A New Approach of Intelligent Physical Health Evaluation based on GRNN and BPNN by Using a Wearable Smart Bracelet System

As physical health gradually becomes more of a concern for people, physical education for students is also more valued by schools and parents. The physical fitness test has become an important part of primary and secondary education evaluation. Traditional physical fitness test for students requires manual measurement and recording, as well as final calculation and verification, which is rather cumbersome. In this paper, we propose a new approach of wearable intelligent physical health evaluation. First, we use self-developed wearable smart bracelets to collect physiological data during exercise. Next, several key features are extracted from the physiological data. Finally, we analyze and model the collected data to obtain the students’ physical health evaluation system by constructing generalized regression neural network (GRNN) and back propagation neural network (BPNN). Experiment results not only show that the neural network based approaches presented in this paper have higher accuracy than multiple linear regression (MLR), but also prove that the proposed wearable intelligent physical health evaluation system can be used as an effective supplement or even an alternative to traditional physical fitness test.