Does a multinational federated learning approach improve the generalizability of ECG and echocardiogram models for detecting HCM compared to single-institution training?
A multinational federated learning approach improves the generalizability of AI models using ECG and echocardiograms for detecting hypertrophic cardiomyopathy compared to single-center training.
Federated learning improved the generalizability of models that use ECGs and echocardiograms to detect and differentiate HCM from other causes of hypertrophy compared with training within a single institution.
Goto et al. (Tue,) studied this question.