Deviations of Functional Connectivity from Normative Maps Identify the Epileptogenic Network in Patients with Drug Resistant Epilepsy
Abstract number :
1.21
Submission category :
2. Translational Research / 2C. Biomarkers
Year :
2024
Submission ID :
1286
Source :
www.aesnet.org
Presentation date :
12/7/2024 12:00:00 AM
Published date :
Authors :
Presenting Author: Ludovica Corona, PhD – Jane and John Justin Institute for Mind Health, Neurosciences Center, Cook Children's Medical Center
Hmayag Partamian, PhD – Cook Children's Health Care System
Samantha Laboy, MS – CookChildren's Health care System
Cynthia Keator, MD – CookChildren's Health care System
Linh Tran, MD – CookChildren's Health care System
Saleem Malik, MD – CookChildren's Health care System
Dave Shahani, MD – CookChildren's Health care System
M. Scott Perry, MD – Jane and John Justin Institute for Mind Health, Neurosciences Center, Cook Children's Medical Center
Christos Papadelis, PhD – Cook Children's Health Care System
Rationale: Functional connectivity (FC) is a widely used method for mapping epileptogenic brain networks. Studies showed increased FC within epileptogenic areas of patients with drug resistant epilepsy (DRE), compared to areas presumed to be non-epileptogenic. It is incorrect to assume areas of patients with DRE as non-epileptogenic or healthy. We aim here to construct normative FC maps from typically developing (TD) children and examine whether FC deviations from these maps can identify the epileptogenic zone (EZ) in patients with DRE. We hypothesize that epileptic brains have higher FC than healthy ones, abnormal regions have higher FC than non-abnormal ones, and that FC is higher inside resected vs. spared abnormal regions.
Methods:
We analyzed resting-state high-density electroencephalography (256 channels) and magnetoencephalography (306 sensors) data from 30 TD (median: 11 yr.; 15 females) and 30 children and young adults with DRE (median: 15 yr.; 19 females) (Fig. 1A). We performed electromagnetic source imaging on artifact-free data portions and reconstructed source time-series in 166 regions of interest (ROIs) defined by an atlas (Fig. 1B). We computed undirected source FC [Amplitude Envelope Correlation (AEC) and corrected imaginary Phase Locking Value (ciPLV)] for different frequency bands (Fig. 1C) and selected FC>90% of maxima across time. For patients with DRE, we computed FC |z-scores| using controls as baseline to identify abnormal subnetworks (Fig. 1D). Each subnetwork defined a graph with nodes equal to abnormal ROIs and edges to their FC. We assessed the properties of each subnetwork with global efficiency, node strength, and Minimum Spanning Tree’s centrality metrics (i.e., betweenness, closeness, degree, and eigenvector) (Fig. 1E). We defined EZ ROIs based on surgical notes. We compared global efficiency between abnormal and non-abnormal subnetworks of patients with focal DRE, and those of TD, as well as regional FC between EZ, non-EZ, and healthy ROIs of TD. We also compared FC between EZ and non-EZ ROIs in patients with focal DRE. Finally, we compared global FC between groups and age, and dominant FC components (generalized eigenvalue decomposition) among epileptogenic and non-epileptogenic hemispheres of patients with focal DRE.
Results:
Global FC was higher in patients with DRE than TD in all bands, as well as when compared across age (p<
Translational Research