An improved deep learning mechanism for EEG recognition in sports health informatics

Abstract

A growing number of studies indicate that concussed athletes may have long-term residual electroencephalography (EEG) defects that can last up to ten years after the injury. With the use of conventional concussion screening techniques, these abnormalities are often ignored. As a result, returning to sports earlier can result in recurrent concussions, raising the risk of recurrent concussions with more severe consequences. This study uses deep learning methods to analyze the EEG signals of athletes. It then proposes and designs a channel attention module connected to the input layer of the convolutional neural network (CNN). The proposed approach automatically learns the EEG signals of different channels for recognizing the contribution of the task. The CNN is then connected to the recurrent neural network (RNN) for further processing. Based on this approach, this study combines the residual unit and the channel attention model to propose a convolutional recurrent neural network (CRNN) structure that is highly effective for EEG signal recognition. In this study, the EEG dataset of the Stanford research project has been used for experimental analysis. The performance of the proposed scheme is evaluated with the help of various performance measures. The experimental result shows that the proposed model improves the recognition accuracy from 82.58% of ResNet13 to 85.68% and attained excellent recognition accuracy of 91.05% by using CAMResNet13 + CRNN architecture.