Brain Tumor Segmentation for MICCAI BraTS (PAUNet FMNet)

[1] Path Aggregation U-Net for Brain Tumor Segmentation Paper link: https://link.springer.com/article/10.1007/s11042-020-08795-9

[2] FMNet: Feature Mining Networks for Brain Tumor Segmentation Paper link: https://ieeexplore.ieee.org/abstract/document/8995303

This repository contains the tensorflow implementation of the model we proposed in our paper

Requirements

• The code has been written in Python (3.5.2) and Tensorflow (1.12.0)
• Make sure to install all the libraries given in requirement.txt (You can do so by the following command)

pip install -r requirement.txt

Data preprocessing

BraTS 2017 BraTS 2018

• Has been finished

Train (To get model)

• Select the GPU you want to use. Add the following at the beginning of training code.

>>> import os
>>> os.environ["CUDA_DEVICE_ORDER"] = "PCI_BUS_ID"
>>> os.environ["CUDA_VISIBLE_DEVICES"] = "0"

• Begin training
• If you want to train all cross validation models. To run:

$ python3 ./train/FCNN.py

$ python3 ./train/DUnet.py

$ python3 ./train/VGG.py

$ python3 ./train/PA+FP.py

$ python3 ./train/PA+FP+ED.py

$ python3 ./train/PA+EFP+ED.py

• If you only want to train one model, To modify the code like following at the end of training code:

if __name__ == '__main__':
    paunet0 = PAUnet0()
    paunet0.train()

    # paunet1 = PAUnet1()
    # paunet1.train()
    #
    # paunet2 = PAUnet2()
    # paunet2.train()
    #
    # paunet3 = PAUnet3()
    # paunet3.train()
    #
    # paunet4 = PAUnet4()
    # paunet4.train()

Test (To get the segmented .npy file)

• If you want to test all cross validation models. To run:

$ python3 ./test/FCNN_test.py

$ python3 ./test/DUnet_test.py

$ python3 ./test/VGG_test.py

$ python3 ./test/PA+FP_test.py

$ python3 ./test/PA+FP+ED_test.py

$ python3 ./test/PA+EFP+ED_test.py

• If you only want to test one model, To modify the code like following at the end of testing code:

if __name__ == '__main__':
    paunet0 = PAUnet0()
    paunet0.test()

    # paunet1 = PAUnet1()
    # paunet1.test()
    #
    # paunet2 = PAUnet2()
    # paunet2.test()
    #
    # paunet3 = PAUnet3()
    # paunet3.test()
    #
    # paunet4 = PAUnet4()
    # paunet4.test()

Calculate Dice

• Modify the path and run:

$ python3 ./utils/read_json.py

• You can choose to calculate dice of cross validation or dice of one model in

./utils/read_json.py

Calculate PPV and Sensitivity

• Modify the path and run:

$ python3 ./utils/cal_dice_ppv_sen_cross_validation.py

• You can also use:

./utils/cal_dice_ppv_sen_onemodel.py

to calculate the PPV and Sensitivity of one model