PREDICTING PLAYER INJURIES USING AI ALGORITHMS BY ANALYZING MOVEMENT PATTERNS, WORKLOAD, AND BIOMETRICS IN UNIVERSITY FOOTBALL

Abstract

Musculoskeletal injuries are a pervasive challenge in university football,
hindering player performance and well-being. This paper explores the
potential of Artificial Intelligence (AI) algorithms to predict player injuries
by analyzing a combination of movement patterns captured through GPS
technology, workload metrics derived from training and match data, and
biometric indicators such as heart rate variability and sleep quality. To
achieve these purposes, sixty Kabaddi players were selected at random to
study. All the subjects were students of Bharathidasan University,
Tiruchirappalli - 620 024, Tamil Nadu, India. The ages of the subjects ranged
between 18 and 23 years. They were simplified into two groups. This study
proposes a methodology for collecting and integrating these data streams,
developing and training AI models (specifically focusing on ANCOVA for
statistical validation), and evaluating their predictive accuracy. This approach
aims to provide coaches and sports medicine staff with actionable insights for
proactive injury prevention strategies, ultimately contributing to improved
player health and performance in university football. The expected outcome
is a statistically validated AI-driven model capable of identifying players at
elevated risk of injury, paving the way for personalized intervention and load
management.