Tony Pan

Image processing for the grid: a toolkit for building grid-enabled image processing applications

Analyzing large and distributed image datasets is a crucial step in understanding the structural and functional characteristics of biological systems. In this paper, we present the design and implementation of a toolkit that allows rapid and efficient development of biomedical image analysis applications in a distributed environment. This toolkit employs the Insight Segmentation and Registration Toolkit (ITK) and Visualization Toolkit (VTK) layered on a component-based framework. We present experimental results on a cluster of workstations.

Image analysis for neuroblastoma classification: segmentation of cell nuclei.

Neuroblastoma is a childhood cancer of the nervous system. Current prognostic classification of this disease partly relies on morphological characteristics of the cells from H&E-stained images. In this work, an automated cell nuclei segmentation method is developed. This method employs morphological top-hat by reconstruction algorithm coupled with hysteresis thresholding to both detect and segment the cell nuclei. Accuracy of the automated cell nuclei segmentation algorithm is measured by comparing its outputs to manual segmentation. The average segmentation accuracy is 90.24plusmn5.14%

Integrated morphologic analysis for the identification and characterization of disease subtypes

Background and objective Morphologic variations of disease are often linked to underlying molecular events and patient outcome, suggesting that quantitative morphometric analysis may provide further insight into disease mechanisms. In this paper a methodology for the subclassification of disease is developed using image analysis techniques. Morphologic signatures that represent patient-specific tumor morphology are derived from the analysis of hundreds of millions of cells in digitized whole slide images. Clustering these signatures aggregates tumors into groups with cohesive morphologic characteristics. This methodology is demonstrated with an analysis of glioblastoma, using data from The Cancer Genome Atlas to identify a prognostically significant morphology-driven subclassification, in which clusters are correlated with transcriptional, genetic, and epigenetic events. Materials and methods Methodology was applied to 162 glioblastomas from The Cancer Genome Atlas to identify morphology-driven clusters and their clinical and molecular correlates. Signatures of patient-specific tumor morphology were generated from analysis of 200 million cells in 462 whole slide images. Morphology-driven clusters were interrogated for associations with patient outcome, response to therapy, molecular classifications, and genetic alterations. An additional layer of deep, genome-wide analysis identified characteristic transcriptional, epigenetic, and copy number variation events. Results and discussion Analysis of glioblastoma identified three prognostically significant patient clusters (median survival 15.3, 10.7, and 13.0 months, log rank p=1.4e-3). Clustering results were validated in a separate dataset. Clusters were characterized by molecular events in nuclear compartment signaling including developmental and cell cycle checkpoint pathways. This analysis demonstrates the potential of high-throughput morphometrics for the subclassification of disease, establishing an approach that complements genomics.

An Integrative Approach for In Silico Glioma Research

The integration of imaging and genomic data is critical to forming a better understanding of disease. Large public datasets, such as The Cancer Genome Atlas, present a unique opportunity to integrate these complementary data types for in silico scientific research. In this letter, we focus on the aspect of pathology image analysis and illustrate the challenges associated with analyzing and integrating large-scale image datasets with molecular characterizations. We present an example study of diffuse glioma brain tumors, where the morphometric analysis of 81 million nuclei is integrated with clinically relevant transcriptomic and genomic characterizations of glioblastoma tumors. The preliminary results demonstrate the potential of combining morphometric and molecular characterizations for in silico research.

ImageMiner: a software system for comparative analysis of tissue microarrays using content-based image retrieval, high-performance computing, and grid technology

OBJECTIVE AND DESIGN The design and implementation of ImageMiner, a software platform for performing comparative analysis of expression patterns in imaged microscopy specimens such as tissue microarrays (TMAs), is described. ImageMiner is a federated system of services that provides a reliable set of analytical and data management capabilities for investigative research applications in pathology. It provides a library of image processing methods, including automated registration, segmentation, feature extraction, and classification, all of which have been tailored, in these studies, to support TMA analysis. The system is designed to leverage high-performance computing machines so that investigators can rapidly analyze large ensembles of imaged TMA specimens. To support deployment in collaborative, multi-institutional projects, ImageMiner features grid-enabled, service-based components so that multiple instances of ImageMiner can be accessed remotely and federated. RESULTS The experimental evaluation shows that: (1) ImageMiner is able to support reliable detection and feature extraction of tumor regions within imaged tissues; (2) images and analysis results managed in ImageMiner can be searched for and retrieved on the basis of image-based features, classification information, and any correlated clinical data, including any metadata that have been generated to describe the specified tissue and TMA; and (3) the system is able to reduce computation time of analyses by exploiting computing clusters, which facilitates analysis of larger sets of tissue samples.

A Data Model and Database for High-resolution Pathology Analytical Image Informatics

Background: The systematic analysis of imaged pathology specimens often results in a vast amount of morphological information at both the cellular and sub-cellular scales. While microscopy scanners and computerized analysis are capable of capturing and analyzing data rapidly, microscopy image data remain underutilized in research and clinical settings. One major obstacle which tends to reduce wider adoption of these new technologies throughout the clinical and scientific communities is the challenge of managing, querying, and integrating the vast amounts of data resulting from the analysis of large digital pathology datasets. This paper presents a data model, which addresses these challenges, and demonstrates its implementation in a relational database system. Context: This paper describes a data model, referred to as Pathology Analytic Imaging Standards (PAIS), and a database implementation, which are designed to support the data management and query requirements of detailed characterization of micro-anatomic morphology through many interrelated analysis pipelines on whole-slide images and tissue microarrays (TMAs). Aims: (1) Development of a data model capable of efficiently representing and storing virtual slide related image, annotation, markup, and feature information. (2) Development of a database, based on the data model, capable of supporting queries for data retrieval based on analysis and image metadata, queries for comparison of results from different analyses, and spatial queries on segmented regions, features, and classified objects. Settings and Design: The work described in this paper is motivated by the challenges associated with characterization of micro-scale features for comparative and correlative analyses involving whole-slides tissue images and TMAs. Technologies for digitizing tissues have advanced significantly in the past decade. Slide scanners are capable of producing high-magnification, high-resolution images from whole slides and TMAs within several minutes. Hence, it is becoming increasingly feasible for basic, clinical, and translational research studies to produce thousands of whole-slide images. Systematic analysis of these large datasets requires efficient data management support for representing and indexing results from hundreds of interrelated analyses generating very large volumes of quantifications such as shape and texture and of classifications of the quantified features. Materials and Methods: We have designed a data model and a database to address the data management requirements of detailed characterization of micro-anatomic morphology through many interrelated analysis pipelines. The data model represents virtual slide related image, annotation, markup and feature information. The database supports a wide range of metadata and spatial queries on images, annotations, markups, and features. Results: We currently have three databases running on a Dell PowerEdge T410 server with CentOS 5.5 Linux operating system. The database server is IBM DB2 Enterprise Edition 9.7.2. The set of databases consists of 1) a TMA database containing image analysis results from 4740 cases of breast cancer, with 641 MB storage size; 2) an algorithm validation database, which stores markups and annotations from two segmentation algorithms and two parameter sets on 18 selected slides, with 66 GB storage size; and 3) an in silico brain tumor study database comprising results from 307 TCGA slides, with 365 GB storage size. The latter two databases also contain human-generated annotations and markups for regions and nuclei. Conclusions: Modeling and managing pathology image analysis results in a database provide immediate benefits on the value and usability of data in a research study. The database provides powerful query capabilities, which are otherwise difficult or cumbersome to support by other approaches such as programming languages. Standardized, semantic annotated data representation and interfaces also make it possible to more efficiently share image data and analysis results.

Integrative, Multimodal Analysis of Glioblastoma Using TCGA Molecular Data, Pathology Images, and Clinical Outcomes

Multimodal, multiscale data synthesis is becoming increasingly critical for successful translational biomedical research. In this letter, we present a large-scale investigative initiative on glioblastoma, a high-grade brain tumor, with complementary data types using in silico approaches. We integrate and analyze data from The Cancer Genome Atlas Project on glioblastoma that includes novel nuclear phenotypic data derived from microscopic slides, genotypic signatures described by transcriptional class and genetic alterations, and clinical outcomes defined by response to therapy and patient survival. Our preliminary results demonstrate numerous clinically and biologically significant correlations across multiple data types, revealing the power of in silico multimodal data integration for cancer research.

Registration and 3D visualization of large microscopy images

Inactivation of the retinoblastoma gene in mouse embryos causes tissue infiltrations into critical sections of the placenta, which has been shown to affect fetal survivability. Our collaborators in cancer genetics are extremely interested in examining the three dimensional nature of these infiltrations given a stack of two dimensional light microscopy images. Three sets of wildtype and mutant placentas was sectioned serially and digitized using a commercial light microscopy scanner. Each individual placenta dataset consisted of approximately 1000 images totaling 700 GB in size, which were registered into a volumetric dataset using National Library of Medicines (NIH/NLM) Insight Segmentation and Registration Toolkit (ITK). This paper describes our method for image registration to aid in volume visualization of tissue level intermixing for both wildtype and Rb- specimens. The registration process faces many challenges arising from the large image sizes, damages during sectioning, staining gradients both within and across sections, and background noise. These issues limit the direct application of standard registration techniques due to frequent convergence to local solutions. In this work, we develop a mixture of automated and semi-automated enhancements with ground-truth validation for the mutual information-based registration algorithm. Our final volume renderings clearly show tissue intermixing differences between both wildtype and Rb- specimens which are not obvious prior to registration.

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

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 4K×4K-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.

Accelerating Large Scale Image Analyses on Parallel, CPU-GPU Equipped Systems

The past decade has witnessed a major paradigm shift in high performance computing with the introduction of accelerators as general purpose processors. These computing devices make available very high parallel computing power at low cost and power consumption, transforming current high performance platforms into heterogeneous CPU-GPU equipped systems. Although the theoretical performance achieved by these hybrid systems is impressive, taking practical advantage of this computing power remains a very challenging problem. Most applications are still deployed to either GPU or CPU, leaving the other resource under- or un-utilized. In this paper, we propose, implement, and evaluate a performance aware scheduling technique along with optimizations to make efficient collaborative use of CPUs and GPUs on a parallel system. In the context of feature computations in large scale image analysis applications, our evaluations show that intelligently co-scheduling CPUs and GPUs can significantly improve performance over GPU-only or multi-core CPU-only approaches.