Versatile Cardiovascular Signal Generation with a Unified Diffusion Transformer
This addresses the challenge of diverse acquisition issues in cardiovascular monitoring for healthcare applications, representing an incremental advance by integrating existing methods into a unified framework.
The paper tackles the problem of joint utilization of cardiovascular signals like PPG, ECG, and BP for real-time monitoring by proposing UniCardio, a multi-modal diffusion transformer that reconstructs low-quality signals and synthesizes unrecorded ones, outperforming task-specific baselines in denoising, imputation, and translation with generated signals matching ground-truth performance in health detection and vital sign estimation.
Cardiovascular signals such as photoplethysmography (PPG), electrocardiography (ECG), and blood pressure (BP) are inherently correlated and complementary, together reflecting the health of cardiovascular system. However, their joint utilization in real-time monitoring is severely limited by diverse acquisition challenges from noisy wearable recordings to burdened invasive procedures. Here we propose UniCardio, a multi-modal diffusion transformer that reconstructs low-quality signals and synthesizes unrecorded signals in a unified generative framework. Its key innovations include a specialized model architecture to manage the signal modalities involved in generation tasks and a continual learning paradigm to incorporate varying modality combinations. By exploiting the complementary nature of cardiovascular signals, UniCardio clearly outperforms recent task-specific baselines in signal denoising, imputation, and translation. The generated signals match the performance of ground-truth signals in detecting abnormal health conditions and estimating vital signs, even in unseen domains, while ensuring interpretability for human experts. These advantages position UniCardio as a promising avenue for advancing AI-assisted healthcare.