Multivariate And Multiscale Complexity Analysis Of Eeg Signals In Mild Cognitive Impairment

Mild Cognitive Impairment(MCI),also known as early dementia and isolated memory disorder,the core symptom of MCI is cognitive decline,most likely to develop Alzheimer’s Disease(AD).Therefore,it is of great significance to study the correlation between MCI EEG signals and cognitive function.In this paper,we study the multi-scale recursive graph quantitative analysis and multi-scale fuzzy entropy method,and further extend the multi-recursive graph cross-analysis and multi-scale multi-scale fuzzy entropy method,and use the coupled model to carry out simulation analysis,and analysis the nonlinear characteristic of MCI EEG Signals.Firstly,the quantitative analysis methods of multi-scale recursive graph are studied:Recurrence Rate(RR),Determinism(DET),Average Diagonal Line Length(DLL)and Multiscale Fuzzy Entropy(MFE)algorithm.Coupling coefficients and thresholds of RR,DET,DLL and Fuzzy En are simulated by using coupling model.The real experiment shows that DET,DLL and FuzzyEn values have good consistency.Sex.Using the DET,DLL and FuzzyEn values to analyze the actual MCI EEG signal analysis,it was found that the DET,DLL and FuzzyEn values of the MCI group and the nMCI group were significantly different at small scales.Pearson linear correlation analysis was performed on the obtained DET,DLL,FuzzyEn value and neuropsychological test scores.The results showed that there was a significant correlation between the two.Secondly,the multi-recursive graph intersection algorithm is studied and verified by the coupled model.It is found that it has good coupling and noise resistance.In the complexity analysis of EEG signals applied to MCI,it was found that the CRPDET value of EEG signals in the MCI group was higher than that in the nMCI group,and the correlation with the neuropsychological test scores was analyzed.It was found that the MCI group and the nMCI group showed Very significant difference.Finally,the multivariate multi-scale fuzzy entropy algorithm is studied,and the coupling model and 1/f noise are used to verify its coupling and anti-noise.It is found that the multivariate multi-scale fuzzy entropy algorithm has good coupling and anti-noise andcan reflect signal correlation.In the complexity analysis of EEG signals applied to MCI,it was found that there were significant differences in the complexity of EEG signals between the two groups.By using the Pearson linear correlation method,it was found that the complexity and cognitive function of EEG signals were found.There is a significant correlation.