Language and Memory Network Alterations in Temporal Lobe Epilepsy: A Functional and Structural Connectivity Study

BACKGROUND AND PURPOSE: This study evaluated preoperative alterations and postoperative reorganization of the joint language-memory network (LMN) from the perspective of resting-state functional and structural connectivity in Temporal lobe epilepsy (TLE). Graph theory and machine learning approaches were employed to explore automatic lateralization.

MATERIALS AND METHODS: Resting-state fMRI and DTI data were obtained from 20 healthy subjects and 35 patients with TLE. Functional and structural connectivity were calculated within the LMN before and after temporal lobectomy. ANOVA was performed to identify significant connectivity differences between groups. Four local graph measures were extracted from functional and structural connectivity matrices. Standard feature selection techniques and genetic algorithm (GA) methods were applied to select the optimal features. Subsequently, the K-nearest neighbor, support vector machine (SVM), Naive Bayes, and logistic regression classification methods were used to classify healthy controls (HCs) and pre-surgical TLE groups, as well as pre-surgical left TLE (LTLE) and right TLE (RTLE) groups. Also, relationships between psychological scores and the selected features were evaluated using a linear regression method.

RESULTS: The results demonstrated increased functional and decreased structural connectivity in TLE patients before surgery. After surgery, significant connections revealed reduced functional connectivity and increased structural connectivity in TLE patients. Functional analysis identified the left parahippocampal region in LTLE and the right temporal regions in RTLE as key areas. Structural connectivity analysis showed that memory-related areas in the bilateral occipital region and the left language-related area were the origins of alterations. The GA method achieved the highest classification performance using SVM for fMRI and DTI graph measures, with accuracy rates of 97% and 88% for distinguishing LTLE from RTLE, and 93% and 87% for distinguishing TLE from HC, respectively. Moreover, a significant relationship was observed between the best-selected features and memory-assisted cognitive tests.

CONCLUSIONS: Pre-surgical functional hyperconnectivity and post-surgical hypoconnectivity and also newly observed bilateral postsurgical structural connectivity, highlighting functional and structural alterations in the LMN network. Additionally, the study underscores the potential of machine learning for TLE diagnosis and lateralization. A limited sample size, particularly in the postsurgical group was one of the constraints of this study.

ABBREVIATIONS: TLE=Temporal lobe epilepsy; LMN=Language-memory network; GA=Genetic algorithm; HC=Healthy controls; LTLE=Left TLE; RTLE=Right TLE; AUC=Area under the curve