TubeTK/Events/2010.07.26: Difference between revisions

From KitwarePublic
< TubeTK‎ | Events
Jump to navigationJump to search
No edit summary
 
(3 intermediate revisions by 2 users not shown)
Line 101: Line 101:
* Primary goal: ultrasound image processing
* Primary goal: ultrasound image processing
* Accomplished
* Accomplished
** Correct a bug in itkAnisotropicHybridDiffusionImageFilter, and prepared figures for Andinet's Insight Journal paper.
** Correct a bug in itkAnisotropicHybridDiffusionImageFilter
** prepared figures for Andinet's Insight Journal paper.
** Improve code coverage for uncovered filters under Application/CLI/** Increase coverage of TubeTK
** Improve code coverage for uncovered filters under Application/CLI/** Increase coverage of TubeTK
* Near term (August 2)
* Near term (August 2)
Line 110: Line 111:
**** http://www.insight-journal.org/midas/item/view/117
**** http://www.insight-journal.org/midas/item/view/117
* Medium term (August 9)
* Medium term (August 9)
[[Category:TubeTK Events and Meetings|2010.07.26]]

Latest revision as of 18:45, 26 July 2013

Topics

  • Dashboards auto update?
    • tubetk/CMake/DashboardScripts tubetk/CMake/CTestCustom...
  • Batch Processing
    • Python vs BatchMake vs Any
  • Slicer
    • TubeNet Viewer
    • Slicer load .tre
    • Slicer Loadable Module
  • Registration
    • Speckle in ultrasound registration
    • Model-based deformation field interpolation
    • Fluid deformation (Marc)
    • Registration metrics based on ultrasound probe orientation
  • Segmentation
    • Unit test VTree code
      • 3D
      • 2D
    • Automated vessel tree extraction
      • Using spatial prior
      • Seed selection
    • Automated distinguishing arteries from veins based on spatial prior
  • Atlas formation
    • Retinal data
      • Email from UIowa (still waiting)
    • Brain data
      • Vessel extractions from Liz
    • Port pipeline to VTree

Status

Patrick

  • Primary goal: Bump and dent identification on IC images
  • Accomplishments
    • Traveled to SSRL to view the acquisition and meet with Greg and Mike.
    • Modified the GenerateFeatures application to handle the input of arbitrary feature images
    • Using the previous features and Casey's new patch-based features, was able to achieve 95% pixel level accuracy in Weka and 23/25 defects found with 0 false positives in image space (after morphology).
    • Explore new features
      • evaluate a variety of standard deviations for intensity and ridge computations
  • Near Term (Aug 2)
    • Receiving code to simulate the tomography directly on GDS Layers
    • Compute dot-product between line (hessian) tangent and normal directions in ES and GDS images
    • Subselect features
    • Product: ~ 5 slides / report to USC illustrating path chosen, strengths, and weaknesses.
      • Real-world tests / workflow
      • Does a trained classifier work on other layers?
      • Does a trained classifier work on other acquisitions?
        • i.e., do we need to insert modifications for training on every slice / acquisition / ?
        • Normalizing for inter-acquisition (or inter-slice) variations?
    • Work with new collaborator at Kitware.
  • Medium term (August 9)
    • Delivery and education
    • Can we get better in simulation?
    • chip-to-chip matching
    • connectivity analysis

Casey

  • Primary goal: Compare populations of vascular networks
  • Near term (Aug 2)
    • Feature extraction
      • Patch-based features (max, median, quantiles)
      • Scales / neighborhood
    • Feature research
      • Width estimate
        • Not the same as Gaussian blur
      • New patch features
      • Location of local max in ridgeness
      • Subsample centerlines
      • Optimal match filter
    • Classifiers
      • Comparison
      • Implementation for transfer to USC
      • Hierarchy (Good/bad. If bad, then add/sub.)
  • Medium term (Aug 9)
    • Connectivity analysis
    • chip-to-chip comparison

Andinet

  • Primary goal: Data from Duke for BWH
  • Accomplishments
    • Data from Duke
      • Contacted several folks to gather information ( we had questions regarding the machine at Duke) and check status.
        • BrainLab: Contacted Pratik. He is working on getting us VV license
        • Duke: Contacted Tanya and learned that SD-5000 ultrasound machine is not integrated with the BrainLab system. The machine is used to acquire ultrasound data independently.
        • Aloka: Contacted John Walsh at Aloka and learned that SD-5000 is very old model and doesn't come with research interface
      • Based upon the conversation I had with you, may be we should probably look into saving data off the BrainLab system itself not bother with the SD-5000.
    • Write IJ article
      • Made progress writing the IJ article. Hua helped me a lot generating results using synthetic data. We have now a solid outline and some write up in most of the sections
      • You can also access the tex, bib, etc files in my Work directory: Work/Andinet/TensorIJ
  • Near term (August 2)
    • IJ Article
      • Move article to TubeTK/Documentation/2010.TensorIJ
      • Add more texts, figures and results and clean it up more.
      • Put together self-contained source code tree containing the classes that we will submit with this paper.
      • Refer to TubeTK
      • Cite grant proposal in article
    • Install VV at Duke
  • Medium term (August 9)

Hua

  • Primary goal: ultrasound image processing
  • Accomplished
    • Correct a bug in itkAnisotropicHybridDiffusionImageFilter
    • prepared figures for Andinet's Insight Journal paper.
    • Improve code coverage for uncovered filters under Application/CLI/** Increase coverage of TubeTK
  • Near term (August 2)
  • Medium term (August 9)