Difference between revisions of "2012 Winter Project Week:TBIValidation"

From NAMIC Wiki
Jump to: navigation, search
 
(17 intermediate revisions by the same user not shown)
Line 3: Line 3:
 
Image:PW-SLC2012.png|[[2012_Winter_Project_Week#Projects|Projects List]]
 
Image:PW-SLC2012.png|[[2012_Winter_Project_Week#Projects|Projects List]]
 
</gallery>
 
</gallery>
 +
 +
<center>
 +
<gallery widths=270px heights=220px perrow=2 caption="Visualization and validation of segmentation ">
 +
File:DefVis DetJacobian TBISeg.png‎|Visualization of the deformation field via determinant of Jacobian
 +
File:DefVis VecMag TBISeg.png‎|Visualization of the deformation field via vector magnitude
 +
File:Manual_seg_2D_X_view.jpg|Manual segmentation - ground truth
 +
File:Semiauto_seg_2D_X_view.jpg|Result of the segmentation algorithm
 +
</gallery>
 +
</center>
 +
  
 
==Key Investigators==
 
==Key Investigators==
Marcel Prastawa, Bo Wang, Andrei Irimia, Micah Chambers, Jack van Horn, Guido Gerig
+
 
 +
*Utah: Bo Wang, Marcel Prastawa, and Guido Gerig
 +
*UCLA: Andrei Irimia, Micah Chambers, Jack Van Horn and Paul M. Vespa
 +
*Kitware: Danielle Pace, Stephen Aylward
  
 
<div style="margin: 20px;">
 
<div style="margin: 20px;">
Line 11: Line 24:
  
 
<h3>Objective</h3>
 
<h3>Objective</h3>
* Defining manual segmentation protocol for MR images of TBI
+
* Define manual segmentation protocol for MR images of TBI
* Comparing the independent segmentation and joint segmentation using Dice test
+
* Validate the segmentation of the algorithm
* Visualizing volume changes, deformation field
+
* Visualize volume changes, deformation field
  
 
</div>
 
</div>
Line 21: Line 34:
 
<h3>Approach, Plan</h3>
 
<h3>Approach, Plan</h3>
 
Our plan for the project week:
 
Our plan for the project week:
* Working with collaborators to define manual segmentation protocol
+
* Work with collaborators to define manual segmentation protocol
* Visualizing volume changes and deformation field according to clinicians' requirements.
+
* Compare the independent segmentation and joint segmentation
 +
* Visualize volume changes and deformation field according to clinicians' requirements.
  
 
</div>
 
</div>
Line 29: Line 43:
  
 
<h3>Progress</h3>
 
<h3>Progress</h3>
To be added.
+
* Summarized the rule the clinician used to do manual segmentation
 +
* Compared the result of the algorithm with clinician's manual segmentation
 +
* Explored different ways to visualize the volume changes/deformation field, we tried the following methods.
 +
** Jacobian determinant
 +
** Vector magnitude
  
 
</div>
 
</div>

Latest revision as of 15:44, 13 January 2012

Home < 2012 Winter Project Week:TBIValidation


Key Investigators

  • Utah: Bo Wang, Marcel Prastawa, and Guido Gerig
  • UCLA: Andrei Irimia, Micah Chambers, Jack Van Horn and Paul M. Vespa
  • Kitware: Danielle Pace, Stephen Aylward

Objective

  • Define manual segmentation protocol for MR images of TBI
  • Validate the segmentation of the algorithm
  • Visualize volume changes, deformation field

Approach, Plan

Our plan for the project week:

  • Work with collaborators to define manual segmentation protocol
  • Compare the independent segmentation and joint segmentation
  • Visualize volume changes and deformation field according to clinicians' requirements.

Progress

  • Summarized the rule the clinician used to do manual segmentation
  • Compared the result of the algorithm with clinician's manual segmentation
  • Explored different ways to visualize the volume changes/deformation field, we tried the following methods.
    • Jacobian determinant
    • Vector magnitude

References

  • Bo Wang, Marcel Prastawa, Andrei Irimia, Micah C. Chambers, Paul M. Vespa, John D. van Horn, Guido Gerig, A Patient-Specific Segmentation Framework for Longitudinal MR Images of Traumatic Brain Injury, SPIE Medical Imaging 2012.
  • Andrei Irimia, Micah C. Chambers, Jeffry R. Alger, Maria Filippou, Marcel W. Prastawa, Bo Wang, David A. Hovda, Guido Gerig, Arthur W. Toga, Ron Kikinis, Paul M. Vespa, John D. van Horn (2011) Comparison of acute and chronic traumatic brain injury using semi-automatic multimodal segmentation of MR volumes. Journal of Neurotrauma