The George B. Moody PhysioNet Challenge 2025 invites teams to develop algorithmic approaches for identifying Chagas disease from standard 12-lead electrocardiograms (ECGs).
Chagas disease is a parasitic infection that is primarily transmitted by triatomine insects in Central and South America, where it is endemic. There is no vaccine for Chagas disease. It affects an estimated 8 million people, and it causes nearly 10,000 deaths annually. After an acute phase, which generally occurs in childhood, Chagas disease enters a life-long chronic phase. In the early stages of infection, Chagas disease has no or mild symptoms, and it can be treated with specific drugs to prevent the progression of the disease. In the later stages of infection, Chagas disease can cause cardiomyopathy, leading to heart failure, cardiac arrhythmias, and thromboembolism. Serological testing has shown the widespread prevalence of Chagas disease in some areas, and such tests can diagnose patients and support treatment, but serological testing capacities are limited. However, Chagas cardiomyopathy often manifests in ECGs, providing a signal for Chagas disease and informing the treatment of Chagas disease-related heart conditions.
For the 2025 Challenge, we ask participants to design and implement working, open-source algorithms for identifying potential cases of Chagas disease from ECGs. The teams with the best scores on the hidden test set will win the Challenge.
This Challenge provides multiple innovations. First, we leveraged several datasets from Brazil with labels from serological testing and from patient reports, providing the opportunity for teams to learn from large datasets with weak labels and smaller datasets with strong labels. Second, we created an evaluation metric that captures the local serological testing capacity and frames the machine learning problem as a triage or pre-screening task. Over 100 teams have participated in the Challenge so far, representing diverse approaches from both academia and industry worldwide.