Research on the application of biosensor technology in the detection and prevention of sports injury in college sports training

Injury detection plays a critical role in minimizing athlete downtime, ensuring safety, optimizing performance, and preventing long-term physical or mental consequences. In college sports, effective injury prevention and detection strategies enhance athlete safety, support peak performance, reduce healthcare costs, and contribute to sustainable athletic development programs. This research evaluates the application of biosensor technology in identifying injury risks, monitoring physiological metrics, and enhancing preventive strategies in college sports training to improve athlete performance and safety. A novel model, Egret Swarm Search-driven Scalable Deep Convolutional Neural Network (ESS-SDCNN), addresses the limitations of traditional approaches by combining SDCNNs with ESS algorithm for optimized feature selection, hyper parameter tuning, and real-time adaptability. Suitable data for injury detection and prevention include real-time physiological readings, motion sensor data, activity patterns, and injury records, with a focus on wearable technology. The Z-score normalization ensures consistent feature scaling. Independent Component Analysis (ICA) is used to extract hidden components from sensor data for improved feature representation. The SDCNN efficiently processes high-dimensional biosensor data, extracting spatial-temporal patterns related to injuries. The ESS algorithm further optimizes feature selection and hyper parameters, enhancing model accuracy, robustness, and adaptability for real-time applications. Results demonstrate that the hybrid ESS-SDCNN model significantly improves injury detection accuracy, enables faster convergence, and provides real-time monitoring and prevention insights. This approach enhances athlete safety, supports injury prevention, and fosters better performance outcomes in college sports training programs.