Functional MRI Study Of Temporal Lobe Epilepsy By Dynamic Functional Connectivity

Background and objectEpilepsy(EP)is a chronic neurological disease that threatens people’s lives and health.It affects at least 65 million people worldwide.The prevalence rate is 4%to 7%in China.Seizures are characterized by sudden,transient and repetitive central nervous system dysfunction caused by abnormal excitability and synchronization of brain neuron discharge.Temporal Lobe Epilepsy(TLE)is the most common type of focal epileptic seizures.The active seizures originate in the temporal lobe and spread through interconnected brain networks confined to or beyond the temporal lobe.Previous studies have shown that TLE is a disease involving large-scale neural network disorders.In recent years,in order to explore more detailed information about the changes in the internal functional connectivity of the brain at resting state,more and more studies have used dynamic functional connectivity(dFC)method to characterize the time-varying characteristics of functional connectivity.Most studies are aimed at exploring local brain networks or the interaction of different brain networks,and few study the time-varying characteristics of the functional connections of large-scale TLE brain networks.In this study,independent component analysis(ICA)was used to divide the whole brain into 7 intrinsic connectivity networks(ICNs).On this basis,sliding time window(STW)technology was used to evaluate the changes in dynamic functional connectivity,to explore the differences of temporal features between TLE and healthy controls and their relationship with cognitive function,and to help to find potential neuroimaging markers.Methods and materials(1)Research subjects:This study used Siemens Prisma 3.0T MR scanner to collect 49 temporal lobe epilepsy(TLE)patients(mean age 29.12±13.02)and 34 volunteers recruited from the community(mean age 28.68±7.55).First,routine MRI scans were performed to exclude subjects with craniocerebral diseases,to ensure that all subjects meet the inclusion and exclusion criteria,and then performed subsequent fMRI scans.(2)Research steps:DPARSFA software was used to preprocess rs-fMRI data on the Matlab platform.Using the ICA method in the GIFT toolbox,the preprocessed rs-fMRI data is separated into 100 independent components(ICs)through data dimensionality reduction,and 40 ICs are selected to be distributed in 7 ICNs;through STW method and K-means clustering algorithm are used to analyze the dynamic functional connectivity in the seven brain networks.(3)Statistical analysis:Firstly,The SPSS 23.0 software was used to first compare the dFC indicators of the TLE group and the HC group,and then the TLE patients were divided into the left TLE(n=26)group and the right TLE(n=23)group according to the origin of the epileptic foci.Subsequently we compared the dFC indicators among the three groups.The SPSS 23.0 software was used to analyze the correlation between the indicators of dFC and the course of disease,the severity of epileptic seizures and the score of the behavioral scale,respectively.Results1.Results of clinical data:There were no significant difference in age,sex ratio and education level between the TLE group and the HC group(P>0.05).The MMSE,MES,MOCA-B scores of TLE group and the N1,N2,N3,N4,and recognition of AVLT in the results of the behavioral scale were lower than those in the HC group(P<0.05).2.dFC indicators for groups comparison:there were significant differences of the mean dwell time and the fraction time in the state 2 and state 4 between the TLE group and the HC group(P<0.05),and the remaining indicators were not statistically significant between the two groups(P>0.05).The mean dwell time and the fraction time in state 2 were statistically different between the left TLE group and HC group(P<0.05);there were no significant differences in three dFC indexes between the right TLE group and HC group,the left TLE group and the right TLE group(P>0.05).3.Comparison of dFC mode between TLE group and HC group:Compared with HC group,TLE patients stayed longer in state 2 and shorter in state 4.Compared with HC group,the functional network connections between the sensorimotor network and salience network,visual network,the functional network connections between executive control network and default mode network were enhanced in TLE group,while the functional network connections between the sensorimotor network and salience network,visual network,language network,the functional network connections between visual network and auditory network、salience network were weakened in TLE group.Compared with the right TLE,the dFC pattern of patients with left TLE has more significant changes.4.Results of the correlation analysis:there was a positive correlation between NHS3 and mean dwell time(r=0.341,P=0.017)and fraction time(r=0.414,P=0.003)of state 2;there was a negative correlation between NHS3 and mean dwell time and fraction time of state 4(r=-0.321,P=0.025);NHS3 was negatively correlated with numbers of transitions(r=-0.305,P=0.033).Conclusion(1)The functional connection of the brain has dynamic properties,and patients with TLE have a longer duration in the state of weak functional connection of the whole brain.(2)The dynamic time-varying characteristics of the brain network of patients with left TLE are more significant than those with right TLE.(3)The more severe the disease,the more time it takes in the sparsely connected state of the whole brain,and the fewer the numbers of transitions between states.