ECG-LLM: foundation model for ECG-based cardiac reasoning

ECG-conditioned LLM trained on 679,112 ECG studies from 186,409 patients across four cohorts, using multimodal-to-language supervision: clinically structured QA pairs derived from ECG, clinical context, CMR and echo. Answers cardiovascular questions from a 12-lead ECG alone: recovers conventional measurements, predicts CMR phenotypes (ventricular/atrial volumes, function) and echo phenotypes (LV wall thickness, aortic stenosis, RV systolic dysfunction). Matches or exceeds baselines on report generation and ECG-QA.

• Form factor: clinical

Electrocardiography (ECG) is an inexpensive, standard-of-care test for cardiac symptoms, but front-line triage often lacks immediate access to definitive imaging such as echocardiography (ECHO) or cardiac magnetic resonance (CMR). Furthermore, most existing ECGAI systems are limited to fixed diagnostic labels or automated reports, constraining their use for patient-specific clinical reasoning. To address this gap, we introduce ECG-LLM, an ECG-conditioned large language model trained across four cohorts comprising 679,112 ECG studies from 186,409 patients. Using a novel multimodal-to-language supervision strategy, ECG-LLM is trained on clinically structured question-answer pairs derived from ECG signals, clinical context, CMR, and ECHO. This unified approach enables the model to answer diverse cardiovascular questions from a 12-lead ECG alone, spanning both conventional interpretation and phenotypes not directly visible on standard ECGs. ECG-LLM successfully recovers conventional ECG measurements, such as heart rate, and strongly predicts complex CMR-derived phenotypes, including ventricular and atrial volumes and ventricular function. Crucially, it detects vital echocardiographic phenotypes, including increased LV wall thickness, aortic stenosis, and right-ventricular systolic dysfunction. On standard ECG understanding tasks, ECG-LLM matches or exceeds existing baselines for diagnostic report generation and the ECG-QA benchmark. By moving beyond fixed-label prediction, this multimodal framework provides clinically valuable, question-driven cardiovascular reasoning to support general practitioner and front-line triage decisions when specialist review is delayed.

Verbatim from DataCite (opens in a new tab) · Licence: https://creativecommons.org/licenses/by/4.0/legalcode

No results from this paper on ChatPPG leaderboards yet. Browse the leaderboards.