MRBrainS18 challenge winning submission

Winning submission of the Grand Challenge on MR Brain Segmentation at MICCAI 2018 by team MISPL (Medical Image and Signal Processing Lab @ DGIST).

• Check the challenge results
• Download pretrained weights
• Check our article: 3D Patchwise U-Net with Transition Layers for MR Brain Segmentation

Usage

Create a python environment able to run the packages Numpy, TensorFlow and SimpleITK. Then you can execute the commands according to the required task as follows.

Training

bash run.sh <run number> train <GPU number> <checkpoint number>

bash run.sh 1 train 0 0

Testing

bash run.sh <run number> test <GPU number> <checkpoint number>

bash run.sh 1 test 0 0

Summaries

bash run.sh <run number> summaries <GPU number>

bash run.sh 1 summaries 0

Update global summary

bash get_summaries.sh

Check the summaries file

cat summary.txt

Citation

If you find the code useful for your research, please consider citing our article:

• MISPL_MRBrainS18:

@inproceedings{mispl_mrbrains18,
  author={Luna, Miguel and Park, Sang Hyun},
  title={3D Patchwise U-Net with Transition Layers for MR Brain Segmentation},
  booktitle={Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries},
  year={2019},
  pages={394-403}
}

• DOI: https://doi.org/10.1007/978-3-030-11723-8_40