Study On The Phase-amplitude Coupling Characteristics Of Epileptic EEG Signals

Epilepsy is a chronic disease characterized by sudden abnormal discharge of neurons in the brain,resulting in transient brain dysfunction.With sudden and recurrent seizures,learning,life and work for patients with a great obstacle,seriously affect the quality of life of patients and their families.Because of the short duration of the attack,clinicians rarely witness the attack and increase the difficulty of diagnosis and treatment.EEG can record the abnormal discharge process of epileptic seizures,which has become the most important tool for the diagnosis and treatment of epilepsy.However,the long-term video EEG is usually recorded for several days or even weeks,which brings a lot of data to the epilepsy center.In this paper,seizure EEG as the research object,analysis of the characteristics of low frequency and high frequency amplitude phase coupling of EEG seizures,study the coupling relation between different low frequency and high frequency rhythms;phase amplitude coupling characteristics of the classification of interictal and ictal EEG based on automatic marking attack data section.Specific work includes:First of all,the Bonn EEG data set based on the coupling relationship between Epilepsy EEG rhythm rhythm amplitude and high frequency low frequency phase between the use of the modulation index(Modulation Index MI)to quantify the coupling intensity of each band,was put forward based on graph partitioning method of high frequency and low frequency range of the MI rhythm.The results showed that the MI value of interictal Gamma rhythm and multiple low frequency rhythm was significantly increased(p<0.01).There was significant difference in MI value between Theta rhythm and Beta rhythm,and the classification accuracy of the MI features of interictal and interictal data was 97%.Secondly,this paper according to the nonlinear characteristics of EEG based on permutation entropy characteristics of EEG seizures,the arrangement of entropy and standard deviation of the characteristics of combined EEG data sets for the Bonn,interictal and ictal data for feature extraction.The results showed that during the process of classification of Epilepsy EEG permutation entropy and standard deviation characteristics are complementary,in the calculation of the permutation entropy symbolic process,there is a loss of scale information,and the standard deviation characteristics can make up for the relevant information,the combination of the two can also cause seizures and epilepsy brain electric recognition rate of 97%.Thirdly,the phase amplitude coupling characteristics of EEG in epileptic dogs were analyzed.The results show that the coupling characteristics can be used for automatic classification of EEG,and the correct rate is 92%,and the coupling characteristics of several low frequency and high frequency rhythms have an impact on the classification results.Finally,the ictal EEG phase-amplitude coupling,such as permutation entropy features are used for data analysis,General Hospital of Shenyang military region in patients with preoperative evaluation of intracranial EEG recording of ictal automatic marker research;based on the single episode of the labeled data,can be very accurately and automatically marked with in other episodes,the classification rate was 95.5%.The clinical application of this research method can effectively reduce the burden of EEG analysis,and has a good application prospect and practical significance.