Synthesis of Standard 12-Lead ECG from Single-Lead ECG Using Shifted Diffusion Models
Abstract
As the primary tool for monitoring cardiac health, a standard 12-lead ECG device is specialized medical equipment that is challenging to integrate into daily life. Meanwhile, existing portable ECG monitoring devices can only capture single-lead ECG, which is insufficient for health diagnosis. To address this issue, we propose a novel shifted diffusion model algorithm that utilizes a single-lead ECG to generate a standard 12-lead ECG. Our algorithm uses the detected single-lead ECG as the condition and employs the diffusion model to synthesize corresponding other 11-lead ECG. The extra shift is utilized in the forward process so that the model can learn better. Our approach has been tested on three datasets, yielding promising results.
Cite
Text
Liu et al. "Synthesis of Standard 12-Lead ECG from Single-Lead ECG Using Shifted Diffusion Models." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2024. doi:10.1007/978-3-031-70378-2_17Markdown
[Liu et al. "Synthesis of Standard 12-Lead ECG from Single-Lead ECG Using Shifted Diffusion Models." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2024.](https://mlanthology.org/ecmlpkdd/2024/liu2024ecmlpkdd-synthesis/) doi:10.1007/978-3-031-70378-2_17BibTeX
@inproceedings{liu2024ecmlpkdd-synthesis,
title = {{Synthesis of Standard 12-Lead ECG from Single-Lead ECG Using Shifted Diffusion Models}},
author = {Liu, Jingwei and Li, Hongyan and Hong, Shenda},
booktitle = {European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases},
year = {2024},
pages = {271-286},
doi = {10.1007/978-3-031-70378-2_17},
url = {https://mlanthology.org/ecmlpkdd/2024/liu2024ecmlpkdd-synthesis/}
}