Abstracts

Inter-signal Ieeg Vector as a Novel Biomarker of Epileptogenic Tissue

Abstract number : 3.197
Submission category : 3. Neurophysiology / 3G. Computational Analysis & Modeling of EEG
Year : 2022
Submission ID : 2204719
Source : www.aesnet.org
Presentation date : 12/5/2022 12:00:00 PM
Published date : Nov 22, 2022, 05:26 AM

Authors :
Jan Cimbalnik, PhD – International Clinical Research Center, St. Anne's University Hospital in Brno; Milan Brazdil, Prof – Behavioral and Social Neuroscience Research Group, CEITEC Central European Institute of Technology; Petr Klimes, PhD – Institute of Scientific Instruments, The Czech Academy of Sciences; Martin Pail, MD – Brno Epilepsy Center, Department of Neurology, St. Anne’s University Hospital; Vojtech Travnicek, MSc – International Clinical Research Center, St. Anne’s University Hospital

Rationale: The patients suffering from focal pharmacoresistant epilepsy are often candidates for epileptic surgery. Despite recent advances in the identification of promising interictal iEEG biomarkers of seizure generating tissue, such as high frequency oscillations (HFO), the precise delineation of the epileptic area remains elusive. In this study we propose a novel bivariate biomarker of epileptogenic tissue derived from a vector which represents delay and information transfer between two iEEG signals.

Methods: We analyzed 67 resting state stereotactic EEG recordings of patients with pharmacoresistant focal epilepsy who underwent monitoring as part of their evaluation for subsequent surgery. The duration of all recordings was 30 minutes with sampling frequency 5 kHz. The seizure onset zone (SOZ) channels and the resected area under individual contacts was determined by epileptologists from coregistred pre-implant MRI and post-implant CT/MRI. The inter-signal iEEG vector (ISEV) was calculated on raw signal, high gamma (55-80 Hz), ripple (80-250 Hz) and fast ripple (250-600 Hz) bands. We evaluated the capability of the inter-signal iEEG vector to differentiate seizure onset zone (SOZ) from the rest of the tissue by comparing values in individual channels aggregated over the whole dataset and in individual patients separately using Wilxocon rank sum test. Subsequently, we carried out a similar analysis on a subset of patients (N=18) who underwent a successful resection of epileptic tissue with Engel IA outcome with minimal follow up 1 year. In this subset the pathologic tissue was defined as overlap between the resected area and SOZ.

Results: On the group level, the algorithm showed best performance in the high gamma band with a significant difference between SOZ and nonSOZ channels with p< < 0.001. In the per patient analysis the most patients with significant differences between SOZ and nonSOZ were in the ripple band (38 out of 67 patients, 57%). The best performance in the subset of patients with excellent outcome was in the ripple band p< < 0.001 on the group level and in the high gamma band where 12 out of 18 patients, 67% showed significant differences between pathological and non-pathological tissue.
Neurophysiology