Difference between revisions of "2010 Summer Project Week Mouse Brain Cortical Thickness Analysis"

From NAMIC Wiki
Jump to: navigation, search
 
(15 intermediate revisions by the same user not shown)
Line 2: Line 2:
 
<gallery>
 
<gallery>
 
Image:PW-MIT2010.png|[[2010_Summer_Project_Week#Projects|Projects List]]
 
Image:PW-MIT2010.png|[[2010_Summer_Project_Week#Projects|Projects List]]
 +
Image:mbrain_scale.png|[[#|Size comparison of mouse and human brains]]
 +
Image:mbrain_segmentation.png|[[#|Segmentation Step]]
 +
Image:mbrain_thickness.png|[[#|Thickness Measurement]]
 +
Image:mbrain_analysis.png|[[#|Statistical Analysis]]
 +
Image:mbrain_module.png|[[Image:mbrain_module.png|Slicer3 Module]]
 
</gallery>
 
</gallery>
  
Line 17: Line 22:
  
 
<h3>Approach, Plan</h3>
 
<h3>Approach, Plan</h3>
−
# Segmentation: The segmentation step provides the neo-cortex separation from other brain structures and thus is a simple preprocessing step for the thickness measurement. Our pipeline uses atlas-based segmentation, such that the structural segmentation as well as source and destination boundaries are manually defined over the atlas by our expert. The source boundary represents the neocortex surface facing the inside of whole brain and the destination boundary represents the surface facing the outside of the brain.
+
* Segmentation
−
# The thickness measurement step: This step computes a Laplacian field from the source boundary toward the destination boundary. By solving forward and backward Laplacian PDE equation, each voxel is assigned two values: forward and backward distance. The forward distance is measured between the voxel and source boundary and the backward distance is measured between the voxel and the destination boundary. The thickness is defined as the sum of these two values on boundary voxels.
+
** Provides the neo-cortex separation from other brain structures
−
# Statistical Analysis: To compare the thickness between different subjects, corresponding points between subjects need to be identified. This is achieved using particle correspondence algorithm minimizing entropy between dynamically moving sample points.
+
** Uses atlas-based segmentation, such that the structural segmentation
 +
** Source and destination boundaries are manually defined over the atlas by our expert.
 +
* The thickness measurement step:  
 +
** Computes a Laplacian field from the source boundary toward the destination boundary.  
 +
** Using measureThicknessFilter by Marc Niethammer
 +
* Statistical Analysis
 +
** Use particle correspondence algorithm minimizing entropy between dynamically moving sample points
 +
** ShapeAnalysis tools by Beatriz Paniagua
 
</div>
 
</div>
  
Line 25: Line 37:
  
 
<h3>Progress</h3>
 
<h3>Progress</h3>
−
Currently all the codes are written in independently running python scripts. Our goal in this project week is to provide fully running python module in Slicer3, from statistical data generation to its visualization. All the script will produce BatchMake script to work with Condor parallel system.
+
* My goal for this project was to convert the pipeline into Slicer3 wizard module
 +
* It was done with need of more improvements (with help of Clement):
 +
** Text editor or CSV editor to process streams of line data
 +
** Result visualization
 +
* I learned here:
 +
** Slicer/QT superbuild: a great playground for better module creation
 +
** CTest: the most impressive one to build robust tools
 +
** GIT: helpful for collaboration and maintenance
 
</div>
 
</div>
 
</div>
 
</div>

Latest revision as of 13:56, 25 June 2010

Home < 2010 Summer Project Week Mouse Brain Cortical Thickness Analysis

Key Investigators

  • UNC: Ipek Oguz, Martin Styner

Objective

Cortical thickness measurement is an important tool in investigating animal models of neurodegenarative diseases. However, the difference between human and rodent models necessitates adaptation of automated image analysis tools to the rodent scale. This project aims to provide tools for mouse and rat brains equivalent to state-of-the-art human brain analysis software.

Approach, Plan

  • Segmentation
    • Provides the neo-cortex separation from other brain structures
    • Uses atlas-based segmentation, such that the structural segmentation
    • Source and destination boundaries are manually defined over the atlas by our expert.
  • The thickness measurement step:
    • Computes a Laplacian field from the source boundary toward the destination boundary.
    • Using measureThicknessFilter by Marc Niethammer
  • Statistical Analysis
    • Use particle correspondence algorithm minimizing entropy between dynamically moving sample points
    • ShapeAnalysis tools by Beatriz Paniagua

Progress

  • My goal for this project was to convert the pipeline into Slicer3 wizard module
  • It was done with need of more improvements (with help of Clement):
    • Text editor or CSV editor to process streams of line data
    • Result visualization
  • I learned here:
    • Slicer/QT superbuild: a great playground for better module creation
    • CTest: the most impressive one to build robust tools
    • GIT: helpful for collaboration and maintenance

Delivery Mechanism

This work will be delivered to the NA-MIC Kit as a (please select the appropriate options by noting YES against them below)

  1. ITK Module
  2. Slicer Module: Thickness measurement filter developed by Marc Niethammer
    1. Built-in
    2. Extension -- python module for whole analysis pipeline
    3. Extension -- loadable
  3. Other (Please specify)

References

Retrieved from "https://www.na-mic.org/w/index.php?title=2010_Summer_Project_Week_Mouse_Brain_Cortical_Thickness_Analysis&oldid=55333"