ECG-MoE: Mixture-of-Expert Electrocardiogram Foundation Model

TL;DR

ECG-MoE introduces a novel hybrid model that effectively captures ECG features for diverse clinical tasks, achieving state-of-the-art accuracy and faster inference.

Contribution

The paper presents ECG-MoE, a mixture-of-experts model that separately models ECG morphology and rhythm, improving performance and efficiency over existing methods.

Findings

Achieves state-of-the-art results on five clinical ECG tasks.

Provides 40% faster inference compared to multi-task baselines.

Effectively models beat morphology and rhythm separately.

Abstract

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.