Author(s)

Yutong Liu

Corresponding Author

Yutong Liu

Abstract

This paper proposes an injury risk warning system for basketball players based on physical modeling and statistical analysis. This paper combines sensor technology and deep learning technology to propose a deep graph convolution model called GCN-INJ. It uses a variety of information such as players' physiological parameters, training intensity, and game data to capture the mutual influence between players through a graph convolutional network (GCN) to more accurately predict the risk of individual injuries. Experimental results show that the model is superior to traditional statistical models in both prediction accuracy and stability. The effectiveness of the model is verified through visual analysis of a large amount of data, and its limitations in practical applications are explored. Future research will focus on further optimizing model performance, introducing more dimensional data sources, improving prediction accuracy, and developing user-friendly applications so that coaches and athletes can monitor and manage injury risks in real time, thereby better supporting sports performance and health maintenance.

Keywords

Physical Modeling, Statistical Analysis, Injury Risk Warning, Graph Convolutional Neural Network, GCN-INJ