Optimization strategies for physical education teaching movements using biomechanical analysis and deep learning techniques

Abstract

Accurately evaluating and improving students’ motor skills and physical quality are essential to physical education’s (PE) efficacy. Biomechanics, which explores the mechanical aspects of movements within biological systems, offers important insights into optimizing PE performance. At the cell molecular level, biomechanics is highly relevant. For example, during physical activities, cells experience mechanical forces. In muscle cells, the interaction between actin and myosin filaments is a fundamental molecular biomechanical process. These molecular events determine muscle contraction and relaxation, which directly affect students' motor performance. DL for optimization and the integration of biomechanical data are lacking from existing methodologies. The suggested technique improves the prediction of movement’s optimization and recognition in PE using the DK-LSTM, an optimized DL model. A comprehensive action dataset that included a variety of physical movements related to PE has been used. This dataset includes a wide range of activities and skill levels, ensuring robust model training and validation. The dataset undergoes preprocessing applying normalization to improve model performance. The Discrete Wavelet Transform (DWT) is used to extract the datasets. Using the processed dataset, the optimization method is utilized for training the DK-LSTM model and adjust its parameters for best results. The DK-LSTM model significantly improves movement recognition accuracy when compared to traditional methods. Recall (96.7%), accuracy (98.3%), precision (95.3%), and score (98%) are performance indicators that show how well a model that distinguishes between various activities can lead to PE teaching movements. Understanding cell molecular biomechanics can help refine training. By knowing how molecular forces impact muscle function, educators can design more suitable PE programs. This integration of knowledge and technology can lead to more precise evaluation and improvement of students' physical abilities in physical education.