Yale School of Medicine (CarDS Lab)
Model weights not public. Contact creators for more information.
EfficientNet-B3 CNN that detects hypertrophic cardiomyopathy directly from images of printed or scanned 12-lead ECGs, rather than from raw digital waveforms, enabling screening from a photo of a paper tracing. Initialized via self-supervised contrastive pretraining on patient identity, then fine-tuned at Yale New Haven Hospital on over 124,000 ECGs from about 67,000 patients, with HCM status confirmed by cardiac MRI or echocardiography. Externally validated on ECG images from MIMIC-IV, Amsterdam UMC, and UK Biobank. Developed by Yale's CarDS Lab.
Architecture
CNN (2D)
EfficientNet-B3 CNN initialized from a self-supervised biocontrastive (patient-identity contrastive) pretraining step on ECG images, then fine-tuned with a weighted binary cross-entropy loss for HCM detection
Framework
PyTorch
Added to catalog
2026-07-13
Yale New Haven Hospital ECG-Image HCM Cohort
124,553 ECG images from 66,987 individuals (2012-2021); HCM defined by concurrent CMR or echocardiography
HCM detection