Combining Interictal Intracranial EEG and Fmri to Compute a Dynamic Resting State Index for Surgical Outcome Validation
Abstract number :
1.183
Submission category :
2. Translational Research / 2A. Human Studies
Year :
2024
Submission ID :
660
Source :
www.aesnet.org
Presentation date :
12/7/2024 12:00:00 AM
Published date :
Authors :
Presenting Author: Varina Boerwinkle, MD – UNC
Kristin Gunnarsdottir, PhD – Johns Hopkins University
Bethany Sussman, PhD – University of California in Los Angeles
Sarah Wyckoff, PhD – Phoenix Children's Hospital
Emilio Cediel, MD – UNC Chapel Hill School of Medicine
Belfin Robinson, PhD – University of North Carolina at Chapel Hill
William Reuther, MS – University of North Carolina at Chapel Hill
Sridevi Sarma, PhD – Johns Hopkins University
Rationale: We created a seizure onset network characterization tool powered by combined dynamic biomarkers of resting state (rs) intracranial stereo-encephalography (rs-iEEG) and rs-functional MRI (rs-fMRI), vetted against surgical outcomes. This may avoid waiting for seizures to plan epilepsy surgery.
Methods: The source-sink index (SSI) is computed for all regions in the iEEG network and for a subset of regions, where clinician experts determined iEEG depth electrodes should be placed due to suspicion from noninvasive SOZ-localizing modalities, in the rs-fMRI network for 17 pediatric drug-resistant patients. The rs-iEEG and the rs-fMRI scores for patients with Engel 1 and 2 surgical outcomes are compared to those with Engel 3 and 4 outcomes.
Results: 17 of 30 patients reviewed met criteria, age 3-15 years. Good and poor outcomes, Engel 1 and 2 were compared to Engel 3 and 4, respectively, by the index combining rs-fMRI and rs-iEEG indices, which showed significant separation, as compared to the individual modality biomarkers alone.
Conclusions: The dynamic network model outperformed the iEEG and fMRI biomarkers alone in terms of predicting good from poor surgical outcome.
Significance: This combined tool may lead to increased surgical candidacy and improved epilepsy-surgery-seizure outcomes. Reducing the reliance on capturing seizures, this SOZ localization tool may reduce invasive monitoring times.
Funding: NINDS 1R01NS125897-01
Translational Research