Basketball is a dynamic sport, characterized by fast moves, powerful hops, and constant changes in direction, which contributes to its great intensity and excitement. However, these same features elevate athletes to a risk of various types, which include the most common ones such as ligament tears, knee problems, and ankle sprains. The knowledge of the risk factors associated with particular movement patterns and environmental conditions is made possible by knowledge of the biomechanical qualities of these injuries is crucial if one has to develop effective preventive and rehabilitation strategies. The purpose of the study is to establish the biomechanical characteristics of basketball player injuries and their application in sports rehabilitation. The study proposed a novel Tunicate Swarm Optimized Flexible Extreme Boosting (TSO-FXGBoost) to predict the injuries of basketball players and their sports rehabilitation. Player’s motion data capture sessions utilize cameras and sensors to record their biomechanics during basketball activities. A Gaussian filter was employed to process the data to eliminate the noise present in the biomechanical data. Principal component analysis (PCA) served as a dimensionality reduction approach to extract relevant features from the pre-processed data. The results demonstrate that certain biomechanical features have a strong correlation with the occurrence of injuries, which indicates great potential in the strategies of prevention of injuries. In a comparative analysis, the suggested approach performs various assessment metrics such as accuracy (98%), recall (96.2%), precision (98.49%), and F1 score (97.8%). The suggested approach and rehabilitation strategies can be customized to each player’s unique biomechanical profile, improving rehabilitation times and lowering the risk of re-injury.
Read full abstract