Adversarial Bias Correction Network for Infant Brain MR Images.
.
- Linux
- NVIDIA GPU with at least 8Gb memory
- CUDA CuDNN
- python (3.6)
- pytorch (0.4.1+)
- torchvision (0.2.1+)
- tensorboardx (1.6+)
Resample and convert the input files into npy format with size of 256x256x256.
Modify the train.py file to match the training data in your own path. Then, run:
python train.py --dataroot /training_data_folder --name project_name --model pix2pix3d --direction AtoB --dataset_mode unaligned --input_nc 1 --output_nc 1 --gpu_ids 0
python test.py --dataroot /data_folder --name project_name --model pix2pix3d --direction AtoB --dataset_mode unaligned --input_nc 1 --output_nc 1 --gpu_ids 0
If you use this code for your research, please cite as:
Chen, Liangjun, et al. "ABCnet: Adversarial bias correction network for infant brain MR images." Medical Image Analysis 72 (2021): 102133.