Study On Characteristics Of Sleep EEG In Depressed Population

As a kind of mental disorder with high incidence,high risk and difficult cure,depression is always accompanied by obvious symptoms such as abnormal brain activity and mood transition.EEG signals,as a physiological electrical signal capable of recording brain activity and emotional state information,are often used in various related studies on diseases of emotional conversion disorders.As a kind of EEG signal,sleep EEG signal has the characteristics of signal stability,interference,real nondisguise and easy collection,which is more conducive to the study of depression and other related diseases.Therefore,based on the sleep EEG signal,this paper puts forward the hypothesis that “the depression group and the normal population have differences in sleep EEG signals”,and the difference of sleep EEG signals between depressed and normal people through machine learning and data analysis.Sex was analyzed to verify the hypothesis.The main research contents and research results are as follows:1.In this paper,based on the data characteristics of sleep EEG signals and the collection method of open database EEG data,a sleep EEG data acquisition system is designed.According to the characteristics of EEG data of depression patients,sleep EEG data of 6 electrodes(F3,F4,C3,C4,O1 and O2)were collected from 36 subjects(18 depressed patients and 18 normal people).Established a sleep-based EEG database based on depression.It laid a data foundation for the study of differences in sleep EEG characteristics in depressed populations.2.This paper deals with and applies sleep EEG data sets for depressed populations.The collected sleep EEG data were divided into awake period,REM period and NREM period according to the AASM sleep staging model,and the appropriate method was selected for data sample selection;reference to the collection method of sleep EEG data and the time complexity of the data model The denoising of original sleep EEG signals based on FIR filter and Kalman filter and the segmentation of data samples based on time window method are realized.According to the size of data samples after segmentation and the data characteristics of sleep EEG signals,the realization is realized.The data feature is extracted,and the feature selection algorithm is used to select the data features.Finally,according to the data characteristics of the feature samples,the data processing based on data standardization and data equalization is realized.Through the data processing flow,the processing and application of sleep EEG signals in depressed people can be realized.3.Based on the feature selection algorithm and data analysis method,this paper verifies the hypothesis that “the depression group and the normal population have differences in sleep EEG signals” and analyzes and studies the influence of feature correlation on the difference performance.The results of the feature selection algorithm to verify the differences indicate that one or some of the characteristic groups of sleep EEG signals in depressed populations may differ from the normal population.Data analysis methods verify the difference results show that the lead,frequency band and feature types of sleep EEG signals can affect the experimental results of the data model,and the correlation between different features will also affect the classification accuracy of the model.In other words,the sleep EEG signals of depressed people and normal people have different differences in different dimensions,which is consistent with the hypothesis that “the depression population and the normal population have differences in sleep EEG data”,further verifying Assumption.In summary,this paper explores the treatment and application of sleep EEG datasets in depressed populations and analyzes and studies the differences in sleep EEG signals between depressed and normal people through this data set and completes the hypothesis.The effective verification has enriched the theoretical basis of the classification model of depression based on sleep EEG signals,and further promoted the development and application of research on mental disorders such as depression based on sleep EEG signals.