A bibliometric analysis of artificial intelligence in sports injury prediction and prevention

Abstract Artificial intelligence (AI) is increasingly being applied to the prediction and prevention of sports injuries, yet research in this area remains dispersed across sports science, biomechanics, sports medicine, data science, and engineering, and its intellectual structure has not been systematically mapped. This study conducted a bibliometric analysis of AI research in sports injury prediction and prevention, together with its performance-related applications. Bibliographic records were retrieved from Scopus on 11 July 2026, with no restrictions on the publication year. After screening, 809 documents (694 articles and 115 reviews) published between 1997 and 2027 were analyzed using the Bibliometrix R-package and VOSviewer. Scientific production grew markedly from 2018 onward, increasing from 13 publications in 2018 to 204 by 2025. The corpus was distributed across 371 sources and 3,226 authors and accumulated 13,295 citations. China and the United States were the leading contributors, with closely comparable affiliation-based output. Keyword co-occurrence analysis produced a network of 44 keywords in 6 clusters, in which machine learning and artificial intelligence were the most central terms, closely linked to injury prevention, injury prediction, biomechanics, and wearable sensing. Thematic mapping identified injury prevention, injury risk, and wearable technology as well-developed central themes, while thematic evolution across three periods (1997–2018, 2019–2022, and 2023–2025) indicated a progression from early injury- and sports-specific research through methodological diversification toward applied sports analytics and athlete monitoring. These

Findings: characterize the structure and trajectory of the published literature rather than the demonstrated effectiveness of AI tools in applied sports settings.