ECG-MoE: Mixture-of-Expert Electrocardiogram Foundation Model
This work provides a more efficient and accurate ECG analysis tool for clinicians, improving cardiac diagnosis by better capturing the complex features of ECG signals.
This paper addresses the limitations of existing ECG foundation models in capturing periodicity and diverse features for various clinical tasks. The proposed ECG-MoE, a hybrid architecture, achieves state-of-the-art performance across five public clinical tasks with 40% faster inference compared to multi-task baselines.
Electrocardiography (ECG) analysis is crucial for cardiac diagnosis, yet existing foundation models often fail to capture the periodicity and diverse features required for varied clinical tasks. We propose ECG-MoE, a hybrid architecture that integrates multi-model temporal features with a cardiac period-aware expert module. Our approach uses a dual-path Mixture-of-Experts to separately model beat-level morphology and rhythm, combined with a hierarchical fusion network using LoRA for efficient inference. Evaluated on five public clinical tasks, ECG-MoE achieves state-of-the-art performance with 40% faster inference than multi-task baselines.