基于前额脑电信号伪迹去除的可穿戴心理健康疲劳检测仪设计研究

Design and research of wearable mental health fatigue detector based on forehead EEG artifact removal

脑电信号常常受到多种伪迹的干扰,而传统的前额伪迹去除方法无法有效去除伪迹,也无法准确分析学生的心理状态和情绪变化。为实现前额脑电伪迹信号的有效去除,并对学生的心理状态进行有效检测,研究对可穿戴检测仪进行改进优化,通过引进快速寻优迭代算法实现脑电信号的有效检测,最终设计出一款面向大学生心理健康疲劳检测的可穿戴检测仪。在对实验环境进行设置后,经实验可知,研究算法的SNR值最高,其最大值为6.1,说明研究提出的方法可以更好地分离出有效的脑电信号。研究方法的MSE值最低,MSE区间为59~69,说明研究方法对电脑信号的重建更为准确,保留了更多的有用信息。第一天考试前γ-VP特征值位于50~80之间,而考试后γ-VP特征值显著下降,处于40左右。综上可知此次基于改进伪迹去除的心理健康疲劳检测仪可以有效去除伪迹,准确分析学生的心理健康和疲劳状态。

Electroencephalogram(EEG)is often disturbed by many artifacts,but the traditional forehead artifact removal method can not effectively remove artifacts,and can not accurately analyze the psychological state and emotional changes of students.In order to effectively remove frontal EEG artifacts and detect the psychological state of students,research has been conducted on improving and optimizing wearable detectors.By introducing a fast optimization iterative algorithm to achieve effective detection of EEG signals,a wearable detector for detecting mental health fatigue in college students has been designed.After setting the experimental environment,it can be seen from the experiment that the SNR value of the research algorithm is the highest,and its maximum value is 6.1,indicating that the proposed method can better separate effective EEG signals.The MSE value of the research method is the lowest,and the MSE range is 59-69,indicating that the research method is more accurate in the reconstruction of computer signals and retains more useful information.Before the first day of the exam,theγ-VP characteristic value was between 50 and 80,while after the exam,theγ-VP characteristic value decreased significantly,and was around 40.In conclusion,the mental health fatigue detector based on improved artifact removal can effectively remove artifacts and accurately analyze students’mental health and fatigue state.