Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
To bridge the diagnostic gap for Chagas disease in resource-constrained settings, the authors align a 1D ECG encoder with a clinically grounded cardiac magnetic resonance (CMR) latent space using 63,193 paired examinations from the UK Biobank. This cross-modal contrastive transfer allows low-cost electrical biosignals to inherit structural representations from expensive, scarce imaging modalities without seeing any Chagas cases during pre-training. Evaluated on Latin American clinical cohorts via a frozen linear probe, the aligned ECG representation achieves an AUROC of 0.851 and top-ranked PhysioNet/CinC 2025 Challenge performance, significantly outperforming unaligned ECG foundation models.
Cheap ECG signals can inherit the structural diagnostic power of scarce cardiac MRIs, transferring zero-shot to detect endemic Chagas cardiomyopathy in Latin America without seeing a single diseased heart during pre-training.
Chagas disease is a major cause of cardiomyopathy in Latin America. Cardiac magnetic resonance (CMR) imaging can characterize its structural abnormalities, but scanners and expert readers remain scarce in endemic regions. Electrocardiography (ECG) is inexpensive and widely available, yet structural disease must be inferred indirectly from electrical signals. We propose to transfer CMR-derived structural knowledge to ECG through contrastive pre-training. Using 63,193 paired ECG-CMR examinations from the UK Biobank, we align an ECG encoder with a clinically grounded CMR embedding space using an asymmetric InfoNCE objective. Despite seeing no Chagas cases during pre-training, the resulting representation improves ECG-based Chagas detection. Across CODE-15% and SaMi-Trop, a frozen linear probe achieves an AUROC of 0.851 and sensitivity at the top 5% of predicted risk (Top5%-TPR) of 0.427 in five-fold cross-validation, compared with 0.827 and 0.377 for an unaligned ECG-FM baseline. On the PhysioNet/CinC 2025 Challenge test set, our model obtains the highest AUROC on SaMi-Trop-3 and the best ELSA-Brasil challenge score among the three top-performing methods, indicating that imaging-supervised ECG representations can generalize to populations and settings beyond the pre-training distribution.