1. NA-MIC National Alliance for Medical Image Computing http://na-mic.org Core 1b – Engineering Computational Platform Jim Miller GE Research

2. National Alliance for Medical Image Computing http://na-mic.org 1 4 2 5 3 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

3. National Alliance for Medical Image Computing http://na-mic.org Computational platform The objective of the Engineering component of Core 1 is to provide software tools and software development processes to deploy innovative technology to clinical researchers, support the scientific algorithm innovation of the Algorithm scientists, and to foster a community to produce high quality software. Computing architecture Desktop Grid computing Stream computing (GPGPU) Cloud computing Computing architecture Desktop Grid computing Stream computing (GPGPU) Cloud computing Analysis platform Feature libraries Multivariate clustering Machine learning Statistical inference Regression analysis Information visualization Informatics Analysis platform Feature libraries Multivariate clustering Machine learning Statistical inference Regression analysis Information visualization Informatics DBP needs Pathology segmentation Longitudinal analysis Patient specific analysis Population analysis DBP needs Pathology segmentation Longitudinal analysis Patient specific analysis Population analysis Algorithm needs Exploratory methods Longitudinal frameworks Interactive frameworks Statistics Machine learning Algorithm needs Exploratory methods Longitudinal frameworks Interactive frameworks Statistics Machine learning

4. National Alliance for Medical Image Computing http://na-mic.org Computing architecture First funding period Second funding period

5. National Alliance for Medical Image Computing http://na-mic.org Analysis platform First funding period Second funding period LIBSVM

6. National Alliance for Medical Image Computing http://na-mic.org Enhanced interactive Python shell Architecture for interactive parallel computing –Same API for computing on multiple cores or on multiple computers –Build an ad hoc cluster on one machine, across machines in your lab, or in the cloud

7. National Alliance for Medical Image Computing http://na-mic.org Scalable Stream Computing OpenCL Runs on iPhones, Laptops, Desktops, Supercomputers and Clouds –Nvidia Tesla 2050 (Fermi) 3G, 1Tflops float $2,499 to buy $2.10 by the hour on AWS Although OpenCL language is common, code must adapt to memory and compute core availability on different hardware NA-MIC will provide development patterns workable across scales

8. National Alliance for Medical Image Computing http://na-mic.org Simplify ITK without templates Accelerate Refactor Smaller Consistent Documented DICOM GPGPU methods Interactivity DICOM networking Modernized Easier to write ITK based Slicer modules ITK v4 is an ARRA funded contract from the National Library of Medicine

9. National Alliance for Medical Image Computing http://na-mic.org Infovis and informatics Figures from Titan presentation at VisWeek 2008 Database vtkSQLDatabas e vtkTable Datastructures Datasources Visualizations and Algorithms vtkSQLQuery

10. National Alliance for Medical Image Computing http://na-mic.org Exploratory image analysis Tying it all together Feature libraries –Intensity, texture, shape, scale Statistics –Regression analysis –Cluster analysis Machine learning Infovis and informatics –Plots, graph layouts –Data selection Interactive frameworks –Fast, flexible –ROIs in image or data space

11. National Alliance for Medical Image Computing http://na-mic.org 1 4 2 5 3 Core 1b – Engineering Computational platform 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

12. National Alliance for Medical Image Computing http://na-mic.org

13. Interactive Segmentation Framework GrowCut Slicer Editor Effect Multiple paint tool interactions Simultaneous viewing of user inputs and segmentation Edit segmentation with additional gestures Framework for other interactive methods User painted gestures Segmentation overlaid with gestures Additional User painted gestures Final Segmentation overlaid with gestures

14. National Alliance for Medical Image Computing http://na-mic.org