Research on classification evaluation model of physical health based on biomechanical parameters for middle school students based on BP neural network

Objective: To construct a classified evaluation model of middle school students’ physical health based on biomechanical parameters and to achieve accurate intervention of adolescents’ physical health through biomechanically informed strategies. Methods: Cluster sampling was used to select 1261 male students’ physical health samples from eight junior middle schools in Jinhua City, Zhejiang Province from 2022 to 2023 as a data set. Second-order clustering and BP neural network algorithm were used to establish a physical health classification evaluation model for middle school students, and the accuracy of the model was verified. Results: The classification model of physical health evaluation of middle school students has an accuracy of about 90%. The integration of biomechanical data improved the model's ability to identify at-risk students and tailor interventions. Conclusion: The classification model has high prediction accuracy and good generalization performance, and provides scientific data support for subsequent accurate intervention of students’ physical health in different categories. By incorporating biomechanical insights, the model not only identifies health risks but also informs targeted, biomechanically sound interventions to improve physical health outcomes. This study bridges the gap between physical health assessment and biomechanics, offering a novel approach to promoting adolescent well-being.