Difference between revisions of "2017 Winter Project Week/MeningiomaSegmentation"

From NAMIC Wiki
Jump to: navigation, search
(Added examples)
Line 8: Line 8:
 
==Key Investigators==
 
==Key Investigators==
 
<!-- Add a bulleted list of investigators and their institutions here -->
 
<!-- Add a bulleted list of investigators and their institutions here -->
 +
* Jakub Kaczmarzyk, MIT
 
* Satrajit Ghosh, MIT
 
* Satrajit Ghosh, MIT
 
* Omar Arnaout, Brigham and Women's Hospital
 
* Omar Arnaout, Brigham and Women's Hospital
Line 39: Line 40:
 
* Get more data, and potentially train a neural network.
 
* Get more data, and potentially train a neural network.
 
|}
 
|}
 +
 +
 +
==Examples==
 +
{| class="wikitable"
 +
|Sometimes, ANTs failed to remove parts of the skull close to the tumor or wrongly removed part of the brain.
 +
[[File:Case_001_ants_brain_failure.png|thumbnail|ANTs brain extraction failure]]
 +
But other times, ANTs extracted the brain very well.
 +
[[File:Case_052_ants_brain.png|thumbnail|Ants brain extraction success]]
 +
Automatic segmentation with ANTs usually classified as CSF or it classified different parts of the tumor as different types of tissue.
 +
[[File:Case_052_ants_brain_seg.png|thumbnail|Brain segmentation with ANTs]]
 +
Semi-automatic segmentation with Slicer was relatively successful. Sometimes the segmentation would bleed outside of the tumor into voxels with similar intensities.
 +
[[File:Case_052_2_slicer_seg.png|thumbnail|Semi-automatic segmentation with Slicer]]
 +
|
 +
We tried FSL's FAST with different numbers of classes. None of these methods could identify the entire tumor mass as one type of tissue in this scan.
 +
[[File:Case_052_fast_3classes.png|thumbnail|FAST brain segmentation with 3 classes]]
 +
[[File:Case_052_fast_4classes.png|thumbnail|FAST brain segmentation with 4 classes]]
 +
[[File:Case_052_fast_5classes.png|thumbnail|FAST brain segmentation with 5 classes]]
 +
|}
 +
 +
 +
 +
  
 
==Background and References==
 
==Background and References==
 
<!-- Use this space for information that may help people better understand your project, like links to papers, source code, or data -->
 
<!-- Use this space for information that may help people better understand your project, like links to papers, source code, or data -->
 
MR images of meningiomas that will be used in this project are available at [http://openneu.ro/metasearch/ OpenNeu.ro].
 
MR images of meningiomas that will be used in this project are available at [http://openneu.ro/metasearch/ OpenNeu.ro].

Revision as of 21:02, 13 January 2017

Home < 2017 Winter Project Week < MeningiomaSegmentation

Key Investigators

  • Jakub Kaczmarzyk, MIT
  • Satrajit Ghosh, MIT
  • Omar Arnaout, Brigham and Women's Hospital

Project Description

Objective Approach and Plan Progress and Next Steps
  • Segment meningiomas in structural MR images.
  • Assess state of brain after surgical removal of meningioma.
  • Evaluate performance of existing segmentation methods.

Progress

  • Segmented with ANTs and FSL.
  • Learned about Slicer segmentation tools, and segmented semi-automatically.
  • Put in contact with people who have segmented meningiomas.

Next steps

  • Continue learning about what has been done in the past.
  • Improve brain-extraction.
  • Continue testing existing segmentation methods.
  • Try Slicer in a Nipype workflow.
  • Apply feature engineering techniques, like manifold learning.
  • Get more data, and potentially train a neural network.


Examples

Sometimes, ANTs failed to remove parts of the skull close to the tumor or wrongly removed part of the brain.
ANTs brain extraction failure

But other times, ANTs extracted the brain very well.

Ants brain extraction success

Automatic segmentation with ANTs usually classified as CSF or it classified different parts of the tumor as different types of tissue.

Brain segmentation with ANTs

Semi-automatic segmentation with Slicer was relatively successful. Sometimes the segmentation would bleed outside of the tumor into voxels with similar intensities.

Semi-automatic segmentation with Slicer

We tried FSL's FAST with different numbers of classes. None of these methods could identify the entire tumor mass as one type of tissue in this scan.

FAST brain segmentation with 3 classes
FAST brain segmentation with 4 classes
FAST brain segmentation with 5 classes



Background and References

MR images of meningiomas that will be used in this project are available at OpenNeu.ro.