A Study of Computerized Techniques for Health Monitoring and Risk Assessment of Athletes in a Big Data Environment in Sports

Abstract

In this study, an athlete health monitoring system based on pulse wave signals is designed to address the problems of low data prediction accuracy and poor real-time performance of traditional athlete health monitoring techniques. The system collects and preprocesses the photoelectric volumetric pulse wave signals of athletes by applying a photoelectric volumetric pulse wave acquisition sensor device. Firstly, the differential thresholding method, the pulse map area method and the waveform fitting method are used to extract the characteristic parameters in the time domain, then the power spectrum analysis and the cepstrum analysis are used to extract the characteristic parameters in the frequency domain, and the HHT method based on the time-frequency hybrid domain is also used to extract the characteristic parameters of the pulse signals. Finally, the Hidden Markov Model is introduced to convert the pulse wave signals of the athletes that need to be evaluated into a sequence of physiological signal features, and the risk assessment of the athletes' health status is carried out. The results of the study show that the system in this paper can significantly differentiate the health status of healthy and diseased populations, and the fitted curve of the recognized healthy athletes' physical health level rises by 0.065, which is in line with their actual health status level. The system in this paper provides a scientific basis for injury prevention and sports performance optimization of athletes, and has important application value in the sports world.