ECGMambaVersionOfJTEHM2020-2021_final

ECG-Mamba and Non-Uniform-Mix

Thanks for reaching out! Here's the explanation for ECG-Mamba and Non-Uniform-Mix.

Quick note: This Markdown file is currently being updated and should be finished very soon. Please check back shortly!

Introduction

Citing

Please cite our paper(s) if you find this repository useful.

@article{jiang2025ecg,
  title={ECG-Mamba: Cardiac Abnormality Classification with Non-Uniform-Mix Augmentation on 12-Lead ECGs},
  author={Jiang, Huawei and Mutahira, Husna and Wei, Shibo and Muhammad, Mannan Saeed},
  journal={IEEE Journal of Translational Engineering in Health and Medicine},
  year={2025},
  publisher={IEEE}
}

If you are interested in this area, you can also cite the below paper on 12-lead ECG multi-label classification via Mamba architecture. This work demonstrated an increase of 3% in AUPRC compared to the ECG-Mamba model.

@misc{jiang2025dimensionalcnnecgmamba,
      title={One Dimensional CNN ECG Mamba for Multilabel Abnormality Classification in 12 Lead ECG}, 
      author={Huawei Jiang and Husna Mutahira and Gan Huang and Mannan Saeed Muhammad},
      year={2025},
      eprint={2510.13046},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2510.13046}, 
}

Getting started

Dataset

In this paper, two challenge datasets are used: the PhysioNet/CinC Challenges of 2020 and 2021. The datasets can be downloaded from the website below:Click

Required package

ECG-Mamba, adapted from Vision Mamba (ViM), requires CUDA 11.8 for compatibility. Follow this link: https://github.com/hustvl/Vim/issues/53 to install the necessary packages

How to run

you can go to the folder of scripts to run the program.

for the scenario challenge 2020

bash ECG_scenario2020.sh

for the scenario challenge 2021

bash ECG_scenario2021.sh

Contact

If you have a question, please start a discussion in the Community section (this is similar to opening an issue on GitHub). I will do my best to help you out, just as I received so much support at the beginning of my own research journey.