Bayesian ECG reconstruction using denoising diffusion generative models
This work addresses cardiac health monitoring and diagnosis by enabling clinical tools like QTc calculation and noise suppression, though it appears incremental as it applies existing DDGM methods to ECG data.
The authors tackled the problem of generating realistic electrocardiogram (ECG) signals and solving linear inverse Bayesian problems using a denoising diffusion generative model (DDGM) trained on healthy ECG data, resulting in applications such as corrected QT interval calculation, noise suppression, lead recovery, and anomaly detection for cardiac health monitoring.
In this work, we propose a denoising diffusion generative model (DDGM) trained with healthy electrocardiogram (ECG) data that focuses on ECG morphology and inter-lead dependence. Our results show that this innovative generative model can successfully generate realistic ECG signals. Furthermore, we explore the application of recent breakthroughs in solving linear inverse Bayesian problems using DDGM. This approach enables the development of several important clinical tools. These include the calculation of corrected QT intervals (QTc), effective noise suppression of ECG signals, recovery of missing ECG leads, and identification of anomalous readings, enabling significant advances in cardiac health monitoring and diagnosis.