metadata
license: cc-by-4.0
language:
- en
tags:
- ecg
- student-teacher
- echocardiograms
EchoingECG (MICCAI 2025)
EchoingECG is a probabilistic student-teacher model designed to improve cardiac function prediction from electrocardiograms (ECGs) by distilling knowledge from echocardiograms (ECHO). This approach leverages uncertainty-aware ECG embeddings and ECHO supervision, integrating Probabilistic Cross-Modal Embeddings (PCME++) and ECHO-CLIP, a vision-language pretrained model, to transfer ECHO knowledge into ECG representations.
Please refer to our github for use: https://github.com/mcintoshML/EchoingECG
Installation
Clone the repository and install dependencies:
git clone https://github.com/mcintoshML/EchoingECG.git
cd EchoingECG
pip install -r requirements.txt
Citation
If you use EchoingECG in your research, please cite:
@InProceedings{GaoYua_EchoingECG_MICCAI2025,
author = { Gao, Yuan and Kim, Sangwook and McIntosh, Chris},
title = { { EchoingECG: An Electrocardiogram Cross-Modal Model for Echocardiogram Tasks } },
booktitle = {proceedings of Medical Image Computing and Computer Assisted Intervention -- MICCAI 2025},
year = {2025},
publisher = {Springer Nature Switzerland},
volume = {LNCS 15964},
month = {September},
page = {175 -- 185}
}