A Signal-Language Foundation Model for Broad-Spectrum Cardiovascular Assessment from Routine Electrocardiography
This work expands routine ECG interpretation beyond common arrhythmias to broad cardiovascular assessment and opportunistic screening of echocardiographic and rare conditions, potentially improving diagnostic capabilities in clinical settings.
ECGCLIP, a signal-language contrastive learning model, was pre-trained on 2.8 million ECG studies and evaluated on 89 tasks across 1.5 million ECGs, achieving strong performance on atrial fibrillation (PRAUC 0.900) and ST-elevation MI (PRAUC 0.383), while also improving detection of rare cardiac diseases like Ebstein anomaly (PRAUC 0.253). The model demonstrated data efficiency, matching full-dataset baselines with only 10% of training data.
Electrocardiography (ECG) is central to cardiovascular care, but conventional AI models are often restricted to common arrhythmias and may generalize poorly across populations or clinically subtle diseases. We developed ECG Contrastive Language-Image Pre-training (ECGCLIP), a signal-language contrastive learning framework that aligns ECG waveforms with expert diagnostic reports. ECGCLIP was pre-trained on 2,837,962 ECG studies from 1,324,856 patients and evaluated on a held-out internal test set plus nine independent external cohorts comprising about 1.5 million ECGs. Evaluation covered 89 downstream tasks, including 45 ECG diagnoses, 39 echocardiographic targets, and 5 rare cardiac diseases, using PRAUC as the primary metric. ECGCLIP consistently improved performance over random initialization and Merl-R18 baselines. On the internal test set, ECGCLIP-R34 achieved strong performance for atrial fibrillation (PRAUC 0.900) and ST-segment elevation myocardial infarction (PRAUC 0.383), with robust generalization across all external cohorts. It also improved low-prevalence and diagnostically elusive diseases, including Ebstein anomaly, constrictive pericarditis, dextrocardia, and cardiac amyloidosis, with internal PRAUC values of 0.253, 0.175, 0.121, and 0.201, respectively. ECGCLIP was data efficient, matching or exceeding full-dataset baseline performance with only 10% of training data. Feature visualization and saliency analysis suggested clinically meaningful representations aligned with established electrocardiographic criteria. These findings indicate that large-scale ECG-report contrastive pre-training can expand routine ECG interpretation beyond common arrhythmias toward broad cardiovascular assessment and opportunistic screening of echocardiographic and rare conditions.