The Study Of Electroencephalograph Network In Patients With Drug-resistant Temporal Lobe Epilepsy

Objective:To explore whether there are differences in electroencephalograph network between drug-resistant temporal lobe epilepsy,drug-responsive temporal lobe epilepsy and normal controls by graph theory analysis.Methods:From January 2016 to February 2019,patients with magnetic resonance negative temporal lobe epilepsy who were hospitalized in the Department of Neurology,Drum Tower Hospital Affiliated to Medical College of Nanjing University were selected.According to the classification of drug treatment effect which is defined by International League Against Epilepsy,the patients without clinical seizures and those who failed in drug treatment were divided into drug-responsive group and drug-resistant group respectively,and the patients with uncertain curative effect were discarded.Normal volunteers matched by sex and age without family history of epilepsy or other neuropsychiatric disorders were recruited as controls.Video electroencephalogram data were collected and preprocessed by MATLAB 2016b software.Then the network is constructed according to the graph theory analysis method.The network parameters(clustering coefficient,characteristic path length,global efficiency and local efficiency)of four bands(delta,theta,alpha and beta)are obtained by computer algorithm.The network parameters of the three groups are analyzed by one-way ANOVA,and the differences among the three groups are compared.Then the network topology map(network P-value connection map)was drawn to visualize the differences and distribution of brain network connections in each group.Results:(1)The brain network attributes of drug-resistant temporal lobe epilepsy patients changed.The network parameters of drug-resistant group were significantly different from those of normal group and drug-responsive group.Compared with the normal group,in the theta and beta bands,the clustering coefficient and local efficiency of drug-resistant group increased,the characteristic path length decreased,and the global efficiency increased,which indicated that the local aggregation and global integration increased,and the attributes of small-world network increased;in the alpha band,the clustering coefficient and local efficiency decreased,the characteristic path length increased,and the global efficiency decreased,which indicated that the local aggregation and global integration decreased,and the attributes of small-world network decreased.Compared with drug-responsive group,the clustering coefficient and local efficiency of drug-resistant group increased in theta,alpha and beta bands,the characteristic path length decreased,and the global efficiency increased,indicating that the local aggregation and global integration increased,and the small world network attributes increased.(2)The brain network topology of drug-resistant temporal lobe epilepsy patients changed.Compared with the normal group,the short-range anterior-head connections were enhanced in the delta,theta and alpha bands in the drug-resistant group;the long-range anterior-head connections were enhanced in the beta band;and the short-range posterior-head connections were weakened in the four bands.Compared with drug-responsive group,the short-range temporal junction was enhanced in drug-resistant group in the four bands,especially in the right temporal region and its junction with adjacent brain regions or contralateral brain regions.Conclusions:The brain network attributes and network topology of patients with temporal lobe epilepsy were changed.The changes of brain network in drug-resistant temporal lobe epilepsy were most significant.The brain network of drug-resistant temporal lobe epilepsy belongs to the small world network.Electroencephalograph network research can help to identify drug resistance in temporal lobe epilepsy.