MD-Dose

This is the codebase for [MD-Dose: A diffusion model based on the Mamba for radiation dose prediction](https://ieeexplore.ieee.org/abstract/document/10822581).

Envs. for Pretraining

• Python 3.10.13

  • conda create -n your_env_name python=3.10.13

• torch 2.1.1 + cu118

  • pip install torch==2.1.1 torchvision==0.16.1 torchaudio==2.1.1 --index-url https://download.pytorch.org/whl/cu118

• Requirements: vim_requirements.txt

  • pip install -r vim/vim_requirements.txt

• Install causal_conv1d and mamba

  • pip install -e causal_conv1d
  • pip install -e mamba

Train Your MD-Dose

python train.py(Training without a structural encoder)

python train_struc.py(Training with a structural encoder)

Acknowledgement ❤️

This project is based on Vim (paper, code), Causal-Conv1d (code). Thanks for their wonderful works.

Citation

If you find MD-Dose is useful in your research or applications, please consider giving us a star 🌟 and citing it by the following BibTeX entry.

@inproceedings{fu2024md,
  title={MD-Dose: A diffusion model based on the Mamba for radiation dose prediction},
  author={Fu, Linjie and Li, Xia and Cai, Xiuding and Wang, Xueyao and Shen, Yali and Yao, Yu},
  booktitle={2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)},
  pages={911--918},
  year={2024},
  organization={IEEE}
}