Injury prevention in sports requires comprehensive monitoring of players' physical and psychological status. While numerous studies have provided valuable insights into specific facets of injury risk, these efforts have often remained fragmented in s...
Injury prevention in sports requires comprehensive monitoring of players' physical and psychological status. While numerous studies have provided valuable insights into specific facets of injury risk, these efforts have often remained fragmented in scope. Considering the multifactorial nature of injuries, integrating diverse perspectives into a unified framework is essential for proactive prevention strategies. To address this, this study proposed a multifaceted analytical approach that combines external load, internal load, and athlete self-report measures. Specifically, the multifaceted analysis in our framework refers to providing decision-support indicators at distinct time points, particularly before and after training. The research questions to guide our study reflect our proposed approach. First, given the observation that the same physical activity may lead to varying fatigue among different players, we ask whether consistently higher perceived fatigue from similar physical activities correlates with greater susceptibility to injury, thereby proposing a model to predict internal load based on external load after training. Second, recognizing the difference between a coach's intent and a player's perception during planning and execution of the training, we question whether predicting the difference could facilitate injury prevention, leading to a model predicting the difference between rate of intended exertion and rate of perceived exertion before training. Finally, considering the critical role of physical and psychological status in injury risk, we inquire if injury probabilities based on athlete self-report measures could support decision-making, thus proposing a model for predicting lost time injuries on the day before training.
To address these questions, we proposed three artificial intelligence-based models contributing towards a data-driven approach to injury prevention in sports. This study makes five major contributions. First, it highlights the importance of integrated monitoring across external load, internal load, and athlete self-report measures. Second, it proposes a model for predicting fatigue corresponding to physical activity. Third, it proposes a model for predicting player responses to intended training intensity. Fourth, it provides a model for injury prediction based on physical and psychological status. Finally, it delivers a comprehensive framework that enables multifaceted analysis of injury prevention and management before and after training.