Research On Affective Disorder Regulatory Mechanisms And Key Technologies Based On EEG Biofeedback

Depression,as one of the affective disorders,has quietly become the fourth largest health risk in the world.The causes of depression are complicated,which may include factors such as genetics and the living environment.Patients with depression often experience low mood,slow thinking,decreased willpower,cognitive impairment,and physical symptoms including fatigue,pain,and sleep disorders.In severe cases,they may even experience delusions,pessimism,and suicidal behavior.In recent years,the number of patients with depression has risen rapidly and causes serious harm not only to individuals,but also to families and society.It is now recognized that the most effective way to treat depression is early detection and early intervention.However,due to the lack of objective and effective diagnostic physiological indicators,current depression diagnosis methods are high cost and has high rate of misdiagnosis.At the same time,drug treatment will may also bring serious side effects.Therefore,to address the above challenge,this study explores a new method of depression diagnosis and intervention using EEG technology.The main contributions and innovations are as follows:1.A pervasive EEG experimental paradigm for depression is proposed.Electrode sites Fp1,Fp2,and Fpz,according to the international 10-20 system are selected as the brain wave acquisition position,since the prefrontal cortex is closely related to human cognition and emotion.The restingstate EEG data and EEG data under audio stimulus were collected using a pervasive three-electrode EEG collector.An EEG database of 256 subjects,which includes 152 depression patients and 113 normal controls were constructed.2.An effective depression EEG feature set was determined.After denoising the original collected data,220 linear and nonlinear features were extracted and used to build the original feature matrix.Subsequently,feature selection was performed using both Wrapper class and Filter class feature selection algorithms.Then,the results of feature selection were classified by six classification algorithms including Bayesian Network,Support Vector Machine,K Nearest Neighbor,Logistic Regression,Random Forest,and Decision Tree,and finally verified by using 10-cross validations.3.A depression EEG classification model based on case retrieval was constructed.In this paper,a case retrieval-based method for depressive EEG recognition and a new case retrieval weight fusion method based on genetic algorithm were proposed.Three common weight setting algorithms PCA,Discrete Coefficient,and Information Entropy were fused by using genetic algorithm,and a new weighted fusion algorithm GA_PCE was proposed.Then it was performed on 23 disease data sets in the UCI public database for verification.The results show that the stability,generalization and accuracy of the GA_PCE weight fusion method are superior to the other three weight setting methods.In addition,we also combined the standard Euclidean distance similarity calculation algorithm with GA_PCE and built a case-based depression EEG classification model.The classification accuracy rate is 91.36%,which is higher than that of commonly used machine learning methods.4.The mechanism of depressive disorder regulation based on EEG feedback was proposed.Based on this,an intervention system for depressive disorder based on EEG feedback was constructed and the effectiveness of the system was verified.First,based on the biofeedback,the principle and working circuit of the depressive disorder intervention mechanism based on EEG feedback is put forward.Then,combining with virtual reality technology,a VR-based depression intervention framework using three-electrode EEG collector is proposed.Finally,experiments were conducted and results show that all patients undergoing intervention had different degrees of improvement in their EEG indicators.The scale-based evaluation also confirmed that the depressive symptoms of the intervention group improved after 3 weeks and significantly improved after 6 weeks(P<0.05)compared with the non-intervention group,further verifying the effectiveness of the depressive disorder intervention system based on EEG feedback.In summary,this article explores the key technologies of EEG feedback-based affective disorder screening methods and regulatory mechanisms and solves the key issues of pervasive EEG experiment design,depressive EEG feature selection,and depressive EEG identification methods.This paper was to study new methods of detection and intervention for depression,and to propose a mechanism of depression regulation based on EEG feedback.This paper also constructed a depression intervention system based on EEG feedback to validate its effectiveness.Experimental verification comprehensively explored and clarified the theoretical mechanism and feasibility of new technologies.This study has enriched the theoretical basis of the mechanism of depressive disorder intervention based on EEG feedback and has made beneficial applications to help promote the further development and application of this method.