MACHINE LEARNING AND DEEP LEARNING APPLICATIONS IN SPORTS BIOMECHANICAL ANALYSIS: A SYSTEMATIC SCOPING REVIEW OF PERFORMANCE ENHANCEMENT AND INJURY PREVENTION STRATEGIES

Authors

Wissem Dhahbi, Research Unit: Sport Sciences, Health and Movement, UR22JS01, High Institute of Sport and Physical Education of Kef, University of Jendouba, Kef, TunisiaFollow
Nidhal Jebabli, Research Unit: Sport Sciences, Health and Movement, UR22JS01, High Institute of Sport and Physical Education of Kef, University of Jendouba, Kef, Tunisia.Follow
Manar Boujabli, Research Unit: Sport Sciences, Health and Movement, UR22JS01, High Institute of Sport and Physical Education of Kef, University of Jendouba, Kef, TunisiaFollow
Marouen Souaifi, Research Unit: Sport Sciences, Health and Movement, UR22JS01, High Institute of Sport and Physical Education of Kef, University of Jendouba, Kef, TunisiaFollow
Ismail Dergaa, Higher Institute of Sport and Physical Education of Ksar-Said, University of Manouba, Manouba, TunisiaFollow
Lamia Ben Ezzdine, Higher Institute of Sport and Physical Education of Ksar-Said, University of Manouba, Manouba, TunisiaFollow

Category

Injury prevention

Document Type

Paper

Abstract

Aim: This scoping review systematically maps current trends in artificial intelligence applications for sports biomechanical analysis, examining performance enhancement and injury prevention utility, methodological approaches, implementation challenges, and ethical considerations. Methods: A comprehensive analysis synthesized evidence on AI methodologies applied to sports biomechanics across various contexts, including machine learning, neural networks, and deep learning approaches integrated with learning management systems. This review examined data collection modes, analytical approaches, and practical implementation in laboratory and field settings. Results: AI applications evolved from basic statistical modeling to sophisticated machine learning configurations, demonstrating superior performance in technique analysis (94% agreement with international judges), individualized training prescription (25% improvement over baseline), and injury risk forecasting (85% pre-competition accuracy). Computer vision technology achieved marker-based joint tracking accuracy within 15mm, while temporal modeling detected biomechanical changes 2.5 training sessions before injury emergence. Learning management systems enhanced knowledge translation, improving coach understanding (45% increase) and athlete adherence (3.4-fold higher). Primary challenges remain data standardization, field validation, model explainability, and integration into established coaching workflows. Conclusion: AI integration into sports biomechanical analysis represents an irreversible paradigm shift, enabling unprecedented capabilities for movement analysis and preservation in sport. Implementation barriers and ethical considerations regarding data ownership, privacy, and equitable access require addressing through interdisciplinary partnerships and technological innovation to extend sophisticated biomechanical analysis across all competitive levels.

Recommended Citation

Dhahbi, Wissem; Jebabli, Nidhal; Boujabli, Manar; Souaifi, Marouen; Dergaa, Ismail; and Ben Ezzdine, Lamia (2025) "MACHINE LEARNING AND DEEP LEARNING APPLICATIONS IN SPORTS BIOMECHANICAL ANALYSIS: A SYSTEMATIC SCOPING REVIEW OF PERFORMANCE ENHANCEMENT AND INJURY PREVENTION STRATEGIES," ISBS Proceedings Archive: Vol. 43: Iss. 1, Article 18.
Available at: https://commons.nmu.edu/isbs/vol43/iss1/18