Robust Multimodal Brain Tumor Segmentation via Feature Disentanglement and Gated Fusion

Tensorflow implementation of our robust multimodal brain tumor segmentation framework.

Paper

Robust Multimodal Brain Tumor Segmentation via Feature Disentanglement and Gated Fusion MICCAI 2019

Installation

• Install TensorFlow 1.10 and CUDA 9.0
• Clone this repo

git clone https://github.com/cchen-cc/Robust-Mseg
cd Robust-Mseg

Data Preparation

• Use nii2tfrecord function in ./preprocessing.py to convert nii data into tfrecord format to be decoded by ./data_loader.py

Train

• Specify the data path in ./main.py
• Run ./main.py to start the training process

Evaluate

• Our trained models can be downloaded from Dropbox.
• Specify the model path and data path in ./evaluate.py
• Run ./evaluate.py to start the evaluation.

Citation

If you find the code useful for your research, please consider cite our paper.

@inproceedings{chen2019robust,
  title={Robust multimodal brain tumor segmentation via feature disentanglement and gated fusion},
  author={Chen, Cheng and Dou, Qi and Jin, Yueming and Chen, Hao and Qin, Jing and Heng, Pheng-Ann},
  booktitle={International Conference on Medical Image Computing and Computer-Assisted Intervention},
  pages={447--456},
  year={2019},
  organization={Springer}
}

Note

• Contact: Cheng Chen (chencheng236@gmail.com)