimation (MLE). The tool is available as a standard Open Geospatial Consortium (OGC) Web Processing Service (WPS), and is accessible through the web using different GIS software clients. We tested the tool to compare Double Pareto and Inverse Gamma models for the probability density of landslide area in different geological, morphological and climatological settings, and to compare landslides shown in inventory maps prepared using different mapping techniques, including (i) field mapping, (ii) visual interpretation of monoscopic and stereoscopic aerial photographs, (iii) visual interpretation of monoscopic and stereoscopic VHR satellite images and (iv) semi-automatic detection and mapping from VHR satellite images. Results show that both models are applicable in different geomorphological settings. In most cases the two models provided very similar results. Non-parametric estimation methods (i.e., HDE and KDE) provided reasonable results for all the tested landslide datasets. For some of the datasets, MLE failed to provide a result, for convergence problems. The two tested models (Double Pareto and Inverse Gamma) resulted in very similar results for large and very large datasets (> 150 samples). Differences in the modeling results were observed for small datasets affected by systematic biases. A distinct rollover was observed in all analyzed landslide datasets, except for a few datasets obtained from landslide inventories prepared through field mapping or by semi-automatic mapping from VHR satellite imagery. The tool can also be used to evaluate the probability density and the frequency density of landslide volume.

Learning-based 3T brain MRI segmentation with guidance from 7T MRI labeling.

PubMed

Deng, Minghui; Yu, Renping; Wang, Li; Shi, Feng; Yap, Pew-Thian; Shen, Dinggang

2016-12-01

Segmentation of brain magnetic resonance (MR) images into white matter (WM), gray matter (GM), and cerebrospinal fluid (CSF) is crucial for brain structural measurement and disease diagnosis. Learning-based segmentation methods depend largely on the availability of good training ground truth. However, the commonly used 3T MR images are of insufficient image quality and often exhibit poor intensity contrast between WM, GM, and CSF. Therefore, they are not ideal for providing good ground truth label data for training learning-based methods. Recent advances in ultrahigh field 7T imaging make it possible to acquire images with excellent intensity contrast and signal-to-noise ratio. In this paper, the authors propose an algorithm based on random forest for segmenting 3T MR images by training a series of classifiers based on reliable labels obtained semiautomatically from 7T MR images. The proposed algorithm iteratively refines the probability maps of WM, GM, and CSF via a cascade of random forest classifiers for improved tissue segmentation. The proposed method was validated on two datasets, i.e., 10 subjects collected at their institution and 797 3T MR images from the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset. Specifically, for the mean Dice ratio of all 10 subjects, the proposed method achieved 94.52% ± 0.9%, 89.49% ± 1.83%, and 79.97% ± 4.32% for WM, GM, and CSF, respectively, which are significantly better than the state-of-the-art methods (p-values < 0.021). For the ADNI dataset, the group difference comparisons indicate that the proposed algorithm outperforms state-of-the-art segmentation methods. The authors have developed and validated a novel fully automated method for 3T brain MR image segmentation. © 2016 American Association of Physicists in Medicine.

Learning-based 3T brain MRI segmentation with guidance from 7T MRI labeling.

PubMed

Deng, Minghui; Yu, Renping; Wang, Li; Shi, Feng; Yap, Pew-Thian; Shen, Dinggang

2016-12-01

Segmentation of brain magnetic resonance (MR) images into white matter (WM), gray matter (GM), and cerebrospinal fluid (CSF) is crucial for brain structural measurement and disease diagnosis. Learning-based segmentation methods depend largely on the availability of good training ground truth. However, the commonly used 3T MR images are of insufficient image quality and often exhibit poor intensity contrast between WM, GM, and CSF. Therefore, they are not ideal for providing good ground truth label data for training learning-based methods. Recent advances in ultrahigh field 7T imaging make it possible to acquire images with excellent intensity contrast and signal-to-noise ratio. In this paper, the authors propose an algorithm based on random forest for segmenting 3T MR images by training a series of classifiers based on reliable labels obtained semiautomatically from 7T MR images. The proposed algorithm iteratively refines the probability maps of WM, GM, and CSF via a cascade of random forest classifiers for improved tissue segmentation. The proposed method was validated on two datasets, i.e., 10 subjects collected at their institution and 797 3T MR images from the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset. Specifically, for the mean Dice ratio of all 10 subjects, the proposed method achieved 94.52% ± 0.9%, 89.49% ± 1.83%, and 79.97% ± 4.32% for WM, GM, and CSF, respectively, which are significantly better than the state-of-the-art methods (p-values < 0.021). For the ADNI dataset, the group difference comparisons indicate that the proposed algorithm outperforms state-of-the-art segmentation methods. The authors have developed and validated a novel fully automated method for 3T brain MR image segmentation.

The 3D Reference Earth Model: Status and Preliminary Results

NASA Astrophysics Data System (ADS)

Moulik, P.; Lekic, V.; Romanowicz, B. A.

2017-12-01

In the 20th century, seismologists constructed models of how average physical properties (e.g. density, rigidity, compressibility, anisotropy) vary with depth in the Earth's interior. These one-dimensional (1D) reference Earth models (e.g. PREM) have proven indispensable in earthquake location, imaging of interior structure, understanding material properties under extreme conditions, and as a reference in other fields, such as particle physics and astronomy. Over the past three decades, new datasets motivated more sophisticated efforts that yielded models of how properties vary both laterally and with depth in the Earth's interior. Though these three-dimensional (3D) models exhibit compelling similarities at large scales, differences in the methodology, representation of structure, and dataset upon which they are based, have prevented the creation of 3D community reference models. As part of the REM-3D project, we are compiling and reconciling reference seismic datasets of body wave travel-time measurements, fundamental mode and overtone surface wave dispersion measurements, and normal mode frequencies and splitting functions. These reference datasets are being inverted for a long-wavelength, 3D reference Earth model that describes the robust long-wavelength features of mantle heterogeneity. As a community reference model with fully quantified uncertainties and tradeoffs and an associated publically available dataset, REM-3D will facilitate Earth imaging studies, earthquake characterization, inferences on temperature and composition in the deep interior, and be of improved utility to emerging scientific endeavors, such as neutrino geoscience. Here, we summarize progress made in the construction of the reference long period dataset and present a preliminary version of REM-3D in the upper-mantle. In order to determine the level of detail warranted for inclusion in REM-3D, we analyze the spectrum of discrepancies between models inverted with different subsets of the reference dataset. This procedure allows us to evaluate the extent of consistency in imaging heterogeneity at various depths and between spatial scales.

Fast Steerable Principal Component Analysis

PubMed Central

Zhao, Zhizhen; Shkolnisky, Yoel; Singer, Amit

2016-01-01

Cryo-electron microscopy nowadays often requires the analysis of hundreds of thousands of 2-D images as large as a few hundred pixels in each direction. Here, we introduce an algorithm that efficiently and accurately performs principal component analysis (PCA) for a large set of 2-D images, and, for each image, the set of its uniform rotations in the plane and their reflections. For a dataset consisting of n images of size L × L pixels, the computational complexity of our algorithm is O(nL3 + L4), while existing algorithms take O(nL4). The new algorithm computes the expansion coefficients of the images in a Fourier–Bessel basis efficiently using the nonuniform fast Fourier transform. We compare the accuracy and efficiency of the new algorithm with traditional PCA and existing algorithms for steerable PCA. PMID:27570801

Simultaneous maximum a posteriori longitudinal PET image reconstruction

NASA Astrophysics Data System (ADS)

Ellis, Sam; Reader, Andrew J.

2017-09-01

Positron emission tomography (PET) is frequently used to monitor functional changes that occur over extended time scales, for example in longitudinal oncology PET protocols that include routine clinical follow-up scans to assess the efficacy of a course of treatment. In these contexts PET datasets are currently reconstructed into images using single-dataset reconstruction methods. Inspired by recently proposed joint PET-MR reconstruction methods, we propose to reconstruct longitudinal datasets simultaneously by using a joint penalty term in order to exploit the high degree of similarity between longitudinal images. We achieved this by penalising voxel-wise differences between pairs of longitudinal PET images in a one-step-late maximum a posteriori (MAP) fashion, resulting in the MAP simultaneous longitudinal reconstruction (SLR) method. The proposed method reduced reconstruction errors and visually improved images relative to standard maximum likelihood expectation-maximisation (ML-EM) in simulated 2D longitudinal brain tumour scans. In reconstructions of split real 3D data with inserted simulated tumours, noise across images reconstructed with MAP-SLR was reduced to levels equivalent to doubling the number of detected counts when using ML-EM. Furthermore, quantification of tumour activities was largely preserved over a variety of longitudinal tumour changes, including changes in size and activity, with larger changes inducing larger biases relative to standard ML-EM reconstructions. Similar improvements were observed for a range of counts levels, demonstrating the robustness of the method when used with a single penalty strength. The results suggest that longitudinal regularisation is a simple but effective method of improving reconstructed PET images without using resolution degrading priors.

Registration uncertainties between 3D cone beam computed tomography and different reference CT datasets in lung stereotactic body radiation therapy.

PubMed

Oechsner, Markus; Chizzali, Barbara; Devecka, Michal; Combs, Stephanie Elisabeth; Wilkens, Jan Jakob; Duma, Marciana Nona

2016-10-26

The aim of this study was to analyze differences in couch shifts (setup errors) resulting from image registration of different CT datasets with free breathing cone beam CTs (FB-CBCT). As well automatic as manual image registrations were performed and registration results were correlated to tumor characteristics. FB-CBCT image registration was performed for 49 patients with lung lesions using slow planning CT (PCT), average intensity projection (AIP), maximum intensity projection (MIP) and mid-ventilation CTs (MidV) as reference images. Both, automatic and manual image registrations were applied. Shift differences were evaluated between the registered CT datasets for automatic and manual registration, respectively. Furthermore, differences between automatic and manual registration were analyzed for the same CT datasets. The registration results were statistically analyzed and correlated to tumor characteristics (3D tumor motion, tumor volume, superior-inferior (SI) distance, tumor environment). Median 3D shift differences over all patients were between 0.5 mm (AIPvsMIP) and 1.9 mm (MIPvsPCT and MidVvsPCT) for the automatic registration and between 1.8 mm (AIPvsPCT) and 2.8 mm (MIPvsPCT and MidVvsPCT) for the manual registration. For some patients, large shift differences (>5.0 mm) were found (maximum 10.5 mm, automatic registration). Comparing automatic vs manual registrations for the same reference CTs, ∆AIP achieved the smallest (1.1 mm) and ∆MIP the largest (1.9 mm) median 3D shift differences. The standard deviation (variability) for the 3D shift differences was also the smallest for ∆AIP (1.1 mm). Significant correlations (p < 0.01) between 3D shift difference and 3D tumor motion (AIPvsMIP, MIPvsMidV) and SI distance (AIPvsMIP) (automatic) and also for 3D tumor motion (∆PCT, ∆MidV; automatic vs manual) were found. Using different CT datasets for image registration with FB-CBCTs can result in different 3D couch shifts. Manual registrations achieved partly different 3D shifts than automatic registrations. AIP CTs yielded the smallest shift differences and might be the most appropriate CT dataset for registration with 3D FB-CBCTs.

Scalable ranked retrieval using document images

NASA Astrophysics Data System (ADS)

Jain, Rajiv; Oard, Douglas W.; Doermann, David

2013-12-01

Despite the explosion of text on the Internet, hard copy documents that have been scanned as images still play a significant role for some tasks. The best method to perform ranked retrieval on a large corpus of document images, however, remains an open research question. The most common approach has been to perform text retrieval using terms generated by optical character recognition. This paper, by contrast, examines whether a scalable segmentation-free image retrieval algorithm, which matches sub-images containing text or graphical objects, can provide additional benefit in satisfying a user's information needs on a large, real world dataset. Results on 7 million scanned pages from the CDIP v1.0 test collection show that content based image retrieval finds a substantial number of documents that text retrieval misses, and that when used as a basis for relevance feedback can yield improvements in retrieval effectiveness.

A Benchmark Dataset and Saliency-guided Stacked Autoencoders for Video-based Salient Object Detection.

PubMed

Li, Jia; Xia, Changqun; Chen, Xiaowu

2017-10-12

Image-based salient object detection (SOD) has been extensively studied in past decades. However, video-based SOD is much less explored due to the lack of large-scale video datasets within which salient objects are unambiguously defined and annotated. Toward this end, this paper proposes a video-based SOD dataset that consists of 200 videos. In constructing the dataset, we manually annotate all objects and regions over 7,650 uniformly sampled keyframes and collect the eye-tracking data of 23 subjects who free-view all videos. From the user data, we find that salient objects in a video can be defined as objects that consistently pop-out throughout the video, and objects with such attributes can be unambiguously annotated by combining manually annotated object/region masks with eye-tracking data of multiple subjects. To the best of our knowledge, it is currently the largest dataset for videobased salient object detection. Based on this dataset, this paper proposes an unsupervised baseline approach for video-based SOD by using saliencyguided stacked autoencoders. In the proposed approach, multiple spatiotemporal saliency cues are first extracted at the pixel, superpixel and object levels. With these saliency cues, stacked autoencoders are constructed in an unsupervised manner that automatically infers a saliency score for each pixel by progressively encoding the high-dimensional saliency cues gathered from the pixel and its spatiotemporal neighbors. In experiments, the proposed unsupervised approach is compared with 31 state-of-the-art models on the proposed dataset and outperforms 30 of them, including 19 imagebased classic (unsupervised or non-deep learning) models, six image-based deep learning models, and five video-based unsupervised models. Moreover, benchmarking results show that the proposed dataset is very challenging and has the potential to boost the development of video-based SOD.

Image fusion using sparse overcomplete feature dictionaries

DOEpatents

Brumby, Steven P.; Bettencourt, Luis; Kenyon, Garrett T.; Chartrand, Rick; Wohlberg, Brendt

2015-10-06

Approaches for deciding what individuals in a population of visual system "neurons" are looking for using sparse overcomplete feature dictionaries are provided. A sparse overcomplete feature dictionary may be learned for an image dataset and a local sparse representation of the image dataset may be built using the learned feature dictionary. A local maximum pooling operation may be applied on the local sparse representation to produce a translation-tolerant representation of the image dataset. An object may then be classified and/or clustered within the translation-tolerant representation of the image dataset using a supervised classification algorithm and/or an unsupervised clustering algorithm.

Toyz: A framework for scientific analysis of large datasets and astronomical images

NASA Astrophysics Data System (ADS)

Moolekamp, F.; Mamajek, E.

2015-11-01

As the size of images and data products derived from astronomical data continues to increase, new tools are needed to visualize and interact with that data in a meaningful way. Motivated by our own astronomical images taken with the Dark Energy Camera (DECam) we present Toyz, an open source Python package for viewing and analyzing images and data stored on a remote server or cluster. Users connect to the Toyz web application via a web browser, making it ​a convenient tool for students to visualize and interact with astronomical data without having to install any software on their local machines. In addition it provides researchers with an easy-to-use tool that allows them to browse the files on a server and quickly view very large images (>2 Gb) taken with DECam and other cameras with a large FOV and create their own visualization tools that can be added on as extensions to the default Toyz framework.

Automated assessment of thigh composition using machine learning for Dixon magnetic resonance images.

PubMed

Yang, Yu Xin; Chong, Mei Sian; Tay, Laura; Yew, Suzanne; Yeo, Audrey; Tan, Cher Heng

2016-10-01

To develop and validate a machine learning based automated segmentation method that jointly analyzes the four contrasts provided by Dixon MRI technique for improved thigh composition segmentation accuracy. The automatic detection of body composition is formulized as a three-class classification issue. Each image voxel in the training dataset is assigned with a correct label. A voxel classifier is trained and subsequently used to predict unseen data. Morphological operations are finally applied to generate volumetric segmented images for different structures. We applied this algorithm on datasets of (1) four contrast images, (2) water and fat images, and (3) unsuppressed images acquired from 190 subjects. The proposed method using four contrasts achieved most accurate and robust segmentation compared to the use of combined fat and water images and the use of unsuppressed image, average Dice coefficients of 0.94 ± 0.03, 0.96 ± 0.03, 0.80 ± 0.03, and 0.97 ± 0.01 has been achieved to bone region, subcutaneous adipose tissue (SAT), inter-muscular adipose tissue (IMAT), and muscle respectively. Our proposed method based on machine learning produces accurate tissue quantification and showed an effective use of large information provided by the four contrast images from Dixon MRI.

A large-scale dataset of solar event reports from automated feature recognition modules

NASA Astrophysics Data System (ADS)

Schuh, Michael A.; Angryk, Rafal A.; Martens, Petrus C.

2016-05-01

The massive repository of images of the Sun captured by the Solar Dynamics Observatory (SDO) mission has ushered in the era of Big Data for Solar Physics. In this work, we investigate the entire public collection of events reported to the Heliophysics Event Knowledgebase (HEK) from automated solar feature recognition modules operated by the SDO Feature Finding Team (FFT). With the SDO mission recently surpassing five years of operations, and over 280,000 event reports for seven types of solar phenomena, we present the broadest and most comprehensive large-scale dataset of the SDO FFT modules to date. We also present numerous statistics on these modules, providing valuable contextual information for better understanding and validating of the individual event reports and the entire dataset as a whole. After extensive data cleaning through exploratory data analysis, we highlight several opportunities for knowledge discovery from data (KDD). Through these important prerequisite analyses presented here, the results of KDD from Solar Big Data will be overall more reliable and better understood. As the SDO mission remains operational over the coming years, these datasets will continue to grow in size and value. Future versions of this dataset will be analyzed in the general framework established in this work and maintained publicly online for easy access by the community.

Data-Oriented Astrophysics at NOAO: The Science Archive & The Data Lab

NASA Astrophysics Data System (ADS)

Juneau, Stephanie; NOAO Data Lab, NOAO Science Archive

2018-06-01

As we keep progressing into an era of increasingly large astronomy datasets, NOAO’s data-oriented mission is growing in prominence. The NOAO Science Archive, which captures and processes the pixel data from mountaintops in Chile and Arizona, now contains holdings at Petabyte scales. Working at the intersection of astronomy and data science, the main goal of the NOAO Data Lab is to provide users with a suite of tools to work close to this data, the catalogs derived from them, as well as externally provided datasets, and thus optimize the scientific productivity of the astronomy community. These tools and services include databases, query tools, virtual storage space, workflows through our Jupyter Notebook server, and scripted analysis. We currently host datasets from NOAO facilities such as the Dark Energy Survey (DES), the DESI imaging Legacy Surveys (LS), the Dark Energy Camera Plane Survey (DECaPS), and the nearly all-sky NOAO Source Catalog (NSC). We are further preparing for large spectroscopy datasets such as DESI. After a brief overview of the Science Archive, the Data Lab and datasets, I will briefly showcase scientific applications showing use of our data holdings. Lastly, I will describe our vision for future developments as we tackle the next technical and scientific challenges.

Mars Global Geologic Mapping: Amazonian Results

NASA Technical Reports Server (NTRS)

Tanaka, K. L.; Dohm, J. M.; Irwin, R.; Kolb, E. J.; Skinner, J. A., Jr.; Hare, T. M.

2008-01-01

We are in the second year of a five-year effort to map the geology of Mars using mainly Mars Global Surveyor, Mars Express, and Mars Odyssey imaging and altimetry datasets. Previously, we have reported on details of project management, mapping datasets (local and regional), initial and anticipated mapping approaches, and tactics of map unit delineation and description [1-2]. For example, we have seen how the multiple types and huge quantity of image data as well as more accurate and detailed altimetry data now available allow for broader and deeper geologic perspectives, based largely on improved landform perception, characterization, and analysis. Here, we describe early mapping results, which include updating of previous northern plains mapping [3], including delineation of mainly Amazonian units and regional fault mapping, as well as other advances.

imzML: Imaging Mass Spectrometry Markup Language: A common data format for mass spectrometry imaging.

PubMed

Römpp, Andreas; Schramm, Thorsten; Hester, Alfons; Klinkert, Ivo; Both, Jean-Pierre; Heeren, Ron M A; Stöckli, Markus; Spengler, Bernhard

2011-01-01

Imaging mass spectrometry is the method of scanning a sample of interest and generating an "image" of the intensity distribution of a specific analyte. The data sets consist of a large number of mass spectra which are usually acquired with identical settings. Existing data formats are not sufficient to describe an MS imaging experiment completely. The data format imzML was developed to allow the flexible and efficient exchange of MS imaging data between different instruments and data analysis software.For this purpose, the MS imaging data is divided in two separate files. The mass spectral data is stored in a binary file to ensure efficient storage. All metadata (e.g., instrumental parameters, sample details) are stored in an XML file which is based on the standard data format mzML developed by HUPO-PSI. The original mzML controlled vocabulary was extended to include specific parameters of imaging mass spectrometry (such as x/y position and spatial resolution). The two files (XML and binary) are connected by offset values in the XML file and are unambiguously linked by a universally unique identifier. The resulting datasets are comparable in size to the raw data and the separate metadata file allows flexible handling of large datasets.Several imaging MS software tools already support imzML. This allows choosing from a (growing) number of processing tools. One is no longer limited to proprietary software, but is able to use the processing software which is best suited for a specific question or application. On the other hand, measurements from different instruments can be compared within one software application using identical settings for data processing. All necessary information for evaluating and implementing imzML can be found at http://www.imzML.org .

The PREP pipeline: standardized preprocessing for large-scale EEG analysis

PubMed Central

Bigdely-Shamlo, Nima; Mullen, Tim; Kothe, Christian; Su, Kyung-Min; Robbins, Kay A.

2015-01-01

The technology to collect brain imaging and physiological measures has become portable and ubiquitous, opening the possibility of large-scale analysis of real-world human imaging. By its nature, such data is large and complex, making automated processing essential. This paper shows how lack of attention to the very early stages of an EEG preprocessing pipeline can reduce the signal-to-noise ratio and introduce unwanted artifacts into the data, particularly for computations done in single precision. We demonstrate that ordinary average referencing improves the signal-to-noise ratio, but that noisy channels can contaminate the results. We also show that identification of noisy channels depends on the reference and examine the complex interaction of filtering, noisy channel identification, and referencing. We introduce a multi-stage robust referencing scheme to deal with the noisy channel-reference interaction. We propose a standardized early-stage EEG processing pipeline (PREP) and discuss the application of the pipeline to more than 600 EEG datasets. The pipeline includes an automatically generated report for each dataset processed. Users can download the PREP pipeline as a freely available MATLAB library from http://eegstudy.org/prepcode. PMID:26150785

Identifying spatially similar gene expression patterns in early stage fruit fly embryo images: binary feature versus invariant moment digital representations

PubMed Central

Gurunathan, Rajalakshmi; Van Emden, Bernard; Panchanathan, Sethuraman; Kumar, Sudhir

2004-01-01

Background Modern developmental biology relies heavily on the analysis of embryonic gene expression patterns. Investigators manually inspect hundreds or thousands of expression patterns to identify those that are spatially similar and to ultimately infer potential gene interactions. However, the rapid accumulation of gene expression pattern data over the last two decades, facilitated by high-throughput techniques, has produced a need for the development of efficient approaches for direct comparison of images, rather than their textual descriptions, to identify spatially similar expression patterns. Results The effectiveness of the Binary Feature Vector (BFV) and Invariant Moment Vector (IMV) based digital representations of the gene expression patterns in finding biologically meaningful patterns was compared for a small (226 images) and a large (1819 images) dataset. For each dataset, an ordered list of images, with respect to a query image, was generated to identify overlapping and similar gene expression patterns, in a manner comparable to what a developmental biologist might do. The results showed that the BFV representation consistently outperforms the IMV representation in finding biologically meaningful matches when spatial overlap of the gene expression pattern and the genes involved are considered. Furthermore, we explored the value of conducting image-content based searches in a dataset where individual expression components (or domains) of multi-domain expression patterns were also included separately. We found that this technique improves performance of both IMV and BFV based searches. Conclusions We conclude that the BFV representation consistently produces a more extensive and better list of biologically useful patterns than the IMV representation. The high quality of results obtained scales well as the search database becomes larger, which encourages efforts to build automated image query and retrieval systems for spatial gene expression patterns. PMID:15603586

A Parallel Adaboost-Backpropagation Neural Network for Massive Image Dataset Classification

NASA Astrophysics Data System (ADS)

Cao, Jianfang; Chen, Lichao; Wang, Min; Shi, Hao; Tian, Yun

2016-12-01

Image classification uses computers to simulate human understanding and cognition of images by automatically categorizing images. This study proposes a faster image classification approach that parallelizes the traditional Adaboost-Backpropagation (BP) neural network using the MapReduce parallel programming model. First, we construct a strong classifier by assembling the outputs of 15 BP neural networks (which are individually regarded as weak classifiers) based on the Adaboost algorithm. Second, we design Map and Reduce tasks for both the parallel Adaboost-BP neural network and the feature extraction algorithm. Finally, we establish an automated classification model by building a Hadoop cluster. We use the Pascal VOC2007 and Caltech256 datasets to train and test the classification model. The results are superior to those obtained using traditional Adaboost-BP neural network or parallel BP neural network approaches. Our approach increased the average classification accuracy rate by approximately 14.5% and 26.0% compared to the traditional Adaboost-BP neural network and parallel BP neural network, respectively. Furthermore, the proposed approach requires less computation time and scales very well as evaluated by speedup, sizeup and scaleup. The proposed approach may provide a foundation for automated large-scale image classification and demonstrates practical value.

A Parallel Adaboost-Backpropagation Neural Network for Massive Image Dataset Classification.

PubMed

Cao, Jianfang; Chen, Lichao; Wang, Min; Shi, Hao; Tian, Yun

2016-12-01

Image classification uses computers to simulate human understanding and cognition of images by automatically categorizing images. This study proposes a faster image classification approach that parallelizes the traditional Adaboost-Backpropagation (BP) neural network using the MapReduce parallel programming model. First, we construct a strong classifier by assembling the outputs of 15 BP neural networks (which are individually regarded as weak classifiers) based on the Adaboost algorithm. Second, we design Map and Reduce tasks for both the parallel Adaboost-BP neural network and the feature extraction algorithm. Finally, we establish an automated classification model by building a Hadoop cluster. We use the Pascal VOC2007 and Caltech256 datasets to train and test the classification model. The results are superior to those obtained using traditional Adaboost-BP neural network or parallel BP neural network approaches. Our approach increased the average classification accuracy rate by approximately 14.5% and 26.0% compared to the traditional Adaboost-BP neural network and parallel BP neural network, respectively. Furthermore, the proposed approach requires less computation time and scales very well as evaluated by speedup, sizeup and scaleup. The proposed approach may provide a foundation for automated large-scale image classification and demonstrates practical value.

A Parallel Adaboost-Backpropagation Neural Network for Massive Image Dataset Classification

PubMed Central

Cao, Jianfang; Chen, Lichao; Wang, Min; Shi, Hao; Tian, Yun

2016-01-01

Image classification uses computers to simulate human understanding and cognition of images by automatically categorizing images. This study proposes a faster image classification approach that parallelizes the traditional Adaboost-Backpropagation (BP) neural network using the MapReduce parallel programming model. First, we construct a strong classifier by assembling the outputs of 15 BP neural networks (which are individually regarded as weak classifiers) based on the Adaboost algorithm. Second, we design Map and Reduce tasks for both the parallel Adaboost-BP neural network and the feature extraction algorithm. Finally, we establish an automated classification model by building a Hadoop cluster. We use the Pascal VOC2007 and Caltech256 datasets to train and test the classification model. The results are superior to those obtained using traditional Adaboost-BP neural network or parallel BP neural network approaches. Our approach increased the average classification accuracy rate by approximately 14.5% and 26.0% compared to the traditional Adaboost-BP neural network and parallel BP neural network, respectively. Furthermore, the proposed approach requires less computation time and scales very well as evaluated by speedup, sizeup and scaleup. The proposed approach may provide a foundation for automated large-scale image classification and demonstrates practical value. PMID:27905520

MRIQC: Advancing the automatic prediction of image quality in MRI from unseen sites

PubMed Central

2017-01-01

Quality control of MRI is essential for excluding problematic acquisitions and avoiding bias in subsequent image processing and analysis. Visual inspection is subjective and impractical for large scale datasets. Although automated quality assessments have been demonstrated on single-site datasets, it is unclear that solutions can generalize to unseen data acquired at new sites. Here, we introduce the MRI Quality Control tool (MRIQC), a tool for extracting quality measures and fitting a binary (accept/exclude) classifier. Our tool can be run both locally and as a free online service via the OpenNeuro.org portal. The classifier is trained on a publicly available, multi-site dataset (17 sites, N = 1102). We perform model selection evaluating different normalization and feature exclusion approaches aimed at maximizing across-site generalization and estimate an accuracy of 76%±13% on new sites, using leave-one-site-out cross-validation. We confirm that result on a held-out dataset (2 sites, N = 265) also obtaining a 76% accuracy. Even though the performance of the trained classifier is statistically above chance, we show that it is susceptible to site effects and unable to account for artifacts specific to new sites. MRIQC performs with high accuracy in intra-site prediction, but performance on unseen sites leaves space for improvement which might require more labeled data and new approaches to the between-site variability. Overcoming these limitations is crucial for a more objective quality assessment of neuroimaging data, and to enable the analysis of extremely large and multi-site samples. PMID:28945803

SU-C-207B-04: Automated Segmentation of Pectoral Muscle in MR Images of Dense Breasts

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

Verburg, E; Waard, SN de; Veldhuis, WB

Purpose: To develop and evaluate a fully automated method for segmentation of the pectoral muscle boundary in Magnetic Resonance Imaging (MRI) of dense breasts. Methods: Segmentation of the pectoral muscle is an important part of automatic breast image analysis methods. Current methods for segmenting the pectoral muscle in breast MRI have difficulties delineating the muscle border correctly in breasts with a large proportion of fibroglandular tissue (i.e., dense breasts). Hence, an automated method based on dynamic programming was developed, incorporating heuristics aimed at shape, location and gradient features.To assess the method, the pectoral muscle was segmented in 91 randomly selectedmore » participants (mean age 56.6 years, range 49.5–75.2 years) from a large MRI screening trial in women with dense breasts (ACR BI-RADS category 4). Each MR dataset consisted of 178 or 179 T1-weighted images with voxel size 0.64 × 0.64 × 1.00 mm3. All images (n=16,287) were reviewed and scored by a radiologist. In contrast to volume overlap coefficients, such as DICE, the radiologist detected deviations in the segmented muscle border and determined whether the result would impact the ability to accurately determine the volume of fibroglandular tissue and detection of breast lesions. Results: According to the radiologist’s scores, 95.5% of the slices did not mask breast tissue in such way that it could affect detection of breast lesions or volume measurements. In 13.1% of the slices a deviation in the segmented muscle border was present which would not impact breast lesion detection. In 70 datasets (78%) at least 95% of the slices were segmented in such a way it would not affect detection of breast lesions, and in 60 (66%) datasets this was 100%. Conclusion: Dynamic programming with dedicated heuristics shows promising potential to segment the pectoral muscle in women with dense breasts.« less

Effective and efficient analysis of spatio-temporal data

NASA Astrophysics Data System (ADS)

Zhang, Zhongnan

Spatio-temporal data mining, i.e., mining knowledge from large amount of spatio-temporal data, is a highly demanding field because huge amounts of spatio-temporal data have been collected in various applications, ranging from remote sensing, to geographical information systems (GIS), computer cartography, environmental assessment and planning, etc. The collection data far exceeded human's ability to analyze which make it crucial to develop analysis tools. Recent studies on data mining have extended to the scope of data mining from relational and transactional datasets to spatial and temporal datasets. Among the various forms of spatio-temporal data, remote sensing images play an important role, due to the growing wide-spreading of outer space satellites. In this dissertation, we proposed two approaches to analyze the remote sensing data. The first one is about applying association rules mining onto images processing. Each image was divided into a number of image blocks. We built a spatial relationship for these blocks during the dividing process. This made a large number of images into a spatio-temporal dataset since each image was shot in time-series. The second one implemented co-occurrence patterns discovery from these images. The generated patterns represent subsets of spatial features that are located together in space and time. A weather analysis is composed of individual analysis of several meteorological variables. These variables include temperature, pressure, dew point, wind, clouds, visibility and so on. Local-scale models provide detailed analysis and forecasts of meteorological phenomena ranging from a few kilometers to about 100 kilometers in size. When some of above meteorological variables have some special change tendency, some kind of severe weather will happen in most cases. Using the discovery of association rules, we found that some special meteorological variables' changing has tight relation with some severe weather situation that will happen very soon. This dissertation is composed of three parts: an introduction, some basic knowledges and relative works, and my own three contributions to the development of approaches for spatio-temporal data mining: DYSTAL algorithm, STARSI algorithm, and COSTCOP+ algorithm.

Converting Static Image Datasets to Spiking Neuromorphic Datasets Using Saccades.

PubMed

Orchard, Garrick; Jayawant, Ajinkya; Cohen, Gregory K; Thakor, Nitish

2015-01-01

Creating datasets for Neuromorphic Vision is a challenging task. A lack of available recordings from Neuromorphic Vision sensors means that data must typically be recorded specifically for dataset creation rather than collecting and labeling existing data. The task is further complicated by a desire to simultaneously provide traditional frame-based recordings to allow for direct comparison with traditional Computer Vision algorithms. Here we propose a method for converting existing Computer Vision static image datasets into Neuromorphic Vision datasets using an actuated pan-tilt camera platform. Moving the sensor rather than the scene or image is a more biologically realistic approach to sensing and eliminates timing artifacts introduced by monitor updates when simulating motion on a computer monitor. We present conversion of two popular image datasets (MNIST and Caltech101) which have played important roles in the development of Computer Vision, and we provide performance metrics on these datasets using spike-based recognition algorithms. This work contributes datasets for future use in the field, as well as results from spike-based algorithms against which future works can compare. Furthermore, by converting datasets already popular in Computer Vision, we enable more direct comparison with frame-based approaches.

Supervised graph hashing for histopathology image retrieval and classification.

PubMed

Shi, Xiaoshuang; Xing, Fuyong; Xu, KaiDi; Xie, Yuanpu; Su, Hai; Yang, Lin

2017-12-01

In pathology image analysis, morphological characteristics of cells are critical to grade many diseases. With the development of cell detection and segmentation techniques, it is possible to extract cell-level information for further analysis in pathology images. However, it is challenging to conduct efficient analysis of cell-level information on a large-scale image dataset because each image usually contains hundreds or thousands of cells. In this paper, we propose a novel image retrieval based framework for large-scale pathology image analysis. For each image, we encode each cell into binary codes to generate image representation using a novel graph based hashing model and then conduct image retrieval by applying a group-to-group matching method to similarity measurement. In order to improve both computational efficiency and memory requirement, we further introduce matrix factorization into the hashing model for scalable image retrieval. The proposed framework is extensively validated with thousands of lung cancer images, and it achieves 97.98% classification accuracy and 97.50% retrieval precision with all cells of each query image used. Copyright © 2017 Elsevier B.V. All rights reserved.

Rapid Global Fitting of Large Fluorescence Lifetime Imaging Microscopy Datasets

PubMed Central

Warren, Sean C.; Margineanu, Anca; Alibhai, Dominic; Kelly, Douglas J.; Talbot, Clifford; Alexandrov, Yuriy; Munro, Ian; Katan, Matilda

2013-01-01

Fluorescence lifetime imaging (FLIM) is widely applied to obtain quantitative information from fluorescence signals, particularly using Förster Resonant Energy Transfer (FRET) measurements to map, for example, protein-protein interactions. Extracting FRET efficiencies or population fractions typically entails fitting data to complex fluorescence decay models but such experiments are frequently photon constrained, particularly for live cell or in vivo imaging, and this leads to unacceptable errors when analysing data on a pixel-wise basis. Lifetimes and population fractions may, however, be more robustly extracted using global analysis to simultaneously fit the fluorescence decay data of all pixels in an image or dataset to a multi-exponential model under the assumption that the lifetime components are invariant across the image (dataset). This approach is often considered to be prohibitively slow and/or computationally expensive but we present here a computationally efficient global analysis algorithm for the analysis of time-correlated single photon counting (TCSPC) or time-gated FLIM data based on variable projection. It makes efficient use of both computer processor and memory resources, requiring less than a minute to analyse time series and multiwell plate datasets with hundreds of FLIM images on standard personal computers. This lifetime analysis takes account of repetitive excitation, including fluorescence photons excited by earlier pulses contributing to the fit, and is able to accommodate time-varying backgrounds and instrument response functions. We demonstrate that this global approach allows us to readily fit time-resolved fluorescence data to complex models including a four-exponential model of a FRET system, for which the FRET efficiencies of the two species of a bi-exponential donor are linked, and polarisation-resolved lifetime data, where a fluorescence intensity and bi-exponential anisotropy decay model is applied to the analysis of live cell homo-FRET data. A software package implementing this algorithm, FLIMfit, is available under an open source licence through the Open Microscopy Environment. PMID:23940626

Assessing microscope image focus quality with deep learning.

PubMed

Yang, Samuel J; Berndl, Marc; Michael Ando, D; Barch, Mariya; Narayanaswamy, Arunachalam; Christiansen, Eric; Hoyer, Stephan; Roat, Chris; Hung, Jane; Rueden, Curtis T; Shankar, Asim; Finkbeiner, Steven; Nelson, Philip

2018-03-15

Large image datasets acquired on automated microscopes typically have some fraction of low quality, out-of-focus images, despite the use of hardware autofocus systems. Identification of these images using automated image analysis with high accuracy is important for obtaining a clean, unbiased image dataset. Complicating this task is the fact that image focus quality is only well-defined in foreground regions of images, and as a result, most previous approaches only enable a computation of the relative difference in quality between two or more images, rather than an absolute measure of quality. We present a deep neural network model capable of predicting an absolute measure of image focus on a single image in isolation, without any user-specified parameters. The model operates at the image-patch level, and also outputs a measure of prediction certainty, enabling interpretable predictions. The model was trained on only 384 in-focus Hoechst (nuclei) stain images of U2OS cells, which were synthetically defocused to one of 11 absolute defocus levels during training. The trained model can generalize on previously unseen real Hoechst stain images, identifying the absolute image focus to within one defocus level (approximately 3 pixel blur diameter difference) with 95% accuracy. On a simpler binary in/out-of-focus classification task, the trained model outperforms previous approaches on both Hoechst and Phalloidin (actin) stain images (F-scores of 0.89 and 0.86, respectively over 0.84 and 0.83), despite only having been presented Hoechst stain images during training. Lastly, we observe qualitatively that the model generalizes to two additional stains, Hoechst and Tubulin, of an unseen cell type (Human MCF-7) acquired on a different instrument. Our deep neural network enables classification of out-of-focus microscope images with both higher accuracy and greater precision than previous approaches via interpretable patch-level focus and certainty predictions. The use of synthetically defocused images precludes the need for a manually annotated training dataset. The model also generalizes to different image and cell types. The framework for model training and image prediction is available as a free software library and the pre-trained model is available for immediate use in Fiji (ImageJ) and CellProfiler.

A method for normalizing pathology images to improve feature extraction for quantitative pathology

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

Tam, Allison; Barker, Jocelyn; Rubin, Daniel

Purpose: With the advent of digital slide scanning technologies and the potential proliferation of large repositories of digital pathology images, many research studies can leverage these data for biomedical discovery and to develop clinical applications. However, quantitative analysis of digital pathology images is impeded by batch effects generated by varied staining protocols and staining conditions of pathological slides. Methods: To overcome this problem, this paper proposes a novel, fully automated stain normalization method to reduce batch effects and thus aid research in digital pathology applications. Their method, intensity centering and histogram equalization (ICHE), normalizes a diverse set of pathology imagesmore » by first scaling the centroids of the intensity histograms to a common point and then applying a modified version of contrast-limited adaptive histogram equalization. Normalization was performed on two datasets of digitized hematoxylin and eosin (H&E) slides of different tissue slices from the same lung tumor, and one immunohistochemistry dataset of digitized slides created by restaining one of the H&E datasets. Results: The ICHE method was evaluated based on image intensity values, quantitative features, and the effect on downstream applications, such as a computer aided diagnosis. For comparison, three methods from the literature were reimplemented and evaluated using the same criteria. The authors found that ICHE not only improved performance compared with un-normalized images, but in most cases showed improvement compared with previous methods for correcting batch effects in the literature. Conclusions: ICHE may be a useful preprocessing step a digital pathology image processing pipeline.« less

A web-based solution for 3D medical image visualization

NASA Astrophysics Data System (ADS)

Hou, Xiaoshuai; Sun, Jianyong; Zhang, Jianguo

2015-03-01

In this presentation, we present a web-based 3D medical image visualization solution which enables interactive large medical image data processing and visualization over the web platform. To improve the efficiency of our solution, we adopt GPU accelerated techniques to process images on the server side while rapidly transferring images to the HTML5 supported web browser on the client side. Compared to traditional local visualization solution, our solution doesn't require the users to install extra software or download the whole volume dataset from PACS server. By designing this web-based solution, it is feasible for users to access the 3D medical image visualization service wherever the internet is available.

Deep Convolutional Neural Networks for breast cancer screening.

PubMed

Chougrad, Hiba; Zouaki, Hamid; Alheyane, Omar

2018-04-01

Radiologists often have a hard time classifying mammography mass lesions which leads to unnecessary breast biopsies to remove suspicions and this ends up adding exorbitant expenses to an already burdened patient and health care system. In this paper we developed a Computer-aided Diagnosis (CAD) system based on deep Convolutional Neural Networks (CNN) that aims to help the radiologist classify mammography mass lesions. Deep learning usually requires large datasets to train networks of a certain depth from scratch. Transfer learning is an effective method to deal with relatively small datasets as in the case of medical images, although it can be tricky as we can easily start overfitting. In this work, we explore the importance of transfer learning and we experimentally determine the best fine-tuning strategy to adopt when training a CNN model. We were able to successfully fine-tune some of the recent, most powerful CNNs and achieved better results compared to other state-of-the-art methods which classified the same public datasets. For instance we achieved 97.35% accuracy and 0.98 AUC on the DDSM database, 95.50% accuracy and 0.97 AUC on the INbreast database and 96.67% accuracy and 0.96 AUC on the BCDR database. Furthermore, after pre-processing and normalizing all the extracted Regions of Interest (ROIs) from the full mammograms, we merged all the datasets to build one large set of images and used it to fine-tune our CNNs. The CNN model which achieved the best results, a 98.94% accuracy, was used as a baseline to build the Breast Cancer Screening Framework. To evaluate the proposed CAD system and its efficiency to classify new images, we tested it on an independent database (MIAS) and got 98.23% accuracy and 0.99 AUC. The results obtained demonstrate that the proposed framework is performant and can indeed be used to predict if the mass lesions are benign or malignant. Copyright © 2018 Elsevier B.V. All rights reserved.

Application of a deep-learning method to the forecast of daily solar flare occurrence using Convolution Neural Network

NASA Astrophysics Data System (ADS)

Shin, Seulki; Moon, Yong-Jae; Chu, Hyoungseok

2017-08-01

As the application of deep-learning methods has been succeeded in various fields, they have a high potential to be applied to space weather forecasting. Convolutional neural network, one of deep learning methods, is specialized in image recognition. In this study, we apply the AlexNet architecture, which is a winner of Imagenet Large Scale Virtual Recognition Challenge (ILSVRC) 2012, to the forecast of daily solar flare occurrence using the MatConvNet software of MATLAB. Our input images are SOHO/MDI, EIT 195Å, and 304Å from January 1996 to December 2010, and output ones are yes or no of flare occurrence. We select training dataset from Jan 1996 to Dec 2000 and from Jan 2003 to Dec 2008. Testing dataset is chosen from Jan 2001 to Dec 2002 and from Jan 2009 to Dec 2010 in order to consider the solar cycle effect. In training dataset, we randomly select one fifth of training data for validation dataset to avoid the overfitting problem. Our model successfully forecasts the flare occurrence with about 0.90 probability of detection (POD) for common flares (C-, M-, and X-class). While POD of major flares (M- and X-class) forecasting is 0.96, false alarm rate (FAR) also scores relatively high(0.60). We also present several statistical parameters such as critical success index (CSI) and true skill statistics (TSS). Our model can immediately be applied to automatic forecasting service when image data are available.

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

Smith, Donald F.; Schulz, Carl; Konijnenburg, Marco

High-resolution Fourier transform ion cyclotron resonance (FT-ICR) mass spectrometry imaging enables the spatial mapping and identification of biomolecules from complex surfaces. The need for long time-domain transients, and thus large raw file sizes, results in a large amount of raw data (“big data”) that must be processed efficiently and rapidly. This can be compounded by largearea imaging and/or high spatial resolution imaging. For FT-ICR, data processing and data reduction must not compromise the high mass resolution afforded by the mass spectrometer. The continuous mode “Mosaic Datacube” approach allows high mass resolution visualization (0.001 Da) of mass spectrometry imaging data, butmore » requires additional processing as compared to featurebased processing. We describe the use of distributed computing for processing of FT-ICR MS imaging datasets with generation of continuous mode Mosaic Datacubes for high mass resolution visualization. An eight-fold improvement in processing time is demonstrated using a Dutch nationally available cloud service.« less

Imaging mass spectrometry data reduction: automated feature identification and extraction.

PubMed

McDonnell, Liam A; van Remoortere, Alexandra; de Velde, Nico; van Zeijl, René J M; Deelder, André M

2010-12-01

Imaging MS now enables the parallel analysis of hundreds of biomolecules, spanning multiple molecular classes, which allows tissues to be described by their molecular content and distribution. When combined with advanced data analysis routines, tissues can be analyzed and classified based solely on their molecular content. Such molecular histology techniques have been used to distinguish regions with differential molecular signatures that could not be distinguished using established histologic tools. However, its potential to provide an independent, complementary analysis of clinical tissues has been limited by the very large file sizes and large number of discrete variables associated with imaging MS experiments. Here we demonstrate data reduction tools, based on automated feature identification and extraction, for peptide, protein, and lipid imaging MS, using multiple imaging MS technologies, that reduce data loads and the number of variables by >100×, and that highlight highly-localized features that can be missed using standard data analysis strategies. It is then demonstrated how these capabilities enable multivariate analysis on large imaging MS datasets spanning multiple tissues. Copyright © 2010 American Society for Mass Spectrometry. Published by Elsevier Inc. All rights reserved.

Phase correction for ALMA. Investigating water vapour radiometer scaling: The long-baseline science verification data case study

NASA Astrophysics Data System (ADS)

Maud, L. T.; Tilanus, R. P. J.; van Kempen, T. A.; Hogerheijde, M. R.; Schmalzl, M.; Yoon, I.; Contreras, Y.; Toribio, M. C.; Asaki, Y.; Dent, W. R. F.; Fomalont, E.; Matsushita, S.

2017-09-01

The Atacama Large millimetre/submillimetre Array (ALMA) makes use of water vapour radiometers (WVR), which monitor the atmospheric water vapour line at 183 GHz along the line of sight above each antenna to correct for phase delays introduced by the wet component of the troposphere. The application of WVR derived phase corrections improve the image quality and facilitate successful observations in weather conditions that were classically marginal or poor. We present work to indicate that a scaling factor applied to the WVR solutions can act to further improve the phase stability and image quality of ALMA data. We find reduced phase noise statistics for 62 out of 75 datasets from the long-baseline science verification campaign after a WVR scaling factor is applied. The improvement of phase noise translates to an expected coherence improvement in 39 datasets. When imaging the bandpass source, we find 33 of the 39 datasets show an improvement in the signal-to-noise ratio (S/N) between a few to 30 percent. There are 23 datasets where the S/N of the science image is improved: 6 by <1%, 11 between 1 and 5%, and 6 above 5%. The higher frequencies studied (band 6 and band 7) are those most improved, specifically datasets with low precipitable water vapour (PWV), <1 mm, where the dominance of the wet component is reduced. Although these improvements are not profound, phase stability improvements via the WVR scaling factor come into play for the higher frequency (>450 GHz) and long-baseline (>5 km) observations. These inherently have poorer phase stability and are taken in low PWV (<1 mm) conditions for which we find the scaling to be most effective. A promising explanation for the scaling factor is the mixing of dry and wet air components, although other origins are discussed. We have produced a python code to allow ALMA users to undertake WVR scaling tests and make improvements to their data.

Automated Topographic Change Detection via Dem Differencing at Large Scales Using The Arcticdem Database

NASA Astrophysics Data System (ADS)

Candela, S. G.; Howat, I.; Noh, M. J.; Porter, C. C.; Morin, P. J.

2016-12-01

In the last decade, high resolution satellite imagery has become an increasingly accessible tool for geoscientists to quantify changes in the Arctic land surface due to geophysical, ecological and anthropomorphic processes. However, the trade off between spatial coverage and spatial-temporal resolution has limited detailed, process-level change detection over large (i.e. continental) scales. The ArcticDEM project utilized over 300,000 Worldview image pairs to produce a nearly 100% coverage elevation model (above 60°N) offering the first polar, high spatial - high resolution (2-8m by region) dataset, often with multiple repeats in areas of particular interest to geo-scientists. A dataset of this size (nearly 250 TB) offers endless new avenues of scientific inquiry, but quickly becomes unmanageable computationally and logistically for the computing resources available to the average scientist. Here we present TopoDiff, a framework for a generalized. automated workflow that requires minimal input from the end user about a study site, and utilizes cloud computing resources to provide a temporally sorted and differenced dataset, ready for geostatistical analysis. This hands-off approach allows the end user to focus on the science, without having to manage thousands of files, or petabytes of data. At the same time, TopoDiff provides a consistent and accurate workflow for image sorting, selection, and co-registration enabling cross-comparisons between research projects.

A curated collection of tissue microarray images and clinical outcome data of prostate cancer patients

PubMed Central

Zhong, Qing; Guo, Tiannan; Rechsteiner, Markus; Rüschoff, Jan H.; Rupp, Niels; Fankhauser, Christian; Saba, Karim; Mortezavi, Ashkan; Poyet, Cédric; Hermanns, Thomas; Zhu, Yi; Moch, Holger; Aebersold, Ruedi; Wild, Peter J.

2017-01-01

Microscopy image data of human cancers provide detailed phenotypes of spatially and morphologically intact tissues at single-cell resolution, thus complementing large-scale molecular analyses, e.g., next generation sequencing or proteomic profiling. Here we describe a high-resolution tissue microarray (TMA) image dataset from a cohort of 71 prostate tissue samples, which was hybridized with bright-field dual colour chromogenic and silver in situ hybridization probes for the tumour suppressor gene PTEN. These tissue samples were digitized and supplemented with expert annotations, clinical information, statistical models of PTEN genetic status, and computer source codes. For validation, we constructed an additional TMA dataset for 424 prostate tissues, hybridized with FISH probes for PTEN, and performed survival analysis on a subset of 339 radical prostatectomy specimens with overall, disease-specific and recurrence-free survival (maximum 167 months). For application, we further produced 6,036 image patches derived from two whole slides. Our curated collection of prostate cancer data sets provides reuse potential for both biomedical and computational studies. PMID:28291248

An Enhanced Algorithm for Automatic Radiometric Harmonization of High-Resolution Optical Satellite Imagery Using Pseudoinvariant Features and Linear Regression

NASA Astrophysics Data System (ADS)

Langheinrich, M.; Fischer, P.; Probeck, M.; Ramminger, G.; Wagner, T.; Krauß, T.

2017-05-01

The growing number of available optical remote sensing data providing large spatial and temporal coverage enables the coherent and gapless observation of the earth's surface on the scale of whole countries or continents. To produce datasets of that size, individual satellite scenes have to be stitched together forming so-called mosaics. Here the problem arises that the different images feature varying radiometric properties depending on the momentary acquisition conditions. The interpretation of optical remote sensing data is to a great extent based on the analysis of the spectral composition of an observed surface reflection. Therefore the normalization of all images included in a large image mosaic is necessary to ensure consistent results concerning the application of procedures to the whole dataset. In this work an algorithm is described which enables the automated spectral harmonization of satellite images to a reference scene. As the stable and satisfying functionality of the proposed algorithm was already put to operational use to process a high number of SPOT-4/-5, IRS LISS-III and Landsat-5 scenes in the frame of the European Environment Agency's Copernicus/GMES Initial Operations (GIO) High-Resolution Layer (HRL) mapping of the HRL Forest for 20 Western, Central and (South)Eastern European countries, it is further evaluated on its reliability concerning the application to newer Sentinel-2 multispectral imaging products. The results show that the algorithm is comparably efficient for the processing of satellite image data from sources other than the sensor configurations it was originally designed for.

Three-dimensional spatiotemporal features for fast content-based retrieval of focal liver lesions.

PubMed

Roy, Sharmili; Chi, Yanling; Liu, Jimin; Venkatesh, Sudhakar K; Brown, Michael S

2014-11-01

Content-based image retrieval systems for 3-D medical datasets still largely rely on 2-D image-based features extracted from a few representative slices of the image stack. Most 2 -D features that are currently used in the literature not only model a 3-D tumor incompletely but are also highly expensive in terms of computation time, especially for high-resolution datasets. Radiologist-specified semantic labels are sometimes used along with image-based 2-D features to improve the retrieval performance. Since radiological labels show large interuser variability, are often unstructured, and require user interaction, their use as lesion characterizing features is highly subjective, tedious, and slow. In this paper, we propose a 3-D image-based spatiotemporal feature extraction framework for fast content-based retrieval of focal liver lesions. All the features are computer generated and are extracted from four-phase abdominal CT images. Retrieval performance and query processing times for the proposed framework is evaluated on a database of 44 hepatic lesions comprising of five pathological types. Bull's eye percentage score above 85% is achieved for three out of the five lesion pathologies and for 98% of query lesions, at least one same type of lesion is ranked among the top two retrieved results. Experiments show that the proposed system's query processing is more than 20 times faster than other already published systems that use 2-D features. With fast computation time and high retrieval accuracy, the proposed system has the potential to be used as an assistant to radiologists for routine hepatic tumor diagnosis.

Automated image based prominent nucleoli detection

PubMed Central

Yap, Choon K.; Kalaw, Emarene M.; Singh, Malay; Chong, Kian T.; Giron, Danilo M.; Huang, Chao-Hui; Cheng, Li; Law, Yan N.; Lee, Hwee Kuan

2015-01-01

Introduction: Nucleolar changes in cancer cells are one of the cytologic features important to the tumor pathologist in cancer assessments of tissue biopsies. However, inter-observer variability and the manual approach to this work hamper the accuracy of the assessment by pathologists. In this paper, we propose a computational method for prominent nucleoli pattern detection. Materials and Methods: Thirty-five hematoxylin and eosin stained images were acquired from prostate cancer, breast cancer, renal clear cell cancer and renal papillary cell cancer tissues. Prostate cancer images were used for the development of a computer-based automated prominent nucleoli pattern detector built on a cascade farm. An ensemble of approximately 1000 cascades was constructed by permuting different combinations of classifiers such as support vector machines, eXclusive component analysis, boosting, and logistic regression. The output of cascades was then combined using the RankBoost algorithm. The output of our prominent nucleoli pattern detector is a ranked set of detected image patches of patterns of prominent nucleoli. Results: The mean number of detected prominent nucleoli patterns in the top 100 ranked detected objects was 58 in the prostate cancer dataset, 68 in the breast cancer dataset, 86 in the renal clear cell cancer dataset, and 76 in the renal papillary cell cancer dataset. The proposed cascade farm performs twice as good as the use of a single cascade proposed in the seminal paper by Viola and Jones. For comparison, a naive algorithm that randomly chooses a pixel as a nucleoli pattern would detect five correct patterns in the first 100 ranked objects. Conclusions: Detection of sparse nucleoli patterns in a large background of highly variable tissue patterns is a difficult challenge our method has overcome. This study developed an accurate prominent nucleoli pattern detector with the potential to be used in the clinical settings. PMID:26167383

Automated image based prominent nucleoli detection.

PubMed

Yap, Choon K; Kalaw, Emarene M; Singh, Malay; Chong, Kian T; Giron, Danilo M; Huang, Chao-Hui; Cheng, Li; Law, Yan N; Lee, Hwee Kuan

2015-01-01

Nucleolar changes in cancer cells are one of the cytologic features important to the tumor pathologist in cancer assessments of tissue biopsies. However, inter-observer variability and the manual approach to this work hamper the accuracy of the assessment by pathologists. In this paper, we propose a computational method for prominent nucleoli pattern detection. Thirty-five hematoxylin and eosin stained images were acquired from prostate cancer, breast cancer, renal clear cell cancer and renal papillary cell cancer tissues. Prostate cancer images were used for the development of a computer-based automated prominent nucleoli pattern detector built on a cascade farm. An ensemble of approximately 1000 cascades was constructed by permuting different combinations of classifiers such as support vector machines, eXclusive component analysis, boosting, and logistic regression. The output of cascades was then combined using the RankBoost algorithm. The output of our prominent nucleoli pattern detector is a ranked set of detected image patches of patterns of prominent nucleoli. The mean number of detected prominent nucleoli patterns in the top 100 ranked detected objects was 58 in the prostate cancer dataset, 68 in the breast cancer dataset, 86 in the renal clear cell cancer dataset, and 76 in the renal papillary cell cancer dataset. The proposed cascade farm performs twice as good as the use of a single cascade proposed in the seminal paper by Viola and Jones. For comparison, a naive algorithm that randomly chooses a pixel as a nucleoli pattern would detect five correct patterns in the first 100 ranked objects. Detection of sparse nucleoli patterns in a large background of highly variable tissue patterns is a difficult challenge our method has overcome. This study developed an accurate prominent nucleoli pattern detector with the potential to be used in the clinical settings.

PANDA: a pipeline toolbox for analyzing brain diffusion images

PubMed Central

Cui, Zaixu; Zhong, Suyu; Xu, Pengfei; He, Yong; Gong, Gaolang

2013-01-01

Diffusion magnetic resonance imaging (dMRI) is widely used in both scientific research and clinical practice in in-vivo studies of the human brain. While a number of post-processing packages have been developed, fully automated processing of dMRI datasets remains challenging. Here, we developed a MATLAB toolbox named “Pipeline for Analyzing braiN Diffusion imAges” (PANDA) for fully automated processing of brain diffusion images. The processing modules of a few established packages, including FMRIB Software Library (FSL), Pipeline System for Octave and Matlab (PSOM), Diffusion Toolkit and MRIcron, were employed in PANDA. Using any number of raw dMRI datasets from different subjects, in either DICOM or NIfTI format, PANDA can automatically perform a series of steps to process DICOM/NIfTI to diffusion metrics [e.g., fractional anisotropy (FA) and mean diffusivity (MD)] that are ready for statistical analysis at the voxel-level, the atlas-level and the Tract-Based Spatial Statistics (TBSS)-level and can finish the construction of anatomical brain networks for all subjects. In particular, PANDA can process different subjects in parallel, using multiple cores either in a single computer or in a distributed computing environment, thus greatly reducing the time cost when dealing with a large number of datasets. In addition, PANDA has a friendly graphical user interface (GUI), allowing the user to be interactive and to adjust the input/output settings, as well as the processing parameters. As an open-source package, PANDA is freely available at http://www.nitrc.org/projects/panda/. This novel toolbox is expected to substantially simplify the image processing of dMRI datasets and facilitate human structural connectome studies. PMID:23439846

Efficient segmentation of 3D fluoroscopic datasets from mobile C-arm

NASA Astrophysics Data System (ADS)

Styner, Martin A.; Talib, Haydar; Singh, Digvijay; Nolte, Lutz-Peter

2004-05-01

The emerging mobile fluoroscopic 3D technology linked with a navigation system combines the advantages of CT-based and C-arm-based navigation. The intra-operative, automatic segmentation of 3D fluoroscopy datasets enables the combined visualization of surgical instruments and anatomical structures for enhanced planning, surgical eye-navigation and landmark digitization. We performed a thorough evaluation of several segmentation algorithms using a large set of data from different anatomical regions and man-made phantom objects. The analyzed segmentation methods include automatic thresholding, morphological operations, an adapted region growing method and an implicit 3D geodesic snake method. In regard to computational efficiency, all methods performed within acceptable limits on a standard Desktop PC (30sec-5min). In general, the best results were obtained with datasets from long bones, followed by extremities. The segmentations of spine, pelvis and shoulder datasets were generally of poorer quality. As expected, the threshold-based methods produced the worst results. The combined thresholding and morphological operations methods were considered appropriate for a smaller set of clean images. The region growing method performed generally much better in regard to computational efficiency and segmentation correctness, especially for datasets of joints, and lumbar and cervical spine regions. The less efficient implicit snake method was able to additionally remove wrongly segmented skin tissue regions. This study presents a step towards efficient intra-operative segmentation of 3D fluoroscopy datasets, but there is room for improvement. Next, we plan to study model-based approaches for datasets from the knee and hip joint region, which would be thenceforth applied to all anatomical regions in our continuing development of an ideal segmentation procedure for 3D fluoroscopic images.

SchizConnect: Mediating neuroimaging databases on schizophrenia and related disorders for large-scale integration.

PubMed

Wang, Lei; Alpert, Kathryn I; Calhoun, Vince D; Cobia, Derin J; Keator, David B; King, Margaret D; Kogan, Alexandr; Landis, Drew; Tallis, Marcelo; Turner, Matthew D; Potkin, Steven G; Turner, Jessica A; Ambite, Jose Luis

2016-01-01

SchizConnect (www.schizconnect.org) is built to address the issues of multiple data repositories in schizophrenia neuroimaging studies. It includes a level of mediation--translating across data sources--so that the user can place one query, e.g. for diffusion images from male individuals with schizophrenia, and find out from across participating data sources how many datasets there are, as well as downloading the imaging and related data. The current version handles the Data Usage Agreements across different studies, as well as interpreting database-specific terminologies into a common framework. New data repositories can also be mediated to bring immediate access to existing datasets. Compared with centralized, upload data sharing models, SchizConnect is a unique, virtual database with a focus on schizophrenia and related disorders that can mediate live data as information is being updated at each data source. It is our hope that SchizConnect can facilitate testing new hypotheses through aggregated datasets, promoting discovery related to the mechanisms underlying schizophrenic dysfunction. Copyright © 2015 Elsevier Inc. All rights reserved.

A multiscale dataset for understanding complex eco-hydrological processes in a heterogeneous oasis system

PubMed Central

Li, Xin; Liu, Shaomin; Xiao, Qin; Ma, Mingguo; Jin, Rui; Che, Tao; Wang, Weizhen; Hu, Xiaoli; Xu, Ziwei; Wen, Jianguang; Wang, Liangxu

2017-01-01

We introduce a multiscale dataset obtained from Heihe Watershed Allied Telemetry Experimental Research (HiWATER) in an oasis-desert area in 2012. Upscaling of eco-hydrological processes on a heterogeneous surface is a grand challenge. Progress in this field is hindered by the poor availability of multiscale observations. HiWATER is an experiment designed to address this challenge through instrumentation on hierarchically nested scales to obtain multiscale and multidisciplinary data. The HiWATER observation system consists of a flux observation matrix of eddy covariance towers, large aperture scintillometers, and automatic meteorological stations; an eco-hydrological sensor network of soil moisture and leaf area index; hyper-resolution airborne remote sensing using LiDAR, imaging spectrometer, multi-angle thermal imager, and L-band microwave radiometer; and synchronical ground measurements of vegetation dynamics, and photosynthesis processes. All observational data were carefully quality controlled throughout sensor calibration, data collection, data processing, and datasets generation. The data are freely available at figshare and the Cold and Arid Regions Science Data Centre. The data should be useful for elucidating multiscale eco-hydrological processes and developing upscaling methods. PMID:28654086

SEGMA: An Automatic SEGMentation Approach for Human Brain MRI Using Sliding Window and Random Forests

PubMed Central

Serag, Ahmed; Wilkinson, Alastair G.; Telford, Emma J.; Pataky, Rozalia; Sparrow, Sarah A.; Anblagan, Devasuda; Macnaught, Gillian; Semple, Scott I.; Boardman, James P.

2017-01-01

Quantitative volumes from brain magnetic resonance imaging (MRI) acquired across the life course may be useful for investigating long term effects of risk and resilience factors for brain development and healthy aging, and for understanding early life determinants of adult brain structure. Therefore, there is an increasing need for automated segmentation tools that can be applied to images acquired at different life stages. We developed an automatic segmentation method for human brain MRI, where a sliding window approach and a multi-class random forest classifier were applied to high-dimensional feature vectors for accurate segmentation. The method performed well on brain MRI data acquired from 179 individuals, analyzed in three age groups: newborns (38–42 weeks gestational age), children and adolescents (4–17 years) and adults (35–71 years). As the method can learn from partially labeled datasets, it can be used to segment large-scale datasets efficiently. It could also be applied to different populations and imaging modalities across the life course. PMID:28163680

Color imaging of Mars by the High Resolution Imaging Science Experiment (HiRISE)

USGS Publications Warehouse

Delamere, W.A.; Tornabene, L.L.; McEwen, A.S.; Becker, K.; Bergstrom, J.W.; Bridges, N.T.; Eliason, E.M.; Gallagher, D.; Herkenhoff, K. E.; Keszthelyi, L.; Mattson, S.; McArthur, G.K.; Mellon, M.T.; Milazzo, M.; Russell, P.S.; Thomas, N.

2010-01-01

HiRISE has been producing a large number of scientifically useful color products of Mars and other planetary objects. The three broad spectral bands, coupled with the highly sensitive 14 bit detectors and time delay integration, enable detection of subtle color differences. The very high spatial resolution of HiRISE can augment the mineralogic interpretations based on multispectral (THEMIS) and hyperspectral datasets (TES, OMEGA and CRISM) and thereby enable detailed geologic and stratigraphic interpretations at meter scales. In addition to providing some examples of color images and their interpretation, we describe the processing techniques used to produce them and note some of the minor artifacts in the output. We also provide an example of how HiRISE color products can be effectively used to expand mineral and lithologic mapping provided by CRISM data products that are backed by other spectral datasets. The utility of high quality color data for understanding geologic processes on Mars has been one of the major successes of HiRISE. ?? 2009 Elsevier Inc.

Investigating Bacterial-Animal Symbioses with Light Sheet Microscopy

PubMed Central

Taormina, Michael J.; Jemielita, Matthew; Stephens, W. Zac; Burns, Adam R.; Troll, Joshua V.; Parthasarathy, Raghuveer; Guillemin, Karen

2014-01-01

SUMMARY Microbial colonization of the digestive tract is a crucial event in vertebrate development, required for maturation of host immunity and establishment of normal digestive physiology. Advances in genomic, proteomic, and metabolomic technologies are providing a more detailed picture of the constituents of the intestinal habitat, but these approaches lack the spatial and temporal resolution needed to characterize the assembly and dynamics of microbial communities in this complex environment. We report the use of light sheet microscopy to provide high resolution imaging of bacterial colonization of the zebrafish intestine. The methodology allows us to characterize bacterial population dynamics across the entire organ and the behaviors of individual bacterial and host cells throughout the colonization process. The large four-dimensional datasets generated by these imaging approaches require new strategies for image analysis. When integrated with other “omics” datasets, information about the spatial and temporal dynamics of microbial cells within the vertebrate intestine will provide new mechanistic insights into how microbial communities assemble and function within hosts. PMID:22983029

Saliency detection by conditional generative adversarial network

NASA Astrophysics Data System (ADS)

Cai, Xiaoxu; Yu, Hui

2018-04-01

Detecting salient objects in images has been a fundamental problem in computer vision. In recent years, deep learning has shown its impressive performance in dealing with many kinds of vision tasks. In this paper, we propose a new method to detect salient objects by using Conditional Generative Adversarial Network (GAN). This type of network not only learns the mapping from RGB images to salient regions, but also learns a loss function for training the mapping. To the best of our knowledge, this is the first time that Conditional GAN has been used in salient object detection. We evaluate our saliency detection method on 2 large publicly available datasets with pixel accurate annotations. The experimental results have shown the significant and consistent improvements over the state-of-the-art method on a challenging dataset, and the testing speed is much faster.

OpenMSI Arrayed Analysis Tools v2.0

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

BOWEN, BENJAMIN; RUEBEL, OLIVER; DE ROND, TRISTAN

2017-02-07

Mass spectrometry imaging (MSI) enables high-resolution spatial mapping of biomolecules in samples and is a valuable tool for the analysis of tissues from plants and animals, microbial interactions, high-throughput screening, drug metabolism, and a host of other applications. This is accomplished by desorbing molecules from the surface on spatially defined locations, using a laser or ion beam. These ions are analyzed by a mass spectrometry and collected into a MSI 'image', a dataset containing unique mass spectra from the sampled spatial locations. MSI is used in a diverse and increasing number of biological applications. The OpenMSI Arrayed Analysis Tool (OMAAT)more » is a new software method that addresses the challenges of analyzing spatially defined samples in large MSI datasets, by providing support for automatic sample position optimization and ion selection.« less

Quantifying Uncertainties in Land Surface Microwave Emissivity Retrievals

NASA Technical Reports Server (NTRS)

Tian, Yudong; Peters-Lidard, Christa D.; Harrison, Kenneth W.; Prigent, Catherine; Norouzi, Hamidreza; Aires, Filipe; Boukabara, Sid-Ahmed; Furuzawa, Fumie A.; Masunaga, Hirohiko

2012-01-01

Uncertainties in the retrievals of microwave land surface emissivities were quantified over two types of land surfaces: desert and tropical rainforest. Retrievals from satellite-based microwave imagers, including SSM/I, TMI and AMSR-E, were studied. Our results show that there are considerable differences between the retrievals from different sensors and from different groups over these two land surface types. In addition, the mean emissivity values show different spectral behavior across the frequencies. With the true emissivity assumed largely constant over both of the two sites throughout the study period, the differences are largely attributed to the systematic and random errors in the retrievals. Generally these retrievals tend to agree better at lower frequencies than at higher ones, with systematic differences ranging 14% (312 K) over desert and 17% (320 K) over rainforest. The random errors within each retrieval dataset are in the range of 0.52% (26 K). In particular, at 85.0/89.0 GHz, there are very large differences between the different retrieval datasets, and within each retrieval dataset itself. Further investigation reveals that these differences are mostly likely caused by rain/cloud contamination, which can lead to random errors up to 1017 K under the most severe conditions.

PANTHER. Pattern ANalytics To support High-performance Exploitation and Reasoning.

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

Czuchlewski, Kristina Rodriguez; Hart, William E.

Sandia has approached the analysis of big datasets with an integrated methodology that uses computer science, image processing, and human factors to exploit critical patterns and relationships in large datasets despite the variety and rapidity of information. The work is part of a three-year LDRD Grand Challenge called PANTHER (Pattern ANalytics To support High-performance Exploitation and Reasoning). To maximize data analysis capability, Sandia pursued scientific advances across three key technical domains: (1) geospatial-temporal feature extraction via image segmentation and classification; (2) geospatial-temporal analysis capabilities tailored to identify and process new signatures more efficiently; and (3) domain- relevant models of humanmore » perception and cognition informing the design of analytic systems. Our integrated results include advances in geographical information systems (GIS) in which we discover activity patterns in noisy, spatial-temporal datasets using geospatial-temporal semantic graphs. We employed computational geometry and machine learning to allow us to extract and predict spatial-temporal patterns and outliers from large aircraft and maritime trajectory datasets. We automatically extracted static and ephemeral features from real, noisy synthetic aperture radar imagery for ingestion into a geospatial-temporal semantic graph. We worked with analysts and investigated analytic workflows to (1) determine how experiential knowledge evolves and is deployed in high-demand, high-throughput visual search workflows, and (2) better understand visual search performance and attention. Through PANTHER, Sandia's fundamental rethinking of key aspects of geospatial data analysis permits the extraction of much richer information from large amounts of data. The project results enable analysts to examine mountains of historical and current data that would otherwise go untouched, while also gaining meaningful, measurable, and defensible insights into overlooked relationships and patterns. The capability is directly relevant to the nation's nonproliferation remote-sensing activities and has broad national security applications for military and intelligence- gathering organizations.« less

NEFI: Network Extraction From Images

PubMed Central

Dirnberger, M.; Kehl, T.; Neumann, A.

2015-01-01

Networks are amongst the central building blocks of many systems. Given a graph of a network, methods from graph theory enable a precise investigation of its properties. Software for the analysis of graphs is widely available and has been applied to study various types of networks. In some applications, graph acquisition is relatively simple. However, for many networks data collection relies on images where graph extraction requires domain-specific solutions. Here we introduce NEFI, a tool that extracts graphs from images of networks originating in various domains. Regarding previous work on graph extraction, theoretical results are fully accessible only to an expert audience and ready-to-use implementations for non-experts are rarely available or insufficiently documented. NEFI provides a novel platform allowing practitioners to easily extract graphs from images by combining basic tools from image processing, computer vision and graph theory. Thus, NEFI constitutes an alternative to tedious manual graph extraction and special purpose tools. We anticipate NEFI to enable time-efficient collection of large datasets. The analysis of these novel datasets may open up the possibility to gain new insights into the structure and function of various networks. NEFI is open source and available at http://nefi.mpi-inf.mpg.de. PMID:26521675

Comprehensive optical and data management infrastructure for high-throughput light-sheet microscopy of whole mouse brains.

PubMed

Müllenbroich, M Caroline; Silvestri, Ludovico; Onofri, Leonardo; Costantini, Irene; Hoff, Marcel Van't; Sacconi, Leonardo; Iannello, Giulio; Pavone, Francesco S

2015-10-01

Comprehensive mapping and quantification of neuronal projections in the central nervous system requires high-throughput imaging of large volumes with microscopic resolution. To this end, we have developed a confocal light-sheet microscope that has been optimized for three-dimensional (3-D) imaging of structurally intact clarified whole-mount mouse brains. We describe the optical and electromechanical arrangement of the microscope and give details on the organization of the microscope management software. The software orchestrates all components of the microscope, coordinates critical timing and synchronization, and has been written in a versatile and modular structure using the LabVIEW language. It can easily be adapted and integrated to other microscope systems and has been made freely available to the light-sheet community. The tremendous amount of data routinely generated by light-sheet microscopy further requires novel strategies for data handling and storage. To complete the full imaging pipeline of our high-throughput microscope, we further elaborate on big data management from streaming of raw images up to stitching of 3-D datasets. The mesoscale neuroanatomy imaged at micron-scale resolution in those datasets allows characterization and quantification of neuronal projections in unsectioned mouse brains.

Deep learning with non-medical training used for chest pathology identification

NASA Astrophysics Data System (ADS)

Bar, Yaniv; Diamant, Idit; Wolf, Lior; Greenspan, Hayit

2015-03-01

In this work, we examine the strength of deep learning approaches for pathology detection in chest radiograph data. Convolutional neural networks (CNN) deep architecture classification approaches have gained popularity due to their ability to learn mid and high level image representations. We explore the ability of a CNN to identify different types of pathologies in chest x-ray images. Moreover, since very large training sets are generally not available in the medical domain, we explore the feasibility of using a deep learning approach based on non-medical learning. We tested our algorithm on a dataset of 93 images. We use a CNN that was trained with ImageNet, a well-known large scale nonmedical image database. The best performance was achieved using a combination of features extracted from the CNN and a set of low-level features. We obtained an area under curve (AUC) of 0.93 for Right Pleural Effusion detection, 0.89 for Enlarged heart detection and 0.79 for classification between healthy and abnormal chest x-ray, where all pathologies are combined into one large class. This is a first-of-its-kind experiment that shows that deep learning with large scale non-medical image databases may be sufficient for general medical image recognition tasks.

Efficient feature extraction from wide-area motion imagery by MapReduce in Hadoop

NASA Astrophysics Data System (ADS)

Cheng, Erkang; Ma, Liya; Blaisse, Adam; Blasch, Erik; Sheaff, Carolyn; Chen, Genshe; Wu, Jie; Ling, Haibin

2014-06-01

Wide-Area Motion Imagery (WAMI) feature extraction is important for applications such as target tracking, traffic management and accident discovery. With the increasing amount of WAMI collections and feature extraction from the data, a scalable framework is needed to handle the large amount of information. Cloud computing is one of the approaches recently applied in large scale or big data. In this paper, MapReduce in Hadoop is investigated for large scale feature extraction tasks for WAMI. Specifically, a large dataset of WAMI images is divided into several splits. Each split has a small subset of WAMI images. The feature extractions of WAMI images in each split are distributed to slave nodes in the Hadoop system. Feature extraction of each image is performed individually in the assigned slave node. Finally, the feature extraction results are sent to the Hadoop File System (HDFS) to aggregate the feature information over the collected imagery. Experiments of feature extraction with and without MapReduce are conducted to illustrate the effectiveness of our proposed Cloud-Enabled WAMI Exploitation (CAWE) approach.

Leveraging the crowd for annotation of retinal images.

PubMed

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

2015-01-01

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

Mutual information based feature selection for medical image retrieval

NASA Astrophysics Data System (ADS)

Zhi, Lijia; Zhang, Shaomin; Li, Yan

2018-04-01

In this paper, authors propose a mutual information based method for lung CT image retrieval. This method is designed to adapt to different datasets and different retrieval task. For practical applying consideration, this method avoids using a large amount of training data. Instead, with a well-designed training process and robust fundamental features and measurements, the method in this paper can get promising performance and maintain economic training computation. Experimental results show that the method has potential practical values for clinical routine application.

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.

Dataset variability leverages white-matter lesion segmentation performance with convolutional neural network

NASA Astrophysics Data System (ADS)

Ravnik, Domen; Jerman, Tim; Pernuš, Franjo; Likar, Boštjan; Å piclin, Žiga

2018-03-01

Performance of a convolutional neural network (CNN) based white-matter lesion segmentation in magnetic resonance (MR) brain images was evaluated under various conditions involving different levels of image preprocessing and augmentation applied and different compositions of the training dataset. On images of sixty multiple sclerosis patients, half acquired on one and half on another scanner of different vendor, we first created a highly accurate multi-rater consensus based lesion segmentations, which were used in several experiments to evaluate the CNN segmentation result. First, the CNN was trained and tested without preprocessing the images and by using various combinations of preprocessing techniques, namely histogram-based intensity standardization, normalization by whitening, and train dataset augmentation by flipping the images across the midsagittal plane. Then, the CNN was trained and tested on images of the same, different or interleaved scanner datasets using a cross-validation approach. The results indicate that image preprocessing has little impact on performance in a same-scanner situation, while between-scanner performance benefits most from intensity standardization and normalization, but also further by incorporating heterogeneous multi-scanner datasets in the training phase. Under such conditions the between-scanner performance of the CNN approaches that of the ideal situation, when the CNN is trained and tested on the same scanner dataset.

Machine learning for neuroimaging with scikit-learn.

PubMed

Abraham, Alexandre; Pedregosa, Fabian; Eickenberg, Michael; Gervais, Philippe; Mueller, Andreas; Kossaifi, Jean; Gramfort, Alexandre; Thirion, Bertrand; Varoquaux, Gaël

2014-01-01

Statistical machine learning methods are increasingly used for neuroimaging data analysis. Their main virtue is their ability to model high-dimensional datasets, e.g., multivariate analysis of activation images or resting-state time series. Supervised learning is typically used in decoding or encoding settings to relate brain images to behavioral or clinical observations, while unsupervised learning can uncover hidden structures in sets of images (e.g., resting state functional MRI) or find sub-populations in large cohorts. By considering different functional neuroimaging applications, we illustrate how scikit-learn, a Python machine learning library, can be used to perform some key analysis steps. Scikit-learn contains a very large set of statistical learning algorithms, both supervised and unsupervised, and its application to neuroimaging data provides a versatile tool to study the brain.

Machine learning for neuroimaging with scikit-learn

PubMed Central

Abraham, Alexandre; Pedregosa, Fabian; Eickenberg, Michael; Gervais, Philippe; Mueller, Andreas; Kossaifi, Jean; Gramfort, Alexandre; Thirion, Bertrand; Varoquaux, Gaël

2014-01-01

Statistical machine learning methods are increasingly used for neuroimaging data analysis. Their main virtue is their ability to model high-dimensional datasets, e.g., multivariate analysis of activation images or resting-state time series. Supervised learning is typically used in decoding or encoding settings to relate brain images to behavioral or clinical observations, while unsupervised learning can uncover hidden structures in sets of images (e.g., resting state functional MRI) or find sub-populations in large cohorts. By considering different functional neuroimaging applications, we illustrate how scikit-learn, a Python machine learning library, can be used to perform some key analysis steps. Scikit-learn contains a very large set of statistical learning algorithms, both supervised and unsupervised, and its application to neuroimaging data provides a versatile tool to study the brain. PMID:24600388

Simulation of bright-field microscopy images depicting pap-smear specimen

PubMed Central

Malm, Patrik; Brun, Anders; Bengtsson, Ewert

2015-01-01

As digital imaging is becoming a fundamental part of medical and biomedical research, the demand for computer-based evaluation using advanced image analysis is becoming an integral part of many research projects. A common problem when developing new image analysis algorithms is the need of large datasets with ground truth on which the algorithms can be tested and optimized. Generating such datasets is often tedious and introduces subjectivity and interindividual and intraindividual variations. An alternative to manually created ground-truth data is to generate synthetic images where the ground truth is known. The challenge then is to make the images sufficiently similar to the real ones to be useful in algorithm development. One of the first and most widely studied medical image analysis tasks is to automate screening for cervical cancer through Pap-smear analysis. As part of an effort to develop a new generation cervical cancer screening system, we have developed a framework for the creation of realistic synthetic bright-field microscopy images that can be used for algorithm development and benchmarking. The resulting framework has been assessed through a visual evaluation by experts with extensive experience of Pap-smear images. The results show that images produced using our described methods are realistic enough to be mistaken for real microscopy images. The developed simulation framework is very flexible and can be modified to mimic many other types of bright-field microscopy images. © 2015 The Authors. Published by Wiley Periodicals, Inc. on behalf of ISAC PMID:25573002

OpenCL based machine learning labeling of biomedical datasets

NASA Astrophysics Data System (ADS)

Amoros, Oscar; Escalera, Sergio; Puig, Anna

2011-03-01

In this paper, we propose a two-stage labeling method of large biomedical datasets through a parallel approach in a single GPU. Diagnostic methods, structures volume measurements, and visualization systems are of major importance for surgery planning, intra-operative imaging and image-guided surgery. In all cases, to provide an automatic and interactive method to label or to tag different structures contained into input data becomes imperative. Several approaches to label or segment biomedical datasets has been proposed to discriminate different anatomical structures in an output tagged dataset. Among existing methods, supervised learning methods for segmentation have been devised to easily analyze biomedical datasets by a non-expert user. However, they still have some problems concerning practical application, such as slow learning and testing speeds. In addition, recent technological developments have led to widespread availability of multi-core CPUs and GPUs, as well as new software languages, such as NVIDIA's CUDA and OpenCL, allowing to apply parallel programming paradigms in conventional personal computers. Adaboost classifier is one of the most widely applied methods for labeling in the Machine Learning community. In a first stage, Adaboost trains a binary classifier from a set of pre-labeled samples described by a set of features. This binary classifier is defined as a weighted combination of weak classifiers. Each weak classifier is a simple decision function estimated on a single feature value. Then, at the testing stage, each weak classifier is independently applied on the features of a set of unlabeled samples. In this work, we propose an alternative representation of the Adaboost binary classifier. We use this proposed representation to define a new GPU-based parallelized Adaboost testing stage using OpenCL. We provide numerical experiments based on large available data sets and we compare our results to CPU-based strategies in terms of time and labeling speeds.

Server-based Approach to Web Visualization of Integrated Three-dimensional Brain Imaging Data

PubMed Central

Poliakov, Andrew V.; Albright, Evan; Hinshaw, Kevin P.; Corina, David P.; Ojemann, George; Martin, Richard F.; Brinkley, James F.

2005-01-01

The authors describe a client-server approach to three-dimensional (3-D) visualization of neuroimaging data, which enables researchers to visualize, manipulate, and analyze large brain imaging datasets over the Internet. All computationally intensive tasks are done by a graphics server that loads and processes image volumes and 3-D models, renders 3-D scenes, and sends the renderings back to the client. The authors discuss the system architecture and implementation and give several examples of client applications that allow visualization and analysis of integrated language map data from single and multiple patients. PMID:15561787

Part-based deep representation for product tagging and search

NASA Astrophysics Data System (ADS)

Chen, Keqing

2017-06-01

Despite previous studies, tagging and indexing the product images remain challenging due to the large inner-class variation of the products. In the traditional methods, the quantized hand-crafted features such as SIFTs are extracted as the representation of the product images, which are not discriminative enough to handle the inner-class variation. For discriminative image representation, this paper firstly presents a novel deep convolutional neural networks (DCNNs) architect true pre-trained on a large-scale general image dataset. Compared to the traditional features, our DCNNs representation is of more discriminative power with fewer dimensions. Moreover, we incorporate the part-based model into the framework to overcome the negative effect of bad alignment and cluttered background and hence the descriptive ability of the deep representation is further enhanced. Finally, we collect and contribute a well-labeled shoe image database, i.e., the TBShoes, on which we apply the part-based deep representation for product image tagging and search, respectively. The experimental results highlight the advantages of the proposed part-based deep representation.

Social Image Captioning: Exploring Visual Attention and User Attention.

PubMed

Wang, Leiquan; Chu, Xiaoliang; Zhang, Weishan; Wei, Yiwei; Sun, Weichen; Wu, Chunlei

2018-02-22

Image captioning with a natural language has been an emerging trend. However, the social image, associated with a set of user-contributed tags, has been rarely investigated for a similar task. The user-contributed tags, which could reflect the user attention, have been neglected in conventional image captioning. Most existing image captioning models cannot be applied directly to social image captioning. In this work, a dual attention model is proposed for social image captioning by combining the visual attention and user attention simultaneously.Visual attention is used to compress a large mount of salient visual information, while user attention is applied to adjust the description of the social images with user-contributed tags. Experiments conducted on the Microsoft (MS) COCO dataset demonstrate the superiority of the proposed method of dual attention.

Social Image Captioning: Exploring Visual Attention and User Attention

PubMed Central

Chu, Xiaoliang; Zhang, Weishan; Wei, Yiwei; Sun, Weichen; Wu, Chunlei

2018-01-01

Image captioning with a natural language has been an emerging trend. However, the social image, associated with a set of user-contributed tags, has been rarely investigated for a similar task. The user-contributed tags, which could reflect the user attention, have been neglected in conventional image captioning. Most existing image captioning models cannot be applied directly to social image captioning. In this work, a dual attention model is proposed for social image captioning by combining the visual attention and user attention simultaneously.Visual attention is used to compress a large mount of salient visual information, while user attention is applied to adjust the description of the social images with user-contributed tags. Experiments conducted on the Microsoft (MS) COCO dataset demonstrate the superiority of the proposed method of dual attention. PMID:29470409

An Evaluation of Data Fusion Products for the Analysis of Dryland Forest Phenology

NASA Astrophysics Data System (ADS)

Walker, J. J.; de Beurs, K.; Wynne, R. H.; Gao, F.

2010-12-01

Semi-arid forest areas cover a significant proportion of the world’s land surface; in the interior western U.S. alone, dryland forests extend across more than 56 million hectares. The scarcity of water in these systems makes them acutely sensitive to sustained weather fluctuations, such as the higher temperatures and altered water regimes predicted under most climate change scenarios. To understand, monitor, and predict the anticipated spatial and temporal changes in these areas, it is vital to characterize current phenological patterns. Phenological analysis of western U.S. drylands is complicated by patchy land cover and mosaics of plant phenology states at a variety of spatial scales. Our aim is to use complementary satellite sensors to mitigate these difficulties and gain greater insight into phenological patterns in dryland forests. In this study we applied the spatial and temporal adaptive reflectance model (STARFM; Gao et al. 2006) to fuse Landsat and MODIS imagery to create synthetic images at Landsat spatial resolution and MODIS temporal resolution. To determine which MODIS dataset is most appropriate for the creation of synthetic images intended for the analysis of dryland forest phenology, we examined the effect of temporal compositing and BRDF function adjustment on the accuracy of STARFM imagery. We assembled seven Landsat 5 scenes (path/row 37/36) and temporally-coincident 500m MODIS datasets (seven daily (MOD09GA), seven 8-day composite (MOD09A1), and fourteen 16-day nadir BRDF-adjusted composite (MCD43A4) images) spanning the 2006 April - October growing season in northern Arizona, which is characterized by large tracts of dryland forest. The STARFM algorithm was applied to each MODIS data series to produce four synthetic images (one daily; one 8-day composite; and two 16-day composites) corresponding to each Landsat image. Validation of the accuracy of the synthetic images was achieved by comparing the reflectance values of a random sample of the identified dryland forest pixels in both images. Preliminary data analysis of the effect of the temporal resolution and dataset parameters indicates that the MODIS 8-day composite image may be a suitable and sufficient dataset for phenological analysis in this dryland forest ecosystem. Overall, this work demonstrates the feasibility of using data fusion products to assemble an imagery dataset at sufficiently high temporal and spatial scales to permit a more detailed examination of the underlying phenological processes and trends in dryland forest areas.

PCA-based approach for subtracting thermal background emission in high-contrast imaging data

NASA Astrophysics Data System (ADS)

Hunziker, S.; Quanz, S. P.; Amara, A.; Meyer, M. R.

2018-03-01

Aims.Ground-based observations at thermal infrared wavelengths suffer from large background radiation due to the sky, telescope and warm surfaces in the instrument. This significantly limits the sensitivity of ground-based observations at wavelengths longer than 3 μm. The main purpose of this work is to analyse this background emission in infrared high-contrast imaging data as illustrative of the problem, show how it can be modelled and subtracted and demonstrate that it can improve the detection of faint sources, such as exoplanets. Methods: We used principal component analysis (PCA) to model and subtract the thermal background emission in three archival high-contrast angular differential imaging datasets in the M' and L' filter. We used an M' dataset of β Pic to describe in detail how the algorithm works and explain how it can be applied. The results of the background subtraction are compared to the results from a conventional mean background subtraction scheme applied to the same dataset. Finally, both methods for background subtraction are compared by performing complete data reductions. We analysed the results from the M' dataset of HD 100546 only qualitatively. For the M' band dataset of β Pic and the L' band dataset of HD 169142, which was obtained with an angular groove phase mask vortex vector coronagraph, we also calculated and analysed the achieved signal-to-noise ratio (S/N). Results: We show that applying PCA is an effective way to remove spatially and temporarily varying thermal background emission down to close to the background limit. The procedure also proves to be very successful at reconstructing the background that is hidden behind the point spread function. In the complete data reductions, we find at least qualitative improvements for HD 100546 and HD 169142, however, we fail to find a significant increase in S/N of β Pic b. We discuss these findings and argue that in particular datasets with strongly varying observing conditions or infrequently sampled sky background will benefit from the new approach.

Image classification using multiscale information fusion based on saliency driven nonlinear diffusion filtering.

PubMed

Hu, Weiming; Hu, Ruiguang; Xie, Nianhua; Ling, Haibin; Maybank, Stephen

2014-04-01

In this paper, we propose saliency driven image multiscale nonlinear diffusion filtering. The resulting scale space in general preserves or even enhances semantically important structures such as edges, lines, or flow-like structures in the foreground, and inhibits and smoothes clutter in the background. The image is classified using multiscale information fusion based on the original image, the image at the final scale at which the diffusion process converges, and the image at a midscale. Our algorithm emphasizes the foreground features, which are important for image classification. The background image regions, whether considered as contexts of the foreground or noise to the foreground, can be globally handled by fusing information from different scales. Experimental tests of the effectiveness of the multiscale space for the image classification are conducted on the following publicly available datasets: 1) the PASCAL 2005 dataset; 2) the Oxford 102 flowers dataset; and 3) the Oxford 17 flowers dataset, with high classification rates.

Interactive brain shift compensation using GPU based programming

NASA Astrophysics Data System (ADS)

van der Steen, Sander; Noordmans, Herke Jan; Verdaasdonk, Rudolf

2009-02-01

Processing large images files or real-time video streams requires intense computational power. Driven by the gaming industry, the processing power of graphic process units (GPUs) has increased significantly. With the pixel shader model 4.0 the GPU can be used for image processing 10x faster than the CPU. Dedicated software was developed to deform 3D MR and CT image sets for real-time brain shift correction during navigated neurosurgery using landmarks or cortical surface traces defined by the navigation pointer. Feedback was given using orthogonal slices and an interactively raytraced 3D brain image. GPU based programming enables real-time processing of high definition image datasets and various applications can be developed in medicine, optics and image sciences.

The Autism Brain Imaging Data Exchange: Towards Large-Scale Evaluation of the Intrinsic Brain Architecture in Autism

PubMed Central

Di Martino, Adriana; Yan, Chao-Gan; Li, Qingyang; Denio, Erin; Castellanos, Francisco X.; Alaerts, Kaat; Anderson, Jeffrey S.; Assaf, Michal; Bookheimer, Susan Y.; Dapretto, Mirella; Deen, Ben; Delmonte, Sonja; Dinstein, Ilan; Ertl-Wagner, Birgit; Fair, Damien A.; Gallagher, Louise; Kennedy, Daniel P.; Keown, Christopher L.; Keysers, Christian; Lainhart, Janet E.; Lord, Catherine; Luna, Beatriz; Menon, Vinod; Minshew, Nancy; Monk, Christopher S.; Mueller, Sophia; Müller, Ralph-Axel; Nebel, Mary Beth; Nigg, Joel T.; O’Hearn, Kirsten; Pelphrey, Kevin A.; Peltier, Scott J.; Rudie, Jeffrey D.; Sunaert, Stefan; Thioux, Marc; Tyszka, J. Michael; Uddin, Lucina Q.; Verhoeven, Judith S.; Wenderoth, Nicole; Wiggins, Jillian L.; Mostofsky, Stewart H.; Milham, Michael P.

2014-01-01

Autism spectrum disorders (ASD) represent a formidable challenge for psychiatry and neuroscience because of their high prevalence, life-long nature, complexity and substantial heterogeneity. Facing these obstacles requires large-scale multidisciplinary efforts. While the field of genetics has pioneered data sharing for these reasons, neuroimaging had not kept pace. In response, we introduce the Autism Brain Imaging Data Exchange (ABIDE) – a grassroots consortium aggregating and openly sharing 1112 existing resting-state functional magnetic resonance imaging (R-fMRI) datasets with corresponding structural MRI and phenotypic information from 539 individuals with ASD and 573 age-matched typical controls (TC; 7–64 years) (http://fcon_1000.projects.nitrc.org/indi/abide/). Here, we present this resource and demonstrate its suitability for advancing knowledge of ASD neurobiology based on analyses of 360 males with ASD and 403 male age-matched TC. We focused on whole-brain intrinsic functional connectivity and also survey a range of voxel-wise measures of intrinsic functional brain architecture. Whole-brain analyses reconciled seemingly disparate themes of both hypo and hyperconnectivity in the ASD literature; both were detected, though hypoconnectivity dominated, particularly for cortico-cortical and interhemispheric functional connectivity. Exploratory analyses using an array of regional metrics of intrinsic brain function converged on common loci of dysfunction in ASD (mid and posterior insula, posterior cingulate cortex), and highlighted less commonly explored regions such as thalamus. The survey of the ABIDE R-fMRI datasets provides unprecedented demonstrations of both replication and novel discovery. By pooling multiple international datasets, ABIDE is expected to accelerate the pace of discovery setting the stage for the next generation of ASD studies. PMID:23774715

Semantic attributes for people's appearance description: an appearance modality for video surveillance applications

NASA Astrophysics Data System (ADS)

Frikha, Mayssa; Fendri, Emna; Hammami, Mohamed

2017-09-01

Using semantic attributes such as gender, clothes, and accessories to describe people's appearance is an appealing modeling method for video surveillance applications. We proposed a midlevel appearance signature based on extracting a list of nameable semantic attributes describing the body in uncontrolled acquisition conditions. Conventional approaches extract the same set of low-level features to learn the semantic classifiers uniformly. Their critical limitation is the inability to capture the dominant visual characteristics for each trait separately. The proposed approach consists of extracting low-level features in an attribute-adaptive way by automatically selecting the most relevant features for each attribute separately. Furthermore, relying on a small training-dataset would easily lead to poor performance due to the large intraclass and interclass variations. We annotated large scale people images collected from different person reidentification benchmarks covering a large attribute sample and reflecting the challenges of uncontrolled acquisition conditions. These annotations were gathered into an appearance semantic attribute dataset that contains 3590 images annotated with 14 attributes. Various experiments prove that carefully designed features for learning the visual characteristics for an attribute provide an improvement of the correct classification accuracy and a reduction of both spatial and temporal complexities against state-of-the-art approaches.

Benchmarking Spike-Based Visual Recognition: A Dataset and Evaluation

PubMed Central

Liu, Qian; Pineda-García, Garibaldi; Stromatias, Evangelos; Serrano-Gotarredona, Teresa; Furber, Steve B.

2016-01-01

Today, increasing attention is being paid to research into spike-based neural computation both to gain a better understanding of the brain and to explore biologically-inspired computation. Within this field, the primate visual pathway and its hierarchical organization have been extensively studied. Spiking Neural Networks (SNNs), inspired by the understanding of observed biological structure and function, have been successfully applied to visual recognition and classification tasks. In addition, implementations on neuromorphic hardware have enabled large-scale networks to run in (or even faster than) real time, making spike-based neural vision processing accessible on mobile robots. Neuromorphic sensors such as silicon retinas are able to feed such mobile systems with real-time visual stimuli. A new set of vision benchmarks for spike-based neural processing are now needed to measure progress quantitatively within this rapidly advancing field. We propose that a large dataset of spike-based visual stimuli is needed to provide meaningful comparisons between different systems, and a corresponding evaluation methodology is also required to measure the performance of SNN models and their hardware implementations. In this paper we first propose an initial NE (Neuromorphic Engineering) dataset based on standard computer vision benchmarksand that uses digits from the MNIST database. This dataset is compatible with the state of current research on spike-based image recognition. The corresponding spike trains are produced using a range of techniques: rate-based Poisson spike generation, rank order encoding, and recorded output from a silicon retina with both flashing and oscillating input stimuli. In addition, a complementary evaluation methodology is presented to assess both model-level and hardware-level performance. Finally, we demonstrate the use of the dataset and the evaluation methodology using two SNN models to validate the performance of the models and their hardware implementations. With this dataset we hope to (1) promote meaningful comparison between algorithms in the field of neural computation, (2) allow comparison with conventional image recognition methods, (3) provide an assessment of the state of the art in spike-based visual recognition, and (4) help researchers identify future directions and advance the field. PMID:27853419

Benchmarking Spike-Based Visual Recognition: A Dataset and Evaluation.

PubMed

Liu, Qian; Pineda-García, Garibaldi; Stromatias, Evangelos; Serrano-Gotarredona, Teresa; Furber, Steve B

2016-01-01

Today, increasing attention is being paid to research into spike-based neural computation both to gain a better understanding of the brain and to explore biologically-inspired computation. Within this field, the primate visual pathway and its hierarchical organization have been extensively studied. Spiking Neural Networks (SNNs), inspired by the understanding of observed biological structure and function, have been successfully applied to visual recognition and classification tasks. In addition, implementations on neuromorphic hardware have enabled large-scale networks to run in (or even faster than) real time, making spike-based neural vision processing accessible on mobile robots. Neuromorphic sensors such as silicon retinas are able to feed such mobile systems with real-time visual stimuli. A new set of vision benchmarks for spike-based neural processing are now needed to measure progress quantitatively within this rapidly advancing field. We propose that a large dataset of spike-based visual stimuli is needed to provide meaningful comparisons between different systems, and a corresponding evaluation methodology is also required to measure the performance of SNN models and their hardware implementations. In this paper we first propose an initial NE (Neuromorphic Engineering) dataset based on standard computer vision benchmarksand that uses digits from the MNIST database. This dataset is compatible with the state of current research on spike-based image recognition. The corresponding spike trains are produced using a range of techniques: rate-based Poisson spike generation, rank order encoding, and recorded output from a silicon retina with both flashing and oscillating input stimuli. In addition, a complementary evaluation methodology is presented to assess both model-level and hardware-level performance. Finally, we demonstrate the use of the dataset and the evaluation methodology using two SNN models to validate the performance of the models and their hardware implementations. With this dataset we hope to (1) promote meaningful comparison between algorithms in the field of neural computation, (2) allow comparison with conventional image recognition methods, (3) provide an assessment of the state of the art in spike-based visual recognition, and (4) help researchers identify future directions and advance the field.

SEGMENTATION OF MITOCHONDRIA IN ELECTRON MICROSCOPY IMAGES USING ALGEBRAIC CURVES.

PubMed

Seyedhosseini, Mojtaba; Ellisman, Mark H; Tasdizen, Tolga

2013-01-01

High-resolution microscopy techniques have been used to generate large volumes of data with enough details for understanding the complex structure of the nervous system. However, automatic techniques are required to segment cells and intracellular structures in these multi-terabyte datasets and make anatomical analysis possible on a large scale. We propose a fully automated method that exploits both shape information and regional statistics to segment irregularly shaped intracellular structures such as mitochondria in electron microscopy (EM) images. The main idea is to use algebraic curves to extract shape features together with texture features from image patches. Then, these powerful features are used to learn a random forest classifier, which can predict mitochondria locations precisely. Finally, the algebraic curves together with regional information are used to segment the mitochondria at the predicted locations. We demonstrate that our method outperforms the state-of-the-art algorithms in segmentation of mitochondria in EM images.

Presenting Big Data in Google Earth with KML

NASA Astrophysics Data System (ADS)

Hagemark, B.

2006-12-01

KML 2.1 and Google Earth 4 provides support to enable streaming of very large datasets, with "smart" loading of data at multiple levels of resolution and incremental update to previously loaded data. This presentation demonstrates this technology for use with the Google Earth KML geometry and image primitives and shows some techniques and tools for creating this KML.

A reference dataset for deformable image registration spatial accuracy evaluation using the COPDgene study archive

NASA Astrophysics Data System (ADS)

Castillo, Richard; Castillo, Edward; Fuentes, David; Ahmad, Moiz; Wood, Abbie M.; Ludwig, Michelle S.; Guerrero, Thomas

2013-05-01

Landmark point-pairs provide a strategy to assess deformable image registration (DIR) accuracy in terms of the spatial registration of the underlying anatomy depicted in medical images. In this study, we propose to augment a publicly available database (www.dir-lab.com) of medical images with large sets of manually identified anatomic feature pairs between breath-hold computed tomography (BH-CT) images for DIR spatial accuracy evaluation. Ten BH-CT image pairs were randomly selected from the COPDgene study cases. Each patient had received CT imaging of the entire thorax in the supine position at one-fourth dose normal expiration and maximum effort full dose inspiration. Using dedicated in-house software, an imaging expert manually identified large sets of anatomic feature pairs between images. Estimates of inter- and intra-observer spatial variation in feature localization were determined by repeat measurements of multiple observers over subsets of randomly selected features. 7298 anatomic landmark features were manually paired between the 10 sets of images. Quantity of feature pairs per case ranged from 447 to 1172. Average 3D Euclidean landmark displacements varied substantially among cases, ranging from 12.29 (SD: 6.39) to 30.90 (SD: 14.05) mm. Repeat registration of uniformly sampled subsets of 150 landmarks for each case yielded estimates of observer localization error, which ranged in average from 0.58 (SD: 0.87) to 1.06 (SD: 2.38) mm for each case. The additions to the online web database (www.dir-lab.com) described in this work will broaden the applicability of the reference data, providing a freely available common dataset for targeted critical evaluation of DIR spatial accuracy performance in multiple clinical settings. Estimates of observer variance in feature localization suggest consistent spatial accuracy for all observers across both four-dimensional CT and COPDgene patient cohorts.

Faster, efficient and secure collection of research images: the utilization of cloud technology to expand the OMI-DB

NASA Astrophysics Data System (ADS)

Patel, M. N.; Young, K.; Halling-Brown, M. D.

2018-03-01

The demand for medical images for research is ever increasing owing to the rapid rise in novel machine learning approaches for early detection and diagnosis. The OPTIMAM Medical Image Database (OMI-DB)1,2 was created to provide a centralized, fully annotated dataset for research. The database contains both processed and unprocessed images, associated data, annotations and expert-determined ground truths. Since the inception of the database in early 2011, the volume of images and associated data collected has dramatically increased owing to automation of the collection pipeline and inclusion of new sites. Currently, these data are stored at each respective collection site and synced periodically to a central store. This leads to a large data footprint at each site, requiring large physical onsite storage, which is expensive. Here, we propose an update to the OMI-DB collection system, whereby the storage of all the data is automatically transferred to the cloud on collection. This change in the data collection paradigm reduces the reliance of physical servers at each site; allows greater scope for future expansion; and removes the need for dedicated backups and improves security. Moreover, with the number of applications to access the data increasing rapidly with the maturity of the dataset cloud technology facilities faster sharing of data and better auditing of data access. Such updates, although may sound trivial; require substantial modification to the existing pipeline to ensure data integrity and security compliance. Here, we describe the extensions to the OMI-DB collection pipeline and discuss the relative merits of the new system.

MIDG-Emerging grid technologies for multi-site preclinical molecular imaging research communities.

PubMed

Lee, Jasper; Documet, Jorge; Liu, Brent; Park, Ryan; Tank, Archana; Huang, H K

2011-03-01

Molecular imaging is the visualization and identification of specific molecules in anatomy for insight into metabolic pathways, tissue consistency, and tracing of solute transport mechanisms. This paper presents the Molecular Imaging Data Grid (MIDG) which utilizes emerging grid technologies in preclinical molecular imaging to facilitate data sharing and discovery between preclinical molecular imaging facilities and their collaborating investigator institutions to expedite translational sciences research. Grid-enabled archiving, management, and distribution of animal-model imaging datasets help preclinical investigators to monitor, access and share their imaging data remotely, and promote preclinical imaging facilities to share published imaging datasets as resources for new investigators. The system architecture of the Molecular Imaging Data Grid is described in a four layer diagram. A data model for preclinical molecular imaging datasets is also presented based on imaging modalities currently used in a molecular imaging center. The MIDG system components and connectivity are presented. And finally, the workflow steps for grid-based archiving, management, and retrieval of preclincial molecular imaging data are described. Initial performance tests of the Molecular Imaging Data Grid system have been conducted at the USC IPILab using dedicated VMware servers. System connectivity, evaluated datasets, and preliminary results are presented. The results show the system's feasibility, limitations, direction of future research. Translational and interdisciplinary research in medicine is increasingly interested in cellular and molecular biology activity at the preclinical levels, utilizing molecular imaging methods on animal models. The task of integrated archiving, management, and distribution of these preclinical molecular imaging datasets at preclinical molecular imaging facilities is challenging due to disparate imaging systems and multiple off-site investigators. A Molecular Imaging Data Grid design, implementation, and initial evaluation is presented to demonstrate the secure and novel data grid solution for sharing preclinical molecular imaging data across the wide-area-network (WAN).

Imaging Planet Formation Inside the Diffraction Limit

NASA Astrophysics Data System (ADS)

Sallum, Stephanie Elise

For decades, astronomers have used observations of mature planetary systems to constrain planet formation theories, beginning with our own solar system and now the thousands of known exoplanets. Recent advances in instrumentation have given us a direct view of some steps in the planet formation process, such as large-scale protostar and protoplanetary disk features and evolution. However, understanding the details of how planets accrete and interact with their environment requires direct observations of protoplanets themselves. Transition disks, protoplanetary disks with inner clearings that may be caused by forming planets, are the best targets for these studies. Their large distances, compared to the stars normally targeted for direct imaging of exoplanets, make protoplanet detection difficult and necessitate novel imaging techniques. In this dissertation, I describe the results of using non-redundant masking (NRM) to search for forming planets in transition disk clearings. I first present a data reduction pipeline that I wrote to this end, using example datasets and simulations to demonstrate reduction and imaging optimizations. I discuss two transition disk NRM case studies: T Cha and LkCa 15. In the case of T Cha, while we detect significant asymmetries, the data cannot be explained by orbiting companions. The fluxes and orbital motion of the LkCa 15 companion signals, however, can be naturally explained by protoplanets in the disk clearing. I use these datasets and simulated observations to illustrate the effects of scattered light from transition disk material on NRM protoplanet searches. I then demonstrate the utility of the dual-aperture Large Binocular Telescope Interferometer's NRM mode on the bright B[e] star MWC 349A. I discuss the implications of this work for planet formation studies as well as future prospects for NRM and related techniques on next generation instruments.

Discriminating Projections for Estimating Face Age in Wild Images

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

Tokola, Ryan A; Bolme, David S; Ricanek, Karl

2014-01-01

We introduce a novel approach to estimating the age of a human from a single uncontrolled image. Current face age estimation algorithms work well in highly controlled images, and some are robust to changes in illumination, but it is usually assumed that images are close to frontal. This bias is clearly seen in the datasets that are commonly used to evaluate age estimation, which either entirely or mostly consist of frontal images. Using pose-specific projections, our algorithm maps image features into a pose-insensitive latent space that is discriminative with respect to age. Age estimation is then performed using a multi-classmore » SVM. We show that our approach outperforms other published results on the Images of Groups dataset, which is the only age-related dataset with a non-trivial number of off-axis face images, and that we are competitive with recent age estimation algorithms on the mostly-frontal FG-NET dataset. We also experimentally demonstrate that our feature projections introduce insensitivity to pose.« less

a Novel Framework for Remote Sensing Image Scene Classification

NASA Astrophysics Data System (ADS)

Jiang, S.; Zhao, H.; Wu, W.; Tan, Q.

2018-04-01

High resolution remote sensing (HRRS) images scene classification aims to label an image with a specific semantic category. HRRS images contain more details of the ground objects and their spatial distribution patterns than low spatial resolution images. Scene classification can bridge the gap between low-level features and high-level semantics. It can be applied in urban planning, target detection and other fields. This paper proposes a novel framework for HRRS images scene classification. This framework combines the convolutional neural network (CNN) and XGBoost, which utilizes CNN as feature extractor and XGBoost as a classifier. Then, this framework is evaluated on two different HRRS images datasets: UC-Merced dataset and NWPU-RESISC45 dataset. Our framework achieved satisfying accuracies on two datasets, which is 95.57 % and 83.35 % respectively. From the experiments result, our framework has been proven to be effective for remote sensing images classification. Furthermore, we believe this framework will be more practical for further HRRS scene classification, since it costs less time on training stage.

A 2D range Hausdorff approach to 3D facial recognition.

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

Koch, Mark William; Russ, Trina Denise; Little, Charles Quentin

2004-11-01

This paper presents a 3D facial recognition algorithm based on the Hausdorff distance metric. The standard 3D formulation of the Hausdorff matching algorithm has been modified to operate on a 2D range image, enabling a reduction in computation from O(N2) to O(N) without large storage requirements. The Hausdorff distance is known for its robustness to data outliers and inconsistent data between two data sets, making it a suitable choice for dealing with the inherent problems in many 3D datasets due to sensor noise and object self-occlusion. For optimal performance, the algorithm assumes a good initial alignment between probe and templatemore » datasets. However, to minimize the error between two faces, the alignment can be iteratively refined. Results from the algorithm are presented using 3D face images from the Face Recognition Grand Challenge database version 1.0.« less

VIEWCACHE: An incremental pointer-based access method for autonomous interoperable databases

NASA Technical Reports Server (NTRS)

Roussopoulos, N.; Sellis, Timos

1993-01-01

One of the biggest problems facing NASA today is to provide scientists efficient access to a large number of distributed databases. Our pointer-based incremental data base access method, VIEWCACHE, provides such an interface for accessing distributed datasets and directories. VIEWCACHE allows database browsing and search performing inter-database cross-referencing with no actual data movement between database sites. This organization and processing is especially suitable for managing Astrophysics databases which are physically distributed all over the world. Once the search is complete, the set of collected pointers pointing to the desired data are cached. VIEWCACHE includes spatial access methods for accessing image datasets, which provide much easier query formulation by referring directly to the image and very efficient search for objects contained within a two-dimensional window. We will develop and optimize a VIEWCACHE External Gateway Access to database management systems to facilitate database search.

Towards dense volumetric pancreas segmentation in CT using 3D fully convolutional networks

NASA Astrophysics Data System (ADS)

Roth, Holger; Oda, Masahiro; Shimizu, Natsuki; Oda, Hirohisa; Hayashi, Yuichiro; Kitasaka, Takayuki; Fujiwara, Michitaka; Misawa, Kazunari; Mori, Kensaku

2018-03-01

Pancreas segmentation in computed tomography imaging has been historically difficult for automated methods because of the large shape and size variations between patients. In this work, we describe a custom-build 3D fully convolutional network (FCN) that can process a 3D image including the whole pancreas and produce an automatic segmentation. We investigate two variations of the 3D FCN architecture; one with concatenation and one with summation skip connections to the decoder part of the network. We evaluate our methods on a dataset from a clinical trial with gastric cancer patients, including 147 contrast enhanced abdominal CT scans acquired in the portal venous phase. Using the summation architecture, we achieve an average Dice score of 89.7 +/- 3.8 (range [79.8, 94.8])% in testing, achieving the new state-of-the-art performance in pancreas segmentation on this dataset.

Imaging samples larger than the field of view: the SLS experience

NASA Astrophysics Data System (ADS)

Vogiatzis Oikonomidis, Ioannis; Lovric, Goran; Cremona, Tiziana P.; Arcadu, Filippo; Patera, Alessandra; Schittny, Johannes C.; Stampanoni, Marco

2017-06-01

Volumetric datasets with micrometer spatial and sub-second temporal resolutions are nowadays routinely acquired using synchrotron X-ray tomographic microscopy (SRXTM). Although SRXTM technology allows the examination of multiple samples with short scan times, many specimens are larger than the field-of-view (FOV) provided by the detector. The extension of the FOV in the direction perpendicular to the rotation axis remains non-trivial. We present a method that can efficiently increase the FOV merging volumetric datasets obtained by region-of-interest tomographies in different 3D positions of the sample with a minimal amount of artefacts and with the ability to handle large amounts of data. The method has been successfully applied for the three-dimensional imaging of a small number of mouse lung acini of intact animals, where pixel sizes down to the micrometer range and short exposure times are required.

Computational optical tomography using 3-D deep convolutional neural networks

NASA Astrophysics Data System (ADS)

Nguyen, Thanh; Bui, Vy; Nehmetallah, George

2018-04-01

Deep convolutional neural networks (DCNNs) offer a promising performance for many image processing areas, such as super-resolution, deconvolution, image classification, denoising, and segmentation, with outstanding results. Here, we develop for the first time, to our knowledge, a method to perform 3-D computational optical tomography using 3-D DCNN. A simulated 3-D phantom dataset was first constructed and converted to a dataset of phase objects imaged on a spatial light modulator. For each phase image in the dataset, the corresponding diffracted intensity image was experimentally recorded on a CCD. We then experimentally demonstrate the ability of the developed 3-D DCNN algorithm to solve the inverse problem by reconstructing the 3-D index of refraction distributions of test phantoms from the dataset from their corresponding diffraction patterns.

Epithelium-Stroma Classification via Convolutional Neural Networks and Unsupervised Domain Adaptation in Histopathological Images.

PubMed

Huang, Yue; Zheng, Han; Liu, Chi; Ding, Xinghao; Rohde, Gustavo K

2017-11-01

Epithelium-stroma classification is a necessary preprocessing step in histopathological image analysis. Current deep learning based recognition methods for histology data require collection of large volumes of labeled data in order to train a new neural network when there are changes to the image acquisition procedure. However, it is extremely expensive for pathologists to manually label sufficient volumes of data for each pathology study in a professional manner, which results in limitations in real-world applications. A very simple but effective deep learning method, that introduces the concept of unsupervised domain adaptation to a simple convolutional neural network (CNN), has been proposed in this paper. Inspired by transfer learning, our paper assumes that the training data and testing data follow different distributions, and there is an adaptation operation to more accurately estimate the kernels in CNN in feature extraction, in order to enhance performance by transferring knowledge from labeled data in source domain to unlabeled data in target domain. The model has been evaluated using three independent public epithelium-stroma datasets by cross-dataset validations. The experimental results demonstrate that for epithelium-stroma classification, the proposed framework outperforms the state-of-the-art deep neural network model, and it also achieves better performance than other existing deep domain adaptation methods. The proposed model can be considered to be a better option for real-world applications in histopathological image analysis, since there is no longer a requirement for large-scale labeled data in each specified domain.

A GPU-Accelerated Approach for Feature Tracking in Time-Varying Imagery Datasets.

PubMed

Peng, Chao; Sahani, Sandip; Rushing, John

2017-10-01

We propose a novel parallel connected component labeling (CCL) algorithm along with efficient out-of-core data management to detect and track feature regions of large time-varying imagery datasets. Our approach contributes to the big data field with parallel algorithms tailored for GPU architectures. We remove the data dependency between frames and achieve pixel-level parallelism. Due to the large size, the entire dataset cannot fit into cached memory. Frames have to be streamed through the memory hierarchy (disk to CPU main memory and then to GPU memory), partitioned, and processed as batches, where each batch is small enough to fit into the GPU. To reconnect the feature regions that are separated due to data partitioning, we present a novel batch merging algorithm to extract the region connection information across multiple batches in a parallel fashion. The information is organized in a memory-efficient structure and supports fast indexing on the GPU. Our experiment uses a commodity workstation equipped with a single GPU. The results show that our approach can efficiently process a weather dataset composed of terabytes of time-varying radar images. The advantages of our approach are demonstrated by comparing to the performance of an efficient CPU cluster implementation which is being used by the weather scientists.

3D/2D model-to-image registration by imitation learning for cardiac procedures.

PubMed

Toth, Daniel; Miao, Shun; Kurzendorfer, Tanja; Rinaldi, Christopher A; Liao, Rui; Mansi, Tommaso; Rhode, Kawal; Mountney, Peter

2018-05-12

In cardiac interventions, such as cardiac resynchronization therapy (CRT), image guidance can be enhanced by involving preoperative models. Multimodality 3D/2D registration for image guidance, however, remains a significant research challenge for fundamentally different image data, i.e., MR to X-ray. Registration methods must account for differences in intensity, contrast levels, resolution, dimensionality, field of view. Furthermore, same anatomical structures may not be visible in both modalities. Current approaches have focused on developing modality-specific solutions for individual clinical use cases, by introducing constraints, or identifying cross-modality information manually. Machine learning approaches have the potential to create more general registration platforms. However, training image to image methods would require large multimodal datasets and ground truth for each target application. This paper proposes a model-to-image registration approach instead, because it is common in image-guided interventions to create anatomical models for diagnosis, planning or guidance prior to procedures. An imitation learning-based method, trained on 702 datasets, is used to register preoperative models to intraoperative X-ray images. Accuracy is demonstrated on cardiac models and artificial X-rays generated from CTs. The registration error was [Formula: see text] on 1000 test cases, superior to that of manual ([Formula: see text]) and gradient-based ([Formula: see text]) registration. High robustness is shown in 19 clinical CRT cases. Besides the proposed methods feasibility in a clinical environment, evaluation has shown good accuracy and high robustness indicating that it could be applied in image-guided interventions.

3D fully convolutional networks for subcortical segmentation in MRI: A large-scale study.

PubMed

Dolz, Jose; Desrosiers, Christian; Ben Ayed, Ismail

2018-04-15

This study investigates a 3D and fully convolutional neural network (CNN) for subcortical brain structure segmentation in MRI. 3D CNN architectures have been generally avoided due to their computational and memory requirements during inference. We address the problem via small kernels, allowing deeper architectures. We further model both local and global context by embedding intermediate-layer outputs in the final prediction, which encourages consistency between features extracted at different scales and embeds fine-grained information directly in the segmentation process. Our model is efficiently trained end-to-end on a graphics processing unit (GPU), in a single stage, exploiting the dense inference capabilities of fully CNNs. We performed comprehensive experiments over two publicly available datasets. First, we demonstrate a state-of-the-art performance on the ISBR dataset. Then, we report a large-scale multi-site evaluation over 1112 unregistered subject datasets acquired from 17 different sites (ABIDE dataset), with ages ranging from 7 to 64 years, showing that our method is robust to various acquisition protocols, demographics and clinical factors. Our method yielded segmentations that are highly consistent with a standard atlas-based approach, while running in a fraction of the time needed by atlas-based methods and avoiding registration/normalization steps. This makes it convenient for massive multi-site neuroanatomical imaging studies. To the best of our knowledge, our work is the first to study subcortical structure segmentation on such large-scale and heterogeneous data. Copyright © 2017 Elsevier Inc. All rights reserved.

A Summary of Large Raindrop Observations from GPM GV Field Campaigns

NASA Technical Reports Server (NTRS)

Gatlin, Patrick N.; Petersen, Walter; Tokay, Ali; Thurai, Merhala; Bringi, V. N.; Carey, Lawrence; Wingo, Matthew

2013-01-01

NASA's Global Precipitation Measurement Mission (GPM) has conducted as series of Ground Validation (GV) studies to assist algorithm development for the GPM core satellite. Characterizing the drop size distribution (DSD) for different types of precipitation systems is critical in order to accurately estimate precipitation across the majority of the planet. Thus far, GV efforts have sampled DSDs in a variety of precipitation systems from Finland to Oklahoma. This dataset consists of over 33 million raindrops sampled by GPM GV's two-dimensional video disdrometers (2DVD) and includes RSD observations from the LPVEx, MC3E, GCPEx, HyMEx and IFloodS campaigns as well as from GV sites in Huntsville, AL and Wallops Island, VA. This study focuses on the larger end of the raindrop size spectrum, which greatly influences radar reflectivity and has implications for moment estimation. Thus knowledge of the maximum diameter is critical to GPM algorithm development. There are over 24,000 raindrops exceeding 5 mm in diameter contained within this disdrometer dataset. The largest raindrops in the 2DVD dataset (>7-8 mm in diameter) are found within intense convective thunderstorms, and their origins are believed to be hailstones. In stratiform rainfall, large raindrops have also been found to fall from lower and thicker melting layers. The 2DVD dataset will be combined with that collected by dual-polarimetric radar and aircraft particle imaging probes to "follow" the vertical evolution of the DSD tail (i.e., retrace the large drops from the surface to their origins aloft).

Individual Brain Charting, a high-resolution fMRI dataset for cognitive mapping.

PubMed

Pinho, Ana Luísa; Amadon, Alexis; Ruest, Torsten; Fabre, Murielle; Dohmatob, Elvis; Denghien, Isabelle; Ginisty, Chantal; Becuwe-Desmidt, Séverine; Roger, Séverine; Laurier, Laurence; Joly-Testault, Véronique; Médiouni-Cloarec, Gaëlle; Doublé, Christine; Martins, Bernadette; Pinel, Philippe; Eger, Evelyn; Varoquaux, Gaël; Pallier, Christophe; Dehaene, Stanislas; Hertz-Pannier, Lucie; Thirion, Bertrand

2018-06-12

Functional Magnetic Resonance Imaging (fMRI) has furthered brain mapping on perceptual, motor, as well as higher-level cognitive functions. However, to date, no data collection has systematically addressed the functional mapping of cognitive mechanisms at a fine spatial scale. The Individual Brain Charting (IBC) project stands for a high-resolution multi-task fMRI dataset that intends to provide the objective basis toward a comprehensive functional atlas of the human brain. The data refer to a cohort of 12 participants performing many different tasks. The large amount of task-fMRI data on the same subjects yields a precise mapping of the underlying functions, free from both inter-subject and inter-site variability. The present article gives a detailed description of the first release of the IBC dataset. It comprises a dozen of tasks, addressing both low- and high- level cognitive functions. This openly available dataset is thus intended to become a reference for cognitive brain mapping.

Joint Blind Source Separation by Multi-set Canonical Correlation Analysis

PubMed Central

Li, Yi-Ou; Adalı, Tülay; Wang, Wei; Calhoun, Vince D

2009-01-01

In this work, we introduce a simple and effective scheme to achieve joint blind source separation (BSS) of multiple datasets using multi-set canonical correlation analysis (M-CCA) [1]. We first propose a generative model of joint BSS based on the correlation of latent sources within and between datasets. We specify source separability conditions, and show that, when the conditions are satisfied, the group of corresponding sources from each dataset can be jointly extracted by M-CCA through maximization of correlation among the extracted sources. We compare source separation performance of the M-CCA scheme with other joint BSS methods and demonstrate the superior performance of the M-CCA scheme in achieving joint BSS for a large number of datasets, group of corresponding sources with heterogeneous correlation values, and complex-valued sources with circular and non-circular distributions. We apply M-CCA to analysis of functional magnetic resonance imaging (fMRI) data from multiple subjects and show its utility in estimating meaningful brain activations from a visuomotor task. PMID:20221319

The Montage architecture for grid-enabled science processing of large, distributed datasets

NASA Technical Reports Server (NTRS)

Jacob, Joseph C.; Katz, Daniel S .; Prince, Thomas; Berriman, Bruce G.; Good, John C.; Laity, Anastasia C.; Deelman, Ewa; Singh, Gurmeet; Su, Mei-Hui

2004-01-01

Montage is an Earth Science Technology Office (ESTO) Computational Technologies (CT) Round III Grand Challenge investigation to deploy a portable, compute-intensive, custom astronomical image mosaicking service for the National Virtual Observatory (NVO). Although Montage is developing a compute- and data-intensive service for the astronomy community, we are also helping to address a problem that spans both Earth and Space science, namely how to efficiently access and process multi-terabyte, distributed datasets. In both communities, the datasets are massive, and are stored in distributed archives that are, in most cases, remote from the available Computational resources. Therefore, state of the art computational grid technologies are a key element of the Montage portal architecture. This paper describes the aspects of the Montage design that are applicable to both the Earth and Space science communities.

Application of neuroanatomical ontologies for neuroimaging data annotation.

PubMed

Turner, Jessica A; Mejino, Jose L V; Brinkley, James F; Detwiler, Landon T; Lee, Hyo Jong; Martone, Maryann E; Rubin, Daniel L

2010-01-01

The annotation of functional neuroimaging results for data sharing and re-use is particularly challenging, due to the diversity of terminologies of neuroanatomical structures and cortical parcellation schemes. To address this challenge, we extended the Foundational Model of Anatomy Ontology (FMA) to include cytoarchitectural, Brodmann area labels, and a morphological cortical labeling scheme (e.g., the part of Brodmann area 6 in the left precentral gyrus). This representation was also used to augment the neuroanatomical axis of RadLex, the ontology for clinical imaging. The resulting neuroanatomical ontology contains explicit relationships indicating which brain regions are "part of" which other regions, across cytoarchitectural and morphological labeling schemas. We annotated a large functional neuroimaging dataset with terms from the ontology and applied a reasoning engine to analyze this dataset in conjunction with the ontology, and achieved successful inferences from the most specific level (e.g., how many subjects showed activation in a subpart of the middle frontal gyrus) to more general (how many activations were found in areas connected via a known white matter tract?). In summary, we have produced a neuroanatomical ontology that harmonizes several different terminologies of neuroanatomical structures and cortical parcellation schemes. This neuroanatomical ontology is publicly available as a view of FMA at the Bioportal website. The ontological encoding of anatomic knowledge can be exploited by computer reasoning engines to make inferences about neuroanatomical relationships described in imaging datasets using different terminologies. This approach could ultimately enable knowledge discovery from large, distributed fMRI studies or medical record mining.

MRI-based intelligence quotient (IQ) estimation with sparse learning.

PubMed

Wang, Liye; Wee, Chong-Yaw; Suk, Heung-Il; Tang, Xiaoying; Shen, Dinggang

2015-01-01

In this paper, we propose a novel framework for IQ estimation using Magnetic Resonance Imaging (MRI) data. In particular, we devise a new feature selection method based on an extended dirty model for jointly considering both element-wise sparsity and group-wise sparsity. Meanwhile, due to the absence of large dataset with consistent scanning protocols for the IQ estimation, we integrate multiple datasets scanned from different sites with different scanning parameters and protocols. In this way, there is large variability in these different datasets. To address this issue, we design a two-step procedure for 1) first identifying the possible scanning site for each testing subject and 2) then estimating the testing subject's IQ by using a specific estimator designed for that scanning site. We perform two experiments to test the performance of our method by using the MRI data collected from 164 typically developing children between 6 and 15 years old. In the first experiment, we use a multi-kernel Support Vector Regression (SVR) for estimating IQ values, and obtain an average correlation coefficient of 0.718 and also an average root mean square error of 8.695 between the true IQs and the estimated ones. In the second experiment, we use a single-kernel SVR for IQ estimation, and achieve an average correlation coefficient of 0.684 and an average root mean square error of 9.166. All these results show the effectiveness of using imaging data for IQ prediction, which is rarely done in the field according to our knowledge.

An automated, high-throughput plant phenotyping system using machine learning-based plant segmentation and image analysis.

PubMed

Lee, Unseok; Chang, Sungyul; Putra, Gian Anantrio; Kim, Hyoungseok; Kim, Dong Hwan

2018-01-01

A high-throughput plant phenotyping system automatically observes and grows many plant samples. Many plant sample images are acquired by the system to determine the characteristics of the plants (populations). Stable image acquisition and processing is very important to accurately determine the characteristics. However, hardware for acquiring plant images rapidly and stably, while minimizing plant stress, is lacking. Moreover, most software cannot adequately handle large-scale plant imaging. To address these problems, we developed a new, automated, high-throughput plant phenotyping system using simple and robust hardware, and an automated plant-imaging-analysis pipeline consisting of machine-learning-based plant segmentation. Our hardware acquires images reliably and quickly and minimizes plant stress. Furthermore, the images are processed automatically. In particular, large-scale plant-image datasets can be segmented precisely using a classifier developed using a superpixel-based machine-learning algorithm (Random Forest), and variations in plant parameters (such as area) over time can be assessed using the segmented images. We performed comparative evaluations to identify an appropriate learning algorithm for our proposed system, and tested three robust learning algorithms. We developed not only an automatic analysis pipeline but also a convenient means of plant-growth analysis that provides a learning data interface and visualization of plant growth trends. Thus, our system allows end-users such as plant biologists to analyze plant growth via large-scale plant image data easily.

Reference-tissue correction of T2-weighted signal intensity for prostate cancer detection

NASA Astrophysics Data System (ADS)

Peng, Yahui; Jiang, Yulei; Oto, Aytekin

2014-03-01

The purpose of this study was to investigate whether correction with respect to reference tissue of T2-weighted MRimage signal intensity (SI) improves its effectiveness for classification of regions of interest (ROIs) as prostate cancer (PCa) or normal prostatic tissue. Two image datasets collected retrospectively were used in this study: 71 cases acquired with GE scanners (dataset A), and 59 cases acquired with Philips scanners (dataset B). Through a consensus histology- MR correlation review, 175 PCa and 108 normal-tissue ROIs were identified and drawn manually. Reference-tissue ROIs were selected in each case from the levator ani muscle, urinary bladder, and pubic bone. T2-weighted image SI was corrected as the ratio of the average T2-weighted image SI within an ROI to that of a reference-tissue ROI. Area under the receiver operating characteristic curve (AUC) was used to evaluate the effectiveness of T2-weighted image SIs for differentiation of PCa from normal-tissue ROIs. AUC (+/- standard error) for uncorrected T2-weighted image SIs was 0.78+/-0.04 (datasets A) and 0.65+/-0.05 (datasets B). AUC for corrected T2-weighted image SIs with respect to muscle, bladder, and bone reference was 0.77+/-0.04 (p=1.0), 0.77+/-0.04 (p=1.0), and 0.75+/-0.04 (p=0.8), respectively, for dataset A; and 0.81+/-0.04 (p=0.002), 0.78+/-0.04 (p<0.001), and 0.79+/-0.04 (p<0.001), respectively, for dataset B. Correction in reference to the levator ani muscle yielded the most consistent results between GE and Phillips images. Correction of T2-weighted image SI in reference to three types of extra-prostatic tissue can improve its effectiveness for differentiation of PCa from normal-tissue ROIs, and correction in reference to the levator ani muscle produces consistent T2-weighted image SIs between GE and Phillips MR images.

Comparison of k-means related clustering methods for nuclear medicine images segmentation

NASA Astrophysics Data System (ADS)

Borys, Damian; Bzowski, Pawel; Danch-Wierzchowska, Marta; Psiuk-Maksymowicz, Krzysztof

2017-03-01

In this paper, we evaluate the performance of SURF descriptor for high resolution satellite imagery (HRSI) retrieval through a BoVW model on a land-use/land-cover (LULC) dataset. Local feature approaches such as SIFT and SURF descriptors can deal with a large variation of scale, rotation and illumination of the images, providing, therefore, a better discriminative power and retrieval efficiency than global features, especially for HRSI which contain a great range of objects and spatial patterns. Moreover, we combine SURF and color features to improve the retrieval accuracy, and we propose to learn a category-specific dictionary for each image category which results in a more discriminative image representation and boosts the image retrieval performance.

High resolution satellite image indexing and retrieval using SURF features and bag of visual words

NASA Astrophysics Data System (ADS)

Bouteldja, Samia; Kourgli, Assia

2017-03-01

In this paper, we evaluate the performance of SURF descriptor for high resolution satellite imagery (HRSI) retrieval through a BoVW model on a land-use/land-cover (LULC) dataset. Local feature approaches such as SIFT and SURF descriptors can deal with a large variation of scale, rotation and illumination of the images, providing, therefore, a better discriminative power and retrieval efficiency than global features, especially for HRSI which contain a great range of objects and spatial patterns. Moreover, we combine SURF and color features to improve the retrieval accuracy, and we propose to learn a category-specific dictionary for each image category which results in a more discriminative image representation and boosts the image retrieval performance.

Image Augmentation for Object Image Classification Based On Combination of Pre-Trained CNN and SVM

NASA Astrophysics Data System (ADS)

Shima, Yoshihiro

2018-04-01

Neural networks are a powerful means of classifying object images. The proposed image category classification method for object images combines convolutional neural networks (CNNs) and support vector machines (SVMs). A pre-trained CNN, called Alex-Net, is used as a pattern-feature extractor. Alex-Net is pre-trained for the large-scale object-image dataset ImageNet. Instead of training, Alex-Net, pre-trained for ImageNet is used. An SVM is used as trainable classifier. The feature vectors are passed to the SVM from Alex-Net. The STL-10 dataset are used as object images. The number of classes is ten. Training and test samples are clearly split. STL-10 object images are trained by the SVM with data augmentation. We use the pattern transformation method with the cosine function. We also apply some augmentation method such as rotation, skewing and elastic distortion. By using the cosine function, the original patterns were left-justified, right-justified, top-justified, or bottom-justified. Patterns were also center-justified and enlarged. Test error rate is decreased by 0.435 percentage points from 16.055% by augmentation with cosine transformation. Error rates are increased by other augmentation method such as rotation, skewing and elastic distortion, compared without augmentation. Number of augmented data is 30 times that of the original STL-10 5K training samples. Experimental test error rate for the test 8k STL-10 object images was 15.620%, which shows that image augmentation is effective for image category classification.

Multi-source Geospatial Data Analysis with Google Earth Engine

NASA Astrophysics Data System (ADS)

Erickson, T.

2014-12-01

The Google Earth Engine platform is a cloud computing environment for data analysis that combines a public data catalog with a large-scale computational facility optimized for parallel processing of geospatial data. The data catalog is a multi-petabyte archive of georeferenced datasets that include images from Earth observing satellite and airborne sensors (examples: USGS Landsat, NASA MODIS, USDA NAIP), weather and climate datasets, and digital elevation models. Earth Engine supports both a just-in-time computation model that enables real-time preview and debugging during algorithm development for open-ended data exploration, and a batch computation mode for applying algorithms over large spatial and temporal extents. The platform automatically handles many traditionally-onerous data management tasks, such as data format conversion, reprojection, and resampling, which facilitates writing algorithms that combine data from multiple sensors and/or models. Although the primary use of Earth Engine, to date, has been the analysis of large Earth observing satellite datasets, the computational platform is generally applicable to a wide variety of use cases that require large-scale geospatial data analyses. This presentation will focus on how Earth Engine facilitates the analysis of geospatial data streams that originate from multiple separate sources (and often communities) and how it enables collaboration during algorithm development and data exploration. The talk will highlight current projects/analyses that are enabled by this functionality.https://earthengine.google.org

Assessment of the NASA-USGS Global Land Survey (GLS) Datasets

USGS Publications Warehouse

Gutman, Garik; Huang, Chengquan; Chander, Gyanesh; Noojipady, Praveen; Masek, Jeffery G.

2013-01-01

The Global Land Survey (GLS) datasets are a collection of orthorectified, cloud-minimized Landsat-type satellite images, providing near complete coverage of the global land area decadally since the early 1970s. The global mosaics are centered on 1975, 1990, 2000, 2005, and 2010, and consist of data acquired from four sensors: Enhanced Thematic Mapper Plus, Thematic Mapper, Multispectral Scanner, and Advanced Land Imager. The GLS datasets have been widely used in land-cover and land-use change studies at local, regional, and global scales. This study evaluates the GLS datasets with respect to their spatial coverage, temporal consistency, geodetic accuracy, radiometric calibration consistency, image completeness, extent of cloud contamination, and residual gaps. In general, the three latest GLS datasets are of a better quality than the GLS-1990 and GLS-1975 datasets, with most of the imagery (85%) having cloud cover of less than 10%, the acquisition years clustered much more tightly around their target years, better co-registration relative to GLS-2000, and better radiometric absolute calibration. Probably, the most significant impediment to scientific use of the datasets is the variability of image phenology (i.e., acquisition day of year). This paper provides end-users with an assessment of the quality of the GLS datasets for specific applications, and where possible, suggestions for mitigating their deficiencies.

Deformable templates guided discriminative models for robust 3D brain MRI segmentation.

PubMed

Liu, Cheng-Yi; Iglesias, Juan Eugenio; Tu, Zhuowen

2013-10-01

Automatically segmenting anatomical structures from 3D brain MRI images is an important task in neuroimaging. One major challenge is to design and learn effective image models accounting for the large variability in anatomy and data acquisition protocols. A deformable template is a type of generative model that attempts to explicitly match an input image with a template (atlas), and thus, they are robust against global intensity changes. On the other hand, discriminative models combine local image features to capture complex image patterns. In this paper, we propose a robust brain image segmentation algorithm that fuses together deformable templates and informative features. It takes advantage of the adaptation capability of the generative model and the classification power of the discriminative models. The proposed algorithm achieves both robustness and efficiency, and can be used to segment brain MRI images with large anatomical variations. We perform an extensive experimental study on four datasets of T1-weighted brain MRI data from different sources (1,082 MRI scans in total) and observe consistent improvement over the state-of-the-art systems.

CImbinator: a web-based tool for drug synergy analysis in small- and large-scale datasets.

PubMed

Flobak, Åsmund; Vazquez, Miguel; Lægreid, Astrid; Valencia, Alfonso

2017-08-01

Drug synergies are sought to identify combinations of drugs particularly beneficial. User-friendly software solutions that can assist analysis of large-scale datasets are required. CImbinator is a web-service that can aid in batch-wise and in-depth analyzes of data from small-scale and large-scale drug combination screens. CImbinator offers to quantify drug combination effects, using both the commonly employed median effect equation, as well as advanced experimental mathematical models describing dose response relationships. CImbinator is written in Ruby and R. It uses the R package drc for advanced drug response modeling. CImbinator is available at http://cimbinator.bioinfo.cnio.es , the source-code is open and available at https://github.com/Rbbt-Workflows/combination_index . A Docker image is also available at https://hub.docker.com/r/mikisvaz/rbbt-ci_mbinator/ . asmund.flobak@ntnu.no or miguel.vazquez@cnio.es. Supplementary data are available at Bioinformatics online. © The Author(s) 2017. Published by Oxford University Press.

A test-retest dataset for assessing long-term reliability of brain morphology and resting-state brain activity.

PubMed

Huang, Lijie; Huang, Taicheng; Zhen, Zonglei; Liu, Jia

2016-03-15

We present a test-retest dataset for evaluation of long-term reliability of measures from structural and resting-state functional magnetic resonance imaging (sMRI and rfMRI) scans. The repeated scan dataset was collected from 61 healthy adults in two sessions using highly similar imaging parameters at an interval of 103-189 days. However, as the imaging parameters were not completely identical, the reliability estimated from this dataset shall reflect the lower bounds of the true reliability of sMRI/rfMRI measures. Furthermore, in conjunction with other test-retest datasets, our dataset may help explore the impact of different imaging parameters on reliability of sMRI/rfMRI measures, which is especially critical for assessing datasets collected from multiple centers. In addition, intelligence quotient (IQ) was measured for each participant using Raven's Advanced Progressive Matrices. The data can thus be used for purposes other than assessing reliability of sMRI/rfMRI alone. For example, data from each single session could be used to associate structural and functional measures of the brain with the IQ metrics to explore brain-IQ association.

Stationary Wavelet Transform and AdaBoost with SVM Based Pathological Brain Detection in MRI Scanning.

PubMed

Nayak, Deepak Ranjan; Dash, Ratnakar; Majhi, Banshidhar

2017-01-01

This paper presents an automatic classification system for segregating pathological brain from normal brains in magnetic resonance imaging scanning. The proposed system employs contrast limited adaptive histogram equalization scheme to enhance the diseased region in brain MR images. Two-dimensional stationary wavelet transform is harnessed to extract features from the preprocessed images. The feature vector is constructed using the energy and entropy values, computed from the level- 2 SWT coefficients. Then, the relevant and uncorrelated features are selected using symmetric uncertainty ranking filter. Subsequently, the selected features are given input to the proposed AdaBoost with support vector machine classifier, where SVM is used as the base classifier of AdaBoost algorithm. To validate the proposed system, three standard MR image datasets, Dataset-66, Dataset-160, and Dataset- 255 have been utilized. The 5 runs of k-fold stratified cross validation results indicate the suggested scheme offers better performance than other existing schemes in terms of accuracy and number of features. The proposed system earns ideal classification over Dataset-66 and Dataset-160; whereas, for Dataset- 255, an accuracy of 99.45% is achieved. Copyright© Bentham Science Publishers; For any queries, please email at epub@benthamscience.org.

Object detection approach using generative sparse, hierarchical networks with top-down and lateral connections for combining texture/color detection and shape/contour detection

DOEpatents

Paiton, Dylan M.; Kenyon, Garrett T.; Brumby, Steven P.; Schultz, Peter F.; George, John S.

2015-07-28

An approach to detecting objects in an image dataset may combine texture/color detection, shape/contour detection, and/or motion detection using sparse, generative, hierarchical models with lateral and top-down connections. A first independent representation of objects in an image dataset may be produced using a color/texture detection algorithm. A second independent representation of objects in the image dataset may be produced using a shape/contour detection algorithm. A third independent representation of objects in the image dataset may be produced using a motion detection algorithm. The first, second, and third independent representations may then be combined into a single coherent output using a combinatorial algorithm.

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

Xia, Kelin; Zhao, Zhixiong; Wei, Guo-Wei, E-mail: wei@math.msu.edu

Although persistent homology has emerged as a promising tool for the topological simplification of complex data, it is computationally intractable for large datasets. We introduce multiresolution persistent homology to handle excessively large datasets. We match the resolution with the scale of interest so as to represent large scale datasets with appropriate resolution. We utilize flexibility-rigidity index to access the topological connectivity of the data set and define a rigidity density for the filtration analysis. By appropriately tuning the resolution of the rigidity density, we are able to focus the topological lens on the scale of interest. The proposed multiresolution topologicalmore » analysis is validated by a hexagonal fractal image which has three distinct scales. We further demonstrate the proposed method for extracting topological fingerprints from DNA molecules. In particular, the topological persistence of a virus capsid with 273 780 atoms is successfully analyzed which would otherwise be inaccessible to the normal point cloud method and unreliable by using coarse-grained multiscale persistent homology. The proposed method has also been successfully applied to the protein domain classification, which is the first time that persistent homology is used for practical protein domain analysis, to our knowledge. The proposed multiresolution topological method has potential applications in arbitrary data sets, such as social networks, biological networks, and graphs.« less

Spectrum image analysis tool - A flexible MATLAB solution to analyze EEL and CL spectrum images.

PubMed

Schmidt, Franz-Philipp; Hofer, Ferdinand; Krenn, Joachim R

2017-02-01

Spectrum imaging techniques, gaining simultaneously structural (image) and spectroscopic data, require appropriate and careful processing to extract information of the dataset. In this article we introduce a MATLAB based software that uses three dimensional data (EEL/CL spectrum image in dm3 format (Gatan Inc.'s DigitalMicrograph ® )) as input. A graphical user interface enables a fast and easy mapping of spectral dependent images and position dependent spectra. First, data processing such as background subtraction, deconvolution and denoising, second, multiple display options including an EEL/CL moviemaker and, third, the applicability on a large amount of data sets with a small work load makes this program an interesting tool to visualize otherwise hidden details. Copyright © 2016 Elsevier Ltd. All rights reserved.

The MIND PALACE: A Multi-Spectral Imaging and Spectroscopy Database for Planetary Science

NASA Astrophysics Data System (ADS)

Eshelman, E.; Doloboff, I.; Hara, E. K.; Uckert, K.; Sapers, H. M.; Abbey, W.; Beegle, L. W.; Bhartia, R.

2017-12-01

The Multi-Instrument Database (MIND) is the web-based home to a well-characterized set of analytical data collected by a suite of deep-UV fluorescence/Raman instruments built at the Jet Propulsion Laboratory (JPL). Samples derive from a growing body of planetary surface analogs, mineral and microbial standards, meteorites, spacecraft materials, and other astrobiologically relevant materials. In addition to deep-UV spectroscopy, datasets stored in MIND are obtained from a variety of analytical techniques obtained over multiple spatial and spectral scales including electron microscopy, optical microscopy, infrared spectroscopy, X-ray fluorescence, and direct fluorescence imaging. Multivariate statistical analysis techniques, primarily Principal Component Analysis (PCA), are used to guide interpretation of these large multi-analytical spectral datasets. Spatial co-referencing of integrated spectral/visual maps is performed using QGIS (geographic information system software). Georeferencing techniques transform individual instrument data maps into a layered co-registered data cube for analysis across spectral and spatial scales. The body of data in MIND is intended to serve as a permanent, reliable, and expanding database of deep-UV spectroscopy datasets generated by this unique suite of JPL-based instruments on samples of broad planetary science interest.

Thermal Structure and Dynamics of Saturn's Northern Springtime Disturbance

NASA Technical Reports Server (NTRS)

Fletcher, Leigh N.; Hesman, Brigette E.; Irwin, Patrick G.; Baines, Kevin H.; Momary, Thomas W.; SanchezLavega, Agustin; Flasar, F. Michael; Read, Peter L.; Orton, Glenn S.; SimonMiller, Amy;

Range and Panoramic Image Fusion Into a Textured Range Image for Culture Heritage Documentation

NASA Astrophysics Data System (ADS)

Bila, Z.; Reznicek, J.; Pavelka, K.

2013-07-01

This paper deals with a fusion of range and panoramic images, where the range image is acquired by a 3D laser scanner and the panoramic image is acquired with a digital still camera mounted on a panoramic head and tripod. The fused resulting dataset, called "textured range image", provides more reliable information about the investigated object for conservators and historians, than using both datasets separately. A simple example of fusion of a range and panoramic images, both obtained in St. Francis Xavier Church in town Opařany, is given here. Firstly, we describe the process of data acquisition, then the processing of both datasets into a proper format for following fusion and the process of fusion. The process of fusion can be divided into a two main parts: transformation and remapping. In the first, transformation, part, both images are related by matching similar features detected on both images with a proper detector, which results in transformation matrix enabling transformation of the range image onto a panoramic image. Then, the range data are remapped from the range image space into a panoramic image space and stored as an additional "range" channel. The process of image fusion is validated by comparing similar features extracted on both datasets.

Knowledge-Guided Robust MRI Brain Extraction for Diverse Large-Scale Neuroimaging Studies on Humans and Non-Human Primates

PubMed Central

Wang, Yaping; Nie, Jingxin; Yap, Pew-Thian; Li, Gang; Shi, Feng; Geng, Xiujuan; Guo, Lei; Shen, Dinggang

2014-01-01

Accurate and robust brain extraction is a critical step in most neuroimaging analysis pipelines. In particular, for the large-scale multi-site neuroimaging studies involving a significant number of subjects with diverse age and diagnostic groups, accurate and robust extraction of the brain automatically and consistently is highly desirable. In this paper, we introduce population-specific probability maps to guide the brain extraction of diverse subject groups, including both healthy and diseased adult human populations, both developing and aging human populations, as well as non-human primates. Specifically, the proposed method combines an atlas-based approach, for coarse skull-stripping, with a deformable-surface-based approach that is guided by local intensity information and population-specific prior information learned from a set of real brain images for more localized refinement. Comprehensive quantitative evaluations were performed on the diverse large-scale populations of ADNI dataset with over 800 subjects (55∼90 years of age, multi-site, various diagnosis groups), OASIS dataset with over 400 subjects (18∼96 years of age, wide age range, various diagnosis groups), and NIH pediatrics dataset with 150 subjects (5∼18 years of age, multi-site, wide age range as a complementary age group to the adult dataset). The results demonstrate that our method consistently yields the best overall results across almost the entire human life span, with only a single set of parameters. To demonstrate its capability to work on non-human primates, the proposed method is further evaluated using a rhesus macaque dataset with 20 subjects. Quantitative comparisons with popularly used state-of-the-art methods, including BET, Two-pass BET, BET-B, BSE, HWA, ROBEX and AFNI, demonstrate that the proposed method performs favorably with superior performance on all testing datasets, indicating its robustness and effectiveness. PMID:24489639

magHD: a new approach to multi-dimensional data storage, analysis, display and exploitation

NASA Astrophysics Data System (ADS)

Angleraud, Christophe

2014-06-01

The ever increasing amount of data and processing capabilities - following the well- known Moore's law - is challenging the way scientists and engineers are currently exploiting large datasets. The scientific visualization tools, although quite powerful, are often too generic and provide abstract views of phenomena, thus preventing cross disciplines fertilization. On the other end, Geographic information Systems allow nice and visually appealing maps to be built but they often get very confused as more layers are added. Moreover, the introduction of time as a fourth analysis dimension to allow analysis of time dependent phenomena such as meteorological or climate models, is encouraging real-time data exploration techniques that allow spatial-temporal points of interests to be detected by integration of moving images by the human brain. Magellium is involved in high performance image processing chains for satellite image processing as well as scientific signal analysis and geographic information management since its creation (2003). We believe that recent work on big data, GPU and peer-to-peer collaborative processing can open a new breakthrough in data analysis and display that will serve many new applications in collaborative scientific computing, environment mapping and understanding. The magHD (for Magellium Hyper-Dimension) project aims at developing software solutions that will bring highly interactive tools for complex datasets analysis and exploration commodity hardware, targeting small to medium scale clusters with expansion capabilities to large cloud based clusters.

Benchmark datasets for 3D MALDI- and DESI-imaging mass spectrometry.

PubMed

Oetjen, Janina; Veselkov, Kirill; Watrous, Jeramie; McKenzie, James S; Becker, Michael; Hauberg-Lotte, Lena; Kobarg, Jan Hendrik; Strittmatter, Nicole; Mróz, Anna K; Hoffmann, Franziska; Trede, Dennis; Palmer, Andrew; Schiffler, Stefan; Steinhorst, Klaus; Aichler, Michaela; Goldin, Robert; Guntinas-Lichius, Orlando; von Eggeling, Ferdinand; Thiele, Herbert; Maedler, Kathrin; Walch, Axel; Maass, Peter; Dorrestein, Pieter C; Takats, Zoltan; Alexandrov, Theodore

2015-01-01

Three-dimensional (3D) imaging mass spectrometry (MS) is an analytical chemistry technique for the 3D molecular analysis of a tissue specimen, entire organ, or microbial colonies on an agar plate. 3D-imaging MS has unique advantages over existing 3D imaging techniques, offers novel perspectives for understanding the spatial organization of biological processes, and has growing potential to be introduced into routine use in both biology and medicine. Owing to the sheer quantity of data generated, the visualization, analysis, and interpretation of 3D imaging MS data remain a significant challenge. Bioinformatics research in this field is hampered by the lack of publicly available benchmark datasets needed to evaluate and compare algorithms. High-quality 3D imaging MS datasets from different biological systems at several labs were acquired, supplied with overview images and scripts demonstrating how to read them, and deposited into MetaboLights, an open repository for metabolomics data. 3D imaging MS data were collected from five samples using two types of 3D imaging MS. 3D matrix-assisted laser desorption/ionization imaging (MALDI) MS data were collected from murine pancreas, murine kidney, human oral squamous cell carcinoma, and interacting microbial colonies cultured in Petri dishes. 3D desorption electrospray ionization (DESI) imaging MS data were collected from a human colorectal adenocarcinoma. With the aim to stimulate computational research in the field of computational 3D imaging MS, selected high-quality 3D imaging MS datasets are provided that could be used by algorithm developers as benchmark datasets.

MitoGen: A Framework for Generating 3D Synthetic Time-Lapse Sequences of Cell Populations in Fluorescence Microscopy.

PubMed

Svoboda, David; Ulman, Vladimir

2017-01-01

The proper analysis of biological microscopy images is an important and complex task. Therefore, it requires verification of all steps involved in the process, including image segmentation and tracking algorithms. It is generally better to verify algorithms with computer-generated ground truth datasets, which, compared to manually annotated data, nowadays have reached high quality and can be produced in large quantities even for 3D time-lapse image sequences. Here, we propose a novel framework, called MitoGen, which is capable of generating ground truth datasets with fully 3D time-lapse sequences of synthetic fluorescence-stained cell populations. MitoGen shows biologically justified cell motility, shape and texture changes as well as cell divisions. Standard fluorescence microscopy phenomena such as photobleaching, blur with real point spread function (PSF), and several types of noise, are simulated to obtain realistic images. The MitoGen framework is scalable in both space and time. MitoGen generates visually plausible data that shows good agreement with real data in terms of image descriptors and mean square displacement (MSD) trajectory analysis. Additionally, it is also shown in this paper that four publicly available segmentation and tracking algorithms exhibit similar performance on both real and MitoGen-generated data. The implementation of MitoGen is freely available.

A Framework to Learn Physics from Atomically Resolved Images

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

Vlcek, L.; Maksov, A.; Pan, M.

Here, we present a generalized framework for physics extraction, i.e., knowledge, from atomically resolved images, and show its utility by applying it to a model system of segregation of chalcogen atoms in an FeSe 0.45Te 0.55 superconductor system. We emphasize that the framework can be used for any imaging data for which a generative physical model exists. Consider that a generative physical model can produce a very large number of configurations, not all of which are observable. By applying a microscope function to a sub-set of this generated data, we form a simulated dataset on which statistics can be computed.

Machine learning for medical images analysis.

PubMed

Criminisi, A

2016-10-01

This article discusses the application of machine learning for the analysis of medical images. Specifically: (i) We show how a special type of learning models can be thought of as automatically optimized, hierarchically-structured, rule-based algorithms, and (ii) We discuss how the issue of collecting large labelled datasets applies to both conventional algorithms as well as machine learning techniques. The size of the training database is a function of model complexity rather than a characteristic of machine learning methods. Crown Copyright © 2016. Published by Elsevier B.V. All rights reserved.

Optimal retinal cyst segmentation from OCT images

NASA Astrophysics Data System (ADS)

Oguz, Ipek; Zhang, Li; Abramoff, Michael D.; Sonka, Milan

2016-03-01

Accurate and reproducible segmentation of cysts and fluid-filled regions from retinal OCT images is an important step allowing quantification of the disease status, longitudinal disease progression, and response to therapy in wet-pathology retinal diseases. However, segmentation of fluid-filled regions from OCT images is a challenging task due to their inhomogeneous appearance, the unpredictability of their number, size and location, as well as the intensity profile similarity between such regions and certain healthy tissue types. While machine learning techniques can be beneficial for this task, they require large training datasets and are often over-fitted to the appearance models of specific scanner vendors. We propose a knowledge-based approach that leverages a carefully designed cost function and graph-based segmentation techniques to provide a vendor-independent solution to this problem. We illustrate the results of this approach on two publicly available datasets with a variety of scanner vendors and retinal disease status. Compared to a previous machine-learning based approach, the volume similarity error was dramatically reduced from 81:3+/-56:4% to 22:2+/-21:3% (paired t-test, p << 0:001).

Diurnal fluctuations in brain volume: Statistical analyses of MRI from large populations.

PubMed

Nakamura, Kunio; Brown, Robert A; Narayanan, Sridar; Collins, D Louis; Arnold, Douglas L

2015-09-01

We investigated fluctuations in brain volume throughout the day using statistical modeling of magnetic resonance imaging (MRI) from large populations. We applied fully automated image analysis software to measure the brain parenchymal fraction (BPF), defined as the ratio of the brain parenchymal volume and intracranial volume, thus accounting for variations in head size. The MRI data came from serial scans of multiple sclerosis (MS) patients in clinical trials (n=755, 3269 scans) and from subjects participating in the Alzheimer's Disease Neuroimaging Initiative (ADNI, n=834, 6114 scans). The percent change in BPF was modeled with a linear mixed effect (LME) model, and the model was applied separately to the MS and ADNI datasets. The LME model for the MS datasets included random subject effects (intercept and slope over time) and fixed effects for the time-of-day, time from the baseline scan, and trial, which accounted for trial-related effects (for example, different inclusion criteria and imaging protocol). The model for ADNI additionally included the demographics (baseline age, sex, subject type [normal, mild cognitive impairment, or Alzheimer's disease], and interaction between subject type and time from baseline). There was a statistically significant effect of time-of-day on the BPF change in MS clinical trial datasets (-0.180 per day, that is, 0.180% of intracranial volume, p=0.019) as well as the ADNI dataset (-0.438 per day, that is, 0.438% of intracranial volume, p<0.0001), showing that the brain volume is greater in the morning. Linearly correcting the BPF values with the time-of-day reduced the required sample size to detect a 25% treatment effect (80% power and 0.05 significance level) on change in brain volume from 2 time-points over a period of 1year by 2.6%. Our results have significant implications for future brain volumetric studies, suggesting that there is a potential acquisition time bias that should be randomized or statistically controlled to account for the day-to-day brain volume fluctuations. Copyright © 2015 Elsevier Inc. All rights reserved.

Web-based visualization of very large scientific astronomy imagery

NASA Astrophysics Data System (ADS)

Bertin, E.; Pillay, R.; Marmo, C.

2015-04-01

Visualizing and navigating through large astronomy images from a remote location with current astronomy display tools can be a frustrating experience in terms of speed and ergonomics, especially on mobile devices. In this paper, we present a high performance, versatile and robust client-server system for remote visualization and analysis of extremely large scientific images. Applications of this work include survey image quality control, interactive data query and exploration, citizen science, as well as public outreach. The proposed software is entirely open source and is designed to be generic and applicable to a variety of datasets. It provides access to floating point data at terabyte scales, with the ability to precisely adjust image settings in real-time. The proposed clients are light-weight, platform-independent web applications built on standard HTML5 web technologies and compatible with both touch and mouse-based devices. We put the system to the test and assess the performance of the system and show that a single server can comfortably handle more than a hundred simultaneous users accessing full precision 32 bit astronomy data.

Querying Patterns in High-Dimensional Heterogenous Datasets

ERIC Educational Resources Information Center

Singh, Vishwakarma

2012-01-01

The recent technological advancements have led to the availability of a plethora of heterogenous datasets, e.g., images tagged with geo-location and descriptive keywords. An object in these datasets is described by a set of high-dimensional feature vectors. For example, a keyword-tagged image is represented by a color-histogram and a…

Atlas Toolkit: Fast registration of 3D morphological datasets in the absence of landmarks

PubMed Central

Grocott, Timothy; Thomas, Paul; Münsterberg, Andrea E.

2016-01-01

Image registration is a gateway technology for Developmental Systems Biology, enabling computational analysis of related datasets within a shared coordinate system. Many registration tools rely on landmarks to ensure that datasets are correctly aligned; yet suitable landmarks are not present in many datasets. Atlas Toolkit is a Fiji/ImageJ plugin collection offering elastic group-wise registration of 3D morphological datasets, guided by segmentation of the interesting morphology. We demonstrate the method by combinatorial mapping of cell signalling events in the developing eyes of chick embryos, and use the integrated datasets to predictively enumerate Gene Regulatory Network states. PMID:26864723

Atlas Toolkit: Fast registration of 3D morphological datasets in the absence of landmarks.

PubMed

Grocott, Timothy; Thomas, Paul; Münsterberg, Andrea E

2016-02-11

Image registration is a gateway technology for Developmental Systems Biology, enabling computational analysis of related datasets within a shared coordinate system. Many registration tools rely on landmarks to ensure that datasets are correctly aligned; yet suitable landmarks are not present in many datasets. Atlas Toolkit is a Fiji/ImageJ plugin collection offering elastic group-wise registration of 3D morphological datasets, guided by segmentation of the interesting morphology. We demonstrate the method by combinatorial mapping of cell signalling events in the developing eyes of chick embryos, and use the integrated datasets to predictively enumerate Gene Regulatory Network states.

Future Directions in Medical Physics: Models, Technology, and Translation to Medicine

NASA Astrophysics Data System (ADS)

Siewerdsen, Jeffrey

The application of physics in medicine has been integral to major advances in diagnostic and therapeutic medicine. Two primary areas represent the mainstay of medical physics research in the last century: in radiation therapy, physicists have propelled advances in conformal radiation treatment and high-precision image guidance; and in diagnostic imaging, physicists have advanced an arsenal of multi-modality imaging that includes CT, MRI, ultrasound, and PET as indispensible tools for noninvasive screening, diagnosis, and assessment of treatment response. In addition to their role in building such technologically rich fields of medicine, physicists have also become integral to daily clinical practice in these areas. The future suggests new opportunities for multi-disciplinary research bridging physics, biology, engineering, and computer science, and collaboration in medical physics carries a strong capacity for identification of significant clinical needs, access to clinical data, and translation of technologies to clinical studies. In radiation therapy, for example, the extraction of knowledge from large datasets on treatment delivery, image-based phenotypes, genomic profile, and treatment outcome will require innovation in computational modeling and connection with medical physics for the curation of large datasets. Similarly in imaging physics, the demand for new imaging technology capable of measuring physical and biological processes over orders of magnitude in scale (from molecules to whole organ systems) and exploiting new contrast mechanisms for greater sensitivity to molecular agents and subtle functional / morphological change will benefit from multi-disciplinary collaboration in physics, biology, and engineering. Also in surgery and interventional radiology, where needs for increased precision and patient safety meet constraints in cost and workflow, development of new technologies for imaging, image registration, and robotic assistance can leverage collaboration in physics, biomedical engineering, and computer science. In each area, there is major opportunity for multi-disciplinary collaboration with medical physics to accelerate the translation of such technologies to clinical use. Research supported by the National Institutes of Health, Siemens Healthcare, and Carestream Health.

Automatic Picking of Foraminifera: Design of the Foraminifera Image Recognition and Sorting Tool (FIRST) Prototype and Results of the Image Classification Scheme

NASA Astrophysics Data System (ADS)

de Garidel-Thoron, T.; Marchant, R.; Soto, E.; Gally, Y.; Beaufort, L.; Bolton, C. T.; Bouslama, M.; Licari, L.; Mazur, J. C.; Brutti, J. M.; Norsa, F.

2017-12-01

Foraminifera tests are the main proxy carriers for paleoceanographic reconstructions. Both geochemical and taxonomical studies require large numbers of tests to achieve statistical relevance. To date, the extraction of foraminifera from the sediment coarse fraction is still done by hand and thus time-consuming. Moreover, the recognition of morphotypes, ecologically relevant, requires some taxonomical skills not easily taught. The automatic recognition and extraction of foraminifera would largely help paleoceanographers to overcome these issues. Recent advances in automatic image classification using machine learning opens the way to automatic extraction of foraminifera. Here we detail progress on the design of an automatic picking machine as part of the FIRST project. The machine handles 30 pre-sieved samples (100-1000µm), separating them into individual particles (including foraminifera) and imaging each in pseudo-3D. The particles are classified and specimens of interest are sorted either for Individual Foraminifera Analyses (44 per slide) and/or for classical multiple analyses (8 morphological classes per slide, up to 1000 individuals per hole). The classification is based on machine learning using Convolutional Neural Networks (CNNs), similar to the approach used in the coccolithophorid imaging system SYRACO. To prove its feasibility, we built two training image datasets of modern planktonic foraminifera containing approximately 2000 and 5000 images each, corresponding to 15 & 25 morphological classes. Using a CNN with a residual topology (ResNet) we achieve over 95% correct classification for each dataset. We tested the network on 160,000 images from 45 depths of a sediment core from the Pacific ocean, for which we have human counts. The current algorithm is able to reproduce the downcore variability in both Globigerinoides ruber and the fragmentation index (r2 = 0.58 and 0.88 respectively). The FIRST prototype yields some promising results for high-resolution paleoceanographic studies and evolutionary studies.

Device and methods for "gold standard" registration of clinical 3D and 2D cerebral angiograms

NASA Astrophysics Data System (ADS)

Madan, Hennadii; Likar, Boštjan; Pernuš, Franjo; Å piclin, Žiga

2015-03-01

Translation of any novel and existing 3D-2D image registration methods into clinical image-guidance systems is limited due to lack of their objective validation on clinical image datasets. The main reason is that, besides the calibration of the 2D imaging system, a reference or "gold standard" registration is very difficult to obtain on clinical image datasets. In the context of cerebral endovascular image-guided interventions (EIGIs), we present a calibration device in the form of a headband with integrated fiducial markers and, secondly, propose an automated pipeline comprising 3D and 2D image processing, analysis and annotation steps, the result of which is a retrospective calibration of the 2D imaging system and an optimal, i.e., "gold standard" registration of 3D and 2D images. The device and methods were used to create the "gold standard" on 15 datasets of 3D and 2D cerebral angiograms, whereas each dataset was acquired on a patient undergoing EIGI for either aneurysm coiling or embolization of arteriovenous malformation. The use of the device integrated seamlessly in the clinical workflow of EIGI. While the automated pipeline eliminated all manual input or interactive image processing, analysis or annotation. In this way, the time to obtain the "gold standard" was reduced from 30 to less than one minute and the "gold standard" of 3D-2D registration on all 15 datasets of cerebral angiograms was obtained with a sub-0.1 mm accuracy.

Topological image texture analysis for quality assessment

NASA Astrophysics Data System (ADS)

Asaad, Aras T.; Rashid, Rasber Dh.; Jassim, Sabah A.

2017-05-01

Image quality is a major factor influencing pattern recognition accuracy and help detect image tampering for forensics. We are concerned with investigating topological image texture analysis techniques to assess different type of degradation. We use Local Binary Pattern (LBP) as a texture feature descriptor. For any image construct simplicial complexes for selected groups of uniform LBP bins and calculate persistent homology invariants (e.g. number of connected components). We investigated image quality discriminating characteristics of these simplicial complexes by computing these models for a large dataset of face images that are affected by the presence of shadows as a result of variation in illumination conditions. Our tests demonstrate that for specific uniform LBP patterns, the number of connected component not only distinguish between different levels of shadow effects but also help detect the infected regions as well.

Image quality of mean temporal arterial and mean temporal portal venous phase images calculated from low dose dynamic volume perfusion CT datasets in patients with hepatocellular carcinoma and pancreatic cancer.

PubMed

Wang, X; Henzler, T; Gawlitza, J; Diehl, S; Wilhelm, T; Schoenberg, S O; Jin, Z Y; Xue, H D; Smakic, A

2016-11-01

Dynamic volume perfusion CT (dVPCT) provides valuable information on tissue perfusion in patients with hepatocellular carcinoma (HCC) and pancreatic cancer. However, currently dVPCT is often performed in addition to conventional CT acquisitions due to the limited morphologic image quality of dose optimized dVPCT protocols. The aim of this study was to prospectively compare objective and subjective image quality, lesion detectability and radiation dose between mean temporal arterial (mTA) and mean temporal portal venous (mTPV) images calculated from low dose dynamic volume perfusion CT (dVPCT) datasets with linearly blended 120-kVp arterial and portal venous datasets in patients with HCC and pancreatic cancer. All patients gave written informed consent for this institutional review board-approved HIPAA compliant study. 27 consecutive patients (18 men, 9 women, mean age, 69.1 years±9.4) with histologically proven HCC or suspected pancreatic cancer were prospectively enrolled. The study CT protocol included a dVPCT protocol performed with 70 or 80kVp tube voltage (18 spiral acquisitions, 71.2s total acquisition times) and standard dual-energy (90/150kVpSn) arterial and portal venous acquisition performed 25min after the dVPCT. The mTA and mTPV images were manually reconstructed from the 3 to 5 best visually selected single arterial and 3 to 5 best single portal venous phases dVPCT dataset. The linearly blended 120-kVp images were calculated from dual-energy CT (DECT) raw data. Image noise, SNR, and CNR of the liver, abdominal aorta (AA) and main portal vein (PV) were compared between the mTA/mTPV and the linearly blended 120-kVp dual-energy arterial and portal venous datasets, respectively. Subjective image quality was evaluated by two radiologists regarding subjective image noise, sharpness and overall diagnostic image quality using a 5-point Likert Scale. In addition, liver lesion detectability was performed for each liver segment by the two radiologists using the linearly blended120-kVp arterial and portal venous datasets as the reference standard. Image noise, SNR and CNR values of the mTA and mTPV were significantly higher when compared to the corresponding linearly blended arterial and portal venous 120-kVp datasets (all p<0.001) except for image noise within the PV in the portal venous phases (p=0.136). image quality of mTA and mTPV were rated significantly better when compared to the linearly blended 120-kVp arterial and portal venous datasets. Both readers were able to detect all liver lesions found on the linearly blended 120-kVp arterial and portal venous datasets using the mTA and mTPV datasets. The effective radiation dose of the dVPCT was 27.6mSv for the 80kVp protocol and 14.5mSv for the 70kVp protocol. The mean effective radiation dose for the linearly blended 120-kVp arterial and portal venous CT protocol together of the upper abdomen was 5.60mSv±1.48mSv. Our preliminary data suggest that subjective and objective image quality of mTA and mTPV datasets calculated from low-kVp dVPCT datasets is non-inferior when compared to linearly blended 120-kVp arterial and portal venous acquisitions in patients with HCC and pancreatic cancer. Thus, dVPCT could be used as a stand-alone imaging technique without additionally performed conventional arterial and portal venous CT acquisitions. Copyright © 2016 Elsevier Ireland Ltd. All rights reserved.

Object detection approach using generative sparse, hierarchical networks with top-down and lateral connections for combining texture/color detection and shape/contour detection

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

Paiton, Dylan M.; Kenyon, Garrett T.; Brumby, Steven P.

An approach to detecting objects in an image dataset may combine texture/color detection, shape/contour detection, and/or motion detection using sparse, generative, hierarchical models with lateral and top-down connections. A first independent representation of objects in an image dataset may be produced using a color/texture detection algorithm. A second independent representation of objects in the image dataset may be produced using a shape/contour detection algorithm. A third independent representation of objects in the image dataset may be produced using a motion detection algorithm. The first, second, and third independent representations may then be combined into a single coherent output using amore » combinatorial algorithm.« less

Imaging mouse cerebellum with serial optical coherence scanner (Conference Presentation)

NASA Astrophysics Data System (ADS)

Liu, Chao J.; Williams, Kristen; Orr, Harry; Taner, Akkin

2017-02-01

We present the serial optical coherence scanner (SOCS), which consists of a polarization sensitive optical coherence tomography and a vibratome with associated controls for serial imaging, to visualize the cerebellum and adjacent brainstem of mouse. The cerebellar cortical layers and white matter are distinguished by using intrinsic optical contrasts. Images from serial scans reveal the large-scale anatomy in detail and map the nerve fiber pathways in the cerebellum and adjacent brainstem. The optical system, which has 5.5 μm axial resolution, utilizes a scan lens or a water-immersion microscope objective resulting in 10 μm or 4 μm lateral resolution, respectively. The large-scale brain imaging at high resolution requires an efficient way to collect large datasets. It is important to improve the SOCS system to deal with large-scale and large number of samples in a reasonable time. The imaging and slicing procedure for a section took about 4 minutes due to a low speed of the vibratome blade to maintain slicing quality. SOCS has potential to investigate pathological changes and monitor the effects of therapeutic drugs in cerebellar diseases such as spinocerebellar ataxia 1 (SCA1). The SCA1 is a neurodegenerative disease characterized by atrophy and eventual loss of Purkinje cells from the cerebellar cortex, and the optical contrasts provided by SOCS is being evaluated for biomarkers of the disease.

FISSA: A neuropil decontamination toolbox for calcium imaging signals.

PubMed

Keemink, Sander W; Lowe, Scott C; Pakan, Janelle M P; Dylda, Evelyn; van Rossum, Mark C W; Rochefort, Nathalie L

2018-02-22

In vivo calcium imaging has become a method of choice to image neuronal population activity throughout the nervous system. These experiments generate large sequences of images. Their analysis is computationally intensive and typically involves motion correction, image segmentation into regions of interest (ROIs), and extraction of fluorescence traces from each ROI. Out of focus fluorescence from surrounding neuropil and other cells can strongly contaminate the signal assigned to a given ROI. In this study, we introduce the FISSA toolbox (Fast Image Signal Separation Analysis) for neuropil decontamination. Given pre-defined ROIs, the FISSA toolbox automatically extracts the surrounding local neuropil and performs blind-source separation with non-negative matrix factorization. Using both simulated and in vivo data, we show that this toolbox performs similarly or better than existing published methods. FISSA requires only little RAM, and allows for fast processing of large datasets even on a standard laptop. The FISSA toolbox is available in Python, with an option for MATLAB format outputs, and can easily be integrated into existing workflows. It is available from Github and the standard Python repositories.

Functional CAR models for large spatially correlated functional datasets.

PubMed

Zhang, Lin; Baladandayuthapani, Veerabhadran; Zhu, Hongxiao; Baggerly, Keith A; Majewski, Tadeusz; Czerniak, Bogdan A; Morris, Jeffrey S

2016-01-01

We develop a functional conditional autoregressive (CAR) model for spatially correlated data for which functions are collected on areal units of a lattice. Our model performs functional response regression while accounting for spatial correlations with potentially nonseparable and nonstationary covariance structure, in both the space and functional domains. We show theoretically that our construction leads to a CAR model at each functional location, with spatial covariance parameters varying and borrowing strength across the functional domain. Using basis transformation strategies, the nonseparable spatial-functional model is computationally scalable to enormous functional datasets, generalizable to different basis functions, and can be used on functions defined on higher dimensional domains such as images. Through simulation studies, we demonstrate that accounting for the spatial correlation in our modeling leads to improved functional regression performance. Applied to a high-throughput spatially correlated copy number dataset, the model identifies genetic markers not identified by comparable methods that ignore spatial correlations.

OASIS: A Data Fusion System Optimized for Access to Distributed Archives

NASA Astrophysics Data System (ADS)

Berriman, G. B.; Kong, M.; Good, J. C.

2002-05-01

The On-Line Archive Science Information Services (OASIS) is accessible as a java applet through the NASA/IPAC Infrared Science Archive home page. It uses Geographical Information System (GIS) technology to provide data fusion and interaction services for astronomers. These services include the ability to process and display arbitrarily large image files, and user-controlled contouring, overlay regeneration and multi-table/image interactions. OASIS has been optimized for access to distributed archives and data sets. Its second release (June 2002) provides a mechanism that enables access to OASIS from "third-party" services and data providers. That is, any data provider who creates a query form to an archive containing a collection of data (images, catalogs, spectra) can direct the result files from the query into OASIS. Similarly, data providers who serve links to datasets or remote services on a web page can access all of these data with one instance of OASIS. In this was any data or service provider is given access to the full suite of capabilites of OASIS. We illustrate the "third-party" access feature with two examples: queries to the high-energy image datasets accessible from GSFC SkyView, and links to data that are returned from a target-based query to the NASA Extragalactic Database (NED). The second release of OASIS also includes a file-transfer manager that reports the status of multiple data downloads from remote sources to the client machine. It is a prototype for a request management system that will ultimately control and manage compute-intensive jobs submitted through OASIS to computing grids, such as request for large scale image mosaics and bulk statistical analysis.

NoRMCorre: An online algorithm for piecewise rigid motion correction of calcium imaging data.

PubMed

Pnevmatikakis, Eftychios A; Giovannucci, Andrea

2017-11-01

Motion correction is a challenging pre-processing problem that arises early in the analysis pipeline of calcium imaging data sequences. The motion artifacts in two-photon microscopy recordings can be non-rigid, arising from the finite time of raster scanning and non-uniform deformations of the brain medium. We introduce an algorithm for fast Non-Rigid Motion Correction (NoRMCorre) based on template matching. NoRMCorre operates by splitting the field of view (FOV) into overlapping spatial patches along all directions. The patches are registered at a sub-pixel resolution for rigid translation against a regularly updated template. The estimated alignments are subsequently up-sampled to create a smooth motion field for each frame that can efficiently approximate non-rigid artifacts in a piecewise-rigid manner. Existing approaches either do not scale well in terms of computational performance or are targeted to non-rigid artifacts arising just from the finite speed of raster scanning, and thus cannot correct for non-rigid motion observable in datasets from a large FOV. NoRMCorre can be run in an online mode resulting in comparable to or even faster than real time motion registration of streaming data. We evaluate its performance with simple yet intuitive metrics and compare against other non-rigid registration methods on simulated data and in vivo two-photon calcium imaging datasets. Open source Matlab and Python code is also made available. The proposed method and accompanying code can be useful for solving large scale image registration problems in calcium imaging, especially in the presence of non-rigid deformations. Copyright © 2017 The Author(s). Published by Elsevier B.V. All rights reserved.

Study on the Classification of GAOFEN-3 Polarimetric SAR Images Using Deep Neural Network

NASA Astrophysics Data System (ADS)

Zhang, J.; Zhang, J.; Zhao, Z.

2018-04-01

Polarimetric Synthetic Aperture Radar (POLSAR) imaging principle determines that the image quality will be affected by speckle noise. So the recognition accuracy of traditional image classification methods will be reduced by the effect of this interference. Since the date of submission, Deep Convolutional Neural Network impacts on the traditional image processing methods and brings the field of computer vision to a new stage with the advantages of a strong ability to learn deep features and excellent ability to fit large datasets. Based on the basic characteristics of polarimetric SAR images, the paper studied the types of the surface cover by using the method of Deep Learning. We used the fully polarimetric SAR features of different scales to fuse RGB images to the GoogLeNet model based on convolution neural network Iterative training, and then use the trained model to test the classification of data validation.First of all, referring to the optical image, we mark the surface coverage type of GF-3 POLSAR image with 8m resolution, and then collect the samples according to different categories. To meet the GoogLeNet model requirements of 256 × 256 pixel image input and taking into account the lack of full-resolution SAR resolution, the original image should be pre-processed in the process of resampling. In this paper, POLSAR image slice samples of different scales with sampling intervals of 2 m and 1 m to be trained separately and validated by the verification dataset. Among them, the training accuracy of GoogLeNet model trained with resampled 2-m polarimetric SAR image is 94.89 %, and that of the trained SAR image with resampled 1 m is 92.65 %.

SCPortalen: human and mouse single-cell centric database

PubMed Central

Noguchi, Shuhei; Böttcher, Michael; Hasegawa, Akira; Kouno, Tsukasa; Kato, Sachi; Tada, Yuhki; Ura, Hiroki; Abe, Kuniya; Shin, Jay W; Plessy, Charles; Carninci, Piero

2018-01-01

Abstract Published single-cell datasets are rich resources for investigators who want to address questions not originally asked by the creators of the datasets. The single-cell datasets might be obtained by different protocols and diverse analysis strategies. The main challenge in utilizing such single-cell data is how we can make the various large-scale datasets to be comparable and reusable in a different context. To challenge this issue, we developed the single-cell centric database ‘SCPortalen’ (http://single-cell.clst.riken.jp/). The current version of the database covers human and mouse single-cell transcriptomics datasets that are publicly available from the INSDC sites. The original metadata was manually curated and single-cell samples were annotated with standard ontology terms. Following that, common quality assessment procedures were conducted to check the quality of the raw sequence. Furthermore, primary data processing of the raw data followed by advanced analyses and interpretation have been performed from scratch using our pipeline. In addition to the transcriptomics data, SCPortalen provides access to single-cell image files whenever available. The target users of SCPortalen are all researchers interested in specific cell types or population heterogeneity. Through the web interface of SCPortalen users are easily able to search, explore and download the single-cell datasets of their interests. PMID:29045713

A hierarchical knowledge-based approach for retrieving similar medical images described with semantic annotations

PubMed Central

Kurtz, Camille; Beaulieu, Christopher F.; Napel, Sandy; Rubin, Daniel L.

2014-01-01

Computer-assisted image retrieval applications could assist radiologist interpretations by identifying similar images in large archives as a means to providing decision support. However, the semantic gap between low-level image features and their high level semantics may impair the system performances. Indeed, it can be challenging to comprehensively characterize the images using low-level imaging features to fully capture the visual appearance of diseases on images, and recently the use of semantic terms has been advocated to provide semantic descriptions of the visual contents of images. However, most of the existing image retrieval strategies do not consider the intrinsic properties of these terms during the comparison of the images beyond treating them as simple binary (presence/absence) features. We propose a new framework that includes semantic features in images and that enables retrieval of similar images in large databases based on their semantic relations. It is based on two main steps: (1) annotation of the images with semantic terms extracted from an ontology, and (2) evaluation of the similarity of image pairs by computing the similarity between the terms using the Hierarchical Semantic-Based Distance (HSBD) coupled to an ontological measure. The combination of these two steps provides a means of capturing the semantic correlations among the terms used to characterize the images that can be considered as a potential solution to deal with the semantic gap problem. We validate this approach in the context of the retrieval and the classification of 2D regions of interest (ROIs) extracted from computed tomographic (CT) images of the liver. Under this framework, retrieval accuracy of more than 0.96 was obtained on a 30-images dataset using the Normalized Discounted Cumulative Gain (NDCG) index that is a standard technique used to measure the effectiveness of information retrieval algorithms when a separate reference standard is available. Classification results of more than 95% were obtained on a 77-images dataset. For comparison purpose, the use of the Earth Mover's Distance (EMD), which is an alternative distance metric that considers all the existing relations among the terms, led to results retrieval accuracy of 0.95 and classification results of 93% with a higher computational cost. The results provided by the presented framework are competitive with the state-of-the-art and emphasize the usefulness of the proposed methodology for radiology image retrieval and classification. PMID:24632078

Brain tumor segmentation using holistically nested neural networks in MRI images.

PubMed

Zhuge, Ying; Krauze, Andra V; Ning, Holly; Cheng, Jason Y; Arora, Barbara C; Camphausen, Kevin; Miller, Robert W

2017-10-01

Gliomas are rapidly progressive, neurologically devastating, largely fatal brain tumors. Magnetic resonance imaging (MRI) is a widely used technique employed in the diagnosis and management of gliomas in clinical practice. MRI is also the standard imaging modality used to delineate the brain tumor target as part of treatment planning for the administration of radiation therapy. Despite more than 20 yr of research and development, computational brain tumor segmentation in MRI images remains a challenging task. We are presenting a novel method of automatic image segmentation based on holistically nested neural networks that could be employed for brain tumor segmentation of MRI images. Two preprocessing techniques were applied to MRI images. The N4ITK method was employed for correction of bias field distortion. A novel landmark-based intensity normalization method was developed so that tissue types have a similar intensity scale in images of different subjects for the same MRI protocol. The holistically nested neural networks (HNN), which extend from the convolutional neural networks (CNN) with a deep supervision through an additional weighted-fusion output layer, was trained to learn the multiscale and multilevel hierarchical appearance representation of the brain tumor in MRI images and was subsequently applied to produce a prediction map of the brain tumor on test images. Finally, the brain tumor was obtained through an optimum thresholding on the prediction map. The proposed method was evaluated on both the Multimodal Brain Tumor Image Segmentation (BRATS) Benchmark 2013 training datasets, and clinical data from our institute. A dice similarity coefficient (DSC) and sensitivity of 0.78 and 0.81 were achieved on 20 BRATS 2013 training datasets with high-grade gliomas (HGG), based on a two-fold cross-validation. The HNN model built on the BRATS 2013 training data was applied to ten clinical datasets with HGG from a locally developed database. DSC and sensitivity of 0.83 and 0.85 were achieved. A quantitative comparison indicated that the proposed method outperforms the popular fully convolutional network (FCN) method. In terms of efficiency, the proposed method took around 10 h for training with 50,000 iterations, and approximately 30 s for testing of a typical MRI image in the BRATS 2013 dataset with a size of 160 × 216 × 176, using a DELL PRECISION workstation T7400, with an NVIDIA Tesla K20c GPU. An effective brain tumor segmentation method for MRI images based on a HNN has been developed. The high level of accuracy and efficiency make this method practical in brain tumor segmentation. It may play a crucial role in both brain tumor diagnostic analysis and in the treatment planning of radiation therapy. Published 2017. This article is a U.S. Government work and is in the public domain in the USA.

ClimateNet: A Machine Learning dataset for Climate Science Research

NASA Astrophysics Data System (ADS)

Prabhat, M.; Biard, J.; Ganguly, S.; Ames, S.; Kashinath, K.; Kim, S. K.; Kahou, S.; Maharaj, T.; Beckham, C.; O'Brien, T. A.; Wehner, M. F.; Williams, D. N.; Kunkel, K.; Collins, W. D.

2017-12-01

Deep Learning techniques have revolutionized commercial applications in Computer vision, speech recognition and control systems. The key for all of these developments was the creation of a curated, labeled dataset ImageNet, for enabling multiple research groups around the world to develop methods, benchmark performance and compete with each other. The success of Deep Learning can be largely attributed to the broad availability of this dataset. Our empirical investigations have revealed that Deep Learning is similarly poised to benefit the task of pattern detection in climate science. Unfortunately, labeled datasets, a key pre-requisite for training, are hard to find. Individual research groups are typically interested in specialized weather patterns, making it hard to unify, and share datasets across groups and institutions. In this work, we are proposing ClimateNet: a labeled dataset that provides labeled instances of extreme weather patterns, as well as associated raw fields in model and observational output. We develop a schema in NetCDF to enumerate weather pattern classes/types, store bounding boxes, and pixel-masks. We are also working on a TensorFlow implementation to natively import such NetCDF datasets, and are providing a reference convolutional architecture for binary classification tasks. Our hope is that researchers in Climate Science, as well as ML/DL, will be able to use (and extend) ClimateNet to make rapid progress in the application of Deep Learning for Climate Science research.

Active appearance model and deep learning for more accurate prostate segmentation on MRI

NASA Astrophysics Data System (ADS)

Cheng, Ruida; Roth, Holger R.; Lu, Le; Wang, Shijun; Turkbey, Baris; Gandler, William; McCreedy, Evan S.; Agarwal, Harsh K.; Choyke, Peter; Summers, Ronald M.; McAuliffe, Matthew J.

2016-03-01

Prostate segmentation on 3D MR images is a challenging task due to image artifacts, large inter-patient prostate shape and texture variability, and lack of a clear prostate boundary specifically at apex and base levels. We propose a supervised machine learning model that combines atlas based Active Appearance Model (AAM) with a Deep Learning model to segment the prostate on MR images. The performance of the segmentation method is evaluated on 20 unseen MR image datasets. The proposed method combining AAM and Deep Learning achieves a mean Dice Similarity Coefficient (DSC) of 0.925 for whole 3D MR images of the prostate using axial cross-sections. The proposed model utilizes the adaptive atlas-based AAM model and Deep Learning to achieve significant segmentation accuracy.

The Wide-Field Imaging Interferometry Testbed: Enabling Techniques for High Angular Resolution Astronomy

NASA Technical Reports Server (NTRS)

Rinehart, S. A.; Armstrong, T.; Frey, Bradley J.; Jung, J.; Kirk, J.; Leisawitz, David T.; Leviton, Douglas B.; Lyon, R.; Maher, Stephen; Martino, Anthony J.;

Image retrieval by information fusion based on scalable vocabulary tree and robust Hausdorff distance

NASA Astrophysics Data System (ADS)

Che, Chang; Yu, Xiaoyang; Sun, Xiaoming; Yu, Boyang

2017-12-01

In recent years, Scalable Vocabulary Tree (SVT) has been shown to be effective in image retrieval. However, for general images where the foreground is the object to be recognized while the background is cluttered, the performance of the current SVT framework is restricted. In this paper, a new image retrieval framework that incorporates a robust distance metric and information fusion is proposed, which improves the retrieval performance relative to the baseline SVT approach. First, the visual words that represent the background are diminished by using a robust Hausdorff distance between different images. Second, image matching results based on three image signature representations are fused, which enhances the retrieval precision. We conducted intensive experiments on small-scale to large-scale image datasets: Corel-9, Corel-48, and PKU-198, where the proposed Hausdorff metric and information fusion outperforms the state-of-the-art methods by about 13, 15, and 15%, respectively.

Image quality classification for DR screening using deep learning.

PubMed

FengLi Yu; Jing Sun; Annan Li; Jun Cheng; Cheng Wan; Jiang Liu

2017-07-01

The quality of input images significantly affects the outcome of automated diabetic retinopathy (DR) screening systems. Unlike the previous methods that only consider simple low-level features such as hand-crafted geometric and structural features, in this paper we propose a novel method for retinal image quality classification (IQC) that performs computational algorithms imitating the working of the human visual system. The proposed algorithm combines unsupervised features from saliency map and supervised features coming from convolutional neural networks (CNN), which are fed to an SVM to automatically detect high quality vs poor quality retinal fundus images. We demonstrate the superior performance of our proposed algorithm on a large retinal fundus image dataset and the method could achieve higher accuracy than other methods. Although retinal images are used in this study, the methodology is applicable to the image quality assessment and enhancement of other types of medical images.

Region Templates: Data Representation and Management for High-Throughput Image Analysis

PubMed Central

Pan, Tony; Kurc, Tahsin; Kong, Jun; Cooper, Lee; Klasky, Scott; Saltz, Joel

2015-01-01

We introduce a region template abstraction and framework for the efficient storage, management and processing of common data types in analysis of large datasets of high resolution images on clusters of hybrid computing nodes. The region template abstraction provides a generic container template for common data structures, such as points, arrays, regions, and object sets, within a spatial and temporal bounding box. It allows for different data management strategies and I/O implementations, while providing a homogeneous, unified interface to applications for data storage and retrieval. A region template application is represented as a hierarchical dataflow in which each computing stage may be represented as another dataflow of finer-grain tasks. The execution of the application is coordinated by a runtime system that implements optimizations for hybrid machines, including performance-aware scheduling for maximizing the utilization of computing devices and techniques to reduce the impact of data transfers between CPUs and GPUs. An experimental evaluation on a state-of-the-art hybrid cluster using a microscopy imaging application shows that the abstraction adds negligible overhead (about 3%) and achieves good scalability and high data transfer rates. Optimizations in a high speed disk based storage implementation of the abstraction to support asynchronous data transfers and computation result in an application performance gain of about 1.13×. Finally, a processing rate of 11,730 4K×4K tiles per minute was achieved for the microscopy imaging application on a cluster with 100 nodes (300 GPUs and 1,200 CPU cores). This computation rate enables studies with very large datasets. PMID:26139953

Deep learning with domain adaptation for accelerated projection-reconstruction MR.

PubMed

Han, Yoseob; Yoo, Jaejun; Kim, Hak Hee; Shin, Hee Jung; Sung, Kyunghyun; Ye, Jong Chul

2018-09-01

The radial k-space trajectory is a well-established sampling trajectory used in conjunction with magnetic resonance imaging. However, the radial k-space trajectory requires a large number of radial lines for high-resolution reconstruction. Increasing the number of radial lines causes longer acquisition time, making it more difficult for routine clinical use. On the other hand, if we reduce the number of radial lines, streaking artifact patterns are unavoidable. To solve this problem, we propose a novel deep learning approach with domain adaptation to restore high-resolution MR images from under-sampled k-space data. The proposed deep network removes the streaking artifacts from the artifact corrupted images. To address the situation given the limited available data, we propose a domain adaptation scheme that employs a pre-trained network using a large number of X-ray computed tomography (CT) or synthesized radial MR datasets, which is then fine-tuned with only a few radial MR datasets. The proposed method outperforms existing compressed sensing algorithms, such as the total variation and PR-FOCUSS methods. In addition, the calculation time is several orders of magnitude faster than the total variation and PR-FOCUSS methods. Moreover, we found that pre-training using CT or MR data from similar organ data is more important than pre-training using data from the same modality for different organ. We demonstrate the possibility of a domain-adaptation when only a limited amount of MR data is available. The proposed method surpasses the existing compressed sensing algorithms in terms of the image quality and computation time. © 2018 International Society for Magnetic Resonance in Medicine.

Deep Learning Methods for Quantifying Invasive Benthic Species in the Great Lakes

NASA Astrophysics Data System (ADS)

Billings, G.; Skinner, K.; Johnson-Roberson, M.

2017-12-01

In recent decades, invasive species such as the round goby and dreissenid mussels have greatly impacted the Great Lakes ecosystem. It is critical to monitor these species, model their distribution, and quantify the impacts on the native fisheries and surrounding ecosystem in order to develop an effective management response. However, data collection in underwater environments is challenging and expensive. Furthermore, the round goby is typically found in rocky habitats, which are inaccessible to standard survey techniques such as bottom trawling. In this work we propose a robotic system for visual data collection to automatically detect and quantify invasive round gobies and mussels in the Great Lakes. Robotic platforms equipped with cameras can perform efficient, cost-effective, low-bias benthic surveys. This data collection can be further optimized through automatic detection and annotation of the target species. Deep learning methods have shown success in image recognition tasks. However, these methods often rely on a labelled training dataset, with up to millions of labelled images. Hand labeling large numbers of images is expensive and often impracticable. Furthermore, data collected in the field may be sparse when only considering images that contain the objects of interest. It is easier to collect dense, clean data in controlled lab settings, but this data is not a realistic representation of real field environments. In this work, we propose a deep learning approach to generate a large set of labelled training data realistic of underwater environments in the field. To generate these images, first we draw random sample images of individual fish and mussels from a library of images captured in a controlled lab environment. Next, these randomly drawn samples will be automatically merged into natural background images. Finally, we will use a generative adversarial network (GAN) that incorporates constraints of the physical model of underwater light propagation to simulate the process of underwater image formation in various water conditions. The output of the GAN will be realistic looking annotated underwater images. This generated dataset of images will be used to train a classifier to identify round gobies and mussels in order to measure the biomass and abundance of these invasive species in the Great Lakes.

Interpolation of diffusion weighted imaging datasets.

PubMed

Dyrby, Tim B; Lundell, Henrik; Burke, Mark W; Reislev, Nina L; Paulson, Olaf B; Ptito, Maurice; Siebner, Hartwig R

2014-12-01

Diffusion weighted imaging (DWI) is used to study white-matter fibre organisation, orientation and structural connectivity by means of fibre reconstruction algorithms and tractography. For clinical settings, limited scan time compromises the possibilities to achieve high image resolution for finer anatomical details and signal-to-noise-ratio for reliable fibre reconstruction. We assessed the potential benefits of interpolating DWI datasets to a higher image resolution before fibre reconstruction using a diffusion tensor model. Simulations of straight and curved crossing tracts smaller than or equal to the voxel size showed that conventional higher-order interpolation methods improved the geometrical representation of white-matter tracts with reduced partial-volume-effect (PVE), except at tract boundaries. Simulations and interpolation of ex-vivo monkey brain DWI datasets revealed that conventional interpolation methods fail to disentangle fine anatomical details if PVE is too pronounced in the original data. As for validation we used ex-vivo DWI datasets acquired at various image resolutions as well as Nissl-stained sections. Increasing the image resolution by a factor of eight yielded finer geometrical resolution and more anatomical details in complex regions such as tract boundaries and cortical layers, which are normally only visualized at higher image resolutions. Similar results were found with typical clinical human DWI dataset. However, a possible bias in quantitative values imposed by the interpolation method used should be considered. The results indicate that conventional interpolation methods can be successfully applied to DWI datasets for mining anatomical details that are normally seen only at higher resolutions, which will aid in tractography and microstructural mapping of tissue compartments. Copyright © 2014. Published by Elsevier Inc.

A framework for automatic creation of gold-standard rigid 3D-2D registration datasets.

PubMed

Madan, Hennadii; Pernuš, Franjo; Likar, Boštjan; Špiclin, Žiga

2017-02-01

Advanced image-guided medical procedures incorporate 2D intra-interventional information into pre-interventional 3D image and plan of the procedure through 3D/2D image registration (32R). To enter clinical use, and even for publication purposes, novel and existing 32R methods have to be rigorously validated. The performance of a 32R method can be estimated by comparing it to an accurate reference or gold standard method (usually based on fiducial markers) on the same set of images (gold standard dataset). Objective validation and comparison of methods are possible only if evaluation methodology is standardized, and the gold standard  dataset is made publicly available. Currently, very few such datasets exist and only one contains images of multiple patients acquired during a procedure. To encourage the creation of gold standard 32R datasets, we propose an automatic framework. The framework is based on rigid registration of fiducial markers. The main novelty is spatial grouping of fiducial markers on the carrier device, which enables automatic marker localization and identification across the 3D and 2D images. The proposed framework was demonstrated on clinical angiograms of 20 patients. Rigid 32R computed by the framework was more accurate than that obtained manually, with the respective target registration error below 0.027 mm compared to 0.040 mm. The framework is applicable for gold standard setup on any rigid anatomy, provided that the acquired images contain spatially grouped fiducial markers. The gold standard datasets and software will be made publicly available.

Development of Multi-perspective Diagnostics and Analysis Algorithms with Applications to Subsonic and Supersonic Combustors

NASA Astrophysics Data System (ADS)

Wickersham, Andrew Joseph

There are two critical research needs for the study of hydrocarbon combustion in high speed flows: 1) combustion diagnostics with adequate temporal and spatial resolution, and 2) mathematical techniques that can extract key information from large datasets. The goal of this work is to address these needs, respectively, by the use of high speed and multi-perspective chemiluminescence and advanced mathematical algorithms. To obtain the measurements, this work explored the application of high speed chemiluminescence diagnostics and the use of fiber-based endoscopes (FBEs) for non-intrusive and multi-perspective chemiluminescence imaging up to 20 kHz. Non-intrusive and full-field imaging measurements provide a wealth of information for model validation and design optimization of propulsion systems. However, it is challenging to obtain such measurements due to various implementation difficulties such as optical access, thermal management, and equipment cost. This work therefore explores the application of FBEs for non-intrusive imaging to supersonic propulsion systems. The FBEs used in this work are demonstrated to overcome many of the aforementioned difficulties and provided datasets from multiple angular positions up to 20 kHz in a supersonic combustor. The combustor operated on ethylene fuel at Mach 2 with an inlet stagnation temperature and pressure of approximately 640 degrees Fahrenheit and 70 psia, respectively. The imaging measurements were obtained from eight perspectives simultaneously, providing full-field datasets under such flow conditions for the first time, allowing the possibility of inferring multi-dimensional measurements. Due to the high speed and multi-perspective nature, such new diagnostic capability generates a large volume of data and calls for analysis algorithms that can process the data and extract key physics effectively. To extract the key combustion dynamics from the measurements, three mathematical methods were investigated in this work: Fourier analysis, proper orthogonal decomposition (POD), and wavelet analysis (WA). These algorithms were first demonstrated and tested on imaging measurements obtained from one perspective in a sub-sonic combustor (up to Mach 0.2). The results show that these algorithms are effective in extracting the key physics from large datasets, including the characteristic frequencies of flow-flame interactions especially during transient processes such as lean blow off and ignition. After these relatively simple tests and demonstrations, these algorithms were applied to process the measurements obtained from multi-perspective in the supersonic combustor. compared to past analyses (which have been limited to data obtained from one perspective only), the availability of data at multiple perspective provide further insights into the flame and flow structures in high speed flows. In summary, this work shows that high speed chemiluminescence is a simple yet powerful combustion diagnostic. Especially when combined with FBEs and the analyses algorithms described in this work, such diagnostics provide full-field imaging at high repetition rate in challenging flows. Based on such measurements, a wealth of information can be obtained from proper analysis algorithms, including characteristic frequency, dominating flame modes, and even multi-dimensional flame and flow structures.

Image stack alignment in full-field X-ray absorption spectroscopy using SIFT_PyOCL.

PubMed

Paleo, Pierre; Pouyet, Emeline; Kieffer, Jérôme

2014-03-01

Full-field X-ray absorption spectroscopy experiments allow the acquisition of millions of spectra within minutes. However, the construction of the hyperspectral image requires an image alignment procedure with sub-pixel precision. While the image correlation algorithm has originally been used for image re-alignment using translations, the Scale Invariant Feature Transform (SIFT) algorithm (which is by design robust versus rotation, illumination change, translation and scaling) presents an additional advantage: the alignment can be limited to a region of interest of any arbitrary shape. In this context, a Python module, named SIFT_PyOCL, has been developed. It implements a parallel version of the SIFT algorithm in OpenCL, providing high-speed image registration and alignment both on processors and graphics cards. The performance of the algorithm allows online processing of large datasets.

Computer assisted analysis of auroral images obtained from high altitude polar satellites

NASA Technical Reports Server (NTRS)

Samadani, Ramin; Flynn, Michael

1993-01-01

Automatic techniques that allow the extraction of physically significant parameters from auroral images were developed. This allows the processing of a much larger number of images than is currently possible with manual techniques. Our techniques were applied to diverse auroral image datasets. These results were made available to geophysicists at NASA and at universities in the form of a software system that performs the analysis. After some feedback from users, an upgraded system was transferred to NASA and to two universities. The feasibility of user-trained search and retrieval of large amounts of data using our automatically derived parameter indices was demonstrated. Techniques based on classification and regression trees (CART) were developed and applied to broaden the types of images to which the automated search and retrieval may be applied. Our techniques were tested with DE-1 auroral images.

TuMore: generation of synthetic brain tumor MRI data for deep learning based segmentation approaches

NASA Astrophysics Data System (ADS)

Lindner, Lydia; Pfarrkirchner, Birgit; Gsaxner, Christina; Schmalstieg, Dieter; Egger, Jan

2018-03-01

Accurate segmentation and measurement of brain tumors plays an important role in clinical practice and research, as it is critical for treatment planning and monitoring of tumor growth. However, brain tumor segmentation is one of the most challenging tasks in medical image analysis. Since manual segmentations are subjective, time consuming and neither accurate nor reliable, there exists a need for objective, robust and fast automated segmentation methods that provide competitive performance. Therefore, deep learning based approaches are gaining interest in the field of medical image segmentation. When the training data set is large enough, deep learning approaches can be extremely effective, but in domains like medicine, only limited data is available in the majority of cases. Due to this reason, we propose a method that allows to create a large dataset of brain MRI (Magnetic Resonance Imaging) images containing synthetic brain tumors - glioblastomas more specifically - and the corresponding ground truth, that can be subsequently used to train deep neural networks.

Advanced Cell Classifier: User-Friendly Machine-Learning-Based Software for Discovering Phenotypes in High-Content Imaging Data.

PubMed

Piccinini, Filippo; Balassa, Tamas; Szkalisity, Abel; Molnar, Csaba; Paavolainen, Lassi; Kujala, Kaisa; Buzas, Krisztina; Sarazova, Marie; Pietiainen, Vilja; Kutay, Ulrike; Smith, Kevin; Horvath, Peter

2017-06-28

High-content, imaging-based screens now routinely generate data on a scale that precludes manual verification and interrogation. Software applying machine learning has become an essential tool to automate analysis, but these methods require annotated examples to learn from. Efficiently exploring large datasets to find relevant examples remains a challenging bottleneck. Here, we present Advanced Cell Classifier (ACC), a graphical software package for phenotypic analysis that addresses these difficulties. ACC applies machine-learning and image-analysis methods to high-content data generated by large-scale, cell-based experiments. It features methods to mine microscopic image data, discover new phenotypes, and improve recognition performance. We demonstrate that these features substantially expedite the training process, successfully uncover rare phenotypes, and improve the accuracy of the analysis. ACC is extensively documented, designed to be user-friendly for researchers without machine-learning expertise, and distributed as a free open-source tool at www.cellclassifier.org. Copyright © 2017 Elsevier Inc. All rights reserved.

Fast large-scale object retrieval with binary quantization

NASA Astrophysics Data System (ADS)

Zhou, Shifu; Zeng, Dan; Shen, Wei; Zhang, Zhijiang; Tian, Qi

2015-11-01

The objective of large-scale object retrieval systems is to search for images that contain the target object in an image database. Where state-of-the-art approaches rely on global image representations to conduct searches, we consider many boxes per image as candidates to search locally in a picture. In this paper, a feature quantization algorithm called binary quantization is proposed. In binary quantization, a scale-invariant feature transform (SIFT) feature is quantized into a descriptive and discriminative bit-vector, which allows itself to adapt to the classic inverted file structure for box indexing. The inverted file, which stores the bit-vector and box ID where the SIFT feature is located inside, is compact and can be loaded into the main memory for efficient box indexing. We evaluate our approach on available object retrieval datasets. Experimental results demonstrate that the proposed approach is fast and achieves excellent search quality. Therefore, the proposed approach is an improvement over state-of-the-art approaches for object retrieval.

A Scalable Cyberinfrastructure for Interactive Visualization of Terascale Microscopy Data

PubMed Central

Venkat, A.; Christensen, C.; Gyulassy, A.; Summa, B.; Federer, F.; Angelucci, A.; Pascucci, V.

2017-01-01

The goal of the recently emerged field of connectomics is to generate a wiring diagram of the brain at different scales. To identify brain circuitry, neuroscientists use specialized microscopes to perform multichannel imaging of labeled neurons at a very high resolution. CLARITY tissue clearing allows imaging labeled circuits through entire tissue blocks, without the need for tissue sectioning and section-to-section alignment. Imaging the large and complex non-human primate brain with sufficient resolution to identify and disambiguate between axons, in particular, produces massive data, creating great computational challenges to the study of neural circuits. Researchers require novel software capabilities for compiling, stitching, and visualizing large imagery. In this work, we detail the image acquisition process and a hierarchical streaming platform, ViSUS, that enables interactive visualization of these massive multi-volume datasets using a standard desktop computer. The ViSUS visualization framework has previously been shown to be suitable for 3D combustion simulation, climate simulation and visualization of large scale panoramic images. The platform is organized around a hierarchical cache oblivious data layout, called the IDX file format, which enables interactive visualization and exploration in ViSUS, scaling to the largest 3D images. In this paper we showcase the VISUS framework used in an interactive setting with the microscopy data. PMID:28638896

A Scalable Cyberinfrastructure for Interactive Visualization of Terascale Microscopy Data.

PubMed

Venkat, A; Christensen, C; Gyulassy, A; Summa, B; Federer, F; Angelucci, A; Pascucci, V

2016-08-01

The goal of the recently emerged field of connectomics is to generate a wiring diagram of the brain at different scales. To identify brain circuitry, neuroscientists use specialized microscopes to perform multichannel imaging of labeled neurons at a very high resolution. CLARITY tissue clearing allows imaging labeled circuits through entire tissue blocks, without the need for tissue sectioning and section-to-section alignment. Imaging the large and complex non-human primate brain with sufficient resolution to identify and disambiguate between axons, in particular, produces massive data, creating great computational challenges to the study of neural circuits. Researchers require novel software capabilities for compiling, stitching, and visualizing large imagery. In this work, we detail the image acquisition process and a hierarchical streaming platform, ViSUS, that enables interactive visualization of these massive multi-volume datasets using a standard desktop computer. The ViSUS visualization framework has previously been shown to be suitable for 3D combustion simulation, climate simulation and visualization of large scale panoramic images. The platform is organized around a hierarchical cache oblivious data layout, called the IDX file format, which enables interactive visualization and exploration in ViSUS, scaling to the largest 3D images. In this paper we showcase the VISUS framework used in an interactive setting with the microscopy data.

MRI-Based Intelligence Quotient (IQ) Estimation with Sparse Learning

PubMed Central

Wang, Liye; Wee, Chong-Yaw; Suk, Heung-Il; Tang, Xiaoying; Shen, Dinggang

2015-01-01

In this paper, we propose a novel framework for IQ estimation using Magnetic Resonance Imaging (MRI) data. In particular, we devise a new feature selection method based on an extended dirty model for jointly considering both element-wise sparsity and group-wise sparsity. Meanwhile, due to the absence of large dataset with consistent scanning protocols for the IQ estimation, we integrate multiple datasets scanned from different sites with different scanning parameters and protocols. In this way, there is large variability in these different datasets. To address this issue, we design a two-step procedure for 1) first identifying the possible scanning site for each testing subject and 2) then estimating the testing subject’s IQ by using a specific estimator designed for that scanning site. We perform two experiments to test the performance of our method by using the MRI data collected from 164 typically developing children between 6 and 15 years old. In the first experiment, we use a multi-kernel Support Vector Regression (SVR) for estimating IQ values, and obtain an average correlation coefficient of 0.718 and also an average root mean square error of 8.695 between the true IQs and the estimated ones. In the second experiment, we use a single-kernel SVR for IQ estimation, and achieve an average correlation coefficient of 0.684 and an average root mean square error of 9.166. All these results show the effectiveness of using imaging data for IQ prediction, which is rarely done in the field according to our knowledge. PMID:25822851

Ultrasound guided fluorescence molecular tomography with improved quantification by an attenuation compensated born-normalization and in vivo preclinical study of cancer

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

Li, Baoqiang; Berti, Romain; Abran, Maxime

2014-05-15

Ultrasound imaging, having the advantages of low-cost and non-invasiveness over MRI and X-ray CT, was reported by several studies as an adequate complement to fluorescence molecular tomography with the perspective of improving localization and quantification of fluorescent molecular targets in vivo. Based on the previous work, an improved dual-modality Fluorescence-Ultrasound imaging system was developed and then validated in imaging study with preclinical tumor model. Ultrasound imaging and a profilometer were used to obtain the anatomical prior information and 3D surface, separately, to precisely extract the tissue boundary on both sides of sample in order to achieve improved fluorescence reconstruction. Furthermore,more » a pattern-based fluorescence reconstruction on the detection side was incorporated to enable dimensional reduction of the dataset while keeping the useful information for reconstruction. Due to its putative role in the current imaging geometry and the chosen reconstruction technique, we developed an attenuation compensated Born-normalization method to reduce the attenuation effects and cancel off experimental factors when collecting quantitative fluorescence datasets over large area. Results of both simulation and phantom study demonstrated that fluorescent targets could be recovered accurately and quantitatively using this reconstruction mechanism. Finally, in vivo experiment confirms that the imaging system associated with the proposed image reconstruction approach was able to extract both functional and anatomical information, thereby improving quantification and localization of molecular targets.« less

A Model-Based Approach for Microvasculature Structure Distortion Correction in Two-Photon Fluorescence Microscopy Images

PubMed Central

Dao, Lam; Glancy, Brian; Lucotte, Bertrand; Chang, Lin-Ching; Balaban, Robert S; Hsu, Li-Yueh

2015-01-01

SUMMARY This paper investigates a post-processing approach to correct spatial distortion in two-photon fluorescence microscopy images for vascular network reconstruction. It is aimed at in vivo imaging of large field-of-view, deep-tissue studies of vascular structures. Based on simple geometric modeling of the object-of-interest, a distortion function is directly estimated from the image volume by deconvolution analysis. Such distortion function is then applied to sub volumes of the image stack to adaptively adjust for spatially varying distortion and reduce the image blurring through blind deconvolution. The proposed technique was first evaluated in phantom imaging of fluorescent microspheres that are comparable in size to the underlying capillary vascular structures. The effectiveness of restoring three-dimensional spherical geometry of the microspheres using the estimated distortion function was compared with empirically measured point-spread function. Next, the proposed approach was applied to in vivo vascular imaging of mouse skeletal muscle to reduce the image distortion of the capillary structures. We show that the proposed method effectively improve the image quality and reduce spatially varying distortion that occurs in large field-of-view deep-tissue vascular dataset. The proposed method will help in qualitative interpretation and quantitative analysis of vascular structures from fluorescence microscopy images. PMID:26224257

Trace: a high-throughput tomographic reconstruction engine for large-scale datasets.

PubMed

Bicer, Tekin; Gürsoy, Doğa; Andrade, Vincent De; Kettimuthu, Rajkumar; Scullin, William; Carlo, Francesco De; Foster, Ian T

2017-01-01

Modern synchrotron light sources and detectors produce data at such scale and complexity that large-scale computation is required to unleash their full power. One of the widely used imaging techniques that generates data at tens of gigabytes per second is computed tomography (CT). Although CT experiments result in rapid data generation, the analysis and reconstruction of the collected data may require hours or even days of computation time with a medium-sized workstation, which hinders the scientific progress that relies on the results of analysis. We present Trace, a data-intensive computing engine that we have developed to enable high-performance implementation of iterative tomographic reconstruction algorithms for parallel computers. Trace provides fine-grained reconstruction of tomography datasets using both (thread-level) shared memory and (process-level) distributed memory parallelization. Trace utilizes a special data structure called replicated reconstruction object to maximize application performance. We also present the optimizations that we apply to the replicated reconstruction objects and evaluate them using tomography datasets collected at the Advanced Photon Source. Our experimental evaluations show that our optimizations and parallelization techniques can provide 158× speedup using 32 compute nodes (384 cores) over a single-core configuration and decrease the end-to-end processing time of a large sinogram (with 4501 × 1 × 22,400 dimensions) from 12.5 h to <5 min per iteration. The proposed tomographic reconstruction engine can efficiently process large-scale tomographic data using many compute nodes and minimize reconstruction times.

Texture Descriptors Ensembles Enable Image-Based Classification of Maturation of Human Stem Cell-Derived Retinal Pigmented Epithelium

PubMed Central

Caetano dos Santos, Florentino Luciano; Skottman, Heli; Juuti-Uusitalo, Kati; Hyttinen, Jari

2016-01-01

Aims A fast, non-invasive and observer-independent method to analyze the homogeneity and maturity of human pluripotent stem cell (hPSC) derived retinal pigment epithelial (RPE) cells is warranted to assess the suitability of hPSC-RPE cells for implantation or in vitro use. The aim of this work was to develop and validate methods to create ensembles of state-of-the-art texture descriptors and to provide a robust classification tool to separate three different maturation stages of RPE cells by using phase contrast microscopy images. The same methods were also validated on a wide variety of biological image classification problems, such as histological or virus image classification. Methods For image classification we used different texture descriptors, descriptor ensembles and preprocessing techniques. Also, three new methods were tested. The first approach was an ensemble of preprocessing methods, to create an additional set of images. The second was the region-based approach, where saliency detection and wavelet decomposition divide each image in two different regions, from which features were extracted through different descriptors. The third method was an ensemble of Binarized Statistical Image Features, based on different sizes and thresholds. A Support Vector Machine (SVM) was trained for each descriptor histogram and the set of SVMs combined by sum rule. The accuracy of the computer vision tool was verified in classifying the hPSC-RPE cell maturation level. Dataset and Results The RPE dataset contains 1862 subwindows from 195 phase contrast images. The final descriptor ensemble outperformed the most recent stand-alone texture descriptors, obtaining, for the RPE dataset, an area under ROC curve (AUC) of 86.49% with the 10-fold cross validation and 91.98% with the leave-one-image-out protocol. The generality of the three proposed approaches was ascertained with 10 more biological image datasets, obtaining an average AUC greater than 97%. Conclusions Here we showed that the developed ensembles of texture descriptors are able to classify the RPE cell maturation stage. Moreover, we proved that preprocessing and region-based decomposition improves many descriptors’ accuracy in biological dataset classification. Finally, we built the first public dataset of stem cell-derived RPE cells, which is publicly available to the scientific community for classification studies. The proposed tool is available at https://www.dei.unipd.it/node/2357 and the RPE dataset at http://www.biomeditech.fi/data/RPE_dataset/. Both are available at https://figshare.com/s/d6fb591f1beb4f8efa6f. PMID:26895509

Image Quality Ranking Method for Microscopy

PubMed Central

Koho, Sami; Fazeli, Elnaz; Eriksson, John E.; Hänninen, Pekka E.

2016-01-01

Automated analysis of microscope images is necessitated by the increased need for high-resolution follow up of events in time. Manually finding the right images to be analyzed, or eliminated from data analysis are common day-to-day problems in microscopy research today, and the constantly growing size of image datasets does not help the matter. We propose a simple method and a software tool for sorting images within a dataset, according to their relative quality. We demonstrate the applicability of our method in finding good quality images in a STED microscope sample preparation optimization image dataset. The results are validated by comparisons to subjective opinion scores, as well as five state-of-the-art blind image quality assessment methods. We also show how our method can be applied to eliminate useless out-of-focus images in a High-Content-Screening experiment. We further evaluate the ability of our image quality ranking method to detect out-of-focus images, by extensive simulations, and by comparing its performance against previously published, well-established microscopy autofocus metrics. PMID:27364703

An Application of Hydraulic Tomography to a Large-Scale Fractured Granite Site, Mizunami, Japan.

PubMed

Zha, Yuanyuan; Yeh, Tian-Chyi J; Illman, Walter A; Tanaka, Tatsuya; Bruines, Patrick; Onoe, Hironori; Saegusa, Hiromitsu; Mao, Deqiang; Takeuchi, Shinji; Wen, Jet-Chau

2016-11-01

While hydraulic tomography (HT) is a mature aquifer characterization technology, its applications to characterize hydrogeology of kilometer-scale fault and fracture zones are rare. This paper sequentially analyzes datasets from two new pumping tests as well as those from two previous pumping tests analyzed by Illman et al. (2009) at a fractured granite site in Mizunami, Japan. Results of this analysis show that datasets from two previous pumping tests at one side of a fault zone as used in the previous study led to inaccurate mapping of fracture and fault zones. Inclusion of the datasets from the two new pumping tests (one of which was conducted on the other side of the fault) yields locations of the fault zone consistent with those based on geological mapping. The new datasets also produce a detailed image of the irregular fault zone, which is not available from geological investigation alone and the previous study. As a result, we conclude that if prior knowledge about geological structures at a field site is considered during the design of HT surveys, valuable non-redundant datasets about the fracture and fault zones can be collected. Only with these non-redundant data sets, can HT then be a viable and robust tool for delineating fracture and fault distributions over kilometer scales, even when only a limited number of boreholes are available. In essence, this paper proves that HT is a new tool for geologists, geophysicists, and engineers for mapping large-scale fracture and fault zone distributions. © 2016, National Ground Water Association.

Topic modeling for cluster analysis of large biological and medical datasets

PubMed Central

2014-01-01

Background The big data moniker is nowhere better deserved than to describe the ever-increasing prodigiousness and complexity of biological and medical datasets. New methods are needed to generate and test hypotheses, foster biological interpretation, and build validated predictors. Although multivariate techniques such as cluster analysis may allow researchers to identify groups, or clusters, of related variables, the accuracies and effectiveness of traditional clustering methods diminish for large and hyper dimensional datasets. Topic modeling is an active research field in machine learning and has been mainly used as an analytical tool to structure large textual corpora for data mining. Its ability to reduce high dimensionality to a small number of latent variables makes it suitable as a means for clustering or overcoming clustering difficulties in large biological and medical datasets. Results In this study, three topic model-derived clustering methods, highest probable topic assignment, feature selection and feature extraction, are proposed and tested on the cluster analysis of three large datasets: Salmonella pulsed-field gel electrophoresis (PFGE) dataset, lung cancer dataset, and breast cancer dataset, which represent various types of large biological or medical datasets. All three various methods are shown to improve the efficacy/effectiveness of clustering results on the three datasets in comparison to traditional methods. A preferable cluster analysis method emerged for each of the three datasets on the basis of replicating known biological truths. Conclusion Topic modeling could be advantageously applied to the large datasets of biological or medical research. The three proposed topic model-derived clustering methods, highest probable topic assignment, feature selection and feature extraction, yield clustering improvements for the three different data types. Clusters more efficaciously represent truthful groupings and subgroupings in the data than traditional methods, suggesting that topic model-based methods could provide an analytic advancement in the analysis of large biological or medical datasets. PMID:25350106

Topic modeling for cluster analysis of large biological and medical datasets.

PubMed

Zhao, Weizhong; Zou, Wen; Chen, James J

2014-01-01

The big data moniker is nowhere better deserved than to describe the ever-increasing prodigiousness and complexity of biological and medical datasets. New methods are needed to generate and test hypotheses, foster biological interpretation, and build validated predictors. Although multivariate techniques such as cluster analysis may allow researchers to identify groups, or clusters, of related variables, the accuracies and effectiveness of traditional clustering methods diminish for large and hyper dimensional datasets. Topic modeling is an active research field in machine learning and has been mainly used as an analytical tool to structure large textual corpora for data mining. Its ability to reduce high dimensionality to a small number of latent variables makes it suitable as a means for clustering or overcoming clustering difficulties in large biological and medical datasets. In this study, three topic model-derived clustering methods, highest probable topic assignment, feature selection and feature extraction, are proposed and tested on the cluster analysis of three large datasets: Salmonella pulsed-field gel electrophoresis (PFGE) dataset, lung cancer dataset, and breast cancer dataset, which represent various types of large biological or medical datasets. All three various methods are shown to improve the efficacy/effectiveness of clustering results on the three datasets in comparison to traditional methods. A preferable cluster analysis method emerged for each of the three datasets on the basis of replicating known biological truths. Topic modeling could be advantageously applied to the large datasets of biological or medical research. The three proposed topic model-derived clustering methods, highest probable topic assignment, feature selection and feature extraction, yield clustering improvements for the three different data types. Clusters more efficaciously represent truthful groupings and subgroupings in the data than traditional methods, suggesting that topic model-based methods could provide an analytic advancement in the analysis of large biological or medical datasets.

Human Vision-Motivated Algorithm Allows Consistent Retinal Vessel Classification Based on Local Color Contrast for Advancing General Diagnostic Exams.

PubMed

Ivanov, Iliya V; Leitritz, Martin A; Norrenberg, Lars A; Völker, Michael; Dynowski, Marek; Ueffing, Marius; Dietter, Johannes

2016-02-01

Abnormalities of blood vessel anatomy, morphology, and ratio can serve as important diagnostic markers for retinal diseases such as AMD or diabetic retinopathy. Large cohort studies demand automated and quantitative image analysis of vascular abnormalities. Therefore, we developed an analytical software tool to enable automated standardized classification of blood vessels supporting clinical reading. A dataset of 61 images was collected from a total of 33 women and 8 men with a median age of 38 years. The pupils were not dilated, and images were taken after dark adaption. In contrast to current methods in which classification is based on vessel profile intensity averages, and similar to human vision, local color contrast was chosen as a discriminator to allow artery vein discrimination and arterial-venous ratio (AVR) calculation without vessel tracking. With 83% ± 1 standard error of the mean for our dataset, we achieved best classification for weighted lightness information from a combination of the red, green, and blue channels. Tested on an independent dataset, our method reached 89% correct classification, which, when benchmarked against conventional ophthalmologic classification, shows significantly improved classification scores. Our study demonstrates that vessel classification based on local color contrast can cope with inter- or intraimage lightness variability and allows consistent AVR calculation. We offer an open-source implementation of this method upon request, which can be integrated into existing tool sets and applied to general diagnostic exams.

Deformable Image Registration based on Similarity-Steered CNN Regression.

PubMed

Cao, Xiaohuan; Yang, Jianhua; Zhang, Jun; Nie, Dong; Kim, Min-Jeong; Wang, Qian; Shen, Dinggang

2017-09-01

Existing deformable registration methods require exhaustively iterative optimization, along with careful parameter tuning, to estimate the deformation field between images. Although some learning-based methods have been proposed for initiating deformation estimation, they are often template-specific and not flexible in practical use. In this paper, we propose a convolutional neural network (CNN) based regression model to directly learn the complex mapping from the input image pair (i.e., a pair of template and subject) to their corresponding deformation field. Specifically, our CNN architecture is designed in a patch-based manner to learn the complex mapping from the input patch pairs to their respective deformation field. First, the equalized active-points guided sampling strategy is introduced to facilitate accurate CNN model learning upon a limited image dataset. Then, the similarity-steered CNN architecture is designed, where we propose to add the auxiliary contextual cue, i.e., the similarity between input patches, to more directly guide the learning process. Experiments on different brain image datasets demonstrate promising registration performance based on our CNN model. Furthermore, it is found that the trained CNN model from one dataset can be successfully transferred to another dataset, although brain appearances across datasets are quite variable.

Primate Brain Anatomy: New Volumetric MRI Measurements for Neuroanatomical Studies.

PubMed

Navarrete, Ana F; Blezer, Erwin L A; Pagnotta, Murillo; de Viet, Elizabeth S M; Todorov, Orlin S; Lindenfors, Patrik; Laland, Kevin N; Reader, Simon M

2018-06-12

Since the publication of the primate brain volumetric dataset of Stephan and colleagues in the early 1980s, no major new comparative datasets covering multiple brain regions and a large number of primate species have become available. However, technological and other advances in the last two decades, particularly magnetic resonance imaging (MRI) and the creation of institutions devoted to the collection and preservation of rare brain specimens, provide opportunities to rectify this situation. Here, we present a new dataset including brain region volumetric measurements of 39 species, including 20 species not previously available in the literature, with measurements of 16 brain areas. These volumes were extracted from MRI of 46 brains of 38 species from the Netherlands Institute of Neuroscience Primate Brain Bank, scanned at high resolution with a 9.4-T scanner, plus a further 7 donated MRI of 4 primate species. Partial measurements were made on an additional 8 brains of 5 species. We make the dataset and MRI scans available online in the hope that they will be of value to researchers conducting comparative studies of primate evolution. © 2018 S. Karger AG, Basel.

Unsupervised tattoo segmentation combining bottom-up and top-down cues

NASA Astrophysics Data System (ADS)

Allen, Josef D.; Zhao, Nan; Yuan, Jiangbo; Liu, Xiuwen

2011-06-01

Tattoo segmentation is challenging due to the complexity and large variance in tattoo structures. We have developed a segmentation algorithm for finding tattoos in an image. Our basic idea is split-merge: split each tattoo image into clusters through a bottom-up process, learn to merge the clusters containing skin and then distinguish tattoo from the other skin via top-down prior in the image itself. Tattoo segmentation with unknown number of clusters is transferred to a figureground segmentation. We have applied our segmentation algorithm on a tattoo dataset and the results have shown that our tattoo segmentation system is efficient and suitable for further tattoo classification and retrieval purpose.

X-ray EM simulation tool for ptychography dataset construction

NASA Astrophysics Data System (ADS)

Stoevelaar, L. Pjotr; Gerini, Giampiero

2018-03-01

In this paper, we present an electromagnetic full-wave modeling framework, as a support EM tool providing data sets for X-ray ptychographic imaging. Modeling the entire scattering problem with Finite Element Method (FEM) tools is, in fact, a prohibitive task, because of the large area illuminated by the beam (due to the poor focusing power at these wavelengths) and the very small features to be imaged. To overcome this problem, the spectrum of the illumination beam is decomposed into a discrete set of plane waves. This allows reducing the electromagnetic modeling volume to the one enclosing the area to be imaged. The total scattered field is reconstructed by superimposing the solutions for each plane wave illumination.

Automated Leaf Tracking using Multi-view Image Sequences of Maize Plants for Leaf-growth Monitoring

NASA Astrophysics Data System (ADS)

Das Choudhury, S.; Awada, T.; Samal, A.; Stoerger, V.; Bashyam, S.

2017-12-01

Extraction of phenotypes with botanical importance by analyzing plant image sequences has the desirable advantages of non-destructive temporal phenotypic measurements of a large number of plants with little or no manual intervention in a relatively short period of time. The health of a plant is best interpreted by the emergence timing and temporal growth of individual leaves. For automated leaf growth monitoring, it is essential to track each leaf throughout the life cycle of the plant. Plants are constantly changing organisms with increasing complexity in architecture due to variations in self-occlusions and phyllotaxy, i.e., arrangements of leaves around the stem. The leaf cross-overs pose challenges to accurately track each leaf using single view image sequence. Thus, we introduce a novel automated leaf tracking algorithm using a graph theoretic approach by multi-view image sequence analysis based on the determination of leaf-tips and leaf-junctions in the 3D space. The basis of the leaf tracking algorithm is: the leaves emerge using bottom-up approach in the case of a maize plant, and the direction of leaf emergence strictly alternates in terms of direction. The algorithm involves labeling of the individual parts of a plant, i.e., leaves and stem, following graphical representation of the plant skeleton, i.e., one-pixel wide connected line obtained from the binary image. The length of the leaf is measured by the number of pixels in the leaf skeleton. To evaluate the performance of the algorithm, a benchmark dataset is indispensable. Thus, we publicly release University of Nebraska-Lincoln Component Plant Phenotyping dataset-2 (UNL-CPPD-2) consisting of images of the 20 maize plants captured by visible light camera of the Lemnatec Scanalyzer 3D high throughout plant phenotyping facility once daily for 60 days from 10 different views. The dataset is aimed to facilitate the development and evaluation of leaf tracking algorithms and their uniform comparisons.

Constructing a Teleseismic Tomographic Image of Taiwan using BATS Recordings

NASA Astrophysics Data System (ADS)

Krajewski, J.; Roecker, S.

2005-12-01

Taiwan is an evolving arc-continent collision located at a complicated part of the plate boundary between the Eurasian and Philippine Sea plates. To better understand the role of the upper mantle in the dynamics of this collision, we reviewed 4 years of data from the Broadband Array in Taiwan for Seismology (BATS) in Taiwan to construct a teleseismic dataset for tomographic imaging of the subsurface of the island. From an initial selection of approximately 300 events, we used waveform correlation to generate a dataset of 4500 relative arrival times. To calculate accurate travel times in three dimensional wavespeed models over the large lateral distances in our model (~800 km), we solve the eikonal equation directly in a spherical coordinate system. We reduce the influence of smearing of crustal heterogeneity into the deeper mantle, we fix the upper 30 km to a previously determined P wavespeed model for the region. Initial resolution tests suggest a spatial limit on the order of 40 km.

Automation process for morphometric analysis of volumetric CT data from pulmonary vasculature in rats.

PubMed

Shingrani, Rahul; Krenz, Gary; Molthen, Robert

2010-01-01

With advances in medical imaging scanners, it has become commonplace to generate large multidimensional datasets. These datasets require tools for a rapid, thorough analysis. To address this need, we have developed an automated algorithm for morphometric analysis incorporating A Visualization Workshop computational and image processing libraries for three-dimensional segmentation, vascular tree generation and structural hierarchical ordering with a two-stage numeric optimization procedure for estimating vessel diameters. We combine this new technique with our mathematical models of pulmonary vascular morphology to quantify structural and functional attributes of lung arterial trees. Our physiological studies require repeated measurements of vascular structure to determine differences in vessel biomechanical properties between animal models of pulmonary disease. Automation provides many advantages including significantly improved speed and minimized operator interaction and biasing. The results are validated by comparison with previously published rat pulmonary arterial micro-CT data analysis techniques, in which vessels were manually mapped and measured using intense operator intervention. Published by Elsevier Ireland Ltd.

Using multimodal information for the segmentation of fluorescent micrographs with application to virology and microbiology.

PubMed

Held, Christian; Wenzel, Jens; Webel, Rike; Marschall, Manfred; Lang, Roland; Palmisano, Ralf; Wittenberg, Thomas

2011-01-01

In order to improve reproducibility and objectivity of fluorescence microscopy based experiments and to enable the evaluation of large datasets, flexible segmentation methods are required which are able to adapt to different stainings and cell types. This adaption is usually achieved by the manual adjustment of the segmentation methods parameters, which is time consuming and challenging for biologists with no knowledge on image processing. To avoid this, parameters of the presented methods automatically adapt to user generated ground truth to determine the best method and the optimal parameter setup. These settings can then be used for segmentation of the remaining images. As robust segmentation methods form the core of such a system, the currently used watershed transform based segmentation routine is replaced by a fast marching level set based segmentation routine which incorporates knowledge on the cell nuclei. Our evaluations reveal that incorporation of multimodal information improves segmentation quality for the presented fluorescent datasets.

Rapid and accurate intraoperative pathological diagnosis by artificial intelligence with deep learning technology.

PubMed

Zhang, Jing; Song, Yanlin; Xia, Fan; Zhu, Chenjing; Zhang, Yingying; Song, Wenpeng; Xu, Jianguo; Ma, Xuelei

2017-09-01

Frozen section is widely used for intraoperative pathological diagnosis (IOPD), which is essential for intraoperative decision making. However, frozen section suffers from some drawbacks, such as time consuming and high misdiagnosis rate. Recently, artificial intelligence (AI) with deep learning technology has shown bright future in medicine. We hypothesize that AI with deep learning technology could help IOPD, with a computer trained by a dataset of intraoperative lesion images. Evidences supporting our hypothesis included the successful use of AI with deep learning technology in diagnosing skin cancer, and the developed method of deep-learning algorithm. Large size of the training dataset is critical to increase the diagnostic accuracy. The performance of the trained machine could be tested by new images before clinical use. Real-time diagnosis, easy to use and potential high accuracy were the advantages of AI for IOPD. In sum, AI with deep learning technology is a promising method to help rapid and accurate IOPD. Copyright © 2017 Elsevier Ltd. All rights reserved.

Helioviewer.org: Browsing Very Large Image Archives Online Using JPEG 2000

NASA Astrophysics Data System (ADS)

Hughitt, V. K.; Ireland, J.; Mueller, D.; Dimitoglou, G.; Garcia Ortiz, J.; Schmidt, L.; Wamsler, B.; Beck, J.; Alexanderian, A.; Fleck, B.

2009-12-01

As the amount of solar data available to scientists continues to increase at faster and faster rates, it is important that there exist simple tools for navigating this data quickly with a minimal amount of effort. By combining heterogeneous solar physics datatypes such as full-disk images and coronagraphs, along with feature and event information, Helioviewer offers a simple and intuitive way to browse multiple datasets simultaneously. Images are stored in a repository using the JPEG 2000 format and tiled dynamically upon a client's request. By tiling images and serving only the portions of the image requested, it is possible for the client to work with very large images without having to fetch all of the data at once. In addition to a focus on intercommunication with other virtual observatories and browsers (VSO, HEK, etc), Helioviewer will offer a number of externally-available application programming interfaces (APIs) to enable easy third party use, adoption and extension. Recent efforts have resulted in increased performance, dynamic movie generation, and improved support for mobile web browsers. Future functionality will include: support for additional data-sources including RHESSI, SDO, STEREO, and TRACE, a navigable timeline of recorded solar events, social annotation, and basic client-side image processing.

ChemNet: A Transferable and Generalizable Deep Neural Network for Small-Molecule Property Prediction

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

Goh, Garrett B.; Siegel, Charles M.; Vishnu, Abhinav

With access to large datasets, deep neural networks through representation learning have been able to identify patterns from raw data, achieving human-level accuracy in image and speech recognition tasks. However, in chemistry, availability of large standardized and labelled datasets is scarce, and with a multitude of chemical properties of interest, chemical data is inherently small and fragmented. In this work, we explore transfer learning techniques in conjunction with the existing Chemception CNN model, to create a transferable and generalizable deep neural network for small-molecule property prediction. Our latest model, ChemNet learns in a semi-supervised manner from inexpensive labels computed frommore » the ChEMBL database. When fine-tuned to the Tox21, HIV and FreeSolv dataset, which are 3 separate chemical tasks that ChemNet was not originally trained on, we demonstrate that ChemNet exceeds the performance of existing Chemception models, contemporary MLP models that trains on molecular fingerprints, and it matches the performance of the ConvGraph algorithm, the current state-of-the-art. Furthermore, as ChemNet has been pre-trained on a large diverse chemical database, it can be used as a universal “plug-and-play” deep neural network, which accelerates the deployment of deep neural networks for the prediction of novel small-molecule chemical properties.« less

ROBIN: a platform for evaluating automatic target recognition algorithms: I. Overview of the project and presentation of the SAGEM DS competition

NASA Astrophysics Data System (ADS)

Duclos, D.; Lonnoy, J.; Guillerm, Q.; Jurie, F.; Herbin, S.; D'Angelo, E.

2008-04-01

The last five years have seen a renewal of Automatic Target Recognition applications, mainly because of the latest advances in machine learning techniques. In this context, large collections of image datasets are essential for training algorithms as well as for their evaluation. Indeed, the recent proliferation of recognition algorithms, generally applied to slightly different problems, make their comparisons through clean evaluation campaigns necessary. The ROBIN project tries to fulfil these two needs by putting unclassified datasets, ground truths, competitions and metrics for the evaluation of ATR algorithms at the disposition of the scientific community. The scope of this project includes single and multi-class generic target detection and generic target recognition, in military and security contexts. From our knowledge, it is the first time that a database of this importance (several hundred thousands of visible and infrared hand annotated images) has been publicly released. Funded by the French Ministry of Defence (DGA) and by the French Ministry of Research, ROBIN is one of the ten Techno-vision projects. Techno-vision is a large and ambitious government initiative for building evaluation means for computer vision technologies, for various application contexts. ROBIN's consortium includes major companies and research centres involved in Computer Vision R&D in the field of defence: Bertin Technologies, CNES, ECA, DGA, EADS, INRIA, ONERA, MBDA, SAGEM, THALES. This paper, which first gives an overview of the whole project, is focused on one of ROBIN's key competitions, the SAGEM Defence Security database. This dataset contains more than eight hundred ground and aerial infrared images of six different vehicles in cluttered scenes including distracters. Two different sets of data are available for each target. The first set includes different views of each vehicle at close range in a "simple" background, and can be used to train algorithms. The second set contains many views of the same vehicle in different contexts and situations simulating operational scenarios.

Dsm Based Orientation of Large Stereo Satellite Image Blocks

NASA Astrophysics Data System (ADS)

d'Angelo, P.; Reinartz, P.

2012-07-01

High resolution stereo satellite imagery is well suited for the creation of digital surface models (DSM). A system for highly automated and operational DSM and orthoimage generation based on CARTOSAT-1 imagery is presented, with emphasis on fully automated georeferencing. The proposed system processes level-1 stereo scenes using the rational polynomial coefficients (RPC) universal sensor model. The RPC are derived from orbit and attitude information and have a much lower accuracy than the ground resolution of approximately 2.5 m. In order to use the images for orthorectification or DSM generation, an affine RPC correction is required. In this paper, GCP are automatically derived from lower resolution reference datasets (Landsat ETM+ Geocover and SRTM DSM). The traditional method of collecting the lateral position from a reference image and interpolating the corresponding height from the DEM ignores the higher lateral accuracy of the SRTM dataset. Our method avoids this drawback by using a RPC correction based on DSM alignment, resulting in improved geolocation of both DSM and ortho images. Scene based method and a bundle block adjustment based correction are developed and evaluated for a test site covering the nothern part of Italy, for which 405 Cartosat-1 Stereopairs are available. Both methods are tested against independent ground truth. Checks against this ground truth indicate a lateral error of 10 meters.

Food Recognition: A New Dataset, Experiments, and Results.

PubMed

Ciocca, Gianluigi; Napoletano, Paolo; Schettini, Raimondo

2017-05-01

We propose a new dataset for the evaluation of food recognition algorithms that can be used in dietary monitoring applications. Each image depicts a real canteen tray with dishes and foods arranged in different ways. Each tray contains multiple instances of food classes. The dataset contains 1027 canteen trays for a total of 3616 food instances belonging to 73 food classes. The food on the tray images has been manually segmented using carefully drawn polygonal boundaries. We have benchmarked the dataset by designing an automatic tray analysis pipeline that takes a tray image as input, finds the regions of interest, and predicts for each region the corresponding food class. We have experimented with three different classification strategies using also several visual descriptors. We achieve about 79% of food and tray recognition accuracy using convolutional-neural-networks-based features. The dataset, as well as the benchmark framework, are available to the research community.

A Robust Post-Processing Workflow for Datasets with Motion Artifacts in Diffusion Kurtosis Imaging

PubMed Central

Li, Xianjun; Yang, Jian; Gao, Jie; Luo, Xue; Zhou, Zhenyu; Hu, Yajie; Wu, Ed X.; Wan, Mingxi

2014-01-01

Purpose The aim of this study was to develop a robust post-processing workflow for motion-corrupted datasets in diffusion kurtosis imaging (DKI). Materials and methods The proposed workflow consisted of brain extraction, rigid registration, distortion correction, artifacts rejection, spatial smoothing and tensor estimation. Rigid registration was utilized to correct misalignments. Motion artifacts were rejected by using local Pearson correlation coefficient (LPCC). The performance of LPCC in characterizing relative differences between artifacts and artifact-free images was compared with that of the conventional correlation coefficient in 10 randomly selected DKI datasets. The influence of rejected artifacts with information of gradient directions and b values for the parameter estimation was investigated by using mean square error (MSE). The variance of noise was used as the criterion for MSEs. The clinical practicality of the proposed workflow was evaluated by the image quality and measurements in regions of interest on 36 DKI datasets, including 18 artifact-free (18 pediatric subjects) and 18 motion-corrupted datasets (15 pediatric subjects and 3 essential tremor patients). Results The relative difference between artifacts and artifact-free images calculated by LPCC was larger than that of the conventional correlation coefficient (p<0.05). It indicated that LPCC was more sensitive in detecting motion artifacts. MSEs of all derived parameters from the reserved data after the artifacts rejection were smaller than the variance of the noise. It suggested that influence of rejected artifacts was less than influence of noise on the precision of derived parameters. The proposed workflow improved the image quality and reduced the measurement biases significantly on motion-corrupted datasets (p<0.05). Conclusion The proposed post-processing workflow was reliable to improve the image quality and the measurement precision of the derived parameters on motion-corrupted DKI datasets. The workflow provided an effective post-processing method for clinical applications of DKI in subjects with involuntary movements. PMID:24727862

A robust post-processing workflow for datasets with motion artifacts in diffusion kurtosis imaging.

PubMed

Li, Xianjun; Yang, Jian; Gao, Jie; Luo, Xue; Zhou, Zhenyu; Hu, Yajie; Wu, Ed X; Wan, Mingxi

2014-01-01

The aim of this study was to develop a robust post-processing workflow for motion-corrupted datasets in diffusion kurtosis imaging (DKI). The proposed workflow consisted of brain extraction, rigid registration, distortion correction, artifacts rejection, spatial smoothing and tensor estimation. Rigid registration was utilized to correct misalignments. Motion artifacts were rejected by using local Pearson correlation coefficient (LPCC). The performance of LPCC in characterizing relative differences between artifacts and artifact-free images was compared with that of the conventional correlation coefficient in 10 randomly selected DKI datasets. The influence of rejected artifacts with information of gradient directions and b values for the parameter estimation was investigated by using mean square error (MSE). The variance of noise was used as the criterion for MSEs. The clinical practicality of the proposed workflow was evaluated by the image quality and measurements in regions of interest on 36 DKI datasets, including 18 artifact-free (18 pediatric subjects) and 18 motion-corrupted datasets (15 pediatric subjects and 3 essential tremor patients). The relative difference between artifacts and artifact-free images calculated by LPCC was larger than that of the conventional correlation coefficient (p<0.05). It indicated that LPCC was more sensitive in detecting motion artifacts. MSEs of all derived parameters from the reserved data after the artifacts rejection were smaller than the variance of the noise. It suggested that influence of rejected artifacts was less than influence of noise on the precision of derived parameters. The proposed workflow improved the image quality and reduced the measurement biases significantly on motion-corrupted datasets (p<0.05). The proposed post-processing workflow was reliable to improve the image quality and the measurement precision of the derived parameters on motion-corrupted DKI datasets. The workflow provided an effective post-processing method for clinical applications of DKI in subjects with involuntary movements.

Direct reconstruction of parametric images for brain PET with event-by-event motion correction: evaluation in two tracers across count levels

NASA Astrophysics Data System (ADS)

Germino, Mary; Gallezot, Jean-Dominque; Yan, Jianhua; Carson, Richard E.

2017-07-01

Parametric images for dynamic positron emission tomography (PET) are typically generated by an indirect method, i.e. reconstructing a time series of emission images, then fitting a kinetic model to each voxel time activity curve. Alternatively, ‘direct reconstruction’, incorporates the kinetic model into the reconstruction algorithm itself, directly producing parametric images from projection data. Direct reconstruction has been shown to achieve parametric images with lower standard error than the indirect method. Here, we present direct reconstruction for brain PET using event-by-event motion correction of list-mode data, applied to two tracers. Event-by-event motion correction was implemented for direct reconstruction in the Parametric Motion-compensation OSEM List-mode Algorithm for Resolution-recovery reconstruction. The direct implementation was tested on simulated and human datasets with tracers [11C]AFM (serotonin transporter) and [11C]UCB-J (synaptic density), which follow the 1-tissue compartment model. Rigid head motion was tracked with the Vicra system. Parametric images of K 1 and distribution volume (V T  =  K 1/k 2) were compared to those generated by the indirect method by regional coefficient of variation (CoV). Performance across count levels was assessed using sub-sampled datasets. For simulated and real datasets at high counts, the two methods estimated K 1 and V T with comparable accuracy. At lower count levels, the direct method was substantially more robust to outliers than the indirect method. Compared to the indirect method, direct reconstruction reduced regional K 1 CoV by 35-48% (simulated dataset), 39-43% ([11C]AFM dataset) and 30-36% ([11C]UCB-J dataset) across count levels (averaged over regions at matched iteration); V T CoV was reduced by 51-58%, 54-60% and 30-46%, respectively. Motion correction played an important role in the dataset with larger motion: correction increased regional V T by 51% on average in the [11C]UCB-J dataset. Direct reconstruction of dynamic brain PET with event-by-event motion correction is achievable and dramatically more robust to noise in V T images than the indirect method.

Dynamic Moss Observed with Hi-C

NASA Technical Reports Server (NTRS)

Alexander, Caroline; Winebarger, Amy; Morton, Richard; Savage, Sabrina

2014-01-01

The High-resolution Coronal Imager (Hi-C), flown on 11 July 2012, has revealed an unprecedented level of detail and substructure within the solar corona. Hi-­-C imaged a large active region (AR11520) with 0.2-0.3'' spatial resolution and 5.5s cadence over a 5 minute period. An additional dataset with a smaller FOV, the same resolution, but with a higher temporal cadence (1s) was also taken during the rocket flight. This dataset was centered on a large patch of 'moss' emission that initially seemed to show very little variability. Image processing revealed this region to be much more dynamic than first thought with numerous bright and dark features observed to appear, move and disappear over the 5 minute observation. Moss is thought to be emission from the upper transition region component of hot loops so studying its dynamics and the relation between the bright/dark features and underlying magnetic features is important to tie the interaction of the different atmospheric layers together. Hi-C allows us to study the coronal emission of the moss at the smallest scales while data from SDO/AIA and HMI is used to give information on these structures at different heights/temperatures. Using the high temporal and spatial resolution of Hi-C the observed moss features were tracked and the distribution of displacements, speeds, and sizes were measured. This allows us to comment on both the physical processes occurring within the dynamic moss and the scales at which these changes are occurring.

Dynamic Moss Observed with Hi-C

NASA Technical Reports Server (NTRS)

Alexander, Caroline; Winebarger, Amy; Morton, Richard; Savage, Sabrina

2014-01-01

The High-resolution Coronal Imager (Hi-C), flown on 11 July 2012, has revealed an unprecedented level of detail and substructure within the solar corona. Hi-C imaged a large active region (AR11520) with 0.2-0.3'' spatial resolution and 5.5s cadence over a 5 minute period. An additional dataset with a smaller FOV, the same resolution, but with a higher temporal cadence (1s) was also taken during the rocket flight. This dataset was centered on a large patch of 'moss' emission that initially seemed to show very little variability. Image processing revealed this region to be much more dynamic than first thought with numerous bright and dark features observed to appear, move and disappear over the 5 minute observation. Moss is thought to be emission from the upper transition region component of hot loops so studying its dynamics and the relation between the bright/dark features and underlying magnetic features is important to tie the interaction of the different atmospheric layers together. Hi-C allows us to study the coronal emission of the moss at the smallest scales while data from SDO/AIA and HMI is used to give information on these structures at different heights/temperatures. Using the high temporal and spatial resolution of Hi-C the observed moss features were tracked and the distribution of displacements, speeds, and sizes were measured. This allows us to comment on both the physical processes occurring within the dynamic moss and the scales at which these changes are occurring.

Brain extraction from normal and pathological images: A joint PCA/Image-Reconstruction approach.

PubMed

Han, Xu; Kwitt, Roland; Aylward, Stephen; Bakas, Spyridon; Menze, Bjoern; Asturias, Alexander; Vespa, Paul; Van Horn, John; Niethammer, Marc

2018-08-01

Brain extraction from 3D medical images is a common pre-processing step. A variety of approaches exist, but they are frequently only designed to perform brain extraction from images without strong pathologies. Extracting the brain from images exhibiting strong pathologies, for example, the presence of a brain tumor or of a traumatic brain injury (TBI), is challenging. In such cases, tissue appearance may substantially deviate from normal tissue appearance and hence violates algorithmic assumptions for standard approaches to brain extraction; consequently, the brain may not be correctly extracted. This paper proposes a brain extraction approach which can explicitly account for pathologies by jointly modeling normal tissue appearance and pathologies. Specifically, our model uses a three-part image decomposition: (1) normal tissue appearance is captured by principal component analysis (PCA), (2) pathologies are captured via a total variation term, and (3) the skull and surrounding tissue is captured by a sparsity term. Due to its convexity, the resulting decomposition model allows for efficient optimization. Decomposition and image registration steps are alternated to allow statistical modeling of normal tissue appearance in a fixed atlas coordinate system. As a beneficial side effect, the decomposition model allows for the identification of potentially pathological areas and the reconstruction of a quasi-normal image in atlas space. We demonstrate the effectiveness of our approach on four datasets: the publicly available IBSR and LPBA40 datasets which show normal image appearance, the BRATS dataset containing images with brain tumors, and a dataset containing clinical TBI images. We compare the performance with other popular brain extraction models: ROBEX, BEaST, MASS, BET, BSE and a recently proposed deep learning approach. Our model performs better than these competing approaches on all four datasets. Specifically, our model achieves the best median (97.11) and mean (96.88) Dice scores over all datasets. The two best performing competitors, ROBEX and MASS, achieve scores of 96.23/95.62 and 96.67/94.25 respectively. Hence, our approach is an effective method for high quality brain extraction for a wide variety of images. Copyright © 2018 Elsevier Inc. All rights reserved.

MMX-I: A data-processing software for multi-modal X-ray imaging and tomography

NASA Astrophysics Data System (ADS)

Bergamaschi, A.; Medjoubi, K.; Messaoudi, C.; Marco, S.; Somogyi, A.

2017-06-01

Scanning hard X-ray imaging allows simultaneous acquisition of multimodal information, including X-ray fluorescence, absorption, phase and dark-field contrasts, providing structural and chemical details of the samples. Combining these scanning techniques with the infrastructure developed for fast data acquisition at Synchrotron Soleil permits to perform multimodal imaging and tomography during routine user experiments at the Nanoscopium beamline. A main challenge of such imaging techniques is the online processing and analysis of the generated very large volume (several hundreds of Giga Bytes) multimodal data-sets. This is especially important for the wide user community foreseen at the user oriented Nanoscopium beamline (e.g. from the fields of Biology, Life Sciences, Geology, Geobiology), having no experience in such data-handling. MMX-I is a new multi-platform open-source freeware for the processing and reconstruction of scanning multi-technique X-ray imaging and tomographic datasets. The MMX-I project aims to offer, both expert users and beginners, the possibility of processing and analysing raw data, either on-site or off-site. Therefore we have developed a multi-platform (Mac, Windows and Linux 64bit) data processing tool, which is easy to install, comprehensive, intuitive, extendable and user-friendly. MMX-I is now routinely used by the Nanoscopium user community and has demonstrated its performance in treating big data.

Exploratory visualization of astronomical data on ultra-high-resolution wall displays

NASA Astrophysics Data System (ADS)

Pietriga, Emmanuel; del Campo, Fernando; Ibsen, Amanda; Primet, Romain; Appert, Caroline; Chapuis, Olivier; Hempel, Maren; Muñoz, Roberto; Eyheramendy, Susana; Jordan, Andres; Dole, Hervé

2016-07-01

Ultra-high-resolution wall displays feature a very high pixel density over a large physical surface, which makes them well-suited to the collaborative, exploratory visualization of large datasets. We introduce FITS-OW, an application designed for such wall displays, that enables astronomers to navigate in large collections of FITS images, query astronomical databases, and display detailed, complementary data and documents about multiple sources simultaneously. We describe how astronomers interact with their data using both the wall's touchsensitive surface and handheld devices. We also report on the technical challenges we addressed in terms of distributed graphics rendering and data sharing over the computer clusters that drive wall displays.

MilxXplore: a web-based system to explore large imaging datasets.

PubMed

Bourgeat, P; Dore, V; Villemagne, V L; Rowe, C C; Salvado, O; Fripp, J

2013-01-01

As large-scale medical imaging studies are becoming more common, there is an increasing reliance on automated software to extract quantitative information from these images. As the size of the cohorts keeps increasing with large studies, there is a also a need for tools that allow results from automated image processing and analysis to be presented in a way that enables fast and efficient quality checking, tagging and reporting on cases in which automatic processing failed or was problematic. MilxXplore is an open source visualization platform, which provides an interface to navigate and explore imaging data in a web browser, giving the end user the opportunity to perform quality control and reporting in a user friendly, collaborative and efficient way. Compared to existing software solutions that often provide an overview of the results at the subject's level, MilxXplore pools the results of individual subjects and time points together, allowing easy and efficient navigation and browsing through the different acquisitions of a subject over time, and comparing the results against the rest of the population. MilxXplore is fast, flexible and allows remote quality checks of processed imaging data, facilitating data sharing and collaboration across multiple locations, and can be easily integrated into a cloud computing pipeline. With the growing trend of open data and open science, such a tool will become increasingly important to share and publish results of imaging analysis.

A new approach to optic disc detection in human retinal images using the firefly algorithm.

PubMed

Rahebi, Javad; Hardalaç, Fırat

2016-03-01

There are various methods and algorithms to detect the optic discs in retinal images. In recent years, much attention has been given to the utilization of the intelligent algorithms. In this paper, we present a new automated method of optic disc detection in human retinal images using the firefly algorithm. The firefly intelligent algorithm is an emerging intelligent algorithm that was inspired by the social behavior of fireflies. The population in this algorithm includes the fireflies, each of which has a specific rate of lighting or fitness. In this method, the insects are compared two by two, and the less attractive insects can be observed to move toward the more attractive insects. Finally, one of the insects is selected as the most attractive, and this insect presents the optimum response to the problem in question. Here, we used the light intensity of the pixels of the retinal image pixels instead of firefly lightings. The movement of these insects due to local fluctuations produces different light intensity values in the images. Because the optic disc is the brightest area in the retinal images, all of the insects move toward brightest area and thus specify the location of the optic disc in the image. The results of implementation show that proposed algorithm could acquire an accuracy rate of 100 % in DRIVE dataset, 95 % in STARE dataset, and 94.38 % in DiaRetDB1 dataset. The results of implementation reveal high capability and accuracy of proposed algorithm in the detection of the optic disc from retinal images. Also, recorded required time for the detection of the optic disc in these images is 2.13 s for DRIVE dataset, 2.81 s for STARE dataset, and 3.52 s for DiaRetDB1 dataset accordingly. These time values are average value.

Evaluation of privacy in high dynamic range video sequences

NASA Astrophysics Data System (ADS)

Řeřábek, Martin; Yuan, Lin; Krasula, Lukáš; Korshunov, Pavel; Fliegel, Karel; Ebrahimi, Touradj

2014-09-01

The ability of high dynamic range (HDR) to capture details in environments with high contrast has a significant impact on privacy in video surveillance. However, the extent to which HDR imaging affects privacy, when compared to a typical low dynamic range (LDR) imaging, is neither well studied nor well understood. To achieve such an objective, a suitable dataset of images and video sequences is needed. Therefore, we have created a publicly available dataset of HDR video for privacy evaluation PEViD-HDR, which is an HDR extension of an existing Privacy Evaluation Video Dataset (PEViD). PEViD-HDR video dataset can help in the evaluations of privacy protection tools, as well as for showing the importance of HDR imaging in video surveillance applications and its influence on the privacy-intelligibility trade-off. We conducted a preliminary subjective experiment demonstrating the usability of the created dataset for evaluation of privacy issues in video. The results confirm that a tone-mapped HDR video contains more privacy sensitive information and details compared to a typical LDR video.

Enhancing studies of the connectome in autism using the autism brain imaging data exchange II

PubMed Central

Di Martino, Adriana; O’Connor, David; Chen, Bosi; Alaerts, Kaat; Anderson, Jeffrey S.; Assaf, Michal; Balsters, Joshua H.; Baxter, Leslie; Beggiato, Anita; Bernaerts, Sylvie; Blanken, Laura M. E.; Bookheimer, Susan Y.; Braden, B. Blair; Byrge, Lisa; Castellanos, F. Xavier; Dapretto, Mirella; Delorme, Richard; Fair, Damien A.; Fishman, Inna; Fitzgerald, Jacqueline; Gallagher, Louise; Keehn, R. Joanne Jao; Kennedy, Daniel P.; Lainhart, Janet E.; Luna, Beatriz; Mostofsky, Stewart H.; Müller, Ralph-Axel; Nebel, Mary Beth; Nigg, Joel T.; O’Hearn, Kirsten; Solomon, Marjorie; Toro, Roberto; Vaidya, Chandan J.; Wenderoth, Nicole; White, Tonya; Craddock, R. Cameron; Lord, Catherine; Leventhal, Bennett; Milham, Michael P.

2017-01-01

The second iteration of the Autism Brain Imaging Data Exchange (ABIDE II) aims to enhance the scope of brain connectomics research in Autism Spectrum Disorder (ASD). Consistent with the initial ABIDE effort (ABIDE I), that released 1112 datasets in 2012, this new multisite open-data resource is an aggregate of resting state functional magnetic resonance imaging (MRI) and corresponding structural MRI and phenotypic datasets. ABIDE II includes datasets from an additional 487 individuals with ASD and 557 controls previously collected across 16 international institutions. The combination of ABIDE I and ABIDE II provides investigators with 2156 unique cross-sectional datasets allowing selection of samples for discovery and/or replication. This sample size can also facilitate the identification of neurobiological subgroups, as well as preliminary examinations of sex differences in ASD. Additionally, ABIDE II includes a range of psychiatric variables to inform our understanding of the neural correlates of co-occurring psychopathology; 284 diffusion imaging datasets are also included. It is anticipated that these enhancements will contribute to unraveling key sources of ASD heterogeneity. PMID:28291247

HEp-2 cell image classification method based on very deep convolutional networks with small datasets

NASA Astrophysics Data System (ADS)

Lu, Mengchi; Gao, Long; Guo, Xifeng; Liu, Qiang; Yin, Jianping

2017-07-01

Human Epithelial-2 (HEp-2) cell images staining patterns classification have been widely used to identify autoimmune diseases by the anti-Nuclear antibodies (ANA) test in the Indirect Immunofluorescence (IIF) protocol. Because manual test is time consuming, subjective and labor intensive, image-based Computer Aided Diagnosis (CAD) systems for HEp-2 cell classification are developing. However, methods proposed recently are mostly manual features extraction with low accuracy. Besides, the scale of available benchmark datasets is small, which does not exactly suitable for using deep learning methods. This issue will influence the accuracy of cell classification directly even after data augmentation. To address these issues, this paper presents a high accuracy automatic HEp-2 cell classification method with small datasets, by utilizing very deep convolutional networks (VGGNet). Specifically, the proposed method consists of three main phases, namely image preprocessing, feature extraction and classification. Moreover, an improved VGGNet is presented to address the challenges of small-scale datasets. Experimental results over two benchmark datasets demonstrate that the proposed method achieves superior performance in terms of accuracy compared with existing methods.

A Neuroelectrical Brain Imaging Study on the Perception of Figurative Paintings against Only their Color or Shape Contents.

PubMed

Maglione, Anton G; Brizi, Ambra; Vecchiato, Giovanni; Rossi, Dario; Trettel, Arianna; Modica, Enrica; Babiloni, Fabio

2017-01-01

In this study, the cortical activity correlated with the perception and appreciation of different set of pictures was estimated by using neuroelectric brain activity and graph theory methodologies in a group of artistic educated persons. The pictures shown to the subjects consisted of original pictures of Titian's and a contemporary artist's paintings (Orig dataset) plus two sets of additional pictures. These additional datasets were obtained from the previous paintings by removing all but the colors or the shapes employed (Color and Style dataset, respectively). Results suggest that the verbal appreciation of Orig dataset when compared to Color and Style ones was mainly correlated to the neuroelectric indexes estimated during the first 10 s of observation of the pictures. Always in the first 10 s of observation: (1) Orig dataset induced more emotion and is perceived with more appreciation than the other two Color and Style datasets; (2) Style dataset is perceived with more attentional effort than the other investigated datasets. During the whole period of observation of 30 s: (1) emotion induced by Color and Style datasets increased across the time while that induced of the Orig dataset remain stable; (2) Color and Style dataset were perceived with more attentional effort than the Orig dataset. During the entire experience, there is evidence of a cortical flow of activity from the parietal and central areas toward the prefrontal and frontal areas during the observation of the images of all the datasets. This is coherent from the notion that active perception of the images with sustained cognitive attention in parietal and central areas caused the generation of the judgment about their aesthetic appreciation in frontal areas.

A Neuroelectrical Brain Imaging Study on the Perception of Figurative Paintings against Only their Color or Shape Contents

PubMed Central

Maglione, Anton G.; Brizi, Ambra; Vecchiato, Giovanni; Rossi, Dario; Trettel, Arianna; Modica, Enrica; Babiloni, Fabio

2017-01-01

In this study, the cortical activity correlated with the perception and appreciation of different set of pictures was estimated by using neuroelectric brain activity and graph theory methodologies in a group of artistic educated persons. The pictures shown to the subjects consisted of original pictures of Titian's and a contemporary artist's paintings (Orig dataset) plus two sets of additional pictures. These additional datasets were obtained from the previous paintings by removing all but the colors or the shapes employed (Color and Style dataset, respectively). Results suggest that the verbal appreciation of Orig dataset when compared to Color and Style ones was mainly correlated to the neuroelectric indexes estimated during the first 10 s of observation of the pictures. Always in the first 10 s of observation: (1) Orig dataset induced more emotion and is perceived with more appreciation than the other two Color and Style datasets; (2) Style dataset is perceived with more attentional effort than the other investigated datasets. During the whole period of observation of 30 s: (1) emotion induced by Color and Style datasets increased across the time while that induced of the Orig dataset remain stable; (2) Color and Style dataset were perceived with more attentional effort than the Orig dataset. During the entire experience, there is evidence of a cortical flow of activity from the parietal and central areas toward the prefrontal and frontal areas during the observation of the images of all the datasets. This is coherent from the notion that active perception of the images with sustained cognitive attention in parietal and central areas caused the generation of the judgment about their aesthetic appreciation in frontal areas. PMID:28790907

Artificial Intelligence and Machine Learning in Radiology: Opportunities, Challenges, Pitfalls, and Criteria for Success.

PubMed

Thrall, James H; Li, Xiang; Li, Quanzheng; Cruz, Cinthia; Do, Synho; Dreyer, Keith; Brink, James

2018-03-01

Worldwide interest in artificial intelligence (AI) applications, including imaging, is high and growing rapidly, fueled by availability of large datasets ("big data"), substantial advances in computing power, and new deep-learning algorithms. Apart from developing new AI methods per se, there are many opportunities and challenges for the imaging community, including the development of a common nomenclature, better ways to share image data, and standards for validating AI program use across different imaging platforms and patient populations. AI surveillance programs may help radiologists prioritize work lists by identifying suspicious or positive cases for early review. AI programs can be used to extract "radiomic" information from images not discernible by visual inspection, potentially increasing the diagnostic and prognostic value derived from image datasets. Predictions have been made that suggest AI will put radiologists out of business. This issue has been overstated, and it is much more likely that radiologists will beneficially incorporate AI methods into their practices. Current limitations in availability of technical expertise and even computing power will be resolved over time and can also be addressed by remote access solutions. Success for AI in imaging will be measured by value created: increased diagnostic certainty, faster turnaround, better outcomes for patients, and better quality of work life for radiologists. AI offers a new and promising set of methods for analyzing image data. Radiologists will explore these new pathways and are likely to play a leading role in medical applications of AI. Copyright © 2017 American College of Radiology. Published by Elsevier Inc. All rights reserved.

An application of cascaded 3D fully convolutional networks for medical image segmentation.

PubMed

Roth, Holger R; Oda, Hirohisa; Zhou, Xiangrong; Shimizu, Natsuki; Yang, Ying; Hayashi, Yuichiro; Oda, Masahiro; Fujiwara, Michitaka; Misawa, Kazunari; Mori, Kensaku

2018-06-01

Recent advances in 3D fully convolutional networks (FCN) have made it feasible to produce dense voxel-wise predictions of volumetric images. In this work, we show that a multi-class 3D FCN trained on manually labeled CT scans of several anatomical structures (ranging from the large organs to thin vessels) can achieve competitive segmentation results, while avoiding the need for handcrafting features or training class-specific models. To this end, we propose a two-stage, coarse-to-fine approach that will first use a 3D FCN to roughly define a candidate region, which will then be used as input to a second 3D FCN. This reduces the number of voxels the second FCN has to classify to ∼10% and allows it to focus on more detailed segmentation of the organs and vessels. We utilize training and validation sets consisting of 331 clinical CT images and test our models on a completely unseen data collection acquired at a different hospital that includes 150 CT scans, targeting three anatomical organs (liver, spleen, and pancreas). In challenging organs such as the pancreas, our cascaded approach improves the mean Dice score from 68.5 to 82.2%, achieving the highest reported average score on this dataset. We compare with a 2D FCN method on a separate dataset of 240 CT scans with 18 classes and achieve a significantly higher performance in small organs and vessels. Furthermore, we explore fine-tuning our models to different datasets. Our experiments illustrate the promise and robustness of current 3D FCN based semantic segmentation of medical images, achieving state-of-the-art results. 1 . Copyright © 2018 Elsevier Ltd. All rights reserved.

Deterministic object tracking using Gaussian ringlet and directional edge features

NASA Astrophysics Data System (ADS)

Krieger, Evan W.; Sidike, Paheding; Aspiras, Theus; Asari, Vijayan K.

2017-10-01

Challenges currently existing for intensity-based histogram feature tracking methods in wide area motion imagery (WAMI) data include object structural information distortions, background variations, and object scale change. These issues are caused by different pavement or ground types and from changing the sensor or altitude. All of these challenges need to be overcome in order to have a robust object tracker, while attaining a computation time appropriate for real-time processing. To achieve this, we present a novel method, Directional Ringlet Intensity Feature Transform (DRIFT), which employs Kirsch kernel filtering for edge features and a ringlet feature mapping for rotational invariance. The method also includes an automatic scale change component to obtain accurate object boundaries and improvements for lowering computation times. We evaluated the DRIFT algorithm on two challenging WAMI datasets, namely Columbus Large Image Format (CLIF) and Large Area Image Recorder (LAIR), to evaluate its robustness and efficiency. Additional evaluations on general tracking video sequences are performed using the Visual Tracker Benchmark and Visual Object Tracking 2014 databases to demonstrate the algorithms ability with additional challenges in long complex sequences including scale change. Experimental results show that the proposed approach yields competitive results compared to state-of-the-art object tracking methods on the testing datasets.

Open Science CBS Neuroimaging Repository: Sharing ultra-high-field MR images of the brain.

PubMed

Tardif, Christine Lucas; Schäfer, Andreas; Trampel, Robert; Villringer, Arno; Turner, Robert; Bazin, Pierre-Louis

2016-01-01

Magnetic resonance imaging at ultra high field opens the door to quantitative brain imaging at sub-millimeter isotropic resolutions. However, novel image processing tools to analyze these new rich datasets are lacking. In this article, we introduce the Open Science CBS Neuroimaging Repository: a unique repository of high-resolution and quantitative images acquired at 7 T. The motivation for this project is to increase interest for high-resolution and quantitative imaging and stimulate the development of image processing tools developed specifically for high-field data. Our growing repository currently includes datasets from MP2RAGE and multi-echo FLASH sequences from 28 and 20 healthy subjects respectively. These datasets represent the current state-of-the-art in in-vivo relaxometry at 7 T, and are now fully available to the entire neuroimaging community. Copyright © 2015 Elsevier Inc. All rights reserved.

Classification of brain MRI with big data and deep 3D convolutional neural networks

NASA Astrophysics Data System (ADS)

Wegmayr, Viktor; Aitharaju, Sai; Buhmann, Joachim

2018-02-01

Our ever-aging society faces the growing problem of neurodegenerative diseases, in particular dementia. Magnetic Resonance Imaging provides a unique tool for non-invasive investigation of these brain diseases. However, it is extremely difficult for neurologists to identify complex disease patterns from large amounts of three-dimensional images. In contrast, machine learning excels at automatic pattern recognition from large amounts of data. In particular, deep learning has achieved impressive results in image classification. Unfortunately, its application to medical image classification remains difficult. We consider two reasons for this difficulty: First, volumetric medical image data is considerably scarcer than natural images. Second, the complexity of 3D medical images is much higher compared to common 2D images. To address the problem of small data set size, we assemble the largest dataset ever used for training a deep 3D convolutional neural network to classify brain images as healthy (HC), mild cognitive impairment (MCI) or Alzheimers disease (AD). We use more than 20.000 images from subjects of these three classes, which is almost 9x the size of the previously largest data set. The problem of high dimensionality is addressed by using a deep 3D convolutional neural network, which is state-of-the-art in large-scale image classification. We exploit its ability to process the images directly, only with standard preprocessing, but without the need for elaborate feature engineering. Compared to other work, our workflow is considerably simpler, which increases clinical applicability. Accuracy is measured on the ADNI+AIBL data sets, and the independent CADDementia benchmark.

Deep Adaptive Log-Demons: Diffeomorphic Image Registration with Very Large Deformations

PubMed Central

Jia, Kebin

2015-01-01

This paper proposes a new framework for capturing large and complex deformation in image registration. Traditionally, this challenging problem relies firstly on a preregistration, usually an affine matrix containing rotation, scale, and translation and afterwards on a nonrigid transformation. According to preregistration, the directly calculated affine matrix, which is obtained by limited pixel information, may misregistrate when large biases exist, thus misleading following registration subversively. To address this problem, for two-dimensional (2D) images, the two-layer deep adaptive registration framework proposed in this paper firstly accurately classifies the rotation parameter through multilayer convolutional neural networks (CNNs) and then identifies scale and translation parameters separately. For three-dimensional (3D) images, affine matrix is located through feature correspondences by a triplanar 2D CNNs. Then deformation removal is done iteratively through preregistration and demons registration. By comparison with the state-of-the-art registration framework, our method gains more accurate registration results on both synthetic and real datasets. Besides, principal component analysis (PCA) is combined with correlation like Pearson and Spearman to form new similarity standards in 2D and 3D registration. Experiment results also show faster convergence speed. PMID:26120356

Deep Adaptive Log-Demons: Diffeomorphic Image Registration with Very Large Deformations.

PubMed

Zhao, Liya; Jia, Kebin

2015-01-01

This paper proposes a new framework for capturing large and complex deformation in image registration. Traditionally, this challenging problem relies firstly on a preregistration, usually an affine matrix containing rotation, scale, and translation and afterwards on a nonrigid transformation. According to preregistration, the directly calculated affine matrix, which is obtained by limited pixel information, may misregistrate when large biases exist, thus misleading following registration subversively. To address this problem, for two-dimensional (2D) images, the two-layer deep adaptive registration framework proposed in this paper firstly accurately classifies the rotation parameter through multilayer convolutional neural networks (CNNs) and then identifies scale and translation parameters separately. For three-dimensional (3D) images, affine matrix is located through feature correspondences by a triplanar 2D CNNs. Then deformation removal is done iteratively through preregistration and demons registration. By comparison with the state-of-the-art registration framework, our method gains more accurate registration results on both synthetic and real datasets. Besides, principal component analysis (PCA) is combined with correlation like Pearson and Spearman to form new similarity standards in 2D and 3D registration. Experiment results also show faster convergence speed.

Towards Exascale Seismic Imaging and Inversion

NASA Astrophysics Data System (ADS)

Tromp, J.; Bozdag, E.; Lefebvre, M. P.; Smith, J. A.; Lei, W.; Ruan, Y.

2015-12-01

Post-petascale supercomputers are now available to solve complex scientific problems that were thought unreachable a few decades ago. They also bring a cohort of concerns tied to obtaining optimum performance. Several issues are currently being investigated by the HPC community. These include energy consumption, fault resilience, scalability of the current parallel paradigms, workflow management, I/O performance and feature extraction with large datasets. In this presentation, we focus on the last three issues. In the context of seismic imaging and inversion, in particular for simulations based on adjoint methods, workflows are well defined.They consist of a few collective steps (e.g., mesh generation or model updates) and of a large number of independent steps (e.g., forward and adjoint simulations of each seismic event, pre- and postprocessing of seismic traces). The greater goal is to reduce the time to solution, that is, obtaining a more precise representation of the subsurface as fast as possible. This brings us to consider both the workflow in its entirety and the parts comprising it. The usual approach is to speedup the purely computational parts based on code optimization in order to reach higher FLOPS and better memory management. This still remains an important concern, but larger scale experiments show that the imaging workflow suffers from severe I/O bottlenecks. Such limitations occur both for purely computational data and seismic time series. The latter are dealt with by the introduction of a new Adaptable Seismic Data Format (ASDF). Parallel I/O libraries, namely HDF5 and ADIOS, are used to drastically reduce the cost of disk access. Parallel visualization tools, such as VisIt, are able to take advantage of ADIOS metadata to extract features and display massive datasets. Because large parts of the workflow are embarrassingly parallel, we are investigating the possibility of automating the imaging process with the integration of scientific workflow management tools, specifically Pegasus.

Combined 60° Wide-Field Choroidal Thickness Maps and High-Definition En Face Vasculature Visualization Using Swept-Source Megahertz OCT at 1050 nm

PubMed Central

Mohler, Kathrin J.; Draxinger, Wolfgang; Klein, Thomas; Kolb, Jan Philip; Wieser, Wolfgang; Haritoglou, Christos; Kampik, Anselm; Fujimoto, James G.; Neubauer, Aljoscha S.; Huber, Robert; Wolf, Armin

2015-01-01

Purpose To demonstrate ultrahigh-speed swept-source optical coherence tomography (SS-OCT) at 1.68 million A-scans/s for choroidal imaging in normal and diseased eyes over a ∼60° field of view. To investigate and correlate wide-field three-dimensional (3D) choroidal thickness (ChT) and vascular patterns using ChT maps and coregistered high-definition en face images extracted from a single densely sampled Megahertz-OCT (MHz-OCT) dataset. Methods High-definition, ∼60° wide-field 3D datasets consisting of 2088 × 1024 A-scans were acquired using a 1.68 MHz prototype SS-OCT system at 1050 nm based on a Fourier-domain mode-locked laser. Nine subjects (nine eyes) with various chorioretinal diseases or without ocular pathology are presented. Coregistered ChT maps, choroidal summation maps, and depth-resolved en face images referenced to either the retinal pigment epithelium or the choroidal–scleral interface were generated using manual segmentation. Results Wide-field ChT maps showed a large inter- and intraindividual variance in peripheral and central ChT. In only four of the nine eyes, the location with the largest ChT was coincident with the fovea. The anatomy of the large lumen vessels of the outer choroid seems to play a major role in determining the global ChT pattern. Focal ChT changes with large thickness gradients were observed in some eyes. Conclusions Different ChT and vascular patterns could be visualized over ∼60° in patients for the first time using OCT. Due to focal ChT changes, a high density of thickness measurements may be favorable. High-definition depth-resolved en face images are complementary to cross sections and thickness maps and enhance the interpretation of different ChT patterns. PMID:26431482

Combined 60° Wide-Field Choroidal Thickness Maps and High-Definition En Face Vasculature Visualization Using Swept-Source Megahertz OCT at 1050 nm.

PubMed

Mohler, Kathrin J; Draxinger, Wolfgang; Klein, Thomas; Kolb, Jan Philip; Wieser, Wolfgang; Haritoglou, Christos; Kampik, Anselm; Fujimoto, James G; Neubauer, Aljoscha S; Huber, Robert; Wolf, Armin

2015-10-01

To demonstrate ultrahigh-speed swept-source optical coherence tomography (SS-OCT) at 1.68 million A-scans/s for choroidal imaging in normal and diseased eyes over a ∼60° field of view. To investigate and correlate wide-field three-dimensional (3D) choroidal thickness (ChT) and vascular patterns using ChT maps and coregistered high-definition en face images extracted from a single densely sampled Megahertz-OCT (MHz-OCT) dataset. High-definition, ∼60° wide-field 3D datasets consisting of 2088 × 1024 A-scans were acquired using a 1.68 MHz prototype SS-OCT system at 1050 nm based on a Fourier-domain mode-locked laser. Nine subjects (nine eyes) with various chorioretinal diseases or without ocular pathology are presented. Coregistered ChT maps, choroidal summation maps, and depth-resolved en face images referenced to either the retinal pigment epithelium or the choroidal-scleral interface were generated using manual segmentation. Wide-field ChT maps showed a large inter- and intraindividual variance in peripheral and central ChT. In only four of the nine eyes, the location with the largest ChT was coincident with the fovea. The anatomy of the large lumen vessels of the outer choroid seems to play a major role in determining the global ChT pattern. Focal ChT changes with large thickness gradients were observed in some eyes. Different ChT and vascular patterns could be visualized over ∼60° in patients for the first time using OCT. Due to focal ChT changes, a high density of thickness measurements may be favorable. High-definition depth-resolved en face images are complementary to cross sections and thickness maps and enhance the interpretation of different ChT patterns.

Weakly supervised image semantic segmentation based on clustering superpixels

NASA Astrophysics Data System (ADS)

Yan, Xiong; Liu, Xiaohua

2018-04-01

In this paper, we propose an image semantic segmentation model which is trained from image-level labeled images. The proposed model starts with superpixel segmenting, and features of the superpixels are extracted by trained CNN. We introduce a superpixel-based graph followed by applying the graph partition method to group correlated superpixels into clusters. For the acquisition of inter-label correlations between the image-level labels in dataset, we not only utilize label co-occurrence statistics but also exploit visual contextual cues simultaneously. At last, we formulate the task of mapping appropriate image-level labels to the detected clusters as a problem of convex minimization. Experimental results on MSRC-21 dataset and LableMe dataset show that the proposed method has a better performance than most of the weakly supervised methods and is even comparable to fully supervised methods.

Accurate Learning with Few Atlases (ALFA): an algorithm for MRI neonatal brain extraction and comparison with 11 publicly available methods.

PubMed

Serag, Ahmed; Blesa, Manuel; Moore, Emma J; Pataky, Rozalia; Sparrow, Sarah A; Wilkinson, A G; Macnaught, Gillian; Semple, Scott I; Boardman, James P

2016-03-24

Accurate whole-brain segmentation, or brain extraction, of magnetic resonance imaging (MRI) is a critical first step in most neuroimage analysis pipelines. The majority of brain extraction algorithms have been developed and evaluated for adult data and their validity for neonatal brain extraction, which presents age-specific challenges for this task, has not been established. We developed a novel method for brain extraction of multi-modal neonatal brain MR images, named ALFA (Accurate Learning with Few Atlases). The method uses a new sparsity-based atlas selection strategy that requires a very limited number of atlases 'uniformly' distributed in the low-dimensional data space, combined with a machine learning based label fusion technique. The performance of the method for brain extraction from multi-modal data of 50 newborns is evaluated and compared with results obtained using eleven publicly available brain extraction methods. ALFA outperformed the eleven compared methods providing robust and accurate brain extraction results across different modalities. As ALFA can learn from partially labelled datasets, it can be used to segment large-scale datasets efficiently. ALFA could also be applied to other imaging modalities and other stages across the life course.

BreakingNews: Article Annotation by Image and Text Processing.

PubMed

Ramisa, Arnau; Yan, Fei; Moreno-Noguer, Francesc; Mikolajczyk, Krystian

2018-05-01

Building upon recent Deep Neural Network architectures, current approaches lying in the intersection of Computer Vision and Natural Language Processing have achieved unprecedented breakthroughs in tasks like automatic captioning or image retrieval. Most of these learning methods, though, rely on large training sets of images associated with human annotations that specifically describe the visual content. In this paper we propose to go a step further and explore the more complex cases where textual descriptions are loosely related to the images. We focus on the particular domain of news articles in which the textual content often expresses connotative and ambiguous relations that are only suggested but not directly inferred from images. We introduce an adaptive CNN architecture that shares most of the structure for multiple tasks including source detection, article illustration and geolocation of articles. Deep Canonical Correlation Analysis is deployed for article illustration, and a new loss function based on Great Circle Distance is proposed for geolocation. Furthermore, we present BreakingNews, a novel dataset with approximately 100K news articles including images, text and captions, and enriched with heterogeneous meta-data (such as GPS coordinates and user comments). We show this dataset to be appropriate to explore all aforementioned problems, for which we provide a baseline performance using various Deep Learning architectures, and different representations of the textual and visual features. We report very promising results and bring to light several limitations of current state-of-the-art in this kind of domain, which we hope will help spur progress in the field.

An Active Patch Model for Real World Texture and Appearance Classification

PubMed Central

Mao, Junhua; Zhu, Jun; Yuille, Alan L.

2014-01-01

This paper addresses the task of natural texture and appearance classification. Our goal is to develop a simple and intuitive method that performs at state of the art on datasets ranging from homogeneous texture (e.g., material texture), to less homogeneous texture (e.g., the fur of animals), and to inhomogeneous texture (the appearance patterns of vehicles). Our method uses a bag-of-words model where the features are based on a dictionary of active patches. Active patches are raw intensity patches which can undergo spatial transformations (e.g., rotation and scaling) and adjust themselves to best match the image regions. The dictionary of active patches is required to be compact and representative, in the sense that we can use it to approximately reconstruct the images that we want to classify. We propose a probabilistic model to quantify the quality of image reconstruction and design a greedy learning algorithm to obtain the dictionary. We classify images using the occurrence frequency of the active patches. Feature extraction is fast (about 100 ms per image) using the GPU. The experimental results show that our method improves the state of the art on a challenging material texture benchmark dataset (KTH-TIPS2). To test our method on less homogeneous or inhomogeneous images, we construct two new datasets consisting of appearance image patches of animals and vehicles cropped from the PASCAL VOC dataset. Our method outperforms competing methods on these datasets. PMID:25531013

Parallel fuzzy connected image segmentation on GPU

PubMed Central

Zhuge, Ying; Cao, Yong; Udupa, Jayaram K.; Miller, Robert W.

2011-01-01

Purpose: Image segmentation techniques using fuzzy connectedness (FC) principles have shown their effectiveness in segmenting a variety of objects in several large applications. However, one challenge in these algorithms has been their excessive computational requirements when processing large image datasets. Nowadays, commodity graphics hardware provides a highly parallel computing environment. In this paper, the authors present a parallel fuzzy connected image segmentation algorithm implementation on NVIDIA’s compute unified device Architecture (cuda) platform for segmenting medical image data sets. Methods: In the FC algorithm, there are two major computational tasks: (i) computing the fuzzy affinity relations and (ii) computing the fuzzy connectedness relations. These two tasks are implemented as cuda kernels and executed on GPU. A dramatic improvement in speed for both tasks is achieved as a result. Results: Our experiments based on three data sets of small, medium, and large data size demonstrate the efficiency of the parallel algorithm, which achieves a speed-up factor of 24.4x, 18.1x, and 10.3x, correspondingly, for the three data sets on the NVIDIA Tesla C1060 over the implementation of the algorithm on CPU, and takes 0.25, 0.72, and 15.04 s, correspondingly, for the three data sets. Conclusions: The authors developed a parallel algorithm of the widely used fuzzy connected image segmentation method on the NVIDIA GPUs, which are far more cost- and speed-effective than both cluster of workstations and multiprocessing systems. A near-interactive speed of segmentation has been achieved, even for the large data set. PMID:21859037

Parallel fuzzy connected image segmentation on GPU.

PubMed

Zhuge, Ying; Cao, Yong; Udupa, Jayaram K; Miller, Robert W

2011-07-01

Image segmentation techniques using fuzzy connectedness (FC) principles have shown their effectiveness in segmenting a variety of objects in several large applications. However, one challenge in these algorithms has been their excessive computational requirements when processing large image datasets. Nowadays, commodity graphics hardware provides a highly parallel computing environment. In this paper, the authors present a parallel fuzzy connected image segmentation algorithm implementation on NVIDIA's compute unified device Architecture (CUDA) platform for segmenting medical image data sets. In the FC algorithm, there are two major computational tasks: (i) computing the fuzzy affinity relations and (ii) computing the fuzzy connectedness relations. These two tasks are implemented as CUDA kernels and executed on GPU. A dramatic improvement in speed for both tasks is achieved as a result. Our experiments based on three data sets of small, medium, and large data size demonstrate the efficiency of the parallel algorithm, which achieves a speed-up factor of 24.4x, 18.1x, and 10.3x, correspondingly, for the three data sets on the NVIDIA Tesla C1060 over the implementation of the algorithm on CPU, and takes 0.25, 0.72, and 15.04 s, correspondingly, for the three data sets. The authors developed a parallel algorithm of the widely used fuzzy connected image segmentation method on the NVIDIA GPUs, which are far more cost- and speed-effective than both cluster of workstations and multiprocessing systems. A near-interactive speed of segmentation has been achieved, even for the large data set.

Discovering Cortical Folding Patterns in Neonatal Cortical Surfaces Using Large-Scale Dataset

PubMed Central

Meng, Yu; Li, Gang; Wang, Li; Lin, Weili; Gilmore, John H.

2017-01-01

The cortical folding of the human brain is highly complex and variable across individuals. Mining the major patterns of cortical folding from modern large-scale neuroimaging datasets is of great importance in advancing techniques for neuroimaging analysis and understanding the inter-individual variations of cortical folding and its relationship with cognitive function and disorders. As the primary cortical folding is genetically influenced and has been established at term birth, neonates with the minimal exposure to the complicated postnatal environmental influence are the ideal candidates for understanding the major patterns of cortical folding. In this paper, for the first time, we propose a novel method for discovering the major patterns of cortical folding in a large-scale dataset of neonatal brain MR images (N = 677). In our method, first, cortical folding is characterized by the distribution of sulcal pits, which are the locally deepest points in cortical sulci. Because deep sulcal pits are genetically related, relatively consistent across individuals, and also stable during brain development, they are well suitable for representing and characterizing cortical folding. Then, the similarities between sulcal pit distributions of any two subjects are measured from spatial, geometrical, and topological points of view. Next, these different measurements are adaptively fused together using a similarity network fusion technique, to preserve their common information and also catch their complementary information. Finally, leveraging the fused similarity measurements, a hierarchical affinity propagation algorithm is used to group similar sulcal folding patterns together. The proposed method has been applied to 677 neonatal brains (the largest neonatal dataset to our knowledge) in the central sulcus, superior temporal sulcus, and cingulate sulcus, and revealed multiple distinct and meaningful folding patterns in each region. PMID:28229131

MultiMap: A Tool to Automatically Extract and Analyse Spatial Microscopic Data From Large Stacks of Confocal Microscopy Images

PubMed Central

Varando, Gherardo; Benavides-Piccione, Ruth; Muñoz, Alberto; Kastanauskaite, Asta; Bielza, Concha; Larrañaga, Pedro; DeFelipe, Javier

2018-01-01

The development of 3D visualization and reconstruction methods to analyse microscopic structures at different levels of resolutions is of great importance to define brain microorganization and connectivity. MultiMap is a new tool that allows the visualization, 3D segmentation and quantification of fluorescent structures selectively in the neuropil from large stacks of confocal microscopy images. The major contribution of this tool is the posibility to easily navigate and create regions of interest of any shape and size within a large brain area that will be automatically 3D segmented and quantified to determine the density of puncta in the neuropil. As a proof of concept, we focused on the analysis of glutamatergic and GABAergic presynaptic axon terminals in the mouse hippocampal region to demonstrate its use as a tool to provide putative excitatory and inhibitory synaptic maps. The segmentation and quantification method has been validated over expert labeled images of the mouse hippocampus and over two benchmark datasets, obtaining comparable results to the expert detections. PMID:29875639

MultiMap: A Tool to Automatically Extract and Analyse Spatial Microscopic Data From Large Stacks of Confocal Microscopy Images.

PubMed

Varando, Gherardo; Benavides-Piccione, Ruth; Muñoz, Alberto; Kastanauskaite, Asta; Bielza, Concha; Larrañaga, Pedro; DeFelipe, Javier

2018-01-01

The development of 3D visualization and reconstruction methods to analyse microscopic structures at different levels of resolutions is of great importance to define brain microorganization and connectivity. MultiMap is a new tool that allows the visualization, 3D segmentation and quantification of fluorescent structures selectively in the neuropil from large stacks of confocal microscopy images. The major contribution of this tool is the posibility to easily navigate and create regions of interest of any shape and size within a large brain area that will be automatically 3D segmented and quantified to determine the density of puncta in the neuropil. As a proof of concept, we focused on the analysis of glutamatergic and GABAergic presynaptic axon terminals in the mouse hippocampal region to demonstrate its use as a tool to provide putative excitatory and inhibitory synaptic maps. The segmentation and quantification method has been validated over expert labeled images of the mouse hippocampus and over two benchmark datasets, obtaining comparable results to the expert detections.

A hybrid approach for fusing 4D-MRI temporal information with 3D-CT for the study of lung and lung tumor motion.

PubMed

Yang, Y X; Teo, S-K; Van Reeth, E; Tan, C H; Tham, I W K; Poh, C L

2015-08-01

Accurate visualization of lung motion is important in many clinical applications, such as radiotherapy of lung cancer. Advancement in imaging modalities [e.g., computed tomography (CT) and MRI] has allowed dynamic imaging of lung and lung tumor motion. However, each imaging modality has its advantages and disadvantages. The study presented in this paper aims at generating synthetic 4D-CT dataset for lung cancer patients by combining both continuous three-dimensional (3D) motion captured by 4D-MRI and the high spatial resolution captured by CT using the authors' proposed approach. A novel hybrid approach based on deformable image registration (DIR) and finite element method simulation was developed to fuse a static 3D-CT volume (acquired under breath-hold) and the 3D motion information extracted from 4D-MRI dataset, creating a synthetic 4D-CT dataset. The study focuses on imaging of lung and lung tumor. Comparing the synthetic 4D-CT dataset with the acquired 4D-CT dataset of six lung cancer patients based on 420 landmarks, accurate results (average error <2 mm) were achieved using the authors' proposed approach. Their hybrid approach achieved a 40% error reduction (based on landmarks assessment) over using only DIR techniques. The synthetic 4D-CT dataset generated has high spatial resolution, has excellent lung details, and is able to show movement of lung and lung tumor over multiple breathing cycles.

Fully automated macular pathology detection in retina optical coherence tomography images using sparse coding and dictionary learning

NASA Astrophysics Data System (ADS)

Sun, Yankui; Li, Shan; Sun, Zhongyang

2017-01-01

We propose a framework for automated detection of dry age-related macular degeneration (AMD) and diabetic macular edema (DME) from retina optical coherence tomography (OCT) images, based on sparse coding and dictionary learning. The study aims to improve the classification performance of state-of-the-art methods. First, our method presents a general approach to automatically align and crop retina regions; then it obtains global representations of images by using sparse coding and a spatial pyramid; finally, a multiclass linear support vector machine classifier is employed for classification. We apply two datasets for validating our algorithm: Duke spectral domain OCT (SD-OCT) dataset, consisting of volumetric scans acquired from 45 subjects-15 normal subjects, 15 AMD patients, and 15 DME patients; and clinical SD-OCT dataset, consisting of 678 OCT retina scans acquired from clinics in Beijing-168, 297, and 213 OCT images for AMD, DME, and normal retinas, respectively. For the former dataset, our classifier correctly identifies 100%, 100%, and 93.33% of the volumes with DME, AMD, and normal subjects, respectively, and thus performs much better than the conventional method; for the latter dataset, our classifier leads to a correct classification rate of 99.67%, 99.67%, and 100.00% for DME, AMD, and normal images, respectively.

Fully automated macular pathology detection in retina optical coherence tomography images using sparse coding and dictionary learning.

PubMed

Sun, Yankui; Li, Shan; Sun, Zhongyang

2017-01-01

We propose a framework for automated detection of dry age-related macular degeneration (AMD) and diabetic macular edema (DME) from retina optical coherence tomography (OCT) images, based on sparse coding and dictionary learning. The study aims to improve the classification performance of state-of-the-art methods. First, our method presents a general approach to automatically align and crop retina regions; then it obtains global representations of images by using sparse coding and a spatial pyramid; finally, a multiclass linear support vector machine classifier is employed for classification. We apply two datasets for validating our algorithm: Duke spectral domain OCT (SD-OCT) dataset, consisting of volumetric scans acquired from 45 subjects—15 normal subjects, 15 AMD patients, and 15 DME patients; and clinical SD-OCT dataset, consisting of 678 OCT retina scans acquired from clinics in Beijing—168, 297, and 213 OCT images for AMD, DME, and normal retinas, respectively. For the former dataset, our classifier correctly identifies 100%, 100%, and 93.33% of the volumes with DME, AMD, and normal subjects, respectively, and thus performs much better than the conventional method; for the latter dataset, our classifier leads to a correct classification rate of 99.67%, 99.67%, and 100.00% for DME, AMD, and normal images, respectively.

Clinical Value of Dual-energy CT in Detection of Pancreatic Adenocarcinoma: Investigation of the Best Pancreatic Tumor Contrast to Noise Ratio.

PubMed

He, Yong-Lan; Zhang, Da-Ming; Xue, Hua-Dan; Jin, Zheng-Yu

2013-01-01

Objective To quantitatively compare and determine the best pancreatic tumor contrast to noise ratio (CNR) in different dual-energy derived datasets. Methods In this retrospective, single center study, 16 patients (9 male, 7 female, average age 59.4±13.2 years) with pathologically diagnosed pancreatic cancer were enrolled. All patients received an abdominal scan using a dual source CT scanner 7 to 31 days before biopsy or surgery. After injection of iodine contrast agent, arterial and pancreatic parenchyma phase were scanned consequently, using a dual-energy scan mode (100 kVp/230 mAs and Sn 140 kVp/178 mAs) in the pancreatic parenchyma phase. A series of derived dual-energy datasets were evaluated including non-liner blending (non-linear blending width 0-500 HU; blending center -500 to 500 HU), mono-energetic (40-190 keV), 100 kVp and 140 kVp. On each datasets, mean CT values of the pancreatic parenchyma and tumor, as well as standard deviation CT values of subcutaneous fat and psoas muscle were measured. Regions of interest of cutaneous fat and major psoas muscle of 100 kVp and 140 kVp images were calculated. Best CNR of subcutaneous fat (CNRF) and CNR of the major psoas muscle (CNRM) of non-liner blending and mono-energetic datasets were calculated with the optimal mono-energetic keV setting and the optimal blending center/width setting for the best CNR. One Way ANOVA test was used for comparison of best CNR between different dual-energy derived datasets. Results The best CNRF (4.48±1.29) was obtained from the non-liner blending datasets at blending center -16.6±103.9 HU and blending width 12.3±10.6 HU. The best CNRF (3.28±0.97) was obtained from the mono-energetic datasets at 73.3±4.3 keV. CNRF in the 100 kVp and 140 kVp were 3.02±0.91 and 1.56±0.56 respectively. Using fat as the noise background, all of these images series showed significant differences (P<0.01) except best CNRF of mono-energetic image sets vs. CNRF of 100 kVp image (P=0.460). Similar results were found using muscle as the noise background (mono-energetic image vs. 100 kVp image: P=0.246; mono-energetic image vs. non-liner blending image: P=0.044; others: P<0.01). Conclusion Compared with mono-energetic datasets and low kVp datasets, non-linear blending image at automatically chosen blending width/window provides better tumor to the pancreas CNR, which might be beneficial for better detection of pancreatic tumors.

Reconstruction of Human Monte Carlo Geometry from Segmented Images

NASA Astrophysics Data System (ADS)

Zhao, Kai; Cheng, Mengyun; Fan, Yanchang; Wang, Wen; Long, Pengcheng; Wu, Yican

2014-06-01

Human computational phantoms have been used extensively for scientific experimental analysis and experimental simulation. This article presented a method for human geometry reconstruction from a series of segmented images of a Chinese visible human dataset. The phantom geometry could actually describe detailed structure of an organ and could be converted into the input file of the Monte Carlo codes for dose calculation. A whole-body computational phantom of Chinese adult female has been established by FDS Team which is named Rad-HUMAN with about 28.8 billion voxel number. For being processed conveniently, different organs on images were segmented with different RGB colors and the voxels were assigned with positions of the dataset. For refinement, the positions were first sampled. Secondly, the large sums of voxels inside the organ were three-dimensional adjacent, however, there were not thoroughly mergence methods to reduce the cell amounts for the description of the organ. In this study, the voxels on the organ surface were taken into consideration of the mergence which could produce fewer cells for the organs. At the same time, an indexed based sorting algorithm was put forward for enhancing the mergence speed. Finally, the Rad-HUMAN which included a total of 46 organs and tissues was described by the cuboids into the Monte Carlo Monte Carlo Geometry for the simulation. The Monte Carlo geometry was constructed directly from the segmented images and the voxels was merged exhaustively. Each organ geometry model was constructed without ambiguity and self-crossing, its geometry information could represent the accuracy appearance and precise interior structure of the organs. The constructed geometry largely retaining the original shape of organs could easily be described into different Monte Carlo codes input file such as MCNP. Its universal property was testified and high-performance was experimentally verified

Multi-site study of additive genetic effects on fractional anisotropy of cerebral white matter: Comparing meta and megaanalytical approaches for data pooling.

PubMed

Kochunov, Peter; Jahanshad, Neda; Sprooten, Emma; Nichols, Thomas E; Mandl, René C; Almasy, Laura; Booth, Tom; Brouwer, Rachel M; Curran, Joanne E; de Zubicaray, Greig I; Dimitrova, Rali; Duggirala, Ravi; Fox, Peter T; Hong, L Elliot; Landman, Bennett A; Lemaitre, Hervé; Lopez, Lorna M; Martin, Nicholas G; McMahon, Katie L; Mitchell, Braxton D; Olvera, Rene L; Peterson, Charles P; Starr, John M; Sussmann, Jessika E; Toga, Arthur W; Wardlaw, Joanna M; Wright, Margaret J; Wright, Susan N; Bastin, Mark E; McIntosh, Andrew M; Boomsma, Dorret I; Kahn, René S; den Braber, Anouk; de Geus, Eco J C; Deary, Ian J; Hulshoff Pol, Hilleke E; Williamson, Douglas E; Blangero, John; van 't Ent, Dennis; Thompson, Paul M; Glahn, David C

2014-07-15

Combining datasets across independent studies can boost statistical power by increasing the numbers of observations and can achieve more accurate estimates of effect sizes. This is especially important for genetic studies where a large number of observations are required to obtain sufficient power to detect and replicate genetic effects. There is a need to develop and evaluate methods for joint-analytical analyses of rich datasets collected in imaging genetics studies. The ENIGMA-DTI consortium is developing and evaluating approaches for obtaining pooled estimates of heritability through meta-and mega-genetic analytical approaches, to estimate the general additive genetic contributions to the intersubject variance in fractional anisotropy (FA) measured from diffusion tensor imaging (DTI). We used the ENIGMA-DTI data harmonization protocol for uniform processing of DTI data from multiple sites. We evaluated this protocol in five family-based cohorts providing data from a total of 2248 children and adults (ages: 9-85) collected with various imaging protocols. We used the imaging genetics analysis tool, SOLAR-Eclipse, to combine twin and family data from Dutch, Australian and Mexican-American cohorts into one large "mega-family". We showed that heritability estimates may vary from one cohort to another. We used two meta-analytical (the sample-size and standard-error weighted) approaches and a mega-genetic analysis to calculate heritability estimates across-population. We performed leave-one-out analysis of the joint estimates of heritability, removing a different cohort each time to understand the estimate variability. Overall, meta- and mega-genetic analyses of heritability produced robust estimates of heritability. Copyright © 2014 Elsevier Inc. All rights reserved.

Binary patterns encoded convolutional neural networks for texture recognition and remote sensing scene classification

NASA Astrophysics Data System (ADS)

Anwer, Rao Muhammad; Khan, Fahad Shahbaz; van de Weijer, Joost; Molinier, Matthieu; Laaksonen, Jorma

2018-04-01

Designing discriminative powerful texture features robust to realistic imaging conditions is a challenging computer vision problem with many applications, including material recognition and analysis of satellite or aerial imagery. In the past, most texture description approaches were based on dense orderless statistical distribution of local features. However, most recent approaches to texture recognition and remote sensing scene classification are based on Convolutional Neural Networks (CNNs). The de facto practice when learning these CNN models is to use RGB patches as input with training performed on large amounts of labeled data (ImageNet). In this paper, we show that Local Binary Patterns (LBP) encoded CNN models, codenamed TEX-Nets, trained using mapped coded images with explicit LBP based texture information provide complementary information to the standard RGB deep models. Additionally, two deep architectures, namely early and late fusion, are investigated to combine the texture and color information. To the best of our knowledge, we are the first to investigate Binary Patterns encoded CNNs and different deep network fusion architectures for texture recognition and remote sensing scene classification. We perform comprehensive experiments on four texture recognition datasets and four remote sensing scene classification benchmarks: UC-Merced with 21 scene categories, WHU-RS19 with 19 scene classes, RSSCN7 with 7 categories and the recently introduced large scale aerial image dataset (AID) with 30 aerial scene types. We demonstrate that TEX-Nets provide complementary information to standard RGB deep model of the same network architecture. Our late fusion TEX-Net architecture always improves the overall performance compared to the standard RGB network on both recognition problems. Furthermore, our final combination leads to consistent improvement over the state-of-the-art for remote sensing scene classification.

Can the usage of human growth hormones affect facial appearance and the accuracy of face recognition systems?

NASA Astrophysics Data System (ADS)

Rose, Jake; Martin, Michael; Bourlai, Thirimachos

2014-06-01

In law enforcement and security applications, the acquisition of face images is critical in producing key trace evidence for the successful identification of potential threats. The goal of the study is to demonstrate that steroid usage significantly affects human facial appearance and hence, the performance of commercial and academic face recognition (FR) algorithms. In this work, we evaluate the performance of state-of-the-art FR algorithms on two unique face image datasets of subjects before (gallery set) and after (probe set) steroid (or human growth hormone) usage. For the purpose of this study, datasets of 73 subjects were created from multiple sources found on the Internet, containing images of men and women before and after steroid usage. Next, we geometrically pre-processed all images of both face datasets. Then, we applied image restoration techniques on the same face datasets, and finally, we applied FR algorithms in order to match the pre-processed face images of our probe datasets against the face images of the gallery set. Experimental results demonstrate that only a specific set of FR algorithms obtain the most accurate results (in terms of the rank-1 identification rate). This is because there are several factors that influence the efficiency of face matchers including (i) the time lapse between the before and after image pre-processing and restoration face photos, (ii) the usage of different drugs (e.g. Dianabol, Winstrol, and Decabolan), (iii) the usage of different cameras to capture face images, and finally, (iv) the variability of standoff distance, illumination and other noise factors (e.g. motion noise). All of the previously mentioned complicated scenarios make clear that cross-scenario matching is a very challenging problem and, thus, further investigation is required.

SAR image dataset of military ground targets with multiple poses for ATR

NASA Astrophysics Data System (ADS)

Belloni, Carole; Balleri, Alessio; Aouf, Nabil; Merlet, Thomas; Le Caillec, Jean-Marc

2017-10-01

Automatic Target Recognition (ATR) is the task of automatically detecting and classifying targets. Recognition using Synthetic Aperture Radar (SAR) images is interesting because SAR images can be acquired at night and under any weather conditions, whereas optical sensors operating in the visible band do not have this capability. Existing SAR ATR algorithms have mostly been evaluated using the MSTAR dataset.1 The problem with the MSTAR is that some of the proposed ATR methods have shown good classification performance even when targets were hidden,2 suggesting the presence of a bias in the dataset. Evaluations of SAR ATR techniques are currently challenging due to the lack of publicly available data in the SAR domain. In this paper, we present a high resolution SAR dataset consisting of images of a set of ground military target models taken at various aspect angles, The dataset can be used for a fair evaluation and comparison of SAR ATR algorithms. We applied the Inverse Synthetic Aperture Radar (ISAR) technique to echoes from targets rotating on a turntable and illuminated with a stepped frequency waveform. The targets in the database consist of four variants of two 1.7m-long models of T-64 and T-72 tanks. The gun, the turret position and the depression angle are varied to form 26 different sequences of images. The emitted signal spanned the frequency range from 13 GHz to 18 GHz to achieve a bandwidth of 5 GHz sampled with 4001 frequency points. The resolution obtained with respect to the size of the model targets is comparable to typical values obtained using SAR airborne systems. Single polarized images (Horizontal-Horizontal) are generated using the backprojection algorithm.3 A total of 1480 images are produced using a 20° integration angle. The images in the dataset are organized in a suggested training and testing set to facilitate a standard evaluation of SAR ATR algorithms.

New Hubble Space Telescope Multi-Wavelength Imaging of the Eagle Nebula

NASA Astrophysics Data System (ADS)

Levay, Zoltan G.; Christian, Carol A.; Mack, Jennifer; Frattare, Lisa M.; Livio, Mario; Meyett, Michele L.; Mutchler, Maximilian J.; Noll, Keith S.; Hubble Heritage

2015-01-01

One of the most iconic images from the Hubble Space Telescope has been the 1995 WFPC2 image of the Eagle Nebula (M16, sometimes known as the "Pillars of Creation"). Nineteen years after those original observations, new images have been obtained with HST's current instrumentation: a small mosaic in visible-light, narrow-band filters with WFC3/UVIS, infrared, broad-band filters with WFC3/IR, and parallel Hα imaging with ACS/WFC. The wider field of view, higher resolution, and broader wavelength coverage of the new images highlight the improved capabilities of HST over its long-lasting operation, made possible by the upgraded instrumentation installed during Space Shuttle servicing missions. Csite images from these datasets are presented to commemorate the 25th anniversary of HST's launch. Carefully combined, aligned and calibrated datasets from the primary WFC3 fields are available as High-Level Science Products in MAST (http://archive.stsci.edu/prepds/heritage/). Color composite images from these datasets are presented to commemorate the 25th anniversary of HST's launch.

NutriNet: A Deep Learning Food and Drink Image Recognition System for Dietary Assessment.

PubMed

Mezgec, Simon; Koroušić Seljak, Barbara

2017-06-27

Automatic food image recognition systems are alleviating the process of food-intake estimation and dietary assessment. However, due to the nature of food images, their recognition is a particularly challenging task, which is why traditional approaches in the field have achieved a low classification accuracy. Deep neural networks have outperformed such solutions, and we present a novel approach to the problem of food and drink image detection and recognition that uses a newly-defined deep convolutional neural network architecture, called NutriNet. This architecture was tuned on a recognition dataset containing 225,953 512 × 512 pixel images of 520 different food and drink items from a broad spectrum of food groups, on which we achieved a classification accuracy of 86 . 72 % , along with an accuracy of 94 . 47 % on a detection dataset containing 130 , 517 images. We also performed a real-world test on a dataset of self-acquired images, combined with images from Parkinson's disease patients, all taken using a smartphone camera, achieving a top-five accuracy of 55 % , which is an encouraging result for real-world images. Additionally, we tested NutriNet on the University of Milano-Bicocca 2016 (UNIMIB2016) food image dataset, on which we improved upon the provided baseline recognition result. An online training component was implemented to continually fine-tune the food and drink recognition model on new images. The model is being used in practice as part of a mobile app for the dietary assessment of Parkinson's disease patients.

PHOG analysis of self-similarity in aesthetic images

NASA Astrophysics Data System (ADS)

Amirshahi, Seyed Ali; Koch, Michael; Denzler, Joachim; Redies, Christoph

2012-03-01

In recent years, there have been efforts in defining the statistical properties of aesthetic photographs and artworks using computer vision techniques. However, it is still an open question how to distinguish aesthetic from non-aesthetic images with a high recognition rate. This is possibly because aesthetic perception is influenced also by a large number of cultural variables. Nevertheless, the search for statistical properties of aesthetic images has not been futile. For example, we have shown that the radially averaged power spectrum of monochrome artworks of Western and Eastern provenance falls off according to a power law with increasing spatial frequency (1/f2 characteristics). This finding implies that this particular subset of artworks possesses a Fourier power spectrum that is self-similar across different scales of spatial resolution. Other types of aesthetic images, such as cartoons, comics and mangas also display this type of self-similarity, as do photographs of complex natural scenes. Since the human visual system is adapted to encode images of natural scenes in a particular efficient way, we have argued that artists imitate these statistics in their artworks. In support of this notion, we presented results that artists portrait human faces with the self-similar Fourier statistics of complex natural scenes although real-world photographs of faces are not self-similar. In view of these previous findings, we investigated other statistical measures of self-similarity to characterize aesthetic and non-aesthetic images. In the present work, we propose a novel measure of self-similarity that is based on the Pyramid Histogram of Oriented Gradients (PHOG). For every image, we first calculate PHOG up to pyramid level 3. The similarity between the histograms of each section at a particular level is then calculated to the parent section at the previous level (or to the histogram at the ground level). The proposed approach is tested on datasets of aesthetic and non-aesthetic categories of monochrome images. The aesthetic image datasets comprise a large variety of artworks of Western provenance. Other man-made aesthetically pleasing images, such as comics, cartoons and mangas, were also studied. For comparison, a database of natural scene photographs is used, as well as datasets of photographs of plants, simple objects and faces that are in general of low aesthetic value. As expected, natural scenes exhibit the highest degree of PHOG self-similarity. Images of artworks also show high selfsimilarity values, followed by cartoons, comics and mangas. On average, other (non-aesthetic) image categories are less self-similar in the PHOG analysis. A measure of scale-invariant self-similarity (PHOG) allows a good separation of the different aesthetic and non-aesthetic image categories. Our results provide further support for the notion that, like complex natural scenes, images of artworks display a higher degree of self-similarity across different scales of resolution than other image categories. Whether the high degree of self-similarity is the basis for the perception of beauty in both complex natural scenery and artworks remains to be investigated.

Extraction of endoscopic images for biomedical figure classification

NASA Astrophysics Data System (ADS)

Xue, Zhiyun; You, Daekeun; Chachra, Suchet; Antani, Sameer; Long, L. R.; Demner-Fushman, Dina; Thoma, George R.

2015-03-01

Modality filtering is an important feature in biomedical image searching systems and may significantly improve the retrieval performance of the system. This paper presents a new method for extracting endoscopic image figures from photograph images in biomedical literature, which are found to have highly diverse content and large variability in appearance. Our proposed method consists of three main stages: tissue image extraction, endoscopic image candidate extraction, and ophthalmic image filtering. For tissue image extraction we use image patch level clustering and MRF relabeling to detect images containing skin/tissue regions. Next, we find candidate endoscopic images by exploiting the round shape characteristics that commonly appear in these images. However, this step needs to compensate for images where endoscopic regions are not entirely round. In the third step we filter out the ophthalmic images which have shape characteristics very similar to the endoscopic images. We do this by using text information, specifically, anatomy terms, extracted from the figure caption. We tested and evaluated our method on a dataset of 115,370 photograph figures, and achieved promising precision and recall rates of 87% and 84%, respectively.

Compression of the Global Land 1-km AVHRR dataset

USGS Publications Warehouse

Kess, B. L.; Steinwand, D.R.; Reichenbach, S.E.

1996-01-01

Large datasets, such as the Global Land 1-km Advanced Very High Resolution Radiometer (AVHRR) Data Set (Eidenshink and Faundeen 1994), require compression methods that provide efficient storage and quick access to portions of the data. A method of lossless compression is described that provides multiresolution decompression within geographic subwindows of multi-spectral, global, 1-km, AVHRR images. The compression algorithm segments each image into blocks and compresses each block in a hierarchical format. Users can access the data by specifying either a geographic subwindow or the whole image and a resolution (1,2,4, 8, or 16 km). The Global Land 1-km AVHRR data are presented in the Interrupted Goode's Homolosine map projection. These images contain masked regions for non-land areas which comprise 80 per cent of the image. A quadtree algorithm is used to compress the masked regions. The compressed region data are stored separately from the compressed land data. Results show that the masked regions compress to 0·143 per cent of the bytes they occupy in the test image and the land areas are compressed to 33·2 per cent of their original size. The entire image is compressed hierarchically to 6·72 per cent of the original image size, reducing the data from 9·05 gigabytes to 623 megabytes. These results are compared to the first order entropy of the residual image produced with lossless Joint Photographic Experts Group predictors. Compression results are also given for Lempel-Ziv-Welch (LZW) and LZ77, the algorithms used by UNIX compress and GZIP respectively. In addition to providing multiresolution decompression of geographic subwindows of the data, the hierarchical approach and the use of quadtrees for storing the masked regions gives a marked improvement over these popular methods.

A peek into the future of radiology using big data applications.

PubMed

Kharat, Amit T; Singhal, Shubham

2017-01-01

Big data is extremely large amount of data which is available in the radiology department. Big data is identified by four Vs - Volume, Velocity, Variety, and Veracity. By applying different algorithmic tools and converting raw data to transformed data in such large datasets, there is a possibility of understanding and using radiology data for gaining new knowledge and insights. Big data analytics consists of 6Cs - Connection, Cloud, Cyber, Content, Community, and Customization. The global technological prowess and per-capita capacity to save digital information has roughly doubled every 40 months since the 1980's. By using big data, the planning and implementation of radiological procedures in radiology departments can be given a great boost. Potential applications of big data in the future are scheduling of scans, creating patient-specific personalized scanning protocols, radiologist decision support, emergency reporting, virtual quality assurance for the radiologist, etc. Targeted use of big data applications can be done for images by supporting the analytic process. Screening software tools designed on big data can be used to highlight a region of interest, such as subtle changes in parenchymal density, solitary pulmonary nodule, or focal hepatic lesions, by plotting its multidimensional anatomy. Following this, we can run more complex applications such as three-dimensional multi planar reconstructions (MPR), volumetric rendering (VR), and curved planar reconstruction, which consume higher system resources on targeted data subsets rather than querying the complete cross-sectional imaging dataset. This pre-emptive selection of dataset can substantially reduce the system requirements such as system memory, server load and provide prompt results. However, a word of caution, "big data should not become "dump data" due to inadequate and poor analysis and non-structured improperly stored data. In the near future, big data can ring in the era of personalized and individualized healthcare.

Processing large remote sensing image data sets on Beowulf clusters

USGS Publications Warehouse

Steinwand, Daniel R.; Maddox, Brian; Beckmann, Tim; Schmidt, Gail

2003-01-01

High-performance computing is often concerned with the speed at which floating- point calculations can be performed. The architectures of many parallel computers and/or their network topologies are based on these investigations. Often, benchmarks resulting from these investigations are compiled with little regard to how a large dataset would move about in these systems. This part of the Beowulf study addresses that concern by looking at specific applications software and system-level modifications. Applications include an implementation of a smoothing filter for time-series data, a parallel implementation of the decision tree algorithm used in the Landcover Characterization project, a parallel Kriging algorithm used to fit point data collected in the field on invasive species to a regular grid, and modifications to the Beowulf project's resampling algorithm to handle larger, higher resolution datasets at a national scale. Systems-level investigations include a feasibility study on Flat Neighborhood Networks and modifications of that concept with Parallel File Systems.

A computational image analysis glossary for biologists.

PubMed

Roeder, Adrienne H K; Cunha, Alexandre; Burl, Michael C; Meyerowitz, Elliot M

2012-09-01

Recent advances in biological imaging have resulted in an explosion in the quality and quantity of images obtained in a digital format. Developmental biologists are increasingly acquiring beautiful and complex images, thus creating vast image datasets. In the past, patterns in image data have been detected by the human eye. Larger datasets, however, necessitate high-throughput objective analysis tools to computationally extract quantitative information from the images. These tools have been developed in collaborations between biologists, computer scientists, mathematicians and physicists. In this Primer we present a glossary of image analysis terms to aid biologists and briefly discuss the importance of robust image analysis in developmental studies.

Creating a seamless 1 km resolution daily land surface temperature dataset for urban and surrounding areas in the conterminous United States

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

Li, Xiaoma; Zhou, Yuyu; Asrar, Ghassem R.

High spatiotemporal land surface temperature (LST) datasets are increasingly needed in a variety of fields such as ecology, hydrology, meteorology, epidemiology, and energy systems. Moderate Resolution Imaging Spectroradiometer (MODIS) LST is one of such high spatiotemporal datasets that are widely used. But, it has large amount of missing values primarily because of clouds. Gapfilling the missing values is an important approach to create high spatiotemporal LST datasets. However current gapfilling methods have limitations in terms of accuracy and time required to assemble the data over large areas (e.g., national and continental levels). In this study, we developed a 3-step hybridmore » method by integrating a combination of daily merging, spatiotemporal gapfilling, and temporal interpolation methods, to create a high spatiotemporal LST dataset using the four daily LST observations from the two MODIS instruments on Terra and Aqua satellites. We applied this method in urban and surrounding areas for the conterminous U.S. in 2010. The evaluation of the gapfilled LST product indicates that its root mean squared error (RMSE) to be 3.3K for mid-daytime (1:30 pm) and 2.7K for mid-13 nighttime (1:30 am) observations. The method can be easily extended to other years and regions and is also applicable to other satellite products. This seamless daily (mid-daytime and mid-nighttime) LST product with 1 km spatial resolution is of great value for studying effects of urbanization (e.g., urban heat island) and the related impacts on people, ecosystems, energy systems and other infrastructure for cities.« less

Evaluation of the sparse coding super-resolution method for improving image quality of up-sampled images in computed tomography

NASA Astrophysics Data System (ADS)

Ota, Junko; Umehara, Kensuke; Ishimaru, Naoki; Ohno, Shunsuke; Okamoto, Kentaro; Suzuki, Takanori; Shirai, Naoki; Ishida, Takayuki

2017-02-01

As the capability of high-resolution displays grows, high-resolution images are often required in Computed Tomography (CT). However, acquiring high-resolution images takes a higher radiation dose and a longer scanning time. In this study, we applied the Sparse-coding-based Super-Resolution (ScSR) method to generate high-resolution images without increasing the radiation dose. We prepared the over-complete dictionary learned the mapping between low- and highresolution patches and seek a sparse representation of each patch of the low-resolution input. These coefficients were used to generate the high-resolution output. For evaluation, 44 CT cases were used as the test dataset. We up-sampled images up to 2 or 4 times and compared the image quality of the ScSR scheme and bilinear and bicubic interpolations, which are the traditional interpolation schemes. We also compared the image quality of three learning datasets. A total of 45 CT images, 91 non-medical images, and 93 chest radiographs were used for dictionary preparation respectively. The image quality was evaluated by measuring peak signal-to-noise ratio (PSNR) and structure similarity (SSIM). The differences of PSNRs and SSIMs between the ScSR method and interpolation methods were statistically significant. Visual assessment confirmed that the ScSR method generated a high-resolution image with sharpness, whereas conventional interpolation methods generated over-smoothed images. To compare three different training datasets, there were no significance between the CT, the CXR and non-medical datasets. These results suggest that the ScSR provides a robust approach for application of up-sampling CT images and yields substantial high image quality of extended images in CT.

WiseView: Visualizing motion and variability of faint WISE sources

NASA Astrophysics Data System (ADS)

Caselden, Dan; Westin, Paul, III; Meisner, Aaron; Kuchner, Marc; Colin, Guillaume

2018-06-01

WiseView renders image blinks of Wide-field Infrared Survey Explorer (WISE) coadds spanning a multi-year time baseline in a browser. The software allows for easy visual identification of motion and variability for sources far beyond the single-frame detection limit, a key threshold not surmounted by many studies. WiseView transparently gathers small image cutouts drawn from many terabytes of unWISE coadds, facilitating access to this large and unique dataset. Users need only input the coordinates of interest and can interactively tune parameters including the image stretch, colormap and blink rate. WiseView was developed in the context of the Backyard Worlds: Planet 9 citizen science project, and has enabled hundreds of brown dwarf candidate discoveries by citizen scientists and professional astronomers.

Topological anomaly detection performance with multispectral polarimetric imagery

NASA Astrophysics Data System (ADS)

Gartley, M. G.; Basener, W.,

2009-05-01

Polarimetric imaging has demonstrated utility for increasing contrast of manmade targets above natural background clutter. Manual detection of manmade targets in multispectral polarimetric imagery can be challenging and a subjective process for large datasets. Analyst exploitation may be improved utilizing conventional anomaly detection algorithms such as RX. In this paper we examine the performance of a relatively new approach to anomaly detection, which leverages topology theory, applied to spectral polarimetric imagery. Detection results for manmade targets embedded in a complex natural background will be presented for both the RX and Topological Anomaly Detection (TAD) approaches. We will also present detailed results examining detection sensitivities relative to: (1) the number of spectral bands, (2) utilization of Stoke's images versus intensity images, and (3) airborne versus spaceborne measurements.

A high-resolution 7-Tesla fMRI dataset from complex natural stimulation with an audio movie.

PubMed

Hanke, Michael; Baumgartner, Florian J; Ibe, Pierre; Kaule, Falko R; Pollmann, Stefan; Speck, Oliver; Zinke, Wolf; Stadler, Jörg

2014-01-01

Here we present a high-resolution functional magnetic resonance (fMRI) dataset - 20 participants recorded at high field strength (7 Tesla) during prolonged stimulation with an auditory feature film ("Forrest Gump"). In addition, a comprehensive set of auxiliary data (T1w, T2w, DTI, susceptibility-weighted image, angiography) as well as measurements to assess technical and physiological noise components have been acquired. An initial analysis confirms that these data can be used to study common and idiosyncratic brain response patterns to complex auditory stimulation. Among the potential uses of this dataset are the study of auditory attention and cognition, language and music perception, and social perception. The auxiliary measurements enable a large variety of additional analysis strategies that relate functional response patterns to structural properties of the brain. Alongside the acquired data, we provide source code and detailed information on all employed procedures - from stimulus creation to data analysis. In order to facilitate replicative and derived works, only free and open-source software was utilized.

Prostate cancer localization with multispectral MRI using cost-sensitive support vector machines and conditional random fields.

PubMed

Artan, Yusuf; Haider, Masoom A; Langer, Deanna L; van der Kwast, Theodorus H; Evans, Andrew J; Yang, Yongyi; Wernick, Miles N; Trachtenberg, John; Yetik, Imam Samil

2010-09-01

Prostate cancer is a leading cause of cancer death for men in the United States. Fortunately, the survival rate for early diagnosed patients is relatively high. Therefore, in vivo imaging plays an important role for the detection and treatment of the disease. Accurate prostate cancer localization with noninvasive imaging can be used to guide biopsy, radiotherapy, and surgery as well as to monitor disease progression. Magnetic resonance imaging (MRI) performed with an endorectal coil provides higher prostate cancer localization accuracy, when compared to transrectal ultrasound (TRUS). However, in general, a single type of MRI is not sufficient for reliable tumor localization. As an alternative, multispectral MRI, i.e., the use of multiple MRI-derived datasets, has emerged as a promising noninvasive imaging technique for the localization of prostate cancer; however almost all studies are with human readers. There is a significant inter and intraobserver variability for human readers, and it is substantially difficult for humans to analyze the large dataset of multispectral MRI. To solve these problems, this study presents an automated localization method using cost-sensitive support vector machines (SVMs) and shows that this method results in improved localization accuracy than classical SVM. Additionally, we develop a new segmentation method by combining conditional random fields (CRF) with a cost-sensitive framework and show that our method further improves cost-sensitive SVM results by incorporating spatial information. We test SVM, cost-sensitive SVM, and the proposed cost-sensitive CRF on multispectral MRI datasets acquired from 21 biopsy-confirmed cancer patients. Our results show that multispectral MRI helps to increase the accuracy of prostate cancer localization when compared to single MR images; and that using advanced methods such as cost-sensitive SVM as well as the proposed cost-sensitive CRF can boost the performance significantly when compared to SVM.

Iterative reconstruction of simulated low count data: a comparison of post-filtering versus regularised OSEM

NASA Astrophysics Data System (ADS)

Karaoglanis, K.; Efthimiou, N.; Tsoumpas, C.

2015-09-01

Low count PET data is a challenge for medical image reconstruction. The statistics of a dataset is a key factor of the quality of the reconstructed images. Reconstruction algorithms which would be able to compensate for low count datasets could provide the means to reduce the patient injected doses and/or reduce the scan times. It has been shown that the use of priors improve the image quality in low count conditions. In this study we compared regularised versus post-filtered OSEM for their performance on challenging simulated low count datasets. Initial visual comparison demonstrated that both algorithms improve the image quality, although the use of regularization does not introduce the undesired blurring as post-filtering.

Demons registration for in vivo and deformable laser scanning confocal endomicroscopy.

PubMed

Chiew, Wei-Ming; Lin, Feng; Seah, Hock Soon

2017-09-01

A critical effect found in noninvasive in vivo endomicroscopic imaging modalities is image distortions due to sporadic movement exhibited by living organisms. In three-dimensional confocal imaging, this effect results in a dataset that is tilted across deeper slices. Apart from that, the sequential flow of the imaging-processing pipeline restricts real-time adjustments due to the unavailability of information obtainable only from subsequent stages. To solve these problems, we propose an approach to render Demons-registered datasets as they are being captured, focusing on the coupling between registration and visualization. To improve the acquisition process, we also propose a real-time visual analytics tool, which complements the imaging pipeline and the Demons registration pipeline with useful visual indicators to provide real-time feedback for immediate adjustments. We highlight the problem of deformation within the visualization pipeline for object-ordered and image-ordered rendering. Visualizations of critical information including registration forces and partial renderings of the captured data are also presented in the analytics system. We demonstrate the advantages of the algorithmic design through experimental results with both synthetically deformed datasets and actual in vivo, time-lapse tissue datasets expressing natural deformations. Remarkably, this algorithm design is for embedded implementation in intelligent biomedical imaging instrumentation with customizable circuitry. (2017) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE).

Demons registration for in vivo and deformable laser scanning confocal endomicroscopy

NASA Astrophysics Data System (ADS)

Chiew, Wei Ming; Lin, Feng; Seah, Hock Soon

2017-09-01

A critical effect found in noninvasive in vivo endomicroscopic imaging modalities is image distortions due to sporadic movement exhibited by living organisms. In three-dimensional confocal imaging, this effect results in a dataset that is tilted across deeper slices. Apart from that, the sequential flow of the imaging-processing pipeline restricts real-time adjustments due to the unavailability of information obtainable only from subsequent stages. To solve these problems, we propose an approach to render Demons-registered datasets as they are being captured, focusing on the coupling between registration and visualization. To improve the acquisition process, we also propose a real-time visual analytics tool, which complements the imaging pipeline and the Demons registration pipeline with useful visual indicators to provide real-time feedback for immediate adjustments. We highlight the problem of deformation within the visualization pipeline for object-ordered and image-ordered rendering. Visualizations of critical information including registration forces and partial renderings of the captured data are also presented in the analytics system. We demonstrate the advantages of the algorithmic design through experimental results with both synthetically deformed datasets and actual in vivo, time-lapse tissue datasets expressing natural deformations. Remarkably, this algorithm design is for embedded implementation in intelligent biomedical imaging instrumentation with customizable circuitry.

Myocardial imaging using ultrahigh-resolution spectral domain optical coherence tomography

PubMed Central

Yao, Xinwen; Gan, Yu; Marboe, Charles C.; Hendon, Christine P.

2016-01-01

Abstract. We present an ultrahigh-resolution spectral domain optical coherence tomography (OCT) system in 800 nm with a low-noise supercontinuum source (SC) optimized for myocardial imaging. The system was demonstrated to have an axial resolution of 2.72  μm with a large imaging depth of 1.78 mm and a 6-dB falloff range of 0.89 mm. The lateral resolution (5.52  μm) was compromised to enhance the image penetration required for myocardial imaging. The noise of the SC source was analyzed extensively and an imaging protocol was proposed for SC-based OCT imaging with appreciable contrast. Three-dimensional datasets were acquired ex vivo on the endocardium side of tissue specimens from different chambers of fresh human and swine hearts. With the increased resolution and contrast, features such as elastic fibers, Purkinje fibers, and collagen fiber bundles were observed. The correlation between the structural information revealed in the OCT images and tissue pathology was discussed as well. PMID:27001162

Myocardial imaging using ultrahigh-resolution spectral domain optical coherence tomography

NASA Astrophysics Data System (ADS)

Yao, Xinwen; Gan, Yu; Marboe, Charles C.; Hendon, Christine P.

2016-06-01

We present an ultrahigh-resolution spectral domain optical coherence tomography (OCT) system in 800 nm with a low-noise supercontinuum source (SC) optimized for myocardial imaging. The system was demonstrated to have an axial resolution of 2.72 μm with a large imaging depth of 1.78 mm and a 6-dB falloff range of 0.89 mm. The lateral resolution (5.52 μm) was compromised to enhance the image penetration required for myocardial imaging. The noise of the SC source was analyzed extensively and an imaging protocol was proposed for SC-based OCT imaging with appreciable contrast. Three-dimensional datasets were acquired ex vivo on the endocardium side of tissue specimens from different chambers of fresh human and swine hearts. With the increased resolution and contrast, features such as elastic fibers, Purkinje fibers, and collagen fiber bundles were observed. The correlation between the structural information revealed in the OCT images and tissue pathology was discussed as well.

Mutual-information-based registration for ultrasound and CT datasets

NASA Astrophysics Data System (ADS)

Firle, Evelyn A.; Wesarg, Stefan; Dold, Christian

2004-05-01

In many applications for minimal invasive surgery the acquisition of intra-operative medical images is helpful if not absolutely necessary. Especially for Brachytherapy imaging is critically important to the safe delivery of the therapy. Modern computed tomography (CT) and magnetic resonance (MR) scanners allow minimal invasive procedures to be performed under direct imaging guidance. However, conventional scanners do not have real-time imaging capability and are expensive technologies requiring a special facility. Ultrasound (U/S) is a much cheaper and one of the most flexible imaging modalities. It can be moved to the application room as required and the physician sees what is happening as it occurs. Nevertheless it may be easier to interpret these 3D intra-operative U/S images if they are used in combination with less noisier preoperative data such as CT. The purpose of our current investigation is to develop a registration tool for automatically combining pre-operative CT volumes with intra-operatively acquired 3D U/S datasets. The applied alignment procedure is based on the information theoretic approach of maximizing the mutual information of two arbitrary datasets from different modalities. Since the CT datasets include a much bigger field of view we introduced a bounding box to narrow down the region of interest within the CT dataset. We conducted a phantom experiment using a CIRS Model 53 U/S Prostate Training Phantom to evaluate the feasibility and accuracy of the proposed method.

Multilevel principal component analysis (mPCA) in shape analysis: A feasibility study in medical and dental imaging.

PubMed

Farnell, D J J; Popat, H; Richmond, S

2016-06-01

Methods used in image processing should reflect any multilevel structures inherent in the image dataset or they run the risk of functioning inadequately. We wish to test the feasibility of multilevel principal components analysis (PCA) to build active shape models (ASMs) for cases relevant to medical and dental imaging. Multilevel PCA was used to carry out model fitting to sets of landmark points and it was compared to the results of "standard" (single-level) PCA. Proof of principle was tested by applying mPCA to model basic peri-oral expressions (happy, neutral, sad) approximated to the junction between the mouth/lips. Monte Carlo simulations were used to create this data which allowed exploration of practical implementation issues such as the number of landmark points, number of images, and number of groups (i.e., "expressions" for this example). To further test the robustness of the method, mPCA was subsequently applied to a dental imaging dataset utilising landmark points (placed by different clinicians) along the boundary of mandibular cortical bone in panoramic radiographs of the face. Changes of expression that varied between groups were modelled correctly at one level of the model and changes in lip width that varied within groups at another for the Monte Carlo dataset. Extreme cases in the test dataset were modelled adequately by mPCA but not by standard PCA. Similarly, variations in the shape of the cortical bone were modelled by one level of mPCA and variations between the experts at another for the panoramic radiographs dataset. Results for mPCA were found to be comparable to those of standard PCA for point-to-point errors via miss-one-out testing for this dataset. These errors reduce with increasing number of eigenvectors/values retained, as expected. We have shown that mPCA can be used in shape models for dental and medical image processing. mPCA was found to provide more control and flexibility when compared to standard "single-level" PCA. Specifically, mPCA is preferable to "standard" PCA when multiple levels occur naturally in the dataset. Copyright © 2016 Elsevier Ireland Ltd. All rights reserved.

Genetically targeted 3D visualisation of Drosophila neurons under Electron Microscopy and X-Ray Microscopy using miniSOG

PubMed Central

Ng, Julian; Browning, Alyssa; Lechner, Lorenz; Terada, Masako; Howard, Gillian; Jefferis, Gregory S. X. E.

2016-01-01

Large dimension, high-resolution imaging is important for neural circuit visualisation as neurons have both long- and short-range patterns: from axons and dendrites to the numerous synapses at terminal endings. Electron Microscopy (EM) is the favoured approach for synaptic resolution imaging but how such structures can be segmented from high-density images within large volume datasets remains challenging. Fluorescent probes are widely used to localise synapses, identify cell-types and in tracing studies. The equivalent EM approach would benefit visualising such labelled structures from within sub-cellular, cellular, tissue and neuroanatomical contexts. Here we developed genetically-encoded, electron-dense markers using miniSOG. We demonstrate their ability in 1) labelling cellular sub-compartments of genetically-targeted neurons, 2) generating contrast under different EM modalities, and 3) segmenting labelled structures from EM volumes using computer-assisted strategies. We also tested non-destructive X-ray imaging on whole Drosophila brains to evaluate contrast staining. This enabled us to target specific regions for EM volume acquisition. PMID:27958322

Semi-automated Neuron Boundary Detection and Nonbranching Process Segmentation in Electron Microscopy Images

PubMed Central

Jurrus, Elizabeth; Watanabe, Shigeki; Giuly, Richard J.; Paiva, Antonio R. C.; Ellisman, Mark H.; Jorgensen, Erik M.; Tasdizen, Tolga

2013-01-01

Neuroscientists are developing new imaging techniques and generating large volumes of data in an effort to understand the complex structure of the nervous system. The complexity and size of this data makes human interpretation a labor-intensive task. To aid in the analysis, new segmentation techniques for identifying neurons in these feature rich datasets are required. This paper presents a method for neuron boundary detection and nonbranching process segmentation in electron microscopy images and visualizing them in three dimensions. It combines both automated segmentation techniques with a graphical user interface for correction of mistakes in the automated process. The automated process first uses machine learning and image processing techniques to identify neuron membranes that deliniate the cells in each two-dimensional section. To segment nonbranching processes, the cell regions in each two-dimensional section are connected in 3D using correlation of regions between sections. The combination of this method with a graphical user interface specially designed for this purpose, enables users to quickly segment cellular processes in large volumes. PMID:22644867

Semi-Automated Neuron Boundary Detection and Nonbranching Process Segmentation in Electron Microscopy Images

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

Jurrus, Elizabeth R.; Watanabe, Shigeki; Giuly, Richard J.

2013-01-01

Neuroscientists are developing new imaging techniques and generating large volumes of data in an effort to understand the complex structure of the nervous system. The complexity and size of this data makes human interpretation a labor-intensive task. To aid in the analysis, new segmentation techniques for identifying neurons in these feature rich datasets are required. This paper presents a method for neuron boundary detection and nonbranching process segmentation in electron microscopy images and visualizing them in three dimensions. It combines both automated segmentation techniques with a graphical user interface for correction of mistakes in the automated process. The automated processmore » first uses machine learning and image processing techniques to identify neuron membranes that deliniate the cells in each two-dimensional section. To segment nonbranching processes, the cell regions in each two-dimensional section are connected in 3D using correlation of regions between sections. The combination of this method with a graphical user interface specially designed for this purpose, enables users to quickly segment cellular processes in large volumes.« less

Hunting for Active Galactic Nuclei in JWST/MIRI Imaging

NASA Astrophysics Data System (ADS)

Lin, Kenneth W.; Pope, Alexandra; Kirkpatrick, Allison

2018-01-01

The mid-infrared is uniquely sensitive to both star formation and active galactic nuclei (AGN) activity in galaxies. While spectra in this range can unambiguously identify these two processes, imaging data from the Spitzer Space Telescope found that the mid-infrared colors are also able to separate AGN from star forming galaxies. With the launch of the James Webb Space Telescope, our access to mid-infrared will be renewed; specifically, MIRI will provide imaging in 9 bands from 5.6-25.5 microns. While predictions show that color diagnostics will be useful with JWST/MIRI, this does not exploit the full dataset of MIRI imaging. In this poster, we discuss a Principal Component Analysis to identify the JWST filters that are most sensitive to the AGN contribution and demonstrate how to use it to identify large samples of AGN from planned MIRI imaging surveys.

On-line 3D motion estimation using low resolution MRI

NASA Astrophysics Data System (ADS)

Glitzner, M.; de Senneville, B. Denis; Lagendijk, J. J. W.; Raaymakers, B. W.; Crijns, S. P. M.

2015-08-01

Image processing such as deformable image registration finds its way into radiotherapy as a means to track non-rigid anatomy. With the advent of magnetic resonance imaging (MRI) guided radiotherapy, intrafraction anatomy snapshots become technically feasible. MRI provides the needed tissue signal for high-fidelity image registration. However, acquisitions, especially in 3D, take a considerable amount of time. Pushing towards real-time adaptive radiotherapy, MRI needs to be accelerated without degrading the quality of information. In this paper, we investigate the impact of image resolution on the quality of motion estimations. Potentially, spatially undersampled images yield comparable motion estimations. At the same time, their acquisition times would reduce greatly due to the sparser sampling. In order to substantiate this hypothesis, exemplary 4D datasets of the abdomen were downsampled gradually. Subsequently, spatiotemporal deformations are extracted consistently using the same motion estimation for each downsampled dataset. Errors between the original and the respectively downsampled version of the dataset are then evaluated. Compared to ground-truth, results show high similarity of deformations estimated from downsampled image data. Using a dataset with {{≤ft(2.5 \\text{mm}\\right)}3} voxel size, deformation fields could be recovered well up to a downsampling factor of 2, i.e. {{≤ft(5 \\text{mm}\\right)}3} . In a therapy guidance scenario MRI, imaging speed could accordingly increase approximately fourfold, with acceptable loss of estimated motion quality.

3-D interactive visualisation tools for Hi spectral line imaging

NASA Astrophysics Data System (ADS)

van der Hulst, J. M.; Punzo, D.; Roerdink, J. B. T. M.

2017-06-01

Upcoming HI surveys will deliver such large datasets that automated processing using the full 3-D information to find and characterize HI objects is unavoidable. Full 3-D visualization is an essential tool for enabling qualitative and quantitative inspection and analysis of the 3-D data, which is often complex in nature. Here we present SlicerAstro, an open-source extension of 3DSlicer, a multi-platform open source software package for visualization and medical image processing, which we developed for the inspection and analysis of HI spectral line data. We describe its initial capabilities, including 3-D filtering, 3-D selection and comparative modelling.

Dependence of subject-specific parameters for a fast helical CT respiratory motion model on breathing rate: an animal study

NASA Astrophysics Data System (ADS)

O'Connell, Dylan; Thomas, David H.; Lamb, James M.; Lewis, John H.; Dou, Tai; Sieren, Jered P.; Saylor, Melissa; Hofmann, Christian; Hoffman, Eric A.; Lee, Percy P.; Low, Daniel A.

2018-02-01

To determine if the parameters relating lung tissue displacement to a breathing surrogate signal in a previously published respiratory motion model vary with the rate of breathing during image acquisition. An anesthetized pig was imaged using multiple fast helical scans to sample the breathing cycle with simultaneous surrogate monitoring. Three datasets were collected while the animal was mechanically ventilated with different respiratory rates: 12 bpm (breaths per minute), 17 bpm, and 24 bpm. Three sets of motion model parameters describing the correspondences between surrogate signals and tissue displacements were determined. The model error was calculated individually for each dataset, as well asfor pairs of parameters and surrogate signals from different experiments. The values of one model parameter, a vector field denoted α which related tissue displacement to surrogate amplitude, determined for each experiment were compared. The mean model error of the three datasets was 1.00  ±  0.36 mm with a 95th percentile value of 1.69 mm. The mean error computed from all combinations of parameters and surrogate signals from different datasets was 1.14  ±  0.42 mm with a 95th percentile of 1.95 mm. The mean difference in α over all pairs of experiments was 4.7%  ±  5.4%, and the 95th percentile was 16.8%. The mean angle between pairs of α was 5.0  ±  4.0 degrees, with a 95th percentile of 13.2 mm. The motion model parameters were largely unaffected by changes in the breathing rate during image acquisition. The mean error associated with mismatched sets of parameters and surrogate signals was 0.14 mm greater than the error achieved when using parameters and surrogate signals acquired with the same breathing rate, while maximum respiratory motion was 23.23 mm on average.

Large-scale time-lapse microscopy of Oct4 expression in human embryonic stem cell colonies.

PubMed

Bhadriraju, Kiran; Halter, Michael; Amelot, Julien; Bajcsy, Peter; Chalfoun, Joe; Vandecreme, Antoine; Mallon, Barbara S; Park, Kye-Yoon; Sista, Subhash; Elliott, John T; Plant, Anne L

2016-07-01

Identification and quantification of the characteristics of stem cell preparations is critical for understanding stem cell biology and for the development and manufacturing of stem cell based therapies. We have developed image analysis and visualization software that allows effective use of time-lapse microscopy to provide spatial and dynamic information from large numbers of human embryonic stem cell colonies. To achieve statistically relevant sampling, we examined >680 colonies from 3 different preparations of cells over 5days each, generating a total experimental dataset of 0.9 terabyte (TB). The 0.5 Giga-pixel images at each time point were represented by multi-resolution pyramids and visualized using the Deep Zoom Javascript library extended to support viewing Giga-pixel images over time and extracting data on individual colonies. We present a methodology that enables quantification of variations in nominally-identical preparations and between colonies, correlation of colony characteristics with Oct4 expression, and identification of rare events. Copyright © 2016. Published by Elsevier B.V.

A hybrid approach for fusing 4D-MRI temporal information with 3D-CT for the study of lung and lung tumor motion

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

Yang, Y. X.; Van Reeth, E.; Poh, C. L., E-mail: clpoh@ntu.edu.sg

2015-08-15

Purpose: Accurate visualization of lung motion is important in many clinical applications, such as radiotherapy of lung cancer. Advancement in imaging modalities [e.g., computed tomography (CT) and MRI] has allowed dynamic imaging of lung and lung tumor motion. However, each imaging modality has its advantages and disadvantages. The study presented in this paper aims at generating synthetic 4D-CT dataset for lung cancer patients by combining both continuous three-dimensional (3D) motion captured by 4D-MRI and the high spatial resolution captured by CT using the authors’ proposed approach. Methods: A novel hybrid approach based on deformable image registration (DIR) and finite elementmore » method simulation was developed to fuse a static 3D-CT volume (acquired under breath-hold) and the 3D motion information extracted from 4D-MRI dataset, creating a synthetic 4D-CT dataset. Results: The study focuses on imaging of lung and lung tumor. Comparing the synthetic 4D-CT dataset with the acquired 4D-CT dataset of six lung cancer patients based on 420 landmarks, accurate results (average error <2 mm) were achieved using the authors’ proposed approach. Their hybrid approach achieved a 40% error reduction (based on landmarks assessment) over using only DIR techniques. Conclusions: The synthetic 4D-CT dataset generated has high spatial resolution, has excellent lung details, and is able to show movement of lung and lung tumor over multiple breathing cycles.« less

A sparsity-based iterative algorithm for reconstruction of micro-CT images from highly undersampled projection datasets obtained with a synchrotron X-ray source

NASA Astrophysics Data System (ADS)

Melli, S. Ali; Wahid, Khan A.; Babyn, Paul; Cooper, David M. L.; Gopi, Varun P.

2016-12-01

Synchrotron X-ray Micro Computed Tomography (Micro-CT) is an imaging technique which is increasingly used for non-invasive in vivo preclinical imaging. However, it often requires a large number of projections from many different angles to reconstruct high-quality images leading to significantly high radiation doses and long scan times. To utilize this imaging technique further for in vivo imaging, we need to design reconstruction algorithms that reduce the radiation dose and scan time without reduction of reconstructed image quality. This research is focused on using a combination of gradient-based Douglas-Rachford splitting and discrete wavelet packet shrinkage image denoising methods to design an algorithm for reconstruction of large-scale reduced-view synchrotron Micro-CT images with acceptable quality metrics. These quality metrics are computed by comparing the reconstructed images with a high-dose reference image reconstructed from 1800 equally spaced projections spanning 180°. Visual and quantitative-based performance assessment of a synthetic head phantom and a femoral cortical bone sample imaged in the biomedical imaging and therapy bending magnet beamline at the Canadian Light Source demonstrates that the proposed algorithm is superior to the existing reconstruction algorithms. Using the proposed reconstruction algorithm to reduce the number of projections in synchrotron Micro-CT is an effective way to reduce the overall radiation dose and scan time which improves in vivo imaging protocols.

Video-based face recognition via convolutional neural networks

NASA Astrophysics Data System (ADS)

Bao, Tianlong; Ding, Chunhui; Karmoshi, Saleem; Zhu, Ming

2017-06-01

Face recognition has been widely studied recently while video-based face recognition still remains a challenging task because of the low quality and large intra-class variation of video captured face images. In this paper, we focus on two scenarios of video-based face recognition: 1)Still-to-Video(S2V) face recognition, i.e., querying a still face image against a gallery of video sequences; 2)Video-to-Still(V2S) face recognition, in contrast to S2V scenario. A novel method was proposed in this paper to transfer still and video face images to an Euclidean space by a carefully designed convolutional neural network, then Euclidean metrics are used to measure the distance between still and video images. Identities of still and video images that group as pairs are used as supervision. In the training stage, a joint loss function that measures the Euclidean distance between the predicted features of training pairs and expanding vectors of still images is optimized to minimize the intra-class variation while the inter-class variation is guaranteed due to the large margin of still images. Transferred features are finally learned via the designed convolutional neural network. Experiments are performed on COX face dataset. Experimental results show that our method achieves reliable performance compared with other state-of-the-art methods.

Sub-word image clustering in Farsi printed books

NASA Astrophysics Data System (ADS)

Soheili, Mohammad Reza; Kabir, Ehsanollah; Stricker, Didier

2015-02-01

Most OCR systems are designed for the recognition of a single page. In case of unfamiliar font faces, low quality papers and degraded prints, the performance of these products drops sharply. However, an OCR system can use redundancy of word occurrences in large documents to improve recognition results. In this paper, we propose a sub-word image clustering method for the applications dealing with large printed documents. We assume that the whole document is printed by a unique unknown font with low quality print. Our proposed method finds clusters of equivalent sub-word images with an incremental algorithm. Due to the low print quality, we propose an image matching algorithm for measuring the distance between two sub-word images, based on Hamming distance and the ratio of the area to the perimeter of the connected components. We built a ground-truth dataset of more than 111000 sub-word images to evaluate our method. All of these images were extracted from an old Farsi book. We cluster all of these sub-words, including isolated letters and even punctuation marks. Then all centers of created clusters are labeled manually. We show that all sub-words of the book can be recognized with more than 99.7% accuracy by assigning the label of each cluster center to all of its members.

MilxXplore: a web-based system to explore large imaging datasets

PubMed Central

Bourgeat, P; Dore, V; Villemagne, V L; Rowe, C C; Salvado, O; Fripp, J

2013-01-01

Objective As large-scale medical imaging studies are becoming more common, there is an increasing reliance on automated software to extract quantitative information from these images. As the size of the cohorts keeps increasing with large studies, there is a also a need for tools that allow results from automated image processing and analysis to be presented in a way that enables fast and efficient quality checking, tagging and reporting on cases in which automatic processing failed or was problematic. Materials and methods MilxXplore is an open source visualization platform, which provides an interface to navigate and explore imaging data in a web browser, giving the end user the opportunity to perform quality control and reporting in a user friendly, collaborative and efficient way. Discussion Compared to existing software solutions that often provide an overview of the results at the subject's level, MilxXplore pools the results of individual subjects and time points together, allowing easy and efficient navigation and browsing through the different acquisitions of a subject over time, and comparing the results against the rest of the population. Conclusions MilxXplore is fast, flexible and allows remote quality checks of processed imaging data, facilitating data sharing and collaboration across multiple locations, and can be easily integrated into a cloud computing pipeline. With the growing trend of open data and open science, such a tool will become increasingly important to share and publish results of imaging analysis. PMID:23775173

Large datasets, logistics, sharing and workflow in screening.

PubMed

Cook, Tessa S

2018-03-29

Cancer screening initiatives exist around the world for different malignancies, most frequently breast, colorectal, and cervical cancer. A number of cancer registries exist to collect relevant data, but while these data may include imaging findings, they rarely, if ever, include actual images. Additionally, the data submitted to the registry are usually correlated with eventual cancer diagnoses and patient outcomes, rather than used with the individual's future screenings. Developing screening programs that allow for images to be submitted to a central location in addition to patient meta data and used for comparison to future screening exams would be very valuable in increasing access to care and ensuring that individuals are effectively screened at appropriate intervals. It would also change the way imaging results and additional patient data are correlated to eventual outcomes. However, it introduces logistical challenges surrounding secure storage and transmission of data to subsequent screening sites. In addition, in the absence of standardized protocols for screening, comparing current and prior imaging, especially from different equipment, can be challenging. Implementing a large-scale screening program with an image-enriched screening registry-effectively, an image-enriched electronic screening record-also requires that incentives exist for screening sites, physicians, and patients to participate; to maximize coverage, participation may have to be supported by government agencies. Workflows will also have to be adjusted to support registry participation for all screening patients in an effort to create a large, robust data set that can be used for future screening efforts as well as research initiatives.center.

SamuROI, a Python-Based Software Tool for Visualization and Analysis of Dynamic Time Series Imaging at Multiple Spatial Scales.

PubMed

Rueckl, Martin; Lenzi, Stephen C; Moreno-Velasquez, Laura; Parthier, Daniel; Schmitz, Dietmar; Ruediger, Sten; Johenning, Friedrich W

2017-01-01

The measurement of activity in vivo and in vitro has shifted from electrical to optical methods. While the indicators for imaging activity have improved significantly over the last decade, tools for analysing optical data have not kept pace. Most available analysis tools are limited in their flexibility and applicability to datasets obtained at different spatial scales. Here, we present SamuROI (Structured analysis of multiple user-defined ROIs), an open source Python-based analysis environment for imaging data. SamuROI simplifies exploratory analysis and visualization of image series of fluorescence changes in complex structures over time and is readily applicable at different spatial scales. In this paper, we show the utility of SamuROI in Ca 2+ -imaging based applications at three spatial scales: the micro-scale (i.e., sub-cellular compartments including cell bodies, dendrites and spines); the meso-scale, (i.e., whole cell and population imaging with single-cell resolution); and the macro-scale (i.e., imaging of changes in bulk fluorescence in large brain areas, without cellular resolution). The software described here provides a graphical user interface for intuitive data exploration and region of interest (ROI) management that can be used interactively within Jupyter Notebook: a publicly available interactive Python platform that allows simple integration of our software with existing tools for automated ROI generation and post-processing, as well as custom analysis pipelines. SamuROI software, source code and installation instructions are publicly available on GitHub and documentation is available online. SamuROI reduces the energy barrier for manual exploration and semi-automated analysis of spatially complex Ca 2+ imaging datasets, particularly when these have been acquired at different spatial scales.

SamuROI, a Python-Based Software Tool for Visualization and Analysis of Dynamic Time Series Imaging at Multiple Spatial Scales

PubMed Central

Rueckl, Martin; Lenzi, Stephen C.; Moreno-Velasquez, Laura; Parthier, Daniel; Schmitz, Dietmar; Ruediger, Sten; Johenning, Friedrich W.

2017-01-01

The measurement of activity in vivo and in vitro has shifted from electrical to optical methods. While the indicators for imaging activity have improved significantly over the last decade, tools for analysing optical data have not kept pace. Most available analysis tools are limited in their flexibility and applicability to datasets obtained at different spatial scales. Here, we present SamuROI (Structured analysis of multiple user-defined ROIs), an open source Python-based analysis environment for imaging data. SamuROI simplifies exploratory analysis and visualization of image series of fluorescence changes in complex structures over time and is readily applicable at different spatial scales. In this paper, we show the utility of SamuROI in Ca2+-imaging based applications at three spatial scales: the micro-scale (i.e., sub-cellular compartments including cell bodies, dendrites and spines); the meso-scale, (i.e., whole cell and population imaging with single-cell resolution); and the macro-scale (i.e., imaging of changes in bulk fluorescence in large brain areas, without cellular resolution). The software described here provides a graphical user interface for intuitive data exploration and region of interest (ROI) management that can be used interactively within Jupyter Notebook: a publicly available interactive Python platform that allows simple integration of our software with existing tools for automated ROI generation and post-processing, as well as custom analysis pipelines. SamuROI software, source code and installation instructions are publicly available on GitHub and documentation is available online. SamuROI reduces the energy barrier for manual exploration and semi-automated analysis of spatially complex Ca2+ imaging datasets, particularly when these have been acquired at different spatial scales. PMID:28706482

Zebra Crossing Spotter: Automatic Population of Spatial Databases for Increased Safety of Blind Travelers

PubMed Central

Ahmetovic, Dragan; Manduchi, Roberto; Coughlan, James M.; Mascetti, Sergio

2016-01-01

In this paper we propose a computer vision-based technique that mines existing spatial image databases for discovery of zebra crosswalks in urban settings. Knowing the location of crosswalks is critical for a blind person planning a trip that includes street crossing. By augmenting existing spatial databases (such as Google Maps or OpenStreetMap) with this information, a blind traveler may make more informed routing decisions, resulting in greater safety during independent travel. Our algorithm first searches for zebra crosswalks in satellite images; all candidates thus found are validated against spatially registered Google Street View images. This cascaded approach enables fast and reliable discovery and localization of zebra crosswalks in large image datasets. While fully automatic, our algorithm could also be complemented by a final crowdsourcing validation stage for increased accuracy. PMID:26824080

Deep 3D convolution neural network for CT brain hemorrhage classification

NASA Astrophysics Data System (ADS)

Jnawali, Kamal; Arbabshirani, Mohammad R.; Rao, Navalgund; Patel, Alpen A.

2018-02-01

Intracranial hemorrhage is a critical conditional with the high mortality rate that is typically diagnosed based on head computer tomography (CT) images. Deep learning algorithms, in particular, convolution neural networks (CNN), are becoming the methodology of choice in medical image analysis for a variety of applications such as computer-aided diagnosis, and segmentation. In this study, we propose a fully automated deep learning framework which learns to detect brain hemorrhage based on cross sectional CT images. The dataset for this work consists of 40,367 3D head CT studies (over 1.5 million 2D images) acquired retrospectively over a decade from multiple radiology facilities at Geisinger Health System. The proposed algorithm first extracts features using 3D CNN and then detects brain hemorrhage using the logistic function as the last layer of the network. Finally, we created an ensemble of three different 3D CNN architectures to improve the classification accuracy. The area under the curve (AUC) of the receiver operator characteristic (ROC) curve of the ensemble of three architectures was 0.87. Their results are very promising considering the fact that the head CT studies were not controlled for slice thickness, scanner type, study protocol or any other settings. Moreover, the proposed algorithm reliably detected various types of hemorrhage within the skull. This work is one of the first applications of 3D CNN trained on a large dataset of cross sectional medical images for detection of a critical radiological condition

Using iterative cluster merging with improved gap statistics to perform online phenotype discovery in the context of high-throughput RNAi screens

PubMed Central

Yin, Zheng; Zhou, Xiaobo; Bakal, Chris; Li, Fuhai; Sun, Youxian; Perrimon, Norbert; Wong, Stephen TC

2008-01-01

Background The recent emergence of high-throughput automated image acquisition technologies has forever changed how cell biologists collect and analyze data. Historically, the interpretation of cellular phenotypes in different experimental conditions has been dependent upon the expert opinions of well-trained biologists. Such qualitative analysis is particularly effective in detecting subtle, but important, deviations in phenotypes. However, while the rapid and continuing development of automated microscope-based technologies now facilitates the acquisition of trillions of cells in thousands of diverse experimental conditions, such as in the context of RNA interference (RNAi) or small-molecule screens, the massive size of these datasets precludes human analysis. Thus, the development of automated methods which aim to identify novel and biological relevant phenotypes online is one of the major challenges in high-throughput image-based screening. Ideally, phenotype discovery methods should be designed to utilize prior/existing information and tackle three challenging tasks, i.e. restoring pre-defined biological meaningful phenotypes, differentiating novel phenotypes from known ones and clarifying novel phenotypes from each other. Arbitrarily extracted information causes biased analysis, while combining the complete existing datasets with each new image is intractable in high-throughput screens. Results Here we present the design and implementation of a novel and robust online phenotype discovery method with broad applicability that can be used in diverse experimental contexts, especially high-throughput RNAi screens. This method features phenotype modelling and iterative cluster merging using improved gap statistics. A Gaussian Mixture Model (GMM) is employed to estimate the distribution of each existing phenotype, and then used as reference distribution in gap statistics. This method is broadly applicable to a number of different types of image-based datasets derived from a wide spectrum of experimental conditions and is suitable to adaptively process new images which are continuously added to existing datasets. Validations were carried out on different dataset, including published RNAi screening using Drosophila embryos [Additional files 1, 2], dataset for cell cycle phase identification using HeLa cells [Additional files 1, 3, 4] and synthetic dataset using polygons, our methods tackled three aforementioned tasks effectively with an accuracy range of 85%–90%. When our method is implemented in the context of a Drosophila genome-scale RNAi image-based screening of cultured cells aimed to identifying the contribution of individual genes towards the regulation of cell-shape, it efficiently discovers meaningful new phenotypes and provides novel biological insight. We also propose a two-step procedure to modify the novelty detection method based on one-class SVM, so that it can be used to online phenotype discovery. In different conditions, we compared the SVM based method with our method using various datasets and our methods consistently outperformed SVM based method in at least two of three tasks by 2% to 5%. These results demonstrate that our methods can be used to better identify novel phenotypes in image-based datasets from a wide range of conditions and organisms. Conclusion We demonstrate that our method can detect various novel phenotypes effectively in complex datasets. Experiment results also validate that our method performs consistently under different order of image input, variation of starting conditions including the number and composition of existing phenotypes, and dataset from different screens. In our findings, the proposed method is suitable for online phenotype discovery in diverse high-throughput image-based genetic and chemical screens. PMID:18534020

A model of traffic signs recognition with convolutional neural network

NASA Astrophysics Data System (ADS)

Hu, Haihe; Li, Yujian; Zhang, Ting; Huo, Yi; Kuang, Wenqing

2016-10-01

In real traffic scenes, the quality of captured images are generally low due to some factors such as lighting conditions, and occlusion on. All of these factors are challengeable for automated recognition algorithms of traffic signs. Deep learning has provided a new way to solve this kind of problems recently. The deep network can automatically learn features from a large number of data samples and obtain an excellent recognition performance. We therefore approach this task of recognition of traffic signs as a general vision problem, with few assumptions related to road signs. We propose a model of Convolutional Neural Network (CNN) and apply the model to the task of traffic signs recognition. The proposed model adopts deep CNN as the supervised learning model, directly takes the collected traffic signs image as the input, alternates the convolutional layer and subsampling layer, and automatically extracts the features for the recognition of the traffic signs images. The proposed model includes an input layer, three convolutional layers, three subsampling layers, a fully-connected layer, and an output layer. To validate the proposed model, the experiments are implemented using the public dataset of China competition of fuzzy image processing. Experimental results show that the proposed model produces a recognition accuracy of 99.01 % on the training dataset, and yield a record of 92% on the preliminary contest within the fourth best.

Challenges in Extracting Information From Large Hydrogeophysical-monitoring Datasets

NASA Astrophysics Data System (ADS)

Day-Lewis, F. D.; Slater, L. D.; Johnson, T.

2012-12-01

Over the last decade, new automated geophysical data-acquisition systems have enabled collection of increasingly large and information-rich geophysical datasets. Concurrent advances in field instrumentation, web services, and high-performance computing have made real-time processing, inversion, and visualization of large three-dimensional tomographic datasets practical. Geophysical-monitoring datasets have provided high-resolution insights into diverse hydrologic processes including groundwater/surface-water exchange, infiltration, solute transport, and bioremediation. Despite the high information content of such datasets, extraction of quantitative or diagnostic hydrologic information is challenging. Visual inspection and interpretation for specific hydrologic processes is difficult for datasets that are large, complex, and (or) affected by forcings (e.g., seasonal variations) unrelated to the target hydrologic process. New strategies are needed to identify salient features in spatially distributed time-series data and to relate temporal changes in geophysical properties to hydrologic processes of interest while effectively filtering unrelated changes. Here, we review recent work using time-series and digital-signal-processing approaches in hydrogeophysics. Examples include applications of cross-correlation, spectral, and time-frequency (e.g., wavelet and Stockwell transforms) approaches to (1) identify salient features in large geophysical time series; (2) examine correlation or coherence between geophysical and hydrologic signals, even in the presence of non-stationarity; and (3) condense large datasets while preserving information of interest. Examples demonstrate analysis of large time-lapse electrical tomography and fiber-optic temperature datasets to extract information about groundwater/surface-water exchange and contaminant transport.

Robust and automated three-dimensional segmentation of densely packed cell nuclei in different biological specimens with Lines-of-Sight decomposition.

PubMed

Mathew, B; Schmitz, A; Muñoz-Descalzo, S; Ansari, N; Pampaloni, F; Stelzer, E H K; Fischer, S C

2015-06-08

Due to the large amount of data produced by advanced microscopy, automated image analysis is crucial in modern biology. Most applications require reliable cell nuclei segmentation. However, in many biological specimens cell nuclei are densely packed and appear to touch one another in the images. Therefore, a major difficulty of three-dimensional cell nuclei segmentation is the decomposition of cell nuclei that apparently touch each other. Current methods are highly adapted to a certain biological specimen or a specific microscope. They do not ensure similarly accurate segmentation performance, i.e. their robustness for different datasets is not guaranteed. Hence, these methods require elaborate adjustments to each dataset. We present an advanced three-dimensional cell nuclei segmentation algorithm that is accurate and robust. Our approach combines local adaptive pre-processing with decomposition based on Lines-of-Sight (LoS) to separate apparently touching cell nuclei into approximately convex parts. We demonstrate the superior performance of our algorithm using data from different specimens recorded with different microscopes. The three-dimensional images were recorded with confocal and light sheet-based fluorescence microscopes. The specimens are an early mouse embryo and two different cellular spheroids. We compared the segmentation accuracy of our algorithm with ground truth data for the test images and results from state-of-the-art methods. The analysis shows that our method is accurate throughout all test datasets (mean F-measure: 91%) whereas the other methods each failed for at least one dataset (F-measure≤69%). Furthermore, nuclei volume measurements are improved for LoS decomposition. The state-of-the-art methods required laborious adjustments of parameter values to achieve these results. Our LoS algorithm did not require parameter value adjustments. The accurate performance was achieved with one fixed set of parameter values. We developed a novel and fully automated three-dimensional cell nuclei segmentation method incorporating LoS decomposition. LoS are easily accessible features that ensure correct splitting of apparently touching cell nuclei independent of their shape, size or intensity. Our method showed superior performance compared to state-of-the-art methods, performing accurately for a variety of test images. Hence, our LoS approach can be readily applied to quantitative evaluation in drug testing, developmental and cell biology.

HYDRA: Revealing heterogeneity of imaging and genetic patterns through a multiple max-margin discriminative analysis framework.

PubMed

Varol, Erdem; Sotiras, Aristeidis; Davatzikos, Christos

2017-01-15

Multivariate pattern analysis techniques have been increasingly used over the past decade to derive highly sensitive and specific biomarkers of diseases on an individual basis. The driving assumption behind the vast majority of the existing methodologies is that a single imaging pattern can distinguish between healthy and diseased populations, or between two subgroups of patients (e.g., progressors vs. non-progressors). This assumption effectively ignores the ample evidence for the heterogeneous nature of brain diseases. Neurodegenerative, neuropsychiatric and neurodevelopmental disorders are largely characterized by high clinical heterogeneity, which likely stems in part from underlying neuroanatomical heterogeneity of various pathologies. Detecting and characterizing heterogeneity may deepen our understanding of disease mechanisms and lead to patient-specific treatments. However, few approaches tackle disease subtype discovery in a principled machine learning framework. To address this challenge, we present a novel non-linear learning algorithm for simultaneous binary classification and subtype identification, termed HYDRA (Heterogeneity through Discriminative Analysis). Neuroanatomical subtypes are effectively captured by multiple linear hyperplanes, which form a convex polytope that separates two groups (e.g., healthy controls from pathologic samples); each face of this polytope effectively defines a disease subtype. We validated HYDRA on simulated and clinical data. In the latter case, we applied the proposed method independently to the imaging and genetic datasets of the Alzheimer's Disease Neuroimaging Initiative (ADNI 1) study. The imaging dataset consisted of T1-weighted volumetric magnetic resonance images of 123 AD patients and 177 controls. The genetic dataset consisted of single nucleotide polymorphism information of 103 AD patients and 139 controls. We identified 3 reproducible subtypes of atrophy in AD relative to controls: (1) diffuse and extensive atrophy, (2) precuneus and extensive temporal lobe atrophy, as well some prefrontal atrophy, (3) atrophy pattern very much confined to the hippocampus and the medial temporal lobe. The genetics dataset yielded two subtypes of AD characterized mainly by the presence/absence of the apolipoprotein E (APOE) ε4 genotype, but also involving differential presence of risk alleles of CD2AP, SPON1 and LOC39095 SNPs that were associated with differences in the respective patterns of brain atrophy, especially in the precuneus. The results demonstrate the potential of the proposed approach to map disease heterogeneity in neuroimaging and genetic studies. Copyright © 2016 Elsevier Inc. All rights reserved.

NutriNet: A Deep Learning Food and Drink Image Recognition System for Dietary Assessment

PubMed Central

Koroušić Seljak, Barbara

2017-01-01

Automatic food image recognition systems are alleviating the process of food-intake estimation and dietary assessment. However, due to the nature of food images, their recognition is a particularly challenging task, which is why traditional approaches in the field have achieved a low classification accuracy. Deep neural networks have outperformed such solutions, and we present a novel approach to the problem of food and drink image detection and recognition that uses a newly-defined deep convolutional neural network architecture, called NutriNet. This architecture was tuned on a recognition dataset containing 225,953 512 × 512 pixel images of 520 different food and drink items from a broad spectrum of food groups, on which we achieved a classification accuracy of 86.72%, along with an accuracy of 94.47% on a detection dataset containing 130,517 images. We also performed a real-world test on a dataset of self-acquired images, combined with images from Parkinson’s disease patients, all taken using a smartphone camera, achieving a top-five accuracy of 55%, which is an encouraging result for real-world images. Additionally, we tested NutriNet on the University of Milano-Bicocca 2016 (UNIMIB2016) food image dataset, on which we improved upon the provided baseline recognition result. An online training component was implemented to continually fine-tune the food and drink recognition model on new images. The model is being used in practice as part of a mobile app for the dietary assessment of Parkinson’s disease patients. PMID:28653995

Normalization of cortical thickness measurements across different T1 magnetic resonance imaging protocols by novel W-Score standardization.

PubMed

Chung, Jinyong; Yoo, Kwangsun; Lee, Peter; Kim, Chan Mi; Roh, Jee Hoon; Park, Ji Eun; Kim, Sang Joon; Seo, Sang Won; Shin, Jeong-Hyeon; Seong, Joon-Kyung; Jeong, Yong

2017-10-01

The use of different 3D T1-weighted magnetic resonance (T1 MR) imaging protocols induces image incompatibility across multicenter studies, negating the many advantages of multicenter studies. A few methods have been developed to address this problem, but significant image incompatibility still remains. Thus, we developed a novel and convenient method to improve image compatibility. W-score standardization creates quality reference values by using a healthy group to obtain normalized disease values. We developed a protocol-specific w-score standardization to control the protocol effect, which is applied to each protocol separately. We used three data sets. In dataset 1, brain T1 MR images of normal controls (NC) and patients with Alzheimer's disease (AD) from two centers, acquired with different T1 MR protocols, were used (Protocol 1 and 2, n = 45/group). In dataset 2, data from six subjects, who underwent MRI with two different protocols (Protocol 1 and 2), were used with different repetition times, echo times, and slice thicknesses. In dataset 3, T1 MR images from a large number of healthy normal controls (Protocol 1: n = 148, Protocol 2: n = 343) were collected for w-score standardization. The protocol effect and disease effect on subjects' cortical thickness were analyzed before and after the application of protocol-specific w-score standardization. As expected, different protocols resulted in differing cortical thickness measurements in both NC and AD subjects. Different measurements were obtained for the same subject when imaged with different protocols. Multivariate pattern difference between measurements was observed between the protocols. Classification accuracy between two protocols was nearly 90%. After applying protocol-specific w-score standardization, the differences between the protocols substantially decreased. Most importantly, protocol-specific w-score standardization reduced both univariate and multivariate differences in the images while maintaining the AD disease effect. Compared to conventional regression methods, our method showed the best performance for in terms of controlling the protocol effect while preserving disease information. Protocol-specific w-score standardization effectively resolved the concerns of conventional regression methods. It showed the best performance for improving the compatibility of a T1 MR post-processed feature, cortical thickness. Copyright © 2017 Elsevier Inc. All rights reserved.

A Deep Learning Algorithm for Prediction of Age-Related Eye Disease Study Severity Scale for Age-Related Macular Degeneration from Color Fundus Photography.

PubMed

Grassmann, Felix; Mengelkamp, Judith; Brandl, Caroline; Harsch, Sebastian; Zimmermann, Martina E; Linkohr, Birgit; Peters, Annette; Heid, Iris M; Palm, Christoph; Weber, Bernhard H F

2018-04-10

Age-related macular degeneration (AMD) is a common threat to vision. While classification of disease stages is critical to understanding disease risk and progression, several systems based on color fundus photographs are known. Most of these require in-depth and time-consuming analysis of fundus images. Herein, we present an automated computer-based classification algorithm. Algorithm development for AMD classification based on a large collection of color fundus images. Validation is performed on a cross-sectional, population-based study. We included 120 656 manually graded color fundus images from 3654 Age-Related Eye Disease Study (AREDS) participants. AREDS participants were >55 years of age, and non-AMD sight-threatening diseases were excluded at recruitment. In addition, performance of our algorithm was evaluated in 5555 fundus images from the population-based Kooperative Gesundheitsforschung in der Region Augsburg (KORA; Cooperative Health Research in the Region of Augsburg) study. We defined 13 classes (9 AREDS steps, 3 late AMD stages, and 1 for ungradable images) and trained several convolution deep learning architectures. An ensemble of network architectures improved prediction accuracy. An independent dataset was used to evaluate the performance of our algorithm in a population-based study. κ Statistics and accuracy to evaluate the concordance between predicted and expert human grader classification. A network ensemble of 6 different neural net architectures predicted the 13 classes in the AREDS test set with a quadratic weighted κ of 92% (95% confidence interval, 89%-92%) and an overall accuracy of 63.3%. In the independent KORA dataset, images wrongly classified as AMD were mainly the result of a macular reflex observed in young individuals. By restricting the KORA analysis to individuals >55 years of age and prior exclusion of other retinopathies, the weighted and unweighted κ increased to 50% and 63%, respectively. Importantly, the algorithm detected 84.2% of all fundus images with definite signs of early or late AMD. Overall, 94.3% of healthy fundus images were classified correctly. Our deep learning algoritm revealed a weighted κ outperforming human graders in the AREDS study and is suitable to classify AMD fundus images in other datasets using individuals >55 years of age. Copyright © 2018 American Academy of Ophthalmology. Published by Elsevier Inc. All rights reserved.

Multi-MHz FDML OCT: snapshot retinal imaging at 6.7 million axial-scans per second

NASA Astrophysics Data System (ADS)

Klein, Thomas; Wieser, Wolfgang; André, Raphael; Pfeiffer, Tom; Eigenwillig, Christoph M.; Huber, Robert

2012-01-01

We demonstrate the acquisition of densely sampled wide-field 3D OCT datasets of the human retina in 0.3s. This performance is achieved with a multi-MHz Fourier domain mode-locked (FDML) laser source operating at 1050nm. A two-beam setup doubles the 3.35MHz laser sweep rate to 6.7MHz, which is 16x faster than results achieved with any non-FDML source used for retinal OCT. We discuss two main benefits of these high line rates: First, large datasets over an ultra-wide field of view can be acquired with a low probability of distortions. Second, even if eye movements occur, now the scan rate is high enough to directly correct even the fastest saccades without loss of information.

Online 3D Ear Recognition by Combining Global and Local Features.

PubMed

Liu, Yahui; Zhang, Bob; Lu, Guangming; Zhang, David

2016-01-01

The three-dimensional shape of the ear has been proven to be a stable candidate for biometric authentication because of its desirable properties such as universality, uniqueness, and permanence. In this paper, a special laser scanner designed for online three-dimensional ear acquisition was described. Based on the dataset collected by our scanner, two novel feature classes were defined from a three-dimensional ear image: the global feature class (empty centers and angles) and local feature class (points, lines, and areas). These features are extracted and combined in an optimal way for three-dimensional ear recognition. Using a large dataset consisting of 2,000 samples, the experimental results illustrate the effectiveness of fusing global and local features, obtaining an equal error rate of 2.2%.

Online 3D Ear Recognition by Combining Global and Local Features

PubMed Central

Liu, Yahui; Zhang, Bob; Lu, Guangming; Zhang, David

2016-01-01

The three-dimensional shape of the ear has been proven to be a stable candidate for biometric authentication because of its desirable properties such as universality, uniqueness, and permanence. In this paper, a special laser scanner designed for online three-dimensional ear acquisition was described. Based on the dataset collected by our scanner, two novel feature classes were defined from a three-dimensional ear image: the global feature class (empty centers and angles) and local feature class (points, lines, and areas). These features are extracted and combined in an optimal way for three-dimensional ear recognition. Using a large dataset consisting of 2,000 samples, the experimental results illustrate the effectiveness of fusing global and local features, obtaining an equal error rate of 2.2%. PMID:27935955

3D shape recovery from image focus using gray level co-occurrence matrix

NASA Astrophysics Data System (ADS)

Mahmood, Fahad; Munir, Umair; Mehmood, Fahad; Iqbal, Javaid

2018-04-01

Recovering a precise and accurate 3-D shape of the target object utilizing robust 3-D shape recovery algorithm is an ultimate objective of computer vision community. Focus measure algorithm plays an important role in this architecture which convert the color values of each pixel of the acquired 2-D image dataset into corresponding focus values. After convolving the focus measure filter with the input 2-D image dataset, a 3-D shape recovery approach is applied which will recover the depth map. In this document, we are concerned with proposing Gray Level Co-occurrence Matrix along with its statistical features for computing the focus information of the image dataset. The Gray Level Co-occurrence Matrix quantifies the texture present in the image using statistical features and then applies joint probability distributive function of the gray level pairs of the input image. Finally, we quantify the focus value of the input image using Gaussian Mixture Model. Due to its little computational complexity, sharp focus measure curve, robust to random noise sources and accuracy, it is considered as superior alternative to most of recently proposed 3-D shape recovery approaches. This algorithm is deeply investigated on real image sequences and synthetic image dataset. The efficiency of the proposed scheme is also compared with the state of art 3-D shape recovery approaches. Finally, by means of two global statistical measures, root mean square error and correlation, we claim that this approach -in spite of simplicity generates accurate results.

Comparison and evaluation of datasets for off-angle iris recognition

NASA Astrophysics Data System (ADS)

Kurtuncu, Osman M.; Cerme, Gamze N.; Karakaya, Mahmut

2016-05-01

In this paper, we investigated the publicly available iris recognition datasets and their data capture procedures in order to determine if they are suitable for the stand-off iris recognition research. Majority of the iris recognition datasets include only frontal iris images. Even if a few datasets include off-angle iris images, the frontal and off-angle iris images are not captured at the same time. The comparison of the frontal and off-angle iris images shows not only differences in the gaze angle but also change in pupil dilation and accommodation as well. In order to isolate the effect of the gaze angle from other challenging issues including dilation and accommodation, the frontal and off-angle iris images are supposed to be captured at the same time by using two different cameras. Therefore, we developed an iris image acquisition platform by using two cameras in this work where one camera captures frontal iris image and the other one captures iris images from off-angle. Based on the comparison of Hamming distance between frontal and off-angle iris images captured with the two-camera- setup and one-camera-setup, we observed that Hamming distance in two-camera-setup is less than one-camera-setup ranging from 0.05 to 0.001. These results show that in order to have accurate results in the off-angle iris recognition research, two-camera-setup is necessary in order to distinguish the challenging issues from each other.

Fast elastic registration of soft tissues under large deformations.

PubMed

Peterlík, Igor; Courtecuisse, Hadrien; Rohling, Robert; Abolmaesumi, Purang; Nguan, Christopher; Cotin, Stéphane; Salcudean, Septimiu

2018-04-01

A fast and accurate fusion of intra-operative images with a pre-operative data is a key component of computer-aided interventions which aim at improving the outcomes of the intervention while reducing the patient's discomfort. In this paper, we focus on the problematic of the intra-operative navigation during abdominal surgery, which requires an accurate registration of tissues undergoing large deformations. Such a scenario occurs in the case of partial hepatectomy: to facilitate the access to the pathology, e.g. a tumor located in the posterior part of the right lobe, the surgery is performed on a patient in lateral position. Due to the change in patient's position, the resection plan based on the pre-operative CT scan acquired in the supine position must be updated to account for the deformations. We suppose that an imaging modality, such as the cone-beam CT, provides the information about the intra-operative shape of an organ, however, due to the reduced radiation dose and contrast, the actual locations of the internal structures necessary to update the planning are not available. To this end, we propose a method allowing for fast registration of the pre-operative data represented by a detailed 3D model of the liver and its internal structure and the actual configuration given by the organ surface extracted from the intra-operative image. The algorithm behind the method combines the iterative closest point technique with a biomechanical model based on a co-rotational formulation of linear elasticity which accounts for large deformations of the tissue. The performance, robustness and accuracy of the method is quantitatively assessed on a control semi-synthetic dataset with known ground truth and a real dataset composed of nine pairs of abdominal CT scans acquired in supine and flank positions. It is shown that the proposed surface-matching method is capable of reducing the target registration error evaluated of the internal structures of the organ from more than 40 mm to less then 10 mm. Moreover, the control data is used to demonstrate the compatibility of the method with intra-operative clinical scenario, while the real datasets are utilized to study the impact of parametrization on the accuracy of the method. The method is also compared to a state-of-the art intensity-based registration technique in terms of accuracy and performance. Copyright © 2017 Elsevier B.V. All rights reserved.

A fully automatic end-to-end method for content-based image retrieval of CT scans with similar liver lesion annotations.

PubMed

Spanier, A B; Caplan, N; Sosna, J; Acar, B; Joskowicz, L

2018-01-01

The goal of medical content-based image retrieval (M-CBIR) is to assist radiologists in the decision-making process by retrieving medical cases similar to a given image. One of the key interests of radiologists is lesions and their annotations, since the patient treatment depends on the lesion diagnosis. Therefore, a key feature of M-CBIR systems is the retrieval of scans with the most similar lesion annotations. To be of value, M-CBIR systems should be fully automatic to handle large case databases. We present a fully automatic end-to-end method for the retrieval of CT scans with similar liver lesion annotations. The input is a database of abdominal CT scans labeled with liver lesions, a query CT scan, and optionally one radiologist-specified lesion annotation of interest. The output is an ordered list of the database CT scans with the most similar liver lesion annotations. The method starts by automatically segmenting the liver in the scan. It then extracts a histogram-based features vector from the segmented region, learns the features' relative importance, and ranks the database scans according to the relative importance measure. The main advantages of our method are that it fully automates the end-to-end querying process, that it uses simple and efficient techniques that are scalable to large datasets, and that it produces quality retrieval results using an unannotated CT scan. Our experimental results on 9 CT queries on a dataset of 41 volumetric CT scans from the 2014 Image CLEF Liver Annotation Task yield an average retrieval accuracy (Normalized Discounted Cumulative Gain index) of 0.77 and 0.84 without/with annotation, respectively. Fully automatic end-to-end retrieval of similar cases based on image information alone, rather that on disease diagnosis, may help radiologists to better diagnose liver lesions.

A survey of GPU-based medical image computing techniques

PubMed Central

Shi, Lin; Liu, Wen; Zhang, Heye; Xie, Yongming

2012-01-01

Medical imaging currently plays a crucial role throughout the entire clinical applications from medical scientific research to diagnostics and treatment planning. However, medical imaging procedures are often computationally demanding due to the large three-dimensional (3D) medical datasets to process in practical clinical applications. With the rapidly enhancing performances of graphics processors, improved programming support, and excellent price-to-performance ratio, the graphics processing unit (GPU) has emerged as a competitive parallel computing platform for computationally expensive and demanding tasks in a wide range of medical image applications. The major purpose of this survey is to provide a comprehensive reference source for the starters or researchers involved in GPU-based medical image processing. Within this survey, the continuous advancement of GPU computing is reviewed and the existing traditional applications in three areas of medical image processing, namely, segmentation, registration and visualization, are surveyed. The potential advantages and associated challenges of current GPU-based medical imaging are also discussed to inspire future applications in medicine. PMID:23256080

An integrative approach for analyzing hundreds of neurons in task performing mice using wide-field calcium imaging.

PubMed

Mohammed, Ali I; Gritton, Howard J; Tseng, Hua-an; Bucklin, Mark E; Yao, Zhaojie; Han, Xue

2016-02-08

Advances in neurotechnology have been integral to the investigation of neural circuit function in systems neuroscience. Recent improvements in high performance fluorescent sensors and scientific CMOS cameras enables optical imaging of neural networks at a much larger scale. While exciting technical advances demonstrate the potential of this technique, further improvement in data acquisition and analysis, especially those that allow effective processing of increasingly larger datasets, would greatly promote the application of optical imaging in systems neuroscience. Here we demonstrate the ability of wide-field imaging to capture the concurrent dynamic activity from hundreds to thousands of neurons over millimeters of brain tissue in behaving mice. This system allows the visualization of morphological details at a higher spatial resolution than has been previously achieved using similar functional imaging modalities. To analyze the expansive data sets, we developed software to facilitate rapid downstream data processing. Using this system, we show that a large fraction of anatomically distinct hippocampal neurons respond to discrete environmental stimuli associated with classical conditioning, and that the observed temporal dynamics of transient calcium signals are sufficient for exploring certain spatiotemporal features of large neural networks.

Learning to rank using user clicks and visual features for image retrieval.

PubMed

Yu, Jun; Tao, Dacheng; Wang, Meng; Rui, Yong

2015-04-01

The inconsistency between textual features and visual contents can cause poor image search results. To solve this problem, click features, which are more reliable than textual information in justifying the relevance between a query and clicked images, are adopted in image ranking model. However, the existing ranking model cannot integrate visual features, which are efficient in refining the click-based search results. In this paper, we propose a novel ranking model based on the learning to rank framework. Visual features and click features are simultaneously utilized to obtain the ranking model. Specifically, the proposed approach is based on large margin structured output learning and the visual consistency is integrated with the click features through a hypergraph regularizer term. In accordance with the fast alternating linearization method, we design a novel algorithm to optimize the objective function. This algorithm alternately minimizes two different approximations of the original objective function by keeping one function unchanged and linearizing the other. We conduct experiments on a large-scale dataset collected from the Microsoft Bing image search engine, and the results demonstrate that the proposed learning to rank models based on visual features and user clicks outperforms state-of-the-art algorithms.

Continental Scale Vegetation Structure Mapping Using Field Calibrated Landsat, ALOS Palsar And GLAS ICESat

NASA Astrophysics Data System (ADS)

Scarth, P.; Phinn, S. R.; Armston, J.; Lucas, R.

2015-12-01

Vertical plant profiles are important descriptors of canopy structure and are used to inform models of biomass, biodiversity and fire risk. In Australia, an approach has been developed to produce large area maps of vertical plant profiles by extrapolating waveform lidar estimates of vertical plant profiles from ICESat/GLAS using large area segmentation of ALOS PALSAR and Landsat satellite image products. The main assumption of this approach is that the vegetation height profiles are consistent across the segments defined from ALOS PALSAR and Landsat image products. More than 1500 field sites were used to develop an index of fractional cover using Landsat data. A time series of the green fraction was used to calculate the persistent green fraction continuously across the landscape. This was fused with ALOS PALSAR L-band Fine Beam Dual polarisation 25m data and used to segment the Australian landscapes. K-means clustering then grouped the segments with similar cover and backscatter into approximately 1000 clusters. Where GLAS-ICESat footprints intersected these clusters, canopy profiles were extracted and aggregated to produce a mean vertical vegetation profile for each cluster that was used to derive mean canopy and understorey height, depth and density. Due to the large number of returns, these retrievals are near continuous across the landscape, enabling them to be used for inventory and modelling applications. To validate this product, a radiative transfer model was adapted to map directional gap probability from airborne waveform lidar datasets to retrieve vertical plant profiles Comparison over several test sites show excellent agreement and work is underway to extend the analysis to improve national biomass mapping. The integration of the three datasets provide options for future operational monitoring of structure and AGB across large areas for quantifying carbon dynamics, structural change and biodiversity.

Trainable multiscript orientation detection

NASA Astrophysics Data System (ADS)

Van Beusekom, Joost; Rangoni, Yves; Breuel, Thomas M.

2010-01-01

Detecting the correct orientation of document images is an important step in large scale digitization processes, as most subsequent document analysis and optical character recognition methods assume upright position of the document page. Many methods have been proposed to solve the problem, most of which base on ascender to descender ratio computation. Unfortunately, this cannot be used for scripts having no descenders nor ascenders. Therefore, we present a trainable method using character similarity to compute the correct orientation. A connected component based distance measure is computed to compare the characters of the document image to characters whose orientation is known. This allows to detect the orientation for which the distance is lowest as the correct orientation. Training is easily achieved by exchanging the reference characters by characters of the script to be analyzed. Evaluation of the proposed approach showed accuracy of above 99% for Latin and Japanese script from the public UW-III and UW-II datasets. An accuracy of 98.9% was obtained for Fraktur on a non-public dataset. Comparison of the proposed method to two methods using ascender / descender ratio based orientation detection shows a significant improvement.

Identifying Autism from Resting-State fMRI Using Long Short-Term Memory Networks.

PubMed

Dvornek, Nicha C; Ventola, Pamela; Pelphrey, Kevin A; Duncan, James S

2017-09-01

Functional magnetic resonance imaging (fMRI) has helped characterize the pathophysiology of autism spectrum disorders (ASD) and carries promise for producing objective biomarkers for ASD. Recent work has focused on deriving ASD biomarkers from resting-state functional connectivity measures. However, current efforts that have identified ASD with high accuracy were limited to homogeneous, small datasets, while classification results for heterogeneous, multi-site data have shown much lower accuracy. In this paper, we propose the use of recurrent neural networks with long short-term memory (LSTMs) for classification of individuals with ASD and typical controls directly from the resting-state fMRI time-series. We used the entire large, multi-site Autism Brain Imaging Data Exchange (ABIDE) I dataset for training and testing the LSTM models. Under a cross-validation framework, we achieved classification accuracy of 68.5%, which is 9% higher than previously reported methods that used fMRI data from the whole ABIDE cohort. Finally, we presented interpretation of the trained LSTM weights, which highlight potential functional networks and regions that are known to be implicated in ASD.

Identifying Autism from Resting-State fMRI Using Long Short-Term Memory Networks

PubMed Central

Dvornek, Nicha C.; Ventola, Pamela; Pelphrey, Kevin A.; Duncan, James S.

2017-01-01

Functional magnetic resonance imaging (fMRI) has helped characterize the pathophysiology of autism spectrum disorders (ASD) and carries promise for producing objective biomarkers for ASD. Recent work has focused on deriving ASD biomarkers from resting-state functional connectivity measures. However, current efforts that have identified ASD with high accuracy were limited to homogeneous, small datasets, while classification results for heterogeneous, multi-site data have shown much lower accuracy. In this paper, we propose the use of recurrent neural networks with long short-term memory (LSTMs) for classification of individuals with ASD and typical controls directly from the resting-state fMRI time-series. We used the entire large, multi-site Autism Brain Imaging Data Exchange (ABIDE) I dataset for training and testing the LSTM models. Under a cross-validation framework, we achieved classification accuracy of 68.5%, which is 9% higher than previously reported methods that used fMRI data from the whole ABIDE cohort. Finally, we presented interpretation of the trained LSTM weights, which highlight potential functional networks and regions that are known to be implicated in ASD. PMID:29104967

A no-reference image and video visual quality metric based on machine learning

NASA Astrophysics Data System (ADS)

Frantc, Vladimir; Voronin, Viacheslav; Semenishchev, Evgenii; Minkin, Maxim; Delov, Aliy

2018-04-01

The paper presents a novel visual quality metric for lossy compressed video quality assessment. High degree of correlation with subjective estimations of quality is due to using of a convolutional neural network trained on a large amount of pairs video sequence-subjective quality score. We demonstrate how our predicted no-reference quality metric correlates with qualitative opinion in a human observer study. Results are shown on the EVVQ dataset with comparison existing approaches.

Residual motion compensation in ECG-gated interventional cardiac vasculature reconstruction

NASA Astrophysics Data System (ADS)

Schwemmer, C.; Rohkohl, C.; Lauritsch, G.; Müller, K.; Hornegger, J.

2013-06-01

Three-dimensional reconstruction of cardiac vasculature from angiographic C-arm CT (rotational angiography) data is a major challenge. Motion artefacts corrupt image quality, reducing usability for diagnosis and guidance. Many state-of-the-art approaches depend on retrospective ECG-gating of projection data for image reconstruction. A trade-off has to be made regarding the size of the ECG-gating window. A large temporal window is desirable to avoid undersampling. However, residual motion will occur in a large window, causing motion artefacts. We present an algorithm to correct for residual motion. Our approach is based on a deformable 2D-2D registration between the forward projection of an initial, ECG-gated reconstruction, and the original projection data. The approach is fully automatic and does not require any complex segmentation of vasculature, or landmarks. The estimated motion is compensated for during the backprojection step of a subsequent reconstruction. We evaluated the method using the publicly available CAVAREV platform and on six human clinical datasets. We found a better visibility of structure, reduced motion artefacts, and increased sharpness of the vessels in the compensated reconstructions compared to the initial reconstructions. At the time of writing, our algorithm outperforms the leading result of the CAVAREV ranking list. For the clinical datasets, we found an average reduction of motion artefacts by 13 ± 6%. Vessel sharpness was improved by 25 ± 12% on average.

Large-Scale Mass Spectrometry Imaging Investigation of Consequences of Cortical Spreading Depression in a Transgenic Mouse Model of Migraine

NASA Astrophysics Data System (ADS)

Carreira, Ricardo J.; Shyti, Reinald; Balluff, Benjamin; Abdelmoula, Walid M.; van Heiningen, Sandra H.; van Zeijl, Rene J.; Dijkstra, Jouke; Ferrari, Michel D.; Tolner, Else A.; McDonnell, Liam A.; van den Maagdenberg, Arn M. J. M.

2015-06-01

Cortical spreading depression (CSD) is the electrophysiological correlate of migraine aura. Transgenic mice carrying the R192Q missense mutation in the Cacna1a gene, which in patients causes familial hemiplegic migraine type 1 (FHM1), exhibit increased propensity to CSD. Herein, mass spectrometry imaging (MSI) was applied for the first time to an animal cohort of transgenic and wild type mice to study the biomolecular changes following CSD in the brain. Ninety-six coronal brain sections from 32 mice were analyzed by MALDI-MSI. All MSI datasets were registered to the Allen Brain Atlas reference atlas of the mouse brain so that the molecular signatures of distinct brain regions could be compared. A number of metabolites and peptides showed substantial changes in the brain associated with CSD. Among those, different mass spectral features showed significant ( t-test, P < 0.05) changes in the cortex, 146 and 377 Da, and in the thalamus, 1820 and 1834 Da, of the CSD-affected hemisphere of FHM1 R192Q mice. Our findings reveal CSD- and genotype-specific molecular changes in the brain of FHM1 transgenic mice that may further our understanding about the role of CSD in migraine pathophysiology. The results also demonstrate the utility of aligning MSI datasets to a common reference atlas for large-scale MSI investigations.

Quantitative monitoring of Arabidopsis thaliana growth and development using high-throughput plant phenotyping

PubMed Central

Arend, Daniel; Lange, Matthias; Pape, Jean-Michel; Weigelt-Fischer, Kathleen; Arana-Ceballos, Fernando; Mücke, Ingo; Klukas, Christian; Altmann, Thomas; Scholz, Uwe; Junker, Astrid

2016-01-01

With the implementation of novel automated, high throughput methods and facilities in the last years, plant phenomics has developed into a highly interdisciplinary research domain integrating biology, engineering and bioinformatics. Here we present a dataset of a non-invasive high throughput plant phenotyping experiment, which uses image- and image analysis- based approaches to monitor the growth and development of 484 Arabidopsis thaliana plants (thale cress). The result is a comprehensive dataset of images and extracted phenotypical features. Such datasets require detailed documentation, standardized description of experimental metadata as well as sustainable data storage and publication in order to ensure the reproducibility of experiments, data reuse and comparability among the scientific community. Therefore the here presented dataset has been annotated using the standardized ISA-Tab format and considering the recently published recommendations for the semantical description of plant phenotyping experiments. PMID:27529152

Quantitative monitoring of Arabidopsis thaliana growth and development using high-throughput plant phenotyping.

PubMed

Arend, Daniel; Lange, Matthias; Pape, Jean-Michel; Weigelt-Fischer, Kathleen; Arana-Ceballos, Fernando; Mücke, Ingo; Klukas, Christian; Altmann, Thomas; Scholz, Uwe; Junker, Astrid

2016-08-16

With the implementation of novel automated, high throughput methods and facilities in the last years, plant phenomics has developed into a highly interdisciplinary research domain integrating biology, engineering and bioinformatics. Here we present a dataset of a non-invasive high throughput plant phenotyping experiment, which uses image- and image analysis- based approaches to monitor the growth and development of 484 Arabidopsis thaliana plants (thale cress). The result is a comprehensive dataset of images and extracted phenotypical features. Such datasets require detailed documentation, standardized description of experimental metadata as well as sustainable data storage and publication in order to ensure the reproducibility of experiments, data reuse and comparability among the scientific community. Therefore the here presented dataset has been annotated using the standardized ISA-Tab format and considering the recently published recommendations for the semantical description of plant phenotyping experiments.

Dynamic analysis, transformation, dissemination and applications of scientific multidimensional data in ArcGIS Platform

NASA Astrophysics Data System (ADS)

Shrestha, S. R.; Collow, T. W.; Rose, B.

2016-12-01

Scientific datasets are generated from various sources and platforms but they are typically produced either by earth observation systems or by modelling systems. These are widely used for monitoring, simulating, or analyzing measurements that are associated with physical, chemical, and biological phenomena over the ocean, atmosphere, or land. A significant subset of scientific datasets stores values directly as rasters or in a form that can be rasterized. This is where a value exists at every cell in a regular grid spanning the spatial extent of the dataset. Government agencies like NOAA, NASA, EPA, USGS produces large volumes of near real-time, forecast, and historical data that drives climatological and meteorological studies, and underpins operations ranging from weather prediction to sea ice loss. Modern science is computationally intensive because of the availability of an enormous amount of scientific data, the adoption of data-driven analysis, and the need to share these dataset and research results with the public. ArcGIS as a platform is sophisticated and capable of handling such complex domain. We'll discuss constructs and capabilities applicable to multidimensional gridded data that can be conceptualized as a multivariate space-time cube. Building on the concept of a two-dimensional raster, a typical multidimensional raster dataset could contain several "slices" within the same spatial extent. We will share a case from the NOAA Climate Forecast Systems Reanalysis (CFSR) multidimensional data as an example of how large collections of rasters can be efficiently organized and managed through a data model within a geodatabase called "Mosaic dataset" and dynamically transformed and analyzed using raster functions. A raster function is a lightweight, raster-valued transformation defined over a mixed set of raster and scalar input. That means, just like any tool, you can provide a raster function with input parameters. It enables dynamic processing of only the data that's being displayed on the screen or requested by an application. We will present the dynamic processing and analysis of CFSR data using the chains of raster function and share it as dynamic multidimensional image service. This workflow and capabilities can be easily applied to any scientific data formats that are supported in mosaic dataset.

Exploratory analysis of TOF-SIMS data from biological surfaces

NASA Astrophysics Data System (ADS)

Vaidyanathan, Seetharaman; Fletcher, John S.; Henderson, Alex; Lockyer, Nicholas P.; Vickerman, John C.

2008-12-01

The application of multivariate analytical tools enables simplification of TOF-SIMS datasets so that useful information can be extracted from complex spectra and images, especially those that do not give readily interpretable results. There is however a challenge in understanding the outputs from such analyses. The problem is complicated when analysing images, given the additional dimensions in the dataset. Here we demonstrate how the application of simple pre-processing routines can enable the interpretation of TOF-SIMS spectra and images. For the spectral data, TOF-SIMS spectra used to discriminate bacterial isolates associated with urinary tract infection were studied. Using different criteria for picking peaks before carrying out PC-DFA enabled identification of the discriminatory information with greater certainty. For the image data, an air-dried salt stressed bacterial sample, discussed in another paper by us in this issue, was studied. Exploration of the image datasets with and without normalisation prior to multivariate analysis by PCA or MAF resulted in different regions of the image being highlighted by the techniques.

Chest x-ray generation and data augmentation for cardiovascular abnormality classification

NASA Astrophysics Data System (ADS)

Madani, Ali; Moradi, Mehdi; Karargyris, Alexandros; Syeda-Mahmood, Tanveer

2018-03-01

Medical imaging datasets are limited in size due to privacy issues and the high cost of obtaining annotations. Augmentation is a widely used practice in deep learning to enrich the data in data-limited scenarios and to avoid overfitting. However, standard augmentation methods that produce new examples of data by varying lighting, field of view, and spatial rigid transformations do not capture the biological variance of medical imaging data and could result in unrealistic images. Generative adversarial networks (GANs) provide an avenue to understand the underlying structure of image data which can then be utilized to generate new realistic samples. In this work, we investigate the use of GANs for producing chest X-ray images to augment a dataset. This dataset is then used to train a convolutional neural network to classify images for cardiovascular abnormalities. We compare our augmentation strategy with traditional data augmentation and show higher accuracy for normal vs abnormal classification in chest X-rays.

Bayesian automated cortical segmentation for neonatal MRI

NASA Astrophysics Data System (ADS)

Chou, Zane; Paquette, Natacha; Ganesh, Bhavana; Wang, Yalin; Ceschin, Rafael; Nelson, Marvin D.; Macyszyn, Luke; Gaonkar, Bilwaj; Panigrahy, Ashok; Lepore, Natasha

2017-11-01

Several attempts have been made in the past few years to develop and implement an automated segmentation of neonatal brain structural MRI. However, accurate automated MRI segmentation remains challenging in this population because of the low signal-to-noise ratio, large partial volume effects and inter-individual anatomical variability of the neonatal brain. In this paper, we propose a learning method for segmenting the whole brain cortical grey matter on neonatal T2-weighted images. We trained our algorithm using a neonatal dataset composed of 3 fullterm and 4 preterm infants scanned at term equivalent age. Our segmentation pipeline combines the FAST algorithm from the FSL library software and a Bayesian segmentation approach to create a threshold matrix that minimizes the error of mislabeling brain tissue types. Our method shows promising results with our pilot training set. In both preterm and full-term neonates, automated Bayesian segmentation generates a smoother and more consistent parcellation compared to FAST, while successfully removing the subcortical structure and cleaning the edges of the cortical grey matter. This method show promising refinement of the FAST segmentation by considerably reducing manual input and editing required from the user, and further improving reliability and processing time of neonatal MR images. Further improvement will include a larger dataset of training images acquired from different manufacturers.

Development of an organ-specific insert phantom generated using a 3D printer for investigations of cardiac computed tomography protocols.

PubMed

Abdullah, Kamarul A; McEntee, Mark F; Reed, Warren; Kench, Peter L

2018-04-30

An ideal organ-specific insert phantom should be able to simulate the anatomical features with appropriate appearances in the resultant computed tomography (CT) images. This study investigated a 3D printing technology to develop a novel and cost-effective cardiac insert phantom derived from volumetric CT image datasets of anthropomorphic chest phantom. Cardiac insert volumes were segmented from CT image datasets, derived from an anthropomorphic chest phantom of Lungman N-01 (Kyoto Kagaku, Japan). These segmented datasets were converted to a virtual 3D-isosurface of heart-shaped shell, while two other removable inserts were included using computer-aided design (CAD) software program. This newly designed cardiac insert phantom was later printed by using a fused deposition modelling (FDM) process via a Creatbot DM Plus 3D printer. Then, several selected filling materials, such as contrast media, oil, water and jelly, were loaded into designated spaces in the 3D-printed phantom. The 3D-printed cardiac insert phantom was positioned within the anthropomorphic chest phantom and 30 repeated CT acquisitions performed using a multi-detector scanner at 120-kVp tube potential. Attenuation (Hounsfield Unit, HU) values were measured and compared to the image datasets of real-patient and Catphan ® 500 phantom. The output of the 3D-printed cardiac insert phantom was a solid acrylic plastic material, which was strong, light in weight and cost-effective. HU values of the filling materials were comparable to the image datasets of real-patient and Catphan ® 500 phantom. A novel and cost-effective cardiac insert phantom for anthropomorphic chest phantom was developed using volumetric CT image datasets with a 3D printer. Hence, this suggested the printing methodology could be applied to generate other phantoms for CT imaging studies. © 2018 The Authors. Journal of Medical Radiation Sciences published by John Wiley & Sons Australia, Ltd on behalf of Australian Society of Medical Imaging and Radiation Therapy and New Zealand Institute of Medical Radiation Technology.

A Two-Stream Deep Fusion Framework for High-Resolution Aerial Scene Classification

PubMed Central

Liu, Fuxian

2018-01-01

One of the challenging problems in understanding high-resolution remote sensing images is aerial scene classification. A well-designed feature representation method and classifier can improve classification accuracy. In this paper, we construct a new two-stream deep architecture for aerial scene classification. First, we use two pretrained convolutional neural networks (CNNs) as feature extractor to learn deep features from the original aerial image and the processed aerial image through saliency detection, respectively. Second, two feature fusion strategies are adopted to fuse the two different types of deep convolutional features extracted by the original RGB stream and the saliency stream. Finally, we use the extreme learning machine (ELM) classifier for final classification with the fused features. The effectiveness of the proposed architecture is tested on four challenging datasets: UC-Merced dataset with 21 scene categories, WHU-RS dataset with 19 scene categories, AID dataset with 30 scene categories, and NWPU-RESISC45 dataset with 45 challenging scene categories. The experimental results demonstrate that our architecture gets a significant classification accuracy improvement over all state-of-the-art references. PMID:29581722

A Two-Stream Deep Fusion Framework for High-Resolution Aerial Scene Classification.

PubMed

Yu, Yunlong; Liu, Fuxian

2018-01-01

One of the challenging problems in understanding high-resolution remote sensing images is aerial scene classification. A well-designed feature representation method and classifier can improve classification accuracy. In this paper, we construct a new two-stream deep architecture for aerial scene classification. First, we use two pretrained convolutional neural networks (CNNs) as feature extractor to learn deep features from the original aerial image and the processed aerial image through saliency detection, respectively. Second, two feature fusion strategies are adopted to fuse the two different types of deep convolutional features extracted by the original RGB stream and the saliency stream. Finally, we use the extreme learning machine (ELM) classifier for final classification with the fused features. The effectiveness of the proposed architecture is tested on four challenging datasets: UC-Merced dataset with 21 scene categories, WHU-RS dataset with 19 scene categories, AID dataset with 30 scene categories, and NWPU-RESISC45 dataset with 45 challenging scene categories. The experimental results demonstrate that our architecture gets a significant classification accuracy improvement over all state-of-the-art references.

Ischemic stroke lesion segmentation in multi-spectral MR images with support vector machine classifiers

NASA Astrophysics Data System (ADS)

Maier, Oskar; Wilms, Matthias; von der Gablentz, Janina; Krämer, Ulrike; Handels, Heinz

2014-03-01

Automatic segmentation of ischemic stroke lesions in magnetic resonance (MR) images is important in clinical practice and for neuroscientific trials. The key problem is to detect largely inhomogeneous regions of varying sizes, shapes and locations. We present a stroke lesion segmentation method based on local features extracted from multi-spectral MR data that are selected to model a human observer's discrimination criteria. A support vector machine classifier is trained on expert-segmented examples and then used to classify formerly unseen images. Leave-one-out cross validation on eight datasets with lesions of varying appearances is performed, showing our method to compare favourably with other published approaches in terms of accuracy and robustness. Furthermore, we compare a number of feature selectors and closely examine each feature's and MR sequence's contribution.

ChiMS: Open-source instrument control software platform on LabVIEW for imaging/depth profiling mass spectrometers.

PubMed

Cui, Yang; Hanley, Luke

2015-06-01

ChiMS is an open-source data acquisition and control software program written within LabVIEW for high speed imaging and depth profiling mass spectrometers. ChiMS can also transfer large datasets from a digitizer to computer memory at high repetition rate, save data to hard disk at high throughput, and perform high speed data processing. The data acquisition mode generally simulates a digital oscilloscope, but with peripheral devices integrated for control as well as advanced data sorting and processing capabilities. Customized user-designed experiments can be easily written based on several included templates. ChiMS is additionally well suited to non-laser based mass spectrometers imaging and various other experiments in laser physics, physical chemistry, and surface science.

ChiMS: Open-source instrument control software platform on LabVIEW for imaging/depth profiling mass spectrometers

PubMed Central

Cui, Yang; Hanley, Luke

2015-01-01

ChiMS is an open-source data acquisition and control software program written within LabVIEW for high speed imaging and depth profiling mass spectrometers. ChiMS can also transfer large datasets from a digitizer to computer memory at high repetition rate, save data to hard disk at high throughput, and perform high speed data processing. The data acquisition mode generally simulates a digital oscilloscope, but with peripheral devices integrated for control as well as advanced data sorting and processing capabilities. Customized user-designed experiments can be easily written based on several included templates. ChiMS is additionally well suited to non-laser based mass spectrometers imaging and various other experiments in laser physics, physical chemistry, and surface science. PMID:26133872

ChiMS: Open-source instrument control software platform on LabVIEW for imaging/depth profiling mass spectrometers

NASA Astrophysics Data System (ADS)

Cui, Yang; Hanley, Luke

2015-06-01

ChiMS is an open-source data acquisition and control software program written within LabVIEW for high speed imaging and depth profiling mass spectrometers. ChiMS can also transfer large datasets from a digitizer to computer memory at high repetition rate, save data to hard disk at high throughput, and perform high speed data processing. The data acquisition mode generally simulates a digital oscilloscope, but with peripheral devices integrated for control as well as advanced data sorting and processing capabilities. Customized user-designed experiments can be easily written based on several included templates. ChiMS is additionally well suited to non-laser based mass spectrometers imaging and various other experiments in laser physics, physical chemistry, and surface science.

Joint optic disc and cup boundary extraction from monocular fundus images.

PubMed

Chakravarty, Arunava; Sivaswamy, Jayanthi

2017-08-01

Accurate segmentation of optic disc and cup from monocular color fundus images plays a significant role in the screening and diagnosis of glaucoma. Though optic cup is characterized by the drop in depth from the disc boundary, most existing methods segment the two structures separately and rely only on color and vessel kink based cues due to the lack of explicit depth information in color fundus images. We propose a novel boundary-based Conditional Random Field formulation that extracts both the optic disc and cup boundaries in a single optimization step. In addition to the color gradients, the proposed method explicitly models the depth which is estimated from the fundus image itself using a coupled, sparse dictionary trained on a set of image-depth map (derived from Optical Coherence Tomography) pairs. The estimated depth achieved a correlation coefficient of 0.80 with respect to the ground truth. The proposed segmentation method outperformed several state-of-the-art methods on five public datasets. The average dice coefficient was in the range of 0.87-0.97 for disc segmentation across three datasets and 0.83 for cup segmentation on the DRISHTI-GS1 test set. The method achieved a good glaucoma classification performance with an average AUC of 0.85 for five fold cross-validation on RIM-ONE v2. We propose a method to jointly segment the optic disc and cup boundaries by modeling the drop in depth between the two structures. Since our method requires a single fundus image per eye during testing it can be employed in the large-scale screening of glaucoma where expensive 3D imaging is unavailable. Copyright © 2017 Elsevier B.V. All rights reserved.

Standardized processing of MALDI imaging raw data for enhancement of weak analyte signals in mouse models of gastric cancer and Alzheimer's disease.

PubMed

Schwartz, Matthias; Meyer, Björn; Wirnitzer, Bernhard; Hopf, Carsten

2015-03-01

Conventional mass spectrometry image preprocessing methods used for denoising, such as the Savitzky-Golay smoothing or discrete wavelet transformation, typically do not only remove noise but also weak signals. Recently, memory-efficient principal component analysis (PCA) in conjunction with random projections (RP) has been proposed for reversible compression and analysis of large mass spectrometry imaging datasets. It considers single-pixel spectra in their local context and consequently offers the prospect of using information from the spectra of adjacent pixels for denoising or signal enhancement. However, little systematic analysis of key RP-PCA parameters has been reported so far, and the utility and validity of this method for context-dependent enhancement of known medically or pharmacologically relevant weak analyte signals in linear-mode matrix-assisted laser desorption/ionization (MALDI) mass spectra has not been explored yet. Here, we investigate MALDI imaging datasets from mouse models of Alzheimer's disease and gastric cancer to systematically assess the importance of selecting the right number of random projections k and of principal components (PCs) L for reconstructing reproducibly denoised images after compression. We provide detailed quantitative data for comparison of RP-PCA-denoising with the Savitzky-Golay and wavelet-based denoising in these mouse models as a resource for the mass spectrometry imaging community. Most importantly, we demonstrate that RP-PCA preprocessing can enhance signals of low-intensity amyloid-β peptide isoforms such as Aβ1-26 even in sparsely distributed Alzheimer's β-amyloid plaques and that it enables enhanced imaging of multiply acetylated histone H4 isoforms in response to pharmacological histone deacetylase inhibition in vivo. We conclude that RP-PCA denoising may be a useful preprocessing step in biomarker discovery workflows.

Avulsion research using flume experiments and highly accurate and temporal-rich SfM datasets

NASA Astrophysics Data System (ADS)

Javernick, L.; Bertoldi, W.; Vitti, A.

2017-12-01

SfM's ability to produce high-quality, large-scale digital elevation models (DEMs) of complicated and rapidly evolving systems has made it a valuable technique for low-budget researchers and practitioners. While SfM has provided valuable datasets that capture single-flood event DEMs, there is an increasing scientific need to capture higher temporal resolution datasets that can quantify the evolutionary processes instead of pre- and post-flood snapshots. However, flood events' dangerous field conditions and image matching challenges (e.g. wind, rain) prevent quality SfM-image acquisition. Conversely, flume experiments offer opportunities to document flood events, but achieving consistent and accurate DEMs to detect subtle changes in dry and inundated areas remains a challenge for SfM (e.g. parabolic error signatures).This research aimed at investigating the impact of naturally occurring and manipulated avulsions on braided river morphology and on the encroachment of floodplain vegetation, using laboratory experiments. This required DEMs with millimeter accuracy and precision and at a temporal resolution to capture the processes. SfM was chosen as it offered the most practical method. Through redundant local network design and a meticulous ground control point (GCP) survey with a Leica Total Station in red laser configuration (reported 2 mm accuracy), the SfM residual errors compared to separate ground truthing data produced mean errors of 1.5 mm (accuracy) and standard deviations of 1.4 mm (precision) without parabolic error signatures. Lighting conditions in the flume were limited to uniform, oblique, and filtered LED strips, which removed glint and thus improved bed elevation mean errors to 4 mm, but errors were further reduced by means of an open source software for refraction correction. The obtained datasets have provided the ability to quantify how small flood events with avulsion can have similar morphologic and vegetation impacts as large flood events without avulsion. Further, this research highlights the potential application of SfM in the laboratory and ability to document physical and biological processes at greater spatial and temporal resolution. Marie Sklodowska-Curie Individual Fellowship: River-HMV, 656917

Three-dimensional estimates of tree canopies: Scaling from high-resolution UAV data to satellite observations

NASA Astrophysics Data System (ADS)

Sankey, T.; Donald, J.; McVay, J.

2015-12-01

High resolution remote sensing images and datasets are typically acquired at a large cost, which poses big a challenge for many scientists. Northern Arizona University recently acquired a custom-engineered, cutting-edge UAV and we can now generate our own images with the instrument. The UAV has a unique capability to carry a large payload including a hyperspectral sensor, which images the Earth surface in over 350 spectral bands at 5 cm resolution, and a lidar scanner, which images the land surface and vegetation in 3-dimensions. Both sensors represent the newest available technology with very high resolution, precision, and accuracy. Using the UAV sensors, we are monitoring the effects of regional forest restoration treatment efforts. Individual tree canopy width and height are measured in the field and via the UAV sensors. The high-resolution UAV images are then used to segment individual tree canopies and to derive 3-dimensional estimates. The UAV image-derived variables are then correlated to the field-based measurements and scaled to satellite-derived tree canopy measurements. The relationships between the field-based and UAV-derived estimates are then extrapolated to a larger area to scale the tree canopy dimensions and to estimate tree density within restored and control forest sites.

T1-weighted in vivo human whole brain MRI dataset with an ultrahigh isotropic resolution of 250 μm.

PubMed

Lüsebrink, Falk; Sciarra, Alessandro; Mattern, Hendrik; Yakupov, Renat; Speck, Oliver

2017-03-14

We present an ultrahigh resolution in vivo human brain magnetic resonance imaging (MRI) dataset. It consists of T 1 -weighted whole brain anatomical data acquired at 7 Tesla with a nominal isotropic resolution of 250 μm of a single young healthy Caucasian subject and was recorded using prospective motion correction. The raw data amounts to approximately 1.2 TB and was acquired in eight hours total scan time. The resolution of this dataset is far beyond any previously published in vivo structural whole brain dataset. Its potential use is to build an in vivo MR brain atlas. Methods for image reconstruction and image restoration can be improved as the raw data is made available. Pre-processing and segmentation procedures can possibly be enhanced for high magnetic field strength and ultrahigh resolution data. Furthermore, potential resolution induced changes in quantitative data analysis can be assessed, e.g., cortical thickness or volumetric measures, as high quality images with an isotropic resolution of 1 and 0.5 mm of the same subject are included in the repository as well.

Sky Detection in Hazy Image.

PubMed

Song, Yingchao; Luo, Haibo; Ma, Junkai; Hui, Bin; Chang, Zheng

2018-04-01

Sky detection plays an essential role in various computer vision applications. Most existing sky detection approaches, being trained on ideal dataset, may lose efficacy when facing unfavorable conditions like the effects of weather and lighting conditions. In this paper, a novel algorithm for sky detection in hazy images is proposed from the perspective of probing the density of haze. We address the problem by an image segmentation and a region-level classification. To characterize the sky of hazy scenes, we unprecedentedly introduce several haze-relevant features that reflect the perceptual hazy density and the scene depth. Based on these features, the sky is separated by two imbalance SVM classifiers and a similarity measurement. Moreover, a sky dataset (named HazySky) with 500 annotated hazy images is built for model training and performance evaluation. To evaluate the performance of our method, we conducted extensive experiments both on our HazySky dataset and the SkyFinder dataset. The results demonstrate that our method performs better on the detection accuracy than previous methods, not only under hazy scenes, but also under other weather conditions.

T1-weighted in vivo human whole brain MRI dataset with an ultrahigh isotropic resolution of 250 μm

NASA Astrophysics Data System (ADS)

Lüsebrink, Falk; Sciarra, Alessandro; Mattern, Hendrik; Yakupov, Renat; Speck, Oliver

2017-03-01

We present an ultrahigh resolution in vivo human brain magnetic resonance imaging (MRI) dataset. It consists of T1-weighted whole brain anatomical data acquired at 7 Tesla with a nominal isotropic resolution of 250 μm of a single young healthy Caucasian subject and was recorded using prospective motion correction. The raw data amounts to approximately 1.2 TB and was acquired in eight hours total scan time. The resolution of this dataset is far beyond any previously published in vivo structural whole brain dataset. Its potential use is to build an in vivo MR brain atlas. Methods for image reconstruction and image restoration can be improved as the raw data is made available. Pre-processing and segmentation procedures can possibly be enhanced for high magnetic field strength and ultrahigh resolution data. Furthermore, potential resolution induced changes in quantitative data analysis can be assessed, e.g., cortical thickness or volumetric measures, as high quality images with an isotropic resolution of 1 and 0.5 mm of the same subject are included in the repository as well.

Sky Detection in Hazy Image

PubMed Central

Song, Yingchao; Luo, Haibo; Ma, Junkai; Hui, Bin; Chang, Zheng

2018-01-01

Sky detection plays an essential role in various computer vision applications. Most existing sky detection approaches, being trained on ideal dataset, may lose efficacy when facing unfavorable conditions like the effects of weather and lighting conditions. In this paper, a novel algorithm for sky detection in hazy images is proposed from the perspective of probing the density of haze. We address the problem by an image segmentation and a region-level classification. To characterize the sky of hazy scenes, we unprecedentedly introduce several haze-relevant features that reflect the perceptual hazy density and the scene depth. Based on these features, the sky is separated by two imbalance SVM classifiers and a similarity measurement. Moreover, a sky dataset (named HazySky) with 500 annotated hazy images is built for model training and performance evaluation. To evaluate the performance of our method, we conducted extensive experiments both on our HazySky dataset and the SkyFinder dataset. The results demonstrate that our method performs better on the detection accuracy than previous methods, not only under hazy scenes, but also under other weather conditions. PMID:29614778