NA-MIC National Alliance for Medical Image Computing Core 1b – Engineering Data Management Stephen R. Aylward Kitware, Inc.
National Alliance for Medical Image Computing Core 1b – Engineering 5 Aims / 5 Platforms Architecture – tools, operating paradigms, reporting mechanisms, integration points End-user platform – interactive methods and information visualization for longitudinal analysis, exploratory data analysis, and translational research Computational platform – stream processing, cloud computing, statistical analysis, informatics, machine learning Data management – non-imaging and derived data, DICOM and cloud services Software engineering and software quality – navigable timeline for revision control, build, test, documentation and release
National Alliance for Medical Image Computing Data Management Key, bi-directional link between clinicians and researchers Foundational to transitioning algorithms from research to clinical practice Essential to quantitative comparison of algorithms January 2011, AHM –11 projects emphasizing data management Registration Case Library MRML XML Schema XNAT for Non-Image Data DICOM-RT
National Alliance for Medical Image Computing Data Management Driving Biological Problems –Data Management challenges: Transfer, Archive, Clinical Interpretation, and Reporting (DICOM) Algorithms Team –Aim 1: Statistical models of anatomy and pathology –Aim 4: Longitudinal and Time-Series Analysis –Milestones: Validation, Optimization, Atlases, Priors
National Alliance for Medical Image Computing Success = Challenge
National Alliance for Medical Image Computing Data Management Technology VTK: InfoVis
National Alliance for Medical Image Computing The Extensible Neuroimaging Archive Toolkit (XNAT) –Management and exploration of neuroimaging and related data. –Secure database backend –Rich web-based user interface –XNAT 1.4 –XNAT Desktop (XND) What’s next: –XNAT for Non-Image Data –Slicer4: FetchMI
National Alliance for Medical Image Computing MIDAS in-use –Insight Journal –NA-MIC: PubDB –RIRE Project –Optical Society of America: Interactive Scientific Publication –Give-a-scan: Patient-created lung CT database What’s Next –Slicer Testing (BrainsFit Data: Uiowa, Kitware) –Server-Side Processing for Algorithm Validation ITK A2D2: Score and Score++ (Kitware, Utah, BWH) NIH R01: CoValic (Kitware, Utah) NIH SBIR: High-Throughput Small Animal Imaging (NAMIC-Kit, Kitware, UNC) –Slicer4: FetchMI
National Alliance for Medical Image Computing 3DSlicer CTK Common Toolkit VTK, QT, DCMTK, DICOM Services (XIP) ITKv4 DICOM Objects and Networking GDCM, DCMTK, Validation DCMTK 286,000 Lines 76 Person Years $4.2M (Ohloh)
National Alliance for Medical Image Computing Non-Image Data: MRML & Informatics Figures from Titan presentation at VisWeek 2008 Database vtkSQLDatabas e vtkTable Datastructures Datasources Visualizations and Algorithms vtkSQLQuery
National Alliance for Medical Image Computing Data Management Technology VTK: InfoVis
National Alliance for Medical Image Computing Core 1b – Engineering 5 Aims / 5 Platforms Architecture – tools, operating paradigms, reporting mechanisms, integration points End-user platform – interactive methods and information visualization for longitudinal analysis, exploratory data analysis, and translational research Computational platform – stream processing, cloud computing, statistical analysis, informatics, machine learning Data management – non-imaging and derived data, DICOM and cloud services Software engineering and software quality – navigable timeline for revision control, build, test, documentation and release