Abstracts

Normative Functional Connectivity Maps to Identify Abnormal Brain Regions in Children with Epilepsy

Abstract number : 1.091
Submission category : 2. Translational Research / 2C. Biomarkers
Year : 2023
Submission ID : 253
Source : www.aesnet.org
Presentation date : 12/2/2023 12:00:00 AM
Published date :

Authors :
Presenting Author: Ludovica Corona, MSc – The University of Texas at Arlington

Hmayag Partamian, Ph.D. – Bioengineering – University of Texas at Arlington; Lorenzo Fabbri, B.Sc – Ph.D. Student, Bioengineering, University of Texas at Arlington; Emily Brock, B.Sc – Research Assistant, Neuroscience Research Center, Cook Children's Health Care System; M Scott Perry, MD – Cook Children's Health Care System; Christos Papadelis, Ph.D – Director of Research, Neuroscience Research Center, Cook Children's Health Care System

Rationale:
Studying epileptogenic networks with functional connectivity (FC) can help identify hubs whose resection may lead patients with drug resistant epilepsy (DRE) to seizure freedom. An effective way to identify these hubs is by constructing normative maps from healthy controls that provide FC standards for all brain regions. Here, we aim to construct noninvasive normative FC maps from typically developing (TD) children and examine whether FC deviations from these maps can identify the epileptogenic zone (EZ) in children with DRE. We hypothesize that high FC is associated with epilepsy and that regions with increased FC deviating from normative maps are more likely to be epileptogenic.


Methods:
We recorded simultaneous resting-state high-density electroencephalography (HD-EEG; 256 channels) and magnetoencephalography (MEG; 306 sensors) data from 20 children with DRE (median: 14 years; 14 females) and 20 healthy controls (median: 10 years; 11 females) (Figure 1A). We performed electromagnetic source imaging (EMSI) on merged artifact-free portions and reconstructed virtual sensors (VSs) time-series in 166 regions of interest of a cortical atlas (Figure1B). We then computed undirected FC [Amplitude Envelope Correlation (AEC) and corrected imaginary Phase Locking Value (ciPLV)] at the source-level from overlapping epochs (3-sec duration each) for different frequency bands (Figure 1C) and selected the top 10% FC values across time. The minimum spanning tree was used to generate a graph with nodes equal to VSs and edges equal to inverted FC values, and its centrality metrics (i.e., betweenness, closeness, degree, and eigenvector) to assess each node’s importance (Figure 1D). For patients with DRE, FC z-scores were computed using controls as baseline. For patients with focal DRE, we defined EZ regions based on presurgical evaluation and as non-EZ those in the contralateral hemisphere. We compared regional FC estimated from EZ and non-EZ regions across patients with focal DRE, and from healthy ones of TD. We also compared FC between EZ and non-EZ regions in patients with focal DRE. Finally, we compared global FC (mean across 166 VSs) between populations and FC between epileptogenic and non-epileptogenic hemispheres (mean across VSs of one hemisphere) of patients with focal DRE.


Results:
We found higher global FC in children with DRE than controls in all bands (p<
Translational Research