Abstract
Temporal lobe epilepsy (TLE) leads to severe neuropsychiatric symptoms and cognitive impairment; however, the underlying mechanism is still not fully understood. Using degree centrality (DC), effective connectivity (EC), and multivariate pattern analysis (MVPA), we aimed to reveal alterations in brain networks and investigate neurofunctional symptoms in TLE patients. A total of fifty TLE patients and forty-seven healthy controls (HCs) were enrolled and underwent resting-state functional magnetic resonance imaging (rsfMRI) and neuropsychological testing. Cerebral functional activity and neuroimaging features were differentiated between the patient and HC groups based on rsfMRI data, DC, EC, and MVPA. Patients showed signs of depression, anxiety, and cognitive impairments in comparison to HCs. DC analysis revealed that patients had greater DC in the right inferior temporal gyrus and lower DC in the left and right cuneus than the HCs did. EC analysis revealed decreased EC from the right cuneus, right praecuneus, right superior occipital gyrus and right lingual gyrus to the left cuneus and increased EC from the left cuneus to the left calcarine sulcus and left middle occipital gyrus in TLE patients compared with HCs. The MoCA and Hamilton Depression Scale scores were correlated with impaired cerebral functional connectivity, according to correlation analysis. The MVPA classifier yielded an area under the curve of 0.91, an accuracy classification rate of 82.47%, a sensitivity of 85.11%, and a specificity of 80.00% according to the granger causality analysis(GCA) maps and receiver operating characteristic curve analysis. According to the current study, patients with TLE may have cognitive and depressive deficits as a result of abnormalities in the hub and related nodes in the default mode network and visual network. Moreover, the GCA might be a good imaging feature for the diagnosis of TLE.