Research on Sports Injury Prediction Model for College Badminton Specialized Students
International Journal of Healthcare Information Systems and Informatics · 2026-07-02 · 21 (1) · 1-19
DOI: 10.4018/ijhisi.414589 ↗Abstract
Aiming at the realistic pain points of “high-intensity training, younger injury risk, and digital management” in badminton in colleges and universities, this study constructs a sports injury prediction and intervention model constructed on real-time load data. Based on a 12-week prospective queue, wearable devices are used to collect sports load data, the light gradient boosting machine-attention model is used to achieve high-precision prediction (area under curve = 0.95), and a double-cycle mechanism of “double threshold adjustment + weekly calibration” is designed for dynamic intervention. The practice shows that the system can reduce the overall damage rate by 32%, and the high-risk damage rate from 29.4% to 17.1%. The research provides a replicable and extensible data-driven early warning scheme for campus sports.
EUVIMED