conda create -n RMAMamba python==3.8.16
conda activate RMAMamba
pip install -r requirements.txt
cd selective_scan && pip install .
Download pretrained checkpoints from Google Drive and move it to the pretrained_pth directory.
Download CirrMRI600+ dataset from this link and move it to the data directory.
python train.py --model RMAMamba_S
python test.py --model RMAMamba_S
Our weight files and result maps are available on Google Drive.
Please cite our paper if you find the work useful:
@article{zeng2025reverse,
title={A Reverse Mamba Attention Network for Pathological Liver Segmentation},
author={Zeng, Jun and Jha, Debesh and Aktas, Ertugrul and Keles, Elif and Medetalibeyoglu, Alpay and Antalek, Matthew and Lewandowski, Robert and Ladner, Daniela and Borhani, Amir A and Durak, Gorkem and others},
journal={arXiv preprint arXiv:2502.18232},
year={2025}
}
Please contact zeng.cqupt@gamil.com for any further questions.