ECG-R1: Protocol-Guided and Modality-Agnostic MLLM for Reliable ECG Interpretation

TL;DR

ECG-R1 is a novel, reliable multimodal large language model specifically designed for ECG interpretation, incorporating protocol-guided data, modality decoupling, and reinforcement learning to improve accuracy and robustness.

Contribution

The paper introduces ECG-R1, the first reasoning ECG MLLM with innovative data generation, a modality-decoupled architecture, and reinforcement learning for evidence-based ECG analysis.

Findings

ECG-R1 demonstrates improved reliability in ECG interpretation.

Public MLLMs often hallucinate, highlighting the need for verification.

Code for ECG-R1 is publicly available at the provided GitHub link.

Abstract

Electrocardiography (ECG) serves as an indispensable diagnostic tool in clinical practice, yet existing multimodal large language models (MLLMs) remain unreliable for ECG interpretation, often producing plausible but clinically incorrect analyses. To address this, we propose ECG-R1, the first reasoning ECG MLLM designed for reliable ECG interpretation via three innovations. First, we construct the interpretation corpus using \textit{Protocol-Guided Instruction Data Generation}, grounding interpretation in measurable ECG features and monograph-defined quantitative thresholds and diagnostic logic. Second, we present a modality-decoupled architecture with \textit{Interleaved Modality Dropout} to improve robustness and cross-modal consistency when either the ECG signal or ECG image is missing. Third, we present \textit{Reinforcement Learning with ECG Diagnostic Evidence Rewards} to strengthen evidence-grounded ECG interpretation. Additionally, we systematically evaluate the ECG interpretation capabilities of proprietary, open-source, and medical MLLMs, and provide the first quantitative evidence that severe hallucinations are widespread, suggesting that the public should not directly trust these outputs without independent verification. Code is available at \href{https://github.com/PKUDigitalHealth/ECG-R1}{here}.