Cortical and Network Reorganization in Glioma-Related Epilepsy: Insights from Structural and Machine Learning Analyses

Motivation: Epilepsy is a debilitating symptom in patients with glioma, yet the underlying cortical alterations associated with Glioma-Related Epilepsy (GRE) remain poorly understood. Goal(s): To identify the reorganization pattern of cortical architecture related to GRE. Approach: High-resolution MRI was used to assess cortical thickness and myelin content in GRE , non-GRE and HC. Structural covariance networks were analyzed. An automated machine learning model was constructed based on cortical features. Results: GRE exhibited cortical thickening and demyelination in contralateral orbitofrontal gyrus, superior frontal gyrus, and contralateral sensorimotor cortex respectively. The machine learning model achieved accuracy of 0.80 and AUC of 0.87 in classifying GRE individuals. Impact: This study reveals significant cortical and network alterations in GRE patients, highlighting cortical reorganization's role in GRE pathophysiology and the potential of machine learning for developing targeted diagnostic and therapeutic strategies.