IoT-enabled real-time health monitoring system for adolescent physical rehabilitation

This study aims to develop an intelligent system leveraging Internet of Thing (IoT) technology to enhance the precision of youth physical training monitoring and improve training outcomes. A wearable device incorporating Micro Electro Mechanical Systems (MEMS) sensors is integrated to collect real-time motion data. Advanced signal processing and filtering techniques are employed to minimize noise interference and improve data accuracy. A particle swarm optimization support vector machine (PSO-SVM) algorithm is utilized to classify motion patterns. To evaluate the system's performance, experiments were conducted to assess motion pattern recognition accuracy, response time, real-time analysis capabilities, and system stability and capacity. The methods we use and the data we collect are from public datasets, do not involve privacy protection for adolescents, and have been approved by the institutional ethics committee. The system demonstrated a motion pattern recognition accuracy exceeding 95% and a response time consistently below 250 ms under various network conditions. Practical applications revealed the system's effectiveness in health monitoring, leading to improved physical fitness and positive rehabilitation outcomes for adolescent patients. This study offers an innovative digital solution for adolescent physical training and health monitoring. The system's strong application potential and valuable insights contribute to the advancement of related research.