Cardiac imaging data improves ECG detection of Chagas disease
Leveraging Cardiac Imaging to Improve ECG-Based Detection of Chagas Disease in Resource-Constrained Settings
Machine LearningArtificial Intelligence
Summary
Chagas disease affects the heart and is common in Latin America. Although detailed heart scans called CMR can show damage clearly, these scans and experts are rare in places where the disease is common. ECGs, which measure the heart’s electrical activity, are easier to get but less direct for detecting damage. The authors trained a computer model to learn from the better heart scans and then used that knowledge to improve disease detection from ECGs alone. This method worked well even in different countries and with patients not seen before.
What this means in practice
- •For clinical diagnostic teams: Improve ECG-based screening for Chagas disease in resource-limited hospitals by using imaging-informed AI models.
- •For medical device developers: Design ECG interpretation software that incorporates cardiac imaging knowledge to better identify heart disease signs in endemic regions.$Commercial implications: Enables creation of improved ECG diagnostic tools tailored for Chagas disease screening in markets with limited imaging access.
Authors
Laura Alvarez-Florez, Daniel Uyterlinde, Samuel Ruipérez-Campillo, Lukas P. A. Arts, Folkert W. Asselbergs, Fleur V. Y. Tjong
Abstract
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.