Application Progress and Research Frontiers of Artificial Intelligence in Strength and Conditioning Training for Basketball Players

Abstract

Artificial intelligence (AI) has emerged as a transformative force in sports science, particularly in strength and conditioning (S&C) training for basketball players. This narrative review synthesizes current evidence on AI applications in basketball S&C, encompassing machine learning (ML)-driven injury risk prediction, computer vision-based biomechanical analysis, wearable sensor-enabled load monitoring, and reinforcement learning for personalized training prescription. A structured literature search of peer-reviewed literature published between 2019 and 2025 identified 15 representative studies for narrative synthesis. Results indicate that AI models, particularly XGBoost and deep learning architectures, achieved a reported area under the receiver operating characteristic curve (AUC-ROC) of 0.840 in predicting lower extremity muscle strains in NBA athletes, while markerless motion capture systems reported centimeter-level agreement with optical reference standards for basketball-specific movement analysis. LSTM neural networks with multi-modal data fusion show promise in personalized exercise volume optimization. However, significant challenges remain, including limited prospective external validation, model interpretability concerns, data privacy issues, and the gap between algorithmic performance and practical coaching implementation. Future research should prioritize explainable AI, digital twin integration, federated learning, and hybrid human-AI coaching frameworks.