Federated Learning Framework for Adverse Drug Reaction Prediction from Multi-Institutional Health Records

Adverse drug reactions are often under-detected when analyses are limited to single institutions, and centralized pooling of electronic health records is frequently constrained by privacy regulations, institutional governance, and technical barriers to data transfer. Current multi-institutional pharmacovigilance typically relies on aggregate summaries rather than collaborative machine learning over distributed individual-level records, leaving complex temporal, clinical, and medication-related p