Resting-state EEG Neural Signal Analysis In Multiple Magnetic Field FMRI

Electroencephalogram(EEG)is a physiological signal with high temporal resolution.Functional Magnetic Resonance Imaging(fMRI)is a high spatial resolution method for indirect imaging of neural activity.EEG-fMRI combines two powerful neural image technology with high temporal resolution and high spatial resolution,compared to Positron Emission Tomography(PET)and Magnetoencephalography(MEG).Microstate is a classical indicator for studying resting-state brain activity.Literature research shows that the current research on the microstate of the brain is still lack of microstate analysis under different magnetic field environments.This research selects EEG-fMRI synchronization technology to study the microstate of resting-state EEG in three different magnetic fields.First,the microstates are clustered differently and then the microstates under different categories are analyzed.The research contents mainly include: EEG artifacts removal under EEG-fMRI synchronous acquisition;microstate of three magnetic field environments;statistics of neurophysiological parameters.This research hopes to obtain the EEG signal interference degree of EEG-fMRI synchronous acquisition,and intensively study the microstate differences in three magnetic field environments.Firstly,this research performs artifact removal analysis on EEG signal in EEG-fMRI synchronous acquisition environment,and analyzes the effects of gradient artifacts,ECG artifacts and EOG artifacts.Specifically,BP Analyzer signal analysis software is selected to segment the EEG signal,measure the interval between different peaks in each piece of data,and determine the optimal peak interval and amplitude of each subject.The optimal base set algorithm in FSTAR plugin is used to remove gradient artifacts.The results show that the gradient artifacts has a great impact on the later data processing.This research uses independent component technology(ICA)to remove for ECG artifacts which is compared with the EEG signals of not completely remove the ECG artifacts.The results show that ECG artifact has a direct impact on the selection of independent components.The EEG topographic map of the independent component of not remove ECG artifacts has a lot of artifact spots.In addition,this research also analyzes the different neural networks corresponding to the microstate of resting-state EEG signals.Using the statistical analysis method,the neurophysiological parameters(average duration,frequency and transfer frequency)were selected to analyze the differences in the activation degree of neural networks under different magnetic field environments.The results show that:(1)average duration: the gradient field environment in the magnetic resonance chamber will affect the duration of the resting-state microstate,which will shorten the duration of various microstates.(2)Frequency: The frequency mean values of the gradient field environment and the interaction environment between the gradient field and the radio-frequency pulse are larger than the non-magnetic field environment in the visual network,the auditory network and the emergent network.The frequency of attention network in the gradient field environment is greater than the other two environments which is no significant difference in the field-free environment and the interaction environment between the gradient field and the radio-frequency pulse.(3)Transition probability: In the field-free environment,the neural network of the brain is mainly transformed by the visual network and the saliency network,and the activation frequency of the visual network is relatively small.In the gradient field environment,the conversion frequency of the three types of microstates is relatively balanced.In the interaction environment between the gradient field and the radio-frequency pulse,the probability of conversion to the auditory network is large.Finally,this research compares the artifact removal and microstate of EEG signals under different magnetic field conditions by EEG-fMRI synchronous acquisition method.This study has a certain role in promoting clinical psychiatric diseases and brain microstate studies.