Research on the application of biomechanics analysis in optimizing physical education movement techniques

College and university students’ general health, fitness, and well-being are greatly enhanced by physical education. As various educational institutions work to improve the efficacy of their physical education programs, more evidence-based techniques are required. Biomechanics, to the movement or structure of student activities, provides insights into the efficiency and effectiveness of physical movements. This study aims to explore how physical activity movement skills can be systematically improved by the use of biomechanics analysis, leading to improved physical results and increased student participation in sports and fitness activities. In this study, a novel synergistic fibroblast-optimized malleable convolutional neural network (SFO-MCNN) is proposed to enhance teaching practices using a biomechanical framework that integrates movement analysis. The data collected from cameras that record students’ movements, capturing joint angles and body positions, as well as data from sensors are gathered from the Kaggle. The data was preprocessed using data cleaning and normalization for the obtained data. A system for assessing instruction quality was created using the suggested model and improved SFO. The findings show that the proposed algorithm has the greatest evaluations for average skill performance, physical fitness, student happiness, and physical education teaching efficiency. By comparing the outcomes with those of conventional approaches, the effectiveness of the proposed framework in improving physical education teaching techniques has been established.