Difference between revisions of "2014 Summer Project Week:Slicer Murin Shape Analysis"

From NAMIC Wiki
Jump to: navigation, search
 
(4 intermediate revisions by the same user not shown)
Line 19: Line 19:
 
<li> Modalities: <b> Optical Projection Tomography</b> [[File:Sample OPT Mouse embryo.zip]]    <br>
 
<li> Modalities: <b> Optical Projection Tomography</b> [[File:Sample OPT Mouse embryo.zip]]    <br>
 
  <B> Micro Computed Tomography </b> [[File:Stained registered sample mCT.zip]] <br>
 
  <B> Micro Computed Tomography </b> [[File:Stained registered sample mCT.zip]] <br>
[[Image:OPT Crossection.PNG|100px]]
+
[[Image:OPT Crossection.PNG|100px]] [[Image:Registered mCT scans.png|100px]]
 
<li> Shape Analysis  
 
<li> Shape Analysis  
 
<ul>
 
<ul>
Line 31: Line 31:
 
<li> Goals for Project Week:  
 
<li> Goals for Project Week:  
 
<ul>
 
<ul>
<li>Meet the community and learn
+
<li> Meet the community and learn from them!
<li>Implement the landmark based Procrustes Analysis in Slicer
+
<li> Raise awareness about issues in using Slicer in high-resolution small animal imaging.
 +
<li> Implement the landmark based Procrustes Analysis in Slicer
 
</ul>
 
</ul>
  
Line 43: Line 44:
 
<ul>
 
<ul>
 
<li>Implement GPA/PCA shape analysis in python</li>
 
<li>Implement GPA/PCA shape analysis in python</li>
<li>Visualize the deformation of a reference volume along the principle components using thin plate splines([[Image:TPS.png|400px]])</li>
+
<li>Provide an interactive tool to visualize the decomposition along the principle components of shape variation using thin plate splines.<br>
 +
[[Image:TPS.png|400px]])</li>
  
<li> <b>Ability to create semi-landmarks to increase spatial coverage. </b>
+
<li> <b>Ability to create semi-landmarks to increase coverage in regions where anatomial landmarks are sparse. </b>
 
<ul>
 
<ul>
 
<li><b>User will  a uniformly sampled point cloud by entering the number of semi-landmarks. Existing “hard” landmarks will be used for their distribution. This will serve as the template to be transferred to all remaining volumes (atlas)
 
<li><b>User will  a uniformly sampled point cloud by entering the number of semi-landmarks. Existing “hard” landmarks will be used for their distribution. This will serve as the template to be transferred to all remaining volumes (atlas)
 
<li>The template will be transferred to a new surface. Existing “hard” landmarks will allow for correspondence. The transferred points will then be moved along the surface of the volume by optimizing the bending energy function.
 
<li>The template will be transferred to a new surface. Existing “hard” landmarks will allow for correspondence. The transferred points will then be moved along the surface of the volume by optimizing the bending energy function.
 
<li>The coordinates of the slid landmarks will be saved into a new fiducial list, from which the GPA analysis can be conducted.</b>
 
<li>The coordinates of the slid landmarks will be saved into a new fiducial list, from which the GPA analysis can be conducted.</b>
 +
<li>These should be accomplished on volume, not surface meshes derived from scans.
 
</ul>
 
</ul>
 
</ul>  
 
</ul>  
Line 56: Line 59:
 
<h3>Progress</h3>
 
<h3>Progress</h3>
 
<ul>
 
<ul>
<li> Generalized Procrustes Alignment</li>
+
<li> Generalized Procrustes Alignment (implemented)</li>
<li> Principal Component and Singular Value Decomposition of the Procrustes aligned coordinates</li>
+
<li> Principal Component and Singular Value Decomposition of the Procrustes aligned coordinates (implemented)</li>
<li> Thin Plate Spline visualization of the shape variables from PCA and/or SVD (by either morphing a reference volume along the shape variable, or visualizing the TPS grid using Transformation Visualizer module).</li>
+
<li> Thin Plate Spline visualization of the shape variables from PCA and/or SVD (implemented).</li>
 +
<li> Transfering and sliding a template of semi-landmarks to the target volume (in progress)
  
 
</ul>
 
</ul>

Latest revision as of 15:46, 24 June 2014

Home < 2014 Summer Project Week:Slicer Murin Shape Analysis

Key Investigators

  • Murat Maga (Seattle Children's Research Institute & University of Washington Dept. of Pediatrics)
  • Ryan Young (Seattle Children's Research Institute)


Project Description

  • Research: Changes in development due to Fetal Alcohol Exposure and how this affects the development of the craniofacial complex.
    • Face is the major diagnostic feature to identify
    • Brain and the CNS are affected primarily.
    • What's the earliest time we begin to detect changes in the face?
    • How does the brain volumes (and gross morphology) relate to changes in the face?
  • Modalities: Optical Projection Tomography File:Sample OPT Mouse embryo.zip
    Micro Computed Tomography File:Stained registered sample mCT.zip
    OPT Crossection.PNG Registered mCT scans.png
  • Shape Analysis
    • We use landmarks to identify the anatomical regions across our developmental series of fetal samples.
    • We want to be able segment brains from about 600 volumes and do a coupled analysis of facial and brain phenotypes.
      Fetus variation picture.PNG
  • Challenges in Slicer with our datasets due to small voxel sizes (6-35 micron). Specifically visualization, recording coordinates of anatomical landmarks, segmentation and registration. (File:Project week question.txt)
  • Goals for Project Week:
    • Meet the community and learn from them!
    • Raise awareness about issues in using Slicer in high-resolution small animal imaging.
    • Implement the landmark based Procrustes Analysis in Slicer


    Objective

    • Create a GPA/PCA shape analysis and visualization module for Slicer.
  • Approach, Plan

    • Implement GPA/PCA shape analysis in python
    • Provide an interactive tool to visualize the decomposition along the principle components of shape variation using thin plate splines.
      TPS.png)
    • Ability to create semi-landmarks to increase coverage in regions where anatomial landmarks are sparse.
      • User will a uniformly sampled point cloud by entering the number of semi-landmarks. Existing “hard” landmarks will be used for their distribution. This will serve as the template to be transferred to all remaining volumes (atlas)
      • The template will be transferred to a new surface. Existing “hard” landmarks will allow for correspondence. The transferred points will then be moved along the surface of the volume by optimizing the bending energy function.
      • The coordinates of the slid landmarks will be saved into a new fiducial list, from which the GPA analysis can be conducted.
      • These should be accomplished on volume, not surface meshes derived from scans.

    Progress

    • Generalized Procrustes Alignment (implemented)
    • Principal Component and Singular Value Decomposition of the Procrustes aligned coordinates (implemented)
    • Thin Plate Spline visualization of the shape variables from PCA and/or SVD (implemented).
    • Transfering and sliding a template of semi-landmarks to the target volume (in progress)

    File:PowerPoint.pdf