EEG Signal Analysis Of Mild Cognitive Impairment Based On Nonlinear Analysis Method

Mild Cognitive Impairment(MCI)is the early stage of Alzheimer’s disease(AD).By analyzing the patient’s EEG signals,the results can serve as theoretical support for the diagnosis of MCI.This article uses three nonlinear data analysis methods to extract the complexity value,determinism value,and entropy value of single channel and multichannel EEG signals,and conducts in-depth analysis on the differences in EEG signals between the two groups of participants and their correlation with cognitive function.These three characteristic indicators are expected to become EEG biomarkers for early diagnosis of MCI,providing assistance for the diagnosis and treatment of MCI.Firstly,an Increment Lempel-Ziv Complexity algorithm is proposed to improve the problem of less classification in the Lempel-Ziv complexity quantization process and improve the algorithm’s anti-noise performance.The algorithm was used to analyze the EEG signals of 19 channels and extract the complexity features of the EEG signals.Analysis shows that the complexity value of the control group is higher than that of the MCI group,and its standard deviation is smaller than that of the MCI group.It shows significant differences in F4 and Fz channels.The complexity value of EEG signals is expected to become the theoretical basis for assisting in the diagnosis of MCI.Secondly,the mapping method in Increment Entropy is introduced,the Increment Recurrence Plot algorithm is proposed,and the information in the Increment Recurrence Plot is quantified through recurrence quantification analysis.This method greatly improves the ability of traditional Recurrence Plot to analyze data certainty.Determinism values of19 channel EEG signals were extracted at both short and long scales.The results showed that the determinism values of the MCI group was higher than that of the control group,and showed significant differences at short scales;At long scales,most channels exhibit correlation with neural scale test items.Finally,the Refined Composite Multivariate Multiscale Dispersion Entropy algorithm was used to analyze multi-channel EEG signals in different brain regions.The entropy changes of six brain regions were extracted at short and long scales,and the relationship between each brain region and brain function was analyzed.The results showed that in the short scale,the entropy value of the MCI group was lower than that of the control group,while in the long scale,the opposite was observed;At long scales,the F,C,and P regions show significant differences.