AI and the World's Hard Problems · Medicine & drug discovery · Nature (computational biology) · 2026-10-22
Athlete-AI: a multi-modal artificial intelligence framework for personalized athlete monitoring and injury risk forecasting
Athlete-AI integrates wearable sensor, motion, pose, and literature-derived variables to evaluate adaptive training and injury-risk modelling in simulations. Removing predictors used to define the simulated labels reduced the F1 score from 86.6% to 72.8%, while benchmark-derived features alone performed comparably to a shuffled-label control. The findings show preliminary computational feasibility and do not constitute real-world, clinical, or prospective athlete validation.