Journal DOI 10.68956/journal.human-performance-futures

Creative Commons source reading · Machine learning for performance

Systematic review of different approaches for performance enhancement in elite sport

Oualid Dehbane; Sara Ouahabi; Sanaa El Filali

Originally published in Front Artif Intell · 2026-03-09

From the article

Selected source text. JournalOS removes markup and shortens excerpts where indicated. The full article provides the complete context.

Abstract

BackgroundElite sport is undergoing rapid technological transformation driven by advanced analytics, artificial intelligence (AI), and immersive systems. While numerous studies address performance enhancement and injury-related applications, evidence remains fragmented across technologies and sport contexts.ObjectiveThis systematic review aimed to examine the prevalence and distribution of advanced analytical technologies across application domains (performance, injury, and emerging objectives) and sport disciplines, and to identify areas of technological maturity in elite sport.MethodsA systematic review was conducted following PRISMA 2020 guidelines. Four databases (Google Scholar, Scopus, Web of Science, IEEE Xplore) were searched for peer-reviewed studies published between January 2019 and March 2025. Fifty-two studies met the…

Principal findings and contribution of this review

This systematic review synthesised evidence from 52 studies published between 2019 and 2025 examining technology-enabled approaches for performance enhancement and injury-related outcomes in elite sport. The results highlight a clear predominance of data-driven and computational approaches, with artificial intelligence (AI)–based methods, including machine learning (ML), deep learning (DL), and generative AI (GAI), representing the largest share of the literature. Statistical analysis and virtual reality (VR) remain important complementary pillars, addressing more established analytical questions and immersive training or rehabilitation contexts, respectively. Across technologies and sport contexts, performance enhancement emerged as the dominant research objective, whereas injury-related applications, although present, were less…

Implications for practice and future research

For practitioners, the findings suggest that AI-based tools offer promising support for performance analysis and training optimisation, but their outputs must be interpreted cautiously, especially in injury prediction contexts. VR appears well-suited for controlled training and rehabilitation environments but requires integration with sport-specific constraints to maximise ecological validity. Future research should prioritise (i) rigorous validation of AI models using clinically and practically meaningful metrics, (ii) expansion of injury-focused applications across sport types, (iii) integration of explainable AI to enhance trust and adoption, and (iv) longitudinal studies assessing real-world effectiveness. Addressing these gaps will be essential for translating technological innovation into sustainable performance and health…

Editorial reading guide

  • What data created and tested the model, and was evaluation independent?
  • What is the cost of a false positive, false negative or out-of-distribution decision?