Synchronous Intensity And Direction Analysis Of Regional EEG Signals

The synchronic phenomenon of EEG is a kind of physiological features which is transmitted,interrelated and coordinated with each other between various regions and it is a phenomenon prevalent in EEG.The simultaneous study of EEG signal in each brain region can help to explore the principle of information transmission between different brain regions and the pathogenesis of neurological diseases(dementia,mild cognitive impairment,etc).In this paper,we mainly study the synchronization intensity and synchronization direction of diabetic impairment by the method of ordinal pattern,permutation mutual information and permutation condition mutual information algorithm.Firstly,this paper introduces the basic knowledge of EEG and the commonly used EEG analysis method,and introduces the neuropsychological scale in detail.Secondly,this paper studies the ordinal pattern,introduces the statistical methods of the ordinal pattern and the probabilistic calculation of the ordinal pattern.Combining the ordinal pattern and the mutual information,the two-channel EEG signal strength analysis method is developed,through the simulation experiment of Henon model,it is shown that the method of permutation mutual information has strong robustness to noise,and can analyze the synchronization intensity of dual channel EEG signal well.In order to analyze the synchronization strength of the multi-channel EEG signal,In this paper,the method of permutation mutual information and S-estimator are combined to develop a global synchronization index algorithm,and the algorithm is applied to the analysis of diabetic impairment EEG signal.Finally,we introduce the Permutation Conditional Mutual Information in detail.Through the simulation experiment of Rossler model,we can explore the influence of each parameter on the algorithm performance,including embedded dimension,data length,delay time and The Influence of ? Value on Algorithm.The simulation results show that the coupling direction index of the mutual information estimation of the permutation condition can accurately reflect the relationship with the coupling coefficient,and the anti-noise performance is prominent.And then applied to analyze the coupling direction analysis of EEG signals of different frequency regions in diabetic patients with aMCI.The results were consistent with pathology,and there are significant differences in each band except the alpha2.