Nonlinear Feature Analysis Methods And Its Application Of Eeg Si Gnals Based On Order Patterns

The human brain is an extremely complex physiological system,and EEG signals contain abundant nonlinear features.Extracting and describing these features,is the key to non-linear analysis methods,which is of great significance for studying the working mechanism of the brain,preventing and diagnosing neurological diseases.The methods based on order patterns is simple,efficient and can reflect the nonlinear dynamic features of the system.It is of great significance to the study of neural signals.This paper mainly focused on the order patterns of EEG signals,proposed and developed new methods,and applied them to the anesthesia monitoring and the coupling direction analysis in epileptic seizures.Firstly,a nonlinear algorithm of estimating depth of anesthesia was purposed,namely order patterns dissimilarity(OPD),which is based on the distance between probability of distribution of order patterns.The results of real data analysis and pharmacokinetics/pharmacodynamics(PK/PD)modeling showed that OPD is a more sensitive index than the traditional permutation entropy(PE),which can track the variation of the anesthesia state and correspond to the pharmacodynamic response of the brain to sevoflurane.Secondly,considering the disadvantage of symbolic transfer entropy(STE),the weighted probability distribution is introduced and developed into weighted symbolic transfer entropy(WSTE).The reliability of this method is numerically assessed with a coupled mass neural model.The simulations show that the directionality index estimated by WSTE can better highlight the coupling strength and direction between signals,and is more suitable for noisy data than that of STE.Finally,in the analysis of epilepsy intracranial EEG,it was found that there was a significant coupling direction from the anterior nucleus of thalamus(ANT)to the seizure onset zone(SOZ)in ictal process,indicating the information flow between the ANT and SOZ.This result is in line with the relevant theory of deep brain stimulation therapy,and has important significance for the diagnosis and treatment of epilepsy.