https://github.com/kaczmarj/meningioma
Segmentation of meningiomas in structural MR images
https://github.com/kaczmarj/meningioma
brain segmentation tumor
Last synced: about 2 months ago
JSON representation
Segmentation of meningiomas in structural MR images
- Host: GitHub
- URL: https://github.com/kaczmarj/meningioma
- Owner: kaczmarj
- Created: 2016-12-19T14:26:28.000Z (over 8 years ago)
- Default Branch: master
- Last Pushed: 2017-05-02T19:28:07.000Z (about 8 years ago)
- Last Synced: 2025-02-10T03:22:22.358Z (4 months ago)
- Topics: brain, segmentation, tumor
- Language: Jupyter Notebook
- Homepage:
- Size: 3.05 MB
- Stars: 1
- Watchers: 3
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: readme.md
Awesome Lists containing this project
README
# Meningioma
Goals
-----
1. To automatically segment meningiomas pre- and post-surgery.
1. To assess brain's spatial recovery post-surgery.
- Data is available on [Brainbox](http://brainbox.pasteur.fr/project/meningioma)
- [Ongoing literature search](/lit_review.md)
Table of contents
-----------------
1. [Ground truth segmentation](#ground-truth-segmentation)
1. [Segmentation with slicer](#segmentation-with-slicer)
1. [ANTs brain extraction](#ants-brain-extraction)
1. [t-SNE on synthetic brain](#t-sne-on-synthetic-brain)
Ground truth segmentation
-------------------------
| case | segmented | reviewed |
|:----:|:---------:|:--------:|
| [001](http://brainbox.pasteur.fr/mri/?url=https://dl.dropbox.com/sh/71jbelduefu41xs/AADysls57HwmJT0pdbbSVI4Na/case_001_2.nii.gz) | ✔ | |
| [002](http://brainbox.pasteur.fr/mri/?url=https://dl.dropbox.com/sh/71jbelduefu41xs/AAAx8y8Mp_jh0yad1CjD2Z4Ra/case_002_2.nii.gz) | ✔ | |
| [003](http://brainbox.pasteur.fr/mri/?url=https://dl.dropbox.com/sh/71jbelduefu41xs/AABhDQJ-GQKMzHdFDGN6MrFBa/case_003_2.nii.gz) | ✔ | |
Segmentation with Slicer
------------------------
An example of semi-automatic segmentation with Slicer's Segmentation Editor module.
ANTs brain extraction
---------------------
We attempted to remove the skull with ANTs `antsBrainExtraction.sh`. Results were mixed. In some cases, the script performed well. In other cases, parts of the brain were removed, or parts of the skull (usually close to the meningioma) remained.
Above: successful brain extraction.
Above: unsuccessful brain extraction.
t-SNE on synthetic brain
------------------------
[[source](/notebooks/dim_reduction.ipynb)]
- blue == background
- green == skull
- red == brain
- cyan == tumor
