This study addresses the high incidence of injuries, particularly lower extremity injuries, among university-level football players and the limitations of traditional injury risk assessment methods that rely on physical examinations and subjective reports. With advancements in wearable technology and biomechanical assessment tools, large volumes of physiological, biomechanical, and performance-related data can be collected; however, these data are not yet effectively utilized for injury prediction. The study aims to develop a machine learning–based injury risk prediction model that integrates diverse data sources to identify key indicators of injury risk. By applying advanced machine learning algorithms, the study seeks to provide accurate, data-driven, and personalized insights that can support proactive injury prevention strategies, assist coaches and medical professionals in targeted decision-making, and ultimately reduce injury incidence while enhancing the performance and overall well-being of university-level football players.