Difference between revisions of "2011 Summer Project Week Customizing EMSegmenter pipelines for brain lesions"

From NAMIC Wiki
Jump to: navigation, search
(Created page with '__NOTOC__ <gallery> Image:PW-MIT2011.png|Projects List Image:notfound.png|Interesting picture to be added... </gallery> '''Full Title of P…')
 
m (Text replacement - "http://www.slicer.org/slicerWiki/index.php/" to "https://www.slicer.org/wiki/")
 
(10 intermediate revisions by one other user not shown)
Line 2: Line 2:
 
<gallery>
 
<gallery>
 
Image:PW-MIT2011.png|[[2011_Summer_Project_Week#Projects|Projects List]]
 
Image:PW-MIT2011.png|[[2011_Summer_Project_Week#Projects|Projects List]]
Image:notfound.png|Interesting picture to be added...
 
 
</gallery>
 
</gallery>
  
'''Full Title of Project'''
+
'''Customizing EMSegmenter pipelines for brain lesions'''
  
 
==Key Investigators==
 
==Key Investigators==
 
* University of Pennsylvania: Dominique Belhachemi
 
* University of Pennsylvania: Dominique Belhachemi
 +
* University of Pennsylvania: Kilian Pohl
 
* Brigham and Women's Hospital: Alex Zaitsev
 
* Brigham and Women's Hospital: Alex Zaitsev
  
Line 18: Line 18:
  
 
<!-- What you plan to achieve during project week goes here -->
 
<!-- What you plan to achieve during project week goes here -->
 
+
The goal is to segment white matter lesions based on T1 and FLAIR brain scans.
  
 
</div>
 
</div>
Line 27: Line 27:
  
 
<!-- What specific things you will be doing during the week: writing code?  trying code on specific datasets?  talking with other attendees?  All of the above? -->
 
<!-- What specific things you will be doing during the week: writing code?  trying code on specific datasets?  talking with other attendees?  All of the above? -->
 
+
We want to use the EMSegmenter module for this project. The plan is to create a two channel pipeline which includes pre-processing (e.g. registration) and applying the EM algorithm to perform the segmentation.
  
 
</div>
 
</div>
Line 36: Line 36:
  
 
<!-- Fill this out before Friday's summary presentations - list what you did and how well it worked. -->
 
<!-- Fill this out before Friday's summary presentations - list what you did and how well it worked. -->
 +
We created a 2 channnel(T1+FLAIR) pipeline for this project. Intensity values of the lesions are calculated automatic but have to manually updated to get a better result.
 +
<b>(see segmentation screenshot)</b>
  
 +
Open: With this approach we detect a lot of false positives outside the white matter region.
  
 
</div>
 
</div>
Line 44: Line 47:
  
 
==References==
 
==References==
 +
 +
<gallery perrow=1: widths=1250px : heights=400px>
 +
Image:EMS_LesionExp_Segmentation.png | Segmentation result
 +
Image:EMS LesionExp FLAIR.png | FLAIR scan
 +
Image:EMS LesionExp T1.png | T1 scan
 +
</gallery>
 +
 +
https://www.slicer.org/wiki/EMSegmenter-Tasks <br>
 +
  
 
<!-- Add links here to background material: where can we download the data?  where is the source code repository?  what papers or web sites will help someone quickly learn what the outstanding issues are? -->
 
<!-- Add links here to background material: where can we download the data?  where is the source code repository?  what papers or web sites will help someone quickly learn what the outstanding issues are? -->
Line 51: Line 63:
 
<!-- List here how you plan to deliver your results to user communities -->
 
<!-- List here how you plan to deliver your results to user communities -->
  
This work will be delivered to the NAMIC Kit as a
+
This work will be delivered to the NAMIC Kit as part of the EMSegmenter module.
 
 
#NITRIC distribution
 
#Slicer Module
 
##Built-in:
 
##Extension -- commandline: 
 
##Extension -- loadable:
 

Latest revision as of 18:07, 10 July 2017

Home < 2011 Summer Project Week Customizing EMSegmenter pipelines for brain lesions

Customizing EMSegmenter pipelines for brain lesions

Key Investigators

  • University of Pennsylvania: Dominique Belhachemi
  • University of Pennsylvania: Kilian Pohl
  • Brigham and Women's Hospital: Alex Zaitsev


Objective

The goal is to segment white matter lesions based on T1 and FLAIR brain scans.

Approach, Plan

We want to use the EMSegmenter module for this project. The plan is to create a two channel pipeline which includes pre-processing (e.g. registration) and applying the EM algorithm to perform the segmentation.

Progress

We created a 2 channnel(T1+FLAIR) pipeline for this project. Intensity values of the lesions are calculated automatic but have to manually updated to get a better result. (see segmentation screenshot)

Open: With this approach we detect a lot of false positives outside the white matter region.

References

https://www.slicer.org/wiki/EMSegmenter-Tasks


Delivery Mechanism

This work will be delivered to the NAMIC Kit as part of the EMSegmenter module.