Analysis Of Magnetoencephalography Based On Symbolic Transfer Entropy And Multiscale Permutation Entropy

Depression is a common and serious medical illness that negatively affects how people feel,the way people think and how people act.At its worst,depression can lead to suicide.Depression has various clinical manifestations and complex pathogenesis.In researches of depression,magnetoencephalography(MEG)receives extensive attention due to its high temporal and spatial resolution.Brain activity is a typical nonlinear complex system,and it is of great significance to study the nonlinear dynamic of brain magnetograph signals.In this paper,experiment apply multiscale permutation entropy(MPE),symbol transfer entropy(STE)and signal frequency band decomposition combined with permutation entropy method to measure the nonlinear dynamics of MEG of depressive patients and healthy subjects.Firstly,in the analysis of MEG with multiscale permutation entropy algorithm,we analyze the influence of different parameters on MEG analysis in patients with depression and healthy people.The results show that the best result is obtained when the scale factor is 8 and the embedding dimension is 4.And the MPE of healthy people is obviously larger than that of depression patients,especially in the frontal area.Secondly,combined with the basic-scale symbolization and transfer entropy,STE for coupling detection,shows satisfied distinctions in the frontal channels in symmetrical information transfer analysis,and in individual analysis,MEG in right frontal region show better results,and all show lower STE in the depressive patients.Thirdly,MPE for nonlinear complexity of MEG in different frequency domains is extracted.For patients with depression,their MEG have low complexity in low frequency while have high permutation entropy in high frequency.From tests of MPE and STE in the brain magnetograph activities,the research also find that depressed patients are significantly influenced by different emotional stimuli and MEG with negative emotional stimuli have low complexity than that with positive emotional stimuli.Through researches on the nonlinear dynamics in different areas of brain and under different emotional stimulations,the significant distinctions between the depressive patients and healthy subjects provide valuable information for clinical diagnosis.