Medical Image Analysis Aerial Image Exploitation Automated Cancer
- Slides: 8
Medical Image Analysis Aerial Image Exploitation • Automated Cancer Detection • Automatic Pathology Quantification • Automatic Segmentation • Image-Guided Surgery • Computer Aided Forensics • Change Detection • Event Detection • Object Tracking • Region Classification Image-Guided Manufacturing Automated Video Analysis • Optical Gauging • Optical Inspection • Large-Part Metrology • Automatic Defect Detection • Complex Scene Decomposition • Market Intelligence • Behavior Recognition • 3 D Reconstruction • Ontology-Based Understanding
GE Research NA-MIC Activities • Dart 2 – released in spring 2006 • ITK – orientations, statistical distributions • Slicer 3 – execution model, example modules National Alliance for Medical Image Computing http: //na-mic. org
Dart 2 • Released in spring 2006 • Enterprise environment under investigation • Now used by GE Healthcare to test imaging platforms (XJT, AW) National Alliance for Medical Image Computing http: //na-mic. org
Execution model Algorithms ITK VTK Slicer Modules Slicer 3. 0 User Desktop Scripts of Slicer Mods VTK Apps Using ITK Batch Programs Non-NAMIC Cmd tools LONI Pipeline Birn Grid Data/Compute Internal application framework using a Slicer 3 plugin National Alliance for Medical Image Computing http: //na-mic. org
Slicer 3 Modules • Command. Line. Module – Add – Grayscale. Model. Maker – Affine. Registration – Histogram. Matching – BSpline. Deformable. Registration – Image. Read. Dicom. Write – Checker. Board – Linear. Registration – Confidence. Connected – Median. Image. Filter – Curvature. Anisotropic. Diffusion – Otsu. Threshold. Image. Filter – GEDicom 2 NRRD – Otsu. Threshold. Segmentation – Gradient. Anisotropic. Diffusion – Resample. Volume – Grayscale. Fill. Hole. Image. Filter – Subtract – Grayscale. Grind. Peak. Image. Filter – Voting. Binary. Hole. Filling. Image. Filter National Alliance for Medical Image Computing http: //na-mic. org
Slicer 3 usage at GE • Medical and biological image analysis • Under consideration for contrast agent development • Support research in explosive detection (baggage and cargo) • May be available as an unsupported research tool on GE MR host platforms. National Alliance for Medical Image Computing http: //na-mic. org
NA-MIC Team • Bill Lorensen, MS (RPI), 39 years, Core 2 PI, • software engineering, Slicer 3 architecture • Jim Miller, Ph. D (RPI), 9 years, Site PI, • software testing, segmentation, Slicer 3 execution model • Dan Blezek, Ph. D (Mayo), 7 years, • software testing, MRI, registration, Slicer 3 execution model • Xiaodong Tao, Ph. D (JHU), 2 years, • MRI, DTI, Slicer 3 data model National Alliance for Medical Image Computing http: //na-mic. org
Publications • • D. Blezek, J. Miller. "Atlas Stratification“. Medical Image Computing and Computer Assisted Intervention 2006 (MICCAI). LNCS 4190, pp. 712– 719. X. Tao and J. V. Miller. "A Method for Registering Diffusion Weighted Magnetic Resonance Images“. Medical Image Computing and Computer Assisted Intervention 2006 (MICCAI). LNCS 4191, pp. 594 -602. National Alliance for Medical Image Computing http: //na-mic. org