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

Jiarui Jin, Haoyu Wang, Xingliang Wu, Xiaocheng Fang, Xiang Lan, Zihan Wang, Deyun Zhang, Bo Liu, Yingying Zhang, Xian Wu, Hongyan Li, Shenda Hong
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:54089-54130, 2026.

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 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 Interleaved Modality Dropout to improve robustness and cross-modal consistency when either the ECG signal or ECG image is missing. Third, we present 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 https://github.com/PKUDigitalHealth/ECG-R1.

Cite this Paper


BibTeX
@InProceedings{pmlr-v306-jin26l, title = {{ECG}-R1: Protocol-Guided and Modality-Agnostic {MLLM} for Reliable {ECG} Interpretation}, author = {Jin, Jiarui and Wang, Haoyu and Wu, Xingliang and Fang, Xiaocheng and Lan, Xiang and Wang, Zihan and Zhang, Deyun and Liu, Bo and Zhang, Yingying and Wu, Xian and Li, Hongyan and Hong, Shenda}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {54089--54130}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/jin26l/jin26l.pdf}, url = {https://proceedings.mlr.press/v306/jin26l.html}, 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 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 Interleaved Modality Dropout to improve robustness and cross-modal consistency when either the ECG signal or ECG image is missing. Third, we present 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 https://github.com/PKUDigitalHealth/ECG-R1.} }
Endnote
%0 Conference Paper %T ECG-R1: Protocol-Guided and Modality-Agnostic MLLM for Reliable ECG Interpretation %A Jiarui Jin %A Haoyu Wang %A Xingliang Wu %A Xiaocheng Fang %A Xiang Lan %A Zihan Wang %A Deyun Zhang %A Bo Liu %A Yingying Zhang %A Xian Wu %A Hongyan Li %A Shenda Hong %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-jin26l %I PMLR %P 54089--54130 %U https://proceedings.mlr.press/v306/jin26l.html %V 306 %X 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 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 Interleaved Modality Dropout to improve robustness and cross-modal consistency when either the ECG signal or ECG image is missing. Third, we present 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 https://github.com/PKUDigitalHealth/ECG-R1.
APA
Jin, J., Wang, H., Wu, X., Fang, X., Lan, X., Wang, Z., Zhang, D., Liu, B., Zhang, Y., Wu, X., Li, H. & Hong, S.. (2026). ECG-R1: Protocol-Guided and Modality-Agnostic MLLM for Reliable ECG Interpretation. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:54089-54130 Available from https://proceedings.mlr.press/v306/jin26l.html.

Related Material