A Traceability Based Study Of EEG Activation And Causality In The Resting State With Eyes Closed And Open

At present,EEG technology is developing rapidly.With the continuous development of neuroimaging technology,we can explore the working principle and physiological mechanism of human brain more deeply.At the same time,as technology continues to evolve,in addition to a lot of different applications,such as electroencephalography(EEG),magnetoencephalography(MEG),and Magnetic Resonance Imaging(MRI)have made great progress.These techniques have played a huge role in the study of brain function and structure.Compared to other technologies,electroencephalography(EEG)is a non-invasive way of recording the activity of the brain's neurons,and it is very good at recording the signals and changes of the brain as it works.EEG signals are time series,which contain a lot of time domain information,so most researchers prefer to study EEG in time domain.In this paper,we combine EEG and inverse problem,introduce graph structure estimation algorithm to supplement the frequency domain analysis method after traceability,and discuss the variation characteristics of brain power in EC and EO from the frequency domain.The data of eyes opening and closing in the resting state have been widely studied as the experimental reference,but many characteristics and differences of eyes opening and closing are still unknown.The main dataset used in this paper is the Cuban dataset——Cuban Human Brain Mapping Project(CHBMP),to explore the causality between brain regions under the condition of eyes-open(EO)and eyes-closed(EC).The main work is as follows :(1)Preprocess the EEG data from the dataset.(2)Construct Head model,source model and transfer matrix.Use the spectral Structured Sparse Bayesian Learning method(s SSBL)to conduct activation imaging of electrophysiological source,and obtain the activation matrix.Finally compare the differences between the two conditions of eyesopen(EO)and eyes-closed(EC).(3)Use the Hidden Gaussian Graphical State-Spacemodel method(HIGGS)to estimate the connectivity of EEG,and obtain the highdimensional voxel matrix.(4)Three different graph structure estimation algorithms were introduced,and we compare their advantages and disadvantages to supplement the analysis method after traceability algorithm.The directed acyclic graph structure estimation algorithm was used to conduct causality estimation on the connectivity matrix of brain regions after partitioning,and the causality estimation matrix was obtained and tested so that the results were analyzed.In this paper,activation imaging and connectivity imaging analysis were innovatively carried out on the data of open and closed eyes in the resting state based on the information in the frequency domain of the EEG after traceability.By innovatively utilizing the algorithm of directed acyclic graph structure estimation,the traceability analysis method is supplemented,and the connectivity imaging analysis is expanded.Through the EEG activation and causal analysis,the mechanism difference between the brain regions under the two conditions of closed eyes and open eyes is revealed.