Comprehensive Analysis of Motion Performance Through Integrated Lower Limb Kinematic and Kinetic Parameters

As advancements in sports science and rehabilitation technologies continue, there is an increasing demand for comprehensive and accurate assessments of human motion performance. Traditional motion capture systems and isolated biomechanical metrics often fall short of providing a holistic perspective on an individual's movement capabilities across diverse activities. To address these limitations, we propose a novel approach to analyzing lower limb motion performance. By integrating a range of kinematic and kinetic parameters, including joint angles, angular velocities, muscle activation patterns, and ground reaction forces, our method comprehensively assesses an individual's lower limb motion capabilities. Using swimming as a case study, we collected IMU-based motion data, which was then processed through feature extraction techniques and analyzed using various machine learning models. Preliminary results demonstrate the effectiveness of this approach, particularly in activities like swimming, where our analysis can categorize athletic performance into different performance levels. Notably, the Random Forest model achieved an accuracy of 99.50% in distinguishing the subject's motion performance in three levels (Excellent, Sound, and Average). This methodology's comprehensive evaluation, adaptability to different environments, and nuanced performance differentiation support personalized training, injury prevention, and performance optimization in sports science and rehabilitation.