Research On The Identification Potential Markers In Magnetoencephalography Of Schizophrenia Patients Based On Nonlinear Analysis

Schizophrenia is a very serious,highly dangerous chronic psychiatric disorder that affects a person’s thought processes,sensory processing,and behavioral expression,and its pathological pathogenesis are unknown.Looking for potential markers in brain signals of schizophrenia patients is a hot topic in its research field.With the advantages of high time resolution and high spatial resolution,Magnetoencephalography(MEG)plays an important role in the study and diagnosis of schizophrenia.The human brain is a very large and complex nonlinear system,so MEG has strong nonlinear properties.In this paper,nonlinear analysis methods were used to explore potential markers in MEG of schizophrenia patients from the aspects of complexity,recurrence properties,functional network,coupling between brain rhythms,and properties of phase space geometry.The results of the study discovered the abnormal nonlinear feature information in the MEG of patients with schizophrenia,and these features can be used as potential markers in the MEG of patients with schizophrenia.The specific work of this paper is as follows:First,the nonlinear complexity of the resting-state MEG in schizophrenia patients was studied using multiscale weighted permutation entropy,the results showed that the MEG of schizophrenia patients had higher nonlinear complexity.The findings of the different brain region studies showed that the complexity differences between schizophrenia patients and healthy controls were the strongest in the frontal and occipital regions,while there were almost no differences in the central regions.The results of the different frequency studies showed differences in nonlinear complexity between schizophrenia patients and healthy controls mainly in the Theta band(4-8 Hz),Alpha band(8-12 Hz),and Beta band(13-30 Hz).Second,multiscale multidimensional recurrence quantification analysis algorithm was designed and the nonlinear recurrence characteristics of resting-state MEG of schizophrenia patients were studied by using this algorithm.The results of the study showed that the multiscale multidimensional recurrence quantification analysis well excavated the characteristics of the nonlinear dynamic characteristics of MEG in schizophrenia patients evolving over time.The recurrence plots of MEG in schizophrenia patients had a significant multiscale structure,and their recurrence values were significantly lower than those of healthy controls.This implies that the nonlinear dynamics of MEG in schizophrenia patients have lower predictability and laminarity,which means that the dynamical properties of the brain during resting state in schizophrenia patients are in a chaotic state.Third,the functional network of MEG in different frequency bands in schizophrenia patients under different stimulation conditions was studied using complex networks.Results revealed that the effects on the topological parameters of the functional network in schizophrenia patients negative stimuli were found to be concentrated in the Alpha 1 band(8-10 Hz)and Beta band,positive stimuli were concentrated in the Alpha 1 band,and neutral and Gray Cross stimuli were concentrated in the Theta and Alpha 1 bands.Subsequently,multidimensional cross recurrence quantification analysis was used to study the dynamical coupling between the MEG in different frequency bands of schizophrenia patients.It was found that the difference of dynamic coupling between different frequency bands of MEG in members of the two groups was mainly in the coupling of Theta and Alpha 1 waves,Theta and Gamma waves,and Delta and Gamma waves.It was also found that the dynamic coupling between different bands of MEG varied more significantly in schizophrenia patients during positive stimuli.Finally,multidimensional recurrence networks were used to study the geometric properties of MEG in phase space in different frequency bands of schizophrenia patients under different stimulation conditions.It was found that in the high-dimensional phase space,the MEG of schizophrenia patients had higher global recurrence rates,higher invariant objective trajectories,higher state transfer efficiency and higher phase space separability,which indicated that the brain activities of schizophrenia patients had highly complex features under stimulation conditions.The results also revealed that the effects on the geometric features of the phase space of the MEG in schizophrenia patients with negative stimuli were concentrated in the Alpha 1 band and Beta band,in positive stimuli and gray cross stimuli were concentrated in the Alpha 1 band,while neutral stimuli had no effect on them.In conclusion,the information of abnormal features in MEG data of schizophrenia patients extracted by nonlinear analysis can be used as markers or auxiliary markers for prediagnosis of schizophrenia,which is helpful for the study of markers in MEG of schizophrenia patients.