2008 Engineering review at Utah

From NAMIC Wiki
Revision as of 18:27, 3 April 2008 by Taox (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search
Home < 2008 Engineering review at Utah

The date is now set at: March 24-26, 2008. Begin: Morning of Monday, March 24. End: Noon on Wednesday, March 26.



Suggested Hotel: Hotel Monaco.

Engineering Attendees: Jim Miller, Xiaodong Tao, Steve Pieper, Will Schroeder, Marco Ruiz (Full list of attendees in the end of page)

Place: SCI Institute, 72 South Campus Drive, Warnock Engineering Building WEB (Directions), 3rd floor Jones conference room 3840 (elevator to 3rd floor, room is to the left).

Proposed discussion items:

Discussion of engineering issues related to the 3 major Utah projects DTI Analysis, Shape Analysis, Segmentation including Pathology. Main aim: Discuss and coordinate Slicer-3 integration efforts of the Utah R&D efforts, in particular w.r.t. pipelines or sequences of processing steps that are necessary in most medical image analysis procedures.


  • Monday 24
      • 9am: Goals, aims and organization of the meeting (Ross / Guido / Will / Jim / Steve)
      • Introductions to kick off discussions:
      • Slicer3 Plugins Presentation [ppt]
      • Segmentation (Marcel) [ppt]
    • 12pm to 1.30pm: Lunch / Possibility to attend SCI distinguished lecture by Andy van Dam titled When is the Pen Mightier Than the Keyboard? [1]
    • 2pm: DTI analysis pipeline (Casey / Tom / Guido) [pdf]
      • Shape analysis (Josh / Tom / Ross )
      • Summary of open issues to be further discussed
    • Dinner (tbd)
  • Wednesday 26
    • 9am Discussion of action items:
      • Strategic plan for software integration (who decides on what, coherence, interoperability)
      • Which modules will be integrated by whom (paired Core-1 / Core-2 effort)
      • Detailed plan with deliverables and timeline
      • Noon: adjourn


Project-specific details to be discussed

  • DTI processing modules/pipeline
    • Infrastructure for DWI/DTI data types
      • How to handle coordinate frames
    • Preprocessing pipeline for DWI->DTI->analysis
      • Filtering
      • Registration
      • Estimation
      • White mater mask
      • User interfaces/interaction
    • Analysis of DTI data
      • Tractography
      • Atlases
      • Derived quantities
      • Hypothesis testing
  • Shape analysis pipeline
    • How to organize data
    • How to break pipeline in to steps
      • Segmentation, preprocessing, parameterization, visualization
    • How to accommodate new modules/techniques
  • Segmentation pipeline
    • Typical sequence of processing steps / procedures
    • Prototyping optimal sequence/parameters via GUI
    • Batch job for large number of image data
    • Segmentation of healthy/pathology
    • What do different procedures have in common (using shared modules)
  • The general problem of registration and atlases
    • Status of spline-based methods
    • More general concept of providing a set of procedures
    • Properties: linear/nonlinear, low vs. highdimensional, invertability, diffeomorphic
    • How to incorporate registration/atlases/atlas building in to slicer3
  • General architecture issues
    • How to provide coherent user interfaces for specific tasks
      • Pipelines for specific data or analyses
      • E.g. see Siemens workstation software with modules/pages
    • Command-line vs built in modules
      • Policies for supporting one versus the other


  • Attendees
    • Ross Whitaker
    • Tom Fletcher
    • Ran Tao
    • Josh Cates
    • Guido Gerig
    • Marcel Prastawa
    • Sylvain Gouttard
    • Casey Goodlett
    • C-F Westin
    • Will Schroeder
    • Steve Pieper
    • Jim Miller
    • Xiaodong Tao
    • Marco Ruiz