Athlete psychological stress assessment method and system based on convolution and recurrent neural network

Abstract

The invention discloses an athlete psychological stress assessment method and system based on a convolution and recurrent neural network. The method comprises the following steps: S1, establishing a psychological stress data set based on heartbeat change of an athlete; s2, constructing and training a psychological stress monitoring model based on the psychological stress data set, wherein the psychological stress monitoring model comprises a convolutional network layer and a cyclic network layer; and S3, predicting a psychological pressure grade result of the athlete according to the psychological pressure monitoring model. The electrocardiosignal features can be automatically extracted through the convolutional neural network and the recurrent neural network, and the limitation of manual feature extraction of a traditional method is overcome to a certain extent.

Description

technical field [0001] The invention relates to the technical field of medical equipment and physiological signal detection, and more specifically relates to a method for assessing psychological stress of athletes based on convolutional neural networks and cyclic neural networks. Background technique [0002] At present, with the development of sports, each athlete has to bear more and more psychological pressure. Excessive psychological pressure has become an important problem affecting athletes' competitive state. Only by continuously monitoring the state of psychological stress in real time can we find excessive psychological pressure in the first place and effectively deal with it to avoid serious consequences. In addition, real-time monitoring of psychological stress can also help sports coaches assess the mental health of athletes and provide timely psychological counseling. Changes in heartbeat activity and changes in psychological stress are determined by the joint...

AI Technical Summary

Problems solved by technology

[0003] However, most of the existing psychological stress monitoring methods based on physiological parameters use manual feature extraction and statistical machine learning model classification. Such methods use short-term ECG signals to estimate psychological stress with low accuracy and are difficult to meet real-time requirements. Monitoring needs; a small number of methods based on deep learning only try to conduct research in other scenarios, which cannot meet the needs of athletes' psychological stress monitoring during training

Images