Wiki page for 2013 Preliminary Draft

 Back to NAMIC_Annual_Reports

Required elements of the New Research Progress Template

1. RESEARCH AND RESOURCE METRICS

1A. Summary of Center Progress

• Brief description of overall objectives of NA-MIC

The National Alliance for Medical Image Computing (NA-MIC) is a multi-institutional, interdisciplinary community of computer scientists, software engineers, and medical investigators who share the common goal of improving healthcare through the development of computational tools for the analysis and visualization of medical image data. The Center maintains a robust and flexible infrastructure for developing and applying advanced imaging technologies across a range of important biomedical research disciplines. Our research and development effort is organized around the Computer Science Core, which includes independent teams for Algorithms and Engineering. The NA-MIC Computer Science Algorithm effort responds to the challenges of the DBPs to expand the horizons of medical image analysis. As a result, the Algorithm activities are typically highly experimental, creating new approaches that are rapidly prototyped, tested, and improved. The Engineering effort supports the needs of the Algorithms effort by creating integrated software platforms that support research and eventual deployment of advanced technology. The Engineering team also develops and maintains processes used to build and sustain a large research community. A separate Core oversees operations and maintenance of The NA-MIC Kit, an integrated set of interoperable free open source software (FOSS) packages; developed, supported and deployed using a collaborative, agile, high quality software process. NA-MIC's current DBPs are investigating solutions to problems in patient-specific data analysis in four clinical areas: Atrial Fibrillation, Huntingdon's Disease, Adaptive Radiotherapy for Head and Neck Cancer, and Traumatic Brain Injury. NA-MIC further provides enabling technology and resources to XX collaborative research projects.

• Brief outline indicating strengths of NA-MIC as a national resource

[take from NA-MIC impact statement, August 2012]

• Summarize progress made in each Research and Core Project.

Algorithms. The NA-MIC Computer Science Algorithm effort responds to the challenges of the DBPs to expand the horizons of medical image analysis. As a result, the Algorithm activities are typically highly experimental, creating new approaches that are rapidly prototyped, tested, and improved. Engineering. The NA-MIC Computer Science Engineering effort supports the needs of the Algorithms effort by creating integrated software platforms that support research and eventual deployment of advanced technology. The Engineering team also develops and maintains processes used to build and sustain a large research community. NA-MIC Kit. The NA-MIC Kit consists of an integrated set of interoperable free open source software (FOSS) packages; developed, supported and deployed using a collaborative, agile, high quality software process. The NA-MIC Kit has been constructed as a layered architecture to provide a spectrum of capabilities, ranging from compute-intensive algorithms to easy-to-use applications. Hence users and developers can choose to engage the NA-MIC Kit at a variety of levels, including developing extensions which can be readily deployed to the broader biomedical imaging community. In the following subsections we highlight the accomplishments from this reporting period for algorithms, engineering, and NA-MIC Kit. 2.1. Algorithms The Algorithms team develops computational methods that support patient-specific analysis of medical images. This effort requires analysis of images that vary significantly from one patient to another, or from one time point to another, presenting distinct challenges to existing state-of-art medical image analysis algorithms. These technical challenges are addressed using four computational approaches: (1) Statistical models of anatomy and pathology; (2) Geometric correspondence; (3) User interactive tools for segmentation; and (4) Longitudinal and time-series analysis. Highlights of these efforts are described in the following sections. Statistical models of anatomy and pathology. A great deal of progress has been made by using modeling approaches that systematically capture the statistics of a problem domain from a collection of examples and then use these statistics to interpret novel images. Some of the approaches include the following:  Non-Parametric Priors for Segmentation are based on nonparametric, probabilistic models for the automatic segmentation of medical images, given a training set of images and corresponding label maps. The resulting inference algorithms rely on pairwise registrations between the test image and individual training images. The training labels are then transferred to the test image and fused to compute the final segmentation of the test subject.  Fast Nearest-Neighbor Lookup in Large Image Databases has been found to improve segmentation quality. Multi-atlases or nonparametric atlas-based techniques for image segmentation require registration of a test image with a small set of very similar images from a database.  Atlases and Registration for DTI Processing are novel methods that enhance the co-registration of DTI data either to a prior image of the same subject or to an existing atlas with predefined fiber tracts or regional white matter parcellation. These are applied in cases of large brain pathology (e.g., TBI). Geometric correspondence. Establishing anatomical correspondences between pairs of patients, groups of patients, patients and templates, and individual patients over time is important for automatic and user-assisted image analysis. The ability to establish geometric correspondences, with and without expert guidance, in challenging clinical circumstances is essential for the DBPs. Progress in two areas was realized this year.  Stochastic Point Set Registration provides non-rigid point set registration algorithms that seek an optimal set of radial basis functions to describe the registration. Preliminary results on 2D and 3D data demonstrate the algorithms' robustness to datasets with noise and with missing information.  Automatic Correspondences For Shape Ensembles has seen improvements in robustness of our entropy-based correspondence system. For example, we have developed a method for particles to interact on surfaces using geodesic distances, which improves the behavior of the system on sharp features or convoluted shapes. User interactive tools for segmentation. The work performed in the past year addresses important aspects of user-interactive segmentation. The patient-specific analysis required by the DBPs has presented images of patients with pathologies and/or injuries that sometimes defy automated approaches. We have focused our research on three principal areas.  Control-Based Interactive Segmentation is a novel contribution based on a modeling formulation that represents interactive segmentation as a feedback system, enabling a principled merging of automated methods and user input.  Globally Optimal Segmentation is a set of methods that rely on global optimization of energy functions via graph cuts. Results on delayed contrast MRI from the Atrial Fibrillation project are quite promising, and this work is currently under review for publication.  Patient-Specific Segmentation Framework for Longitudinal MR Images of Traumatic Brain Injury addresses the need for robust, reproducible segmentations of MR images of TBI and is crucial for quantitative analysis of recovery and treatment efficacy. Validation of this new automatic segmentation compared to expert segmentations of acute and chronic images was provided on 3 longitudinal TBI datasets, demonstrating that joint segmentation of 4D multi-time point data is superior to individual segmentations. Longitudinal and time-series analysis. An important component of patient-specific data analysis is the ability to analyze multiple images from the same patient over time as a disease or injury progresses or responds to treatment, or to assess neurodevelopment or neurodegeneration. Longitudinal image analysis is important for all four DBPs in this project; we have focused in the past year on the areas described below.  Connectivity Changes in Disease demonstrates a novel probabilistic framework to merge information from diffusion weighted imaging tractography and resting-state functional magnetic resonance imaging correlations to identify connectivity patterns in the brain. The method simultaneously infers the templates of latent connectivity for each population and the differences in connectivity between the groups.  Modeling Pathology Evolution is used in brain tumor patients to monitor the state of disease and to evaluate therapeutic options. This work investigated a joint generative model of tumor growth and of image observation that naturally handles multimodal and longitudinal data, important for TBI.  Longitudinal Analysis of DTI Change Trajectories develops models that represent the growth trajectories of individual subjects to study and understand white matter changes in neurodevelopment, neurodegeneration, and disease progress. Application of this methodology to study early brain development in a longitudinal neuroimaging study, including validation of reproducibility, has been shown.  Analysis of Longitudinal Shape Variability via Subject-Specific Growth Modeling are statistical analyses of longitudinal imaging data which are crucial for understanding normal anatomical development as well as disease progression. We have developed a new type of growth model parameterized by acceleration, whereas standard methods typically control the velocity. This mimics the behavior of biological tissue; cross validation experiments show that our method is robust to missing observations, is less sensitive to noise, and is therefore more likely to capture the underlying biological growth.  Longitudinal and Time Series Analysis are novel methods for longitudinal registration and time series regression. These methods enable compact approximation of an image time-series through an initial image and an initial momentum, resulting in dramatically simplified computations. 2.2 Engineering The Engineering Team builds bridges between the various NA-MIC cores and ultimately to the wider biomedical computing community. Working with the Algorithms Team, it deploys leading edge biomedical computing tools back to the DBPs, which are then used to perform impactful health research. In addition, the tools developed by the Engineering Team are used to train and disseminate technologies across the research community. The Team places particular focus on developing sustainable communities through the creation of open platforms, quality-inducing software processes, and integration to a broad variety of computational tools and databases. The following describes some of the highlights of the past year's work. The flagship product from the Engineering Team is the 3D Slicer application. It is the delivery platform for much current work, and it is an enabling technology for the wider biomedical computing community. This past year saw the release of Slicer version 4.0 (Slicer4) which represents a significant advance in capabilities and underlying technologies. Since its release at RSNA in November, Slicer4 has been downloaded over 45,000 times, at a rate of over 100 downloads per day, from users and research groups located around the world. Slicer4 is now a modern, stable platform built with the Qt GUI system (eliminated the previous KWWidgets GUI), and rewritten for simplicity, enabling simpler approaches to debugging, faster startup, and more responsive behavior. Beyond the core Slicer4 platform, several important features were also added to the application. These include:  The Slicer Extension Manager is now called the "Slicer Catalog" (an App Store) and will enable the community to create compact modules which extend the core functionality.  Python has been adopted as the preferred scripting language, a preferred programming language in the scientific computing community. Hence a variety of computational packages are now available to extend Slicer capabilities at run-time.  Slicer4 includes a DICOM listener and DICOM Query/Retrieve capabilities for integration with standard clinical image management environments and workflows.  Compatibility with ITK version 4 was developed and continuously maintained over the past year as ITKv4 matured. Slicer will officially switch to ITKv4 in the coming months.  Slicer Execution Model modules (also known as Command Line Modules) are now available as Nipype tools, enabling local and distributed scripted execution of processing pipelines. Such methods for distributed computing are essential to tackling the Big Data and complex algorithms that current research is producing.  Finally, a whole host of application improvements have been made including an improved flexible view layout system; a revised implementation of the Expectation Maximization (EM) Segmenter; faster hardware-accelerated volume rendering; improved markups and annotations; improved atlas and model hierarchy support; and a streamlined and revised diffusion MRI implementation. Community support for NA-MIC and the various NA-MIC Kit tools continues. The goals of this effort are to transition new technologies to the wider community, to enable community members to contribute back to Slicer and the NA-MIC Kit, and to ensure high-quality systems. Beyond some of the support activities mentioned previously, the following are other accomplishments.  We have begun integrating the SimpleITK module of ITKv4 into Slicer to ensure simple integration capabilities with emerging algorithms.  Additional open data support has been added to Slicer such as ultrasound (e.g., video) and 4D (e.g., gated CT) data.  We have integrated the extension writing and the documentation generation processes. The documentation created when an extension is written is now automatically ported to a web host for easier access from within and outside of Slicer, ensuring that documentation resources keep up with the rapid pace of development. 2.3 NA-MIC Kit The NA-MIC Kit is designed to accelerate the pace of research and facilitate clinical evaluation. Along these lines, the past year realized significant milestones toward the creation of a stable research platform, supporting the ability to easily extend and disseminate novel additions, all in the context of a world-wide, broad research community. Beyond the major highlights related to the Slicer4 application platform described in the previous section, the following are a few of the highlights of the past year.  CMake and its associated software process tools (CTest, CDash, and CPack) are used to build, test and deploy software in a cross-platform manner. CMake continues one of the most well-known pieces of the NA-MIC Kit, with more than 2,000 known downloads per day (as well as being included by various Linux distributions). CMake 2.8.7 was released with NA-MIC support.  CDash Package Manager (CDash 2.0.2) was released with support from NA-MIC. One of the most significant contributions to CDash from NA-MIC was the package upload process. This process enables the many Slicer testing machines to upload the executables and packages created during testing to the main CDash server. This, in turn, allows users to download those testing packages and run additional tests or use them in their research. This complete automation of the test-release cycle is a massive time-saver for the Service core and has greatly reduced the time to discover and resolve bugs and to improve the stability of Slicer.  Significant data integration efforts were completed over the past year. XNAT was greatly improved in its usability and interfaces. DICOM support was greatly enhanced, including the ability to embed Slicer MRML scene files as DICOM lollipops, meaning that Slicer data exchange across the DICOM standard is now possible. In addition, DCMTK was integrated into the NA-MIC Kit, meaning that DICOM support and functionality was greatly increased.  NA-MIC supports and nurtures an extensive biomedical research community. Along these lines it develops integration tools and interfaces with other communities. CTK, supported by NA-MIC funding, is one such community and interfaces with other open-source toolkits (e.g., MITK from the German Cancer Research Center in Heidelberg, XIP from Siemens, GIMIAS from UPF in Spain, and OpenMAF from U of Bologna). CTK now provides several innovative GUI and DICOM elements that specifically save GUI space, user-time, and developer effort when building custom medical applications. The NAMIC Kit also integrated the BRAINSFit system, a collection of programs for registering images with mutual information based metric. BRAINSFit uses the Slicer execution model framework to define the command line arguments and is fully integrated with Slicer using the module discovery capabilities.  Recent developments are in the process of being integrated into the NA-MIC Kit and the Slicer application platform.  The Slicer Catalog allows users to install, uninstall, search, browse, and rank Slicer extensions. This user experience is available from within Slicer and over the web, much like the Android and Apple App Stores. Developers can contribute, update, document, and post screenshots on their modules and receive community feedback.  The analysis infrastructure for Diffusion Weighted MRI (DWI) IO and visualization has been generalized to be used for other time varying acquisitions like multivolume analysis, dynamic contrast enhanced MRI (DCE), and gated cardiac CT.  To cover the use of Qt and newer versions of VTK (both part of the NA-MIC Kit), advanced charting and analytics options have been demonstrated in Slicer4, and will be fleshed out in the coming year.

• Driving Biological Projects

1The Center worked synergistically with the Driving Biological Projects (DBPs) to achieve fundamental advances in shape representation, shape analysis, groupwise registration, diffusion estimation, segmentation and quantification, functional estimation, distortion correction, and clustering.

• Discuss at least 3 Collaborative Research Projects (these may include collaborating R01/R21s or other projects not directly funded by the Center's NCBC grant, but using Center tools or algorithms in a substantial and enabling manner.

Collaboration 1: TBA Collaboration 2: TBA Collaboration 3: TBA

• Brief description of new training and outreach activities conducted during reporting interval (7/1/2012 - 6/30/2013). Provide web-links if available.

Sonia Pujol: Please update and revised. This year NA-MIC hosted XX workshops and courses at national universities and international venues, providing training and exposure to medical researchers in 3D Slicer and other NA-MIC technologies. NA-MIC also xxxx launched the first DTI Tractography Challenge for Neurosurgical Planning at the XXth International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2011) conference in Toronto, Canada, demonstrating its continued commitment to validation. The purpose of the validation effort is to assess the performance of NA-MIC algorithms in a variety of settings.

• Impact of Center (i) on biomedical research and research training and outreach at our institution (BWH) and (i) broader scientific community. Institutional benefits might include, organization of special courses and meetings, attraction of students, and faculty participation. Scientific community benefits may include software released, workshops organized, collaborations established, service performed, technology developed, and technology disseminated through patents, publications, peer-reviewed citations of center collaborations by non-center investigators, and personnel trained.

Provide a Center Summary Table

 Progress made by innovation and image analysis , and scientific CoresResearch CoresThe scientific development is driven by 4 DBPs.  In

addition to activities that sustain the NA-MIC Kit and integrity of the Center’s software infrastructure, NA-MIC has an impressive outreach program that delivers software, data, and innovative science to the broader biomedical community through its publications and training venues. NA-MIC also has instituted a unique validation effort where software developers and end-users participate in hands-on workshops to measure and improve medical image algorithms.

Required elements:

Finally, this year saw the release of Slicer version 4.0 and 4.1 (Slicer4) which represents a

significant advance in capabilities and underlying technologies. The software was released at RSNA 2011 in November. As in past years, a detailed presentation of current work was made at the All Hands Meeting in Salt Lake City, Utah, January 9-13, 2012, and can be viewed in detail on the NA-MIC Wiki [http://wiki.namic. org/Wiki/index.php/ 2012_Winter_Project_Week]. This represents the 8th Annual Progress Report and second year of the second cycle of funding. The report includes Highlights and Impact statements, individual progress reports from the four DBPs (Atrial Fibrillation, Huntington’s Disease, Adaptive Radiotherapy for Head and Neck Cancer, and Traumatic Brain Injury), a science and technology summary from the Computer Science Core (Algorithms, Engineering, and NA-MIC Kit), and a review of Training activities, including the validation effort. The report concludes with a bibliography of 33 peer-reviewed journal articles and 21 peer-reviewed conference reports and the annual recommendations of the External Advisory Board, which met on January 12, 2012 in Salt Lake City, coincident with Winter Project Week.