DiffECG: A Versatile Probabilistic Diffusion Model for ECG Signals Synthesis

TL;DR

DiffECG introduces a versatile diffusion probabilistic model for ECG signal synthesis, capable of heartbeat generation, signal imputation, and forecasting, outperforming existing models and improving classifier performance.

Contribution

It presents the first generalized conditional diffusion model for ECG synthesis, addressing multiple tasks with superior results.

Findings

Outperforms state-of-the-art ECG generative models.

Enhances classifier performance on ECG tasks.

Effective for heartbeat generation, imputation, and forecasting.

Abstract

Within cardiovascular disease detection using deep learning applied to ECG signals, the complexities of handling physiological signals have sparked growing interest in leveraging deep generative models for effective data augmentation. In this paper, we introduce a novel versatile approach based on denoising diffusion probabilistic models for ECG synthesis, addressing three scenarios: (i) heartbeat generation, (ii) partial signal imputation, and (iii) full heartbeat forecasting. Our approach presents the first generalized conditional approach for ECG synthesis, and our experimental results demonstrate its effectiveness for various ECG-related tasks. Moreover, we show that our approach outperforms other state-of-the-art ECG generative models and can enhance the performance of state-of-the-art classifiers.