-
Download dataset BRATS 2018 competition page. unzip and place in your computer add the path of the folder in new_train.py folder
-
Install Python 3 and dependencies:
nibabel,
keras,
pytables,
nilearn,
SimpleITK,
nipype
(nipype is required for preprocessing only)
- preprocess.py give the path of your data-set folder to preprocess the image or you can download preprocessed data directly from Here
- Run the training:
To Start training the model:
$ python New_train.py
you can also download pretrained model from Here
$ python new_model_prediction.py
The predictions of our pretrained model is in the prtV
folder along with the input data and ground truth labels for
comparison.
I have also uploded the pretrained model prediction outputs and each dice score in the pretv folder.
you can download all outputs from
Here
For Evaluation Run evaluate.py
- The source code was mainly inspired by the author David G Ellis, from https://github.com/ellisdg/3DUnetCNN.
GBM Data Citation:
- Spyridon Bakas, Hamed Akbari, Aristeidis Sotiras, Michel Bilello, Martin Rozycki, Justin Kirby, John Freymann, Keyvan Farahani, and Christos Davatzikos. (2017) Segmentation Labels and Radiomic Features for the Pre-operative Scans of the TCGA-GBM collection. The Cancer Imaging Archive. https://doi.org/10.7937/K9/TCIA.2017.KLXWJJ1Q
LGG Data Citation:
- Spyridon Bakas, Hamed Akbari, Aristeidis Sotiras, Michel Bilello, Martin Rozycki, Justin Kirby, John Freymann, Keyvan Farahani, and Christos Davatzikos. (2017) Segmentation Labels and Radiomic Features for the Pre-operative Scans of the TCGA-LGG collection. The Cancer Imaging Archive. https://doi.org/10.7937/K9/TCIA.2017.GJQ7R0EF