Artificial Intelligence and Machine Learning in Athlete Monitoring, Injury Prediction and Performance Optimization: A Narrative Review

Abstract

Artificial intelligence (AI) and machine learning (ML) are playing an increasingly prominent role in sports medicine, reshaping approaches to athlete monitoring, injury risk assessment, and performance enhancement. This narrative review summarizes current evidence on the application of AI and ML in three closely related areas: training load monitoring, injury prediction and prevention, and the individualization of training. A PubMed search was performed to identify studies published in English between 2019 and 2025. Wearable technologies generate data on internal and external training loads, as well as heart rate variability, providing the foundation for ML-based analyses. Decision trees, random forests, support vector machines, and artificial neural networks are among the methods most frequently used for injury prediction, although their performance varies across sporting disciplines. AI is also being applied to support individualized training, biomechanical feedback, nutritional planning, and return-to-play decisions. Emerging fields of application include sports cardiology, digital twin technology, and 3D-printed sensors. Current limitations include issues related to data quality and interoperability, limited multicenter validation, and concerns regarding ethics and data privacy. AI is increasingly integrated into athlete care, although broader clinical adoption will depend on further validation studies.