Boundary-Guided Transformer

This repo is the implementation of our paper "a boundary-guided transformer based method for measuring distance from rectal tumor to anal verge on magnetic resonance images" published in Patterns.

DATA_Dataset: https://pan.baidu.com/s/1Rndj5ympeC2yRp2RF8Sv9w Extraction code：1234

Requirements

conda install pytorch torchvision cudatoolkit=11.0 -c pytorch

timm

pip install timm

opencv

pip install opencv-python

##data

see Boundary-Guided-Transformer-for-Automatic-DTAV-Measurement-in-MR-Images/Boundary-Guided-Transformer/data_example/

Usage

Train model

python Boundary-Guided-Transformer-for-Automatic-DTAV-Measurement-in-MR-Images/Boundary-Guided-Transformer/train.py 
optional arguments:
--data_path                   Data path for dataset £¨default='/home/shenjj/BGTR/data/'£©
--crop_h                      Crop height for training images 
--crop_w                      Crop width for training images 
--batch_size                  Number of data for each batch to train 
--epochs                      Number of sweeps over the dataset to train 
--save_path                   Save path for results

Eval model

python Boundary-Guided-Transformer-for-Automatic-DTAV-Measurement-in-MR-Images/Boundary-Guided-Transformer/viewer.py --model_weight model.pth
optional arguments:
--data_path                   Data path for dataset 
--model_weight                Pretrained model weight 
The model are trained using PyTorch 1.10.1 with a NVIDIA GeForce RTX 3090 GPU. The size of the input image is set to 512 X 512 and batch size set to 4.

DTAV measurememt based on segmentation results

python Boundary-Guided-Transformer-for-Automatic-DTAV-Measurement-in-MR-Images/DTAV Measurement/DTAV.py
An example of a test:
python Boundary-Guided-Transformer-for-Automatic-DTAV-Measurement-in-MR-Images/DTAV Measurement/test.py

Citation

If this code is helpful for your study, please cite:

@article{shen2023boundary,
  title={A boundary-guided transformer for measuring distance from rectal tumor to anal verge on magnetic resonance images},
  author={Shen, Jianjun and Lu, Siyi and Qu, Ruize and Zhao, Hao and Zhang, Li and Chang, An and Zhang, Yu and Fu, Wei and Zhang, Zhipeng},
  journal={Patterns},
  year={2023},
  publisher={Elsevier}
}

Acknowledgements

Part of codes are reused from the GatedSCNN and Cerberus. Thanks to Xiaoxue Chen and Towaki Takikawa for the codes of GatedSCNN and Cerberus.

uzeful/BGTR