Evaluating AI and traditional models for injury prediction and prevention in high-contact sports

Fikruzzaman, Muhammad and Sazali, Razif and Md Yusoff, Yusandra and Zulqarnain, Muhammad and Haziq, Amrun and Adnan, Aizzat and Linoby, Adam (2025) Evaluating AI and traditional models for injury prediction and prevention in high-contact sports. In: International Graduate Colloquium: Sports and Physical Exercise Assembly of Knowledge Sharing i-SPEAK 2025 Series 2. Universiti Teknologi MARA, Negeri Sembilan, pp. 96-97. ISBN 978-629-95953-5-9
Abstract

This systematic review and meta-analysis, registered under PROSPERO and following PRISMA 2020 guidelines, evaluates the efficacy of artificial intelligence in predicting and preventing injuries within high-contact sports. By synthesizing data from prospective cohort and observational studies across major databases, the research compares machine learning-based models against traditional predictive methods regarding injury incidence, type, and recovery outcomes. The study utilizes a structured PICO framework to analyze model accuracy, robustness, and interpretability, aiming to bridge the methodological fragmentation in sports medicine. Ultimately, the qualitative and quantitative findings provide insights into individualized risk factor identification and the practical utility of AI for enhancing athlete safety.

Item Details
Edit Item
Edit Item
Downloads & Files
[thumbnail of 135202.pdf]
Text
135202.pdf
Download (868kB)
Indexing & Metrics