Intracranial Disease-Region Composite-Interpretation Technology for Enhanced Source Localization in Pediatric Epilepsy Surgery

Abstract

Electroencephalography (EEG) based source localization (ESL) is a useful method to localize the epileptogenic zone in epilepsy surgery. However, previous techniques only perform 3-dimensional (3D) reconstruction, and do not conduct delineation on the cortex surface as a resection guidance, and there is very little data on intracranial EEG and pediatric cases. This study proposes an Intracranial Disease-region Composite-interpretation (IDC) EEG-based source localization (ESL) scheme that uses 3D extended reality (XR) edge computing to enhance visualization and comprehensive interpretation of intracranial EEG-based source localization (iESL) for patients with pediatric epilepsy. The proposed IDC-ESL method was effective in predicting the surgical outcome in patients with focal epilepsy, which can be effectively used for epilepsy surgery. Seizure freedom was clearly associated with complete resection of combined EEG features of interictal spike, high-frequency oscillation (HFO), and seizure onset zone (SOZ), and it had the highest significance in localizing the epileptogenic zone. However, for patients with Lennox-Gastaut syndrome (LGS), IDC-ESL was not performed effectively because of a deeply seated lesion and multifocal abnormalities. It could only roughly estimate the affected area, mainly because of insular involvement. Cautious interpretation based on intraoperative electrocorticography (ECoG) is required for accurate insular resection, particularly for LGS cases.