Difference between revisions of "ProjectWeek200706:vtkITKWrapperForRuleBasedSegmentation"

From NAMIC Wiki
Jump to: navigation, search
 
 
(18 intermediate revisions by 3 users not shown)
Line 1: Line 1:
 
{|
 
{|
 
|[[Image:ProjectWeek-2007.png|thumb|320px|Return to [[2007_Programming/Project_Week_MIT|Project Week Main Page]] ]]
 
|[[Image:ProjectWeek-2007.png|thumb|320px|Return to [[2007_Programming/Project_Week_MIT|Project Week Main Page]] ]]
|[[Image:genuFAp.jpg|thumb|320px|Scatter plot of the original FA data through the genu of the corpus callosum of a normal brain.]]
+
|[[Image:Rulebasedseg1.jpg|thumb|200px|The Slicer2 RuleBasedSegmentation module.]]
|[[Image:genuFA.jpg|thumb|320px|Regression of FA data; solid line represents the mean and dotted lines the standard deviation.]]
 
 
|}
 
|}
  
 
__NOTOC__
 
__NOTOC__
 
===Key Investigators===
 
===Key Investigators===
 +
* Georgia Tech: Tauseef Rehman, John Melonakos
 
* Kitware: Brad Davis
 
* Kitware: Brad Davis
* BWH: Sylvain Bouix
+
* BWH: Nicole Aucoin, Marek Kubicki
  
 
<div style="margin: 20px;">
 
<div style="margin: 20px;">
  
 
<div style="width: 27%; float: left; padding-right: 3%;">
 
<div style="width: 27%; float: left; padding-right: 3%;">
 +
 
<h1>Objective</h1>
 
<h1>Objective</h1>
We are developing methods for analyzing diffusion tensor data along fiber tracts. The goal is to be able to make statistical group comparisons with fiber tracts as a common reference frame for comparison.
+
We are developing rule-based segmentation techniques which speedup the process and improve the accuracy for delineating the DLPFC in brain MRI scans. Our objective is to develop Slicer modules to facilitate clinical use of these techniques.
 
 
  
 +
A functional Slicer2 module has been developed and now needs to be tested and used by our Core 3 partners.  In this project, we are supporting our Core 3 partners by making enhancements to improve the user friendliness of the module.
 
</div>
 
</div>
  
 
<div style="width: 27%; float: left; padding-right: 3%;">
 
<div style="width: 27%; float: left; padding-right: 3%;">
  
<h1>Approaches and Challenges </h1>
+
<h1>Approach, Plan </h1>
 
+
Our approach for segmenting the DLPFC is described in the references below.  The challenge is to make this software user-friendly to enable clinical use of the tool. Our plan for the week is to fix/add several features to improve the user experience.
Our approach for analyzing diffusion tensors is summarized in the IPMI 2007 reference below.  The main challenge to this approach is <foo>.
 
 
</div>
 
</div>
  
Line 29: Line 29:
  
 
<h1>Progress</h1>
 
<h1>Progress</h1>
 +
====June 2007 Project Week====
 +
 +
The Slicer2 RuleBasedSegmentation Module was troubleshooted to locate problems in the interface. The following problems were identified:
 +
 +
1. The sliders to select the ROI bounds were not linked to the respective "Pick" buttons.
 +
 +
2. There was no visual feedback to the user for completion of "ApplyBoundaryConditions" and "ApplyBayesianSegmentation"
 +
commands.
 +
 +
3. The 3D cube to initialize the ROI needed to be deselected after ROI selection.
 +
 +
4. User could not select filter parameters such as number of classes, label number of the ROI mask, and the output filename.
 +
 +
5. The filter output was not getting displayed correctly.
  
 +
6. The procedure could not be run multiple times.
  
====June 2007 Project Week====
 
This is where you put in progress made in Project Week 2007.
 
  
====January 2007 Project Half Week====
+
All the above problems were fixed. The filter parameters (number of classes,label number of ROI mask) were exposed to the user through the GUI and set to common default values. The visual confirmation for the "ApplyBoundaryCondtions" was set to disabling the Bounding Cube and for the "ApplyBayesianSegmentation" the segmentation results were displayed correctly. The help tab and tooltip were updated to incorporate the UI changes. Multiple test runs were performed to confirm consistent behavior.
Software for the fiber tracking and statistical analysis along the tracts has been implemented. The statistical methods for diffusion tensors are implemented as ITK code as part of the [[NA-MIC/Projects/Diffusion_Image_Analysis/DTI_Software_and_Algorithm_Infrastructure|DTI Software Infrastructure]] project. The methods have been validated on a repeated scan of a healthy individual. This work has been published as a conference paper (MICCAI 2005) and a journal version (MEDIA 2006). Our recent IPMI 2007 paper includes a nonparametric regression method for analyzing data along a fiber tract.
+
 
 +
====2005-2007====
 +
This code was developed between 2005-2007.  First is was developed and tested in Matlab. Then the sub-volume creation rules were ported to Slicer2 while the Bayesian segmentation was ported to ITK (see the references below for more detail). Finally, in early 2007, a vtk wrapper of the ITK Bayesian code was developed, thus completing the Slicer2 RuleBasedSegmentation module.
  
 
</div>
 
</div>
Line 44: Line 59:
  
  
===Publications===
+
===References===
* Fletcher, P.T., Tao, R., Jeong, W.-K., Whitaker, R.T., "A Volumetric Approach to Quantifying Region-to-Region White Matter Connectivity in Diffusion Tensor MRI," to appear Information Processing in Medical Imaging (IPMI) 2007.
+
* Ramsey Al-Hakim, James Fallon, Delphine Nain, John Melonakos, and Allen Tannenbaum. A dorsolateral prefrontal cortex semi-automatic segmenter. In SPIE Medical Imaging, 2006.
* Corouge, I., Fletcher, P.T., Joshi, S., Gilmore, J.H., and Gerig, G., "Fiber Tract-Oriented Statistics for Quantitative Diffusion Tensor MRI Analysis," Medical Image Analysis 10 (2006), 786--798.
+
* J. Melonakos, K. Krishnan, and A. Tannenbaum. An ITK Filter for Bayesian Segmentation: itkBayesianClassifierImageFilter. Insight Journal, 2006.
* Corouge, I., Fletcher, P.T., Joshi, S., Gilmore J.H., and Gerig, G., Fiber Tract-Oriented Statistics for Quantitative Diffusion Tensor MRI Analysis, Lecture Notes in Computer Science LNCS, James S. Duncan and Guido Gerig, editors, Springer Verlag, Vol. 3749, Oct. 2005, pp. 131 -- 138
+
* J. Melonakos, R. Al-Hakim, J. Fallon, and A. Tannenbaum. Knowledge-Based Segmentation of Brain MRI Scans Using the Insight Toolkit. Insight Journal, 2005.
* C. Goodlett, I. Corouge, M. Jomier, and G. Gerig, A Quantitative DTI Fiber Tract Analysis Suite, The Insight Journal, vol. ISC/NAMIC/ MICCAI Workshop on Open-Source Software, 2005, Online publication: http://hdl.handle.net/1926/39 .
 

Latest revision as of 14:52, 4 December 2007

Home < ProjectWeek200706:vtkITKWrapperForRuleBasedSegmentation
The Slicer2 RuleBasedSegmentation module.


Key Investigators

  • Georgia Tech: Tauseef Rehman, John Melonakos
  • Kitware: Brad Davis
  • BWH: Nicole Aucoin, Marek Kubicki

Objective

We are developing rule-based segmentation techniques which speedup the process and improve the accuracy for delineating the DLPFC in brain MRI scans. Our objective is to develop Slicer modules to facilitate clinical use of these techniques.

A functional Slicer2 module has been developed and now needs to be tested and used by our Core 3 partners. In this project, we are supporting our Core 3 partners by making enhancements to improve the user friendliness of the module.

Approach, Plan

Our approach for segmenting the DLPFC is described in the references below. The challenge is to make this software user-friendly to enable clinical use of the tool. Our plan for the week is to fix/add several features to improve the user experience.

Progress

June 2007 Project Week

The Slicer2 RuleBasedSegmentation Module was troubleshooted to locate problems in the interface. The following problems were identified:

1. The sliders to select the ROI bounds were not linked to the respective "Pick" buttons.

2. There was no visual feedback to the user for completion of "ApplyBoundaryConditions" and "ApplyBayesianSegmentation" commands.

3. The 3D cube to initialize the ROI needed to be deselected after ROI selection.

4. User could not select filter parameters such as number of classes, label number of the ROI mask, and the output filename.

5. The filter output was not getting displayed correctly.

6. The procedure could not be run multiple times.


All the above problems were fixed. The filter parameters (number of classes,label number of ROI mask) were exposed to the user through the GUI and set to common default values. The visual confirmation for the "ApplyBoundaryCondtions" was set to disabling the Bounding Cube and for the "ApplyBayesianSegmentation" the segmentation results were displayed correctly. The help tab and tooltip were updated to incorporate the UI changes. Multiple test runs were performed to confirm consistent behavior.

2005-2007

This code was developed between 2005-2007. First is was developed and tested in Matlab. Then the sub-volume creation rules were ported to Slicer2 while the Bayesian segmentation was ported to ITK (see the references below for more detail). Finally, in early 2007, a vtk wrapper of the ITK Bayesian code was developed, thus completing the Slicer2 RuleBasedSegmentation module.



References

  • Ramsey Al-Hakim, James Fallon, Delphine Nain, John Melonakos, and Allen Tannenbaum. A dorsolateral prefrontal cortex semi-automatic segmenter. In SPIE Medical Imaging, 2006.
  • J. Melonakos, K. Krishnan, and A. Tannenbaum. An ITK Filter for Bayesian Segmentation: itkBayesianClassifierImageFilter. Insight Journal, 2006.
  • J. Melonakos, R. Al-Hakim, J. Fallon, and A. Tannenbaum. Knowledge-Based Segmentation of Brain MRI Scans Using the Insight Toolkit. Insight Journal, 2005.