Machine learning prediction of pediatric adverse drug reactions using consensus-derived scarce data

Adverse drug reactions (ADRs) are a significant cause of morbidity and mortality in children, whose distinct physiological development creates pharmacological vulnerabilities poorly served by adult-focused studies and scarce clinical data. This work develops a comprehensive computational approach for pediatric pharmacovigilance that integrates consensus-driven signal detection, multi-level biological features, and interpretable machine learning. Using 1.4 million FDA Adverse Event Reporting System (FAERS) reports, the authors construct the largest curated pediatric drug–ADR dataset; severity-specific thresholds and voting across four algorithms (PRR, ROR, BCPNN, EBGM) optimize ADR identification, while multi-level biological fingerprints combined with XGBoost improve predictive performance, especially in imbalanced scenarios.