Quantitative determination of concordance in localizing epileptic focus by component-based EEG-fMRI. (August 2019)

• Record Type: Journal Article
• Authors: Ebrahimzadeh, Elias; Shams, Mohammad; Fayaz, Farahnaz; Rajabion, Lila; Mirbagheri, Mahya; Nadjar Araabi, Babak; Soltanian-Zadeh, Hamid
• Is Part Of: Computer methods and programs in biomedicine. Volume 177(2019)
• Page Start: 231
• Page End: 241
• Publication Date: 2019-08
• Subjects: Simultaneous EEG-fMRI -- Epilepsy -- Independent component analysis (ICA) -- Blood-oxygen-level dependent imaging (BOLD) -- Generalized linear model (GLM); Medicine -- Computer programs -- Periodicals; Biology -- Computer programs -- Periodicals; Computers -- Periodicals; Medicine -- Periodicals; Médecine -- Logiciels -- Périodiques; Biologie -- Logiciels -- Périodiques; Biology -- Computer programs; Medicine -- Computer programs; Periodicals; Electronic journals; 610.28
• DOI: 10.1016/j.cmpb.2019.06.003
• Languages: English
• ISSNs: 0169-2607

Abstract

Highlights: A novel systematic, quantitative approach is presented to evaluate concordance between the Component-related BOLD clusters and IED-location using simultaneous EEG-fMRI. This study demonstrated that BOLD changes are related to epileptic spikes in different brain structures. The component-based EEG-fMRI is considered as a reliable predictor of the spike onset zone. The proposed method improves localization accuracy to 97% which marks a dramatic rise compared to the conventional works. The concordance level is determined based on the distance between the center of the cluster with the maximum BOLD and the dipole using our component-based EEG-fMRI method. Abstract: Background and objective: Accurate seizure onset zone (SOZ) localization is an essential step in pre-surgical assessment of patients with refractory focal epilepsy. Complex pathophysiology of epileptic cerebral structures, seizure types and frequencies have not been considered as influential features for accurate identification of SOZ using EEG-fMRI. There is a crucial need to quantitatively measure concordance between presumed SOZ and IED-related BOLD response in different brain regions to improve SOZ delineation. Methods: A novel component-based EEG-fMRI approach is proposed to measure physical distance between BOLD clusters and selected component dipole location using patient-specific high resolution anatomical images. The method is applied on 18 patients with refractory focal epilepsy to localize epileptic focus and determine concordance quantitatively and compare between maximum BOLD cluster with identified component dipole. To measure concordance, distance from a voxel with maximal z-score of maximum BOLD to center of extracted component dipole is measured. Results: BOLD clusters to spikes distances for concordant (<25 mm), partially concordant (25–50 mm), and discordant (>50 mm) groups were significantly different ( p < 0.0001). The results showed full concordance in 17 IED types (17.85 ± 4.69 mm), partial concordance in 4 (36.47 ± 8.84 mm), and nodiscordance, which is a significant rise compared to the existing literature. The proposed method is premised on the cross-correlation between the spike template outside the scanner and the highly-ranked extracted components. It successfully surpasses the limitations of conventional EEG-fMRI studies which are largely dependent on inside-scanner spikes. More significantly, the proposed method improves localization accuracy to 97% which marks a dramatic rise compared to conventional works. Conclusions: This study demonstrated that BOLD changes were related to epileptic spikes in different brain regions in patients with refractory focal epilepsy. In a systematic quantitative approach, concordance levels based on the distance between center of maximum BOLD cluster and dipole were determined by component-based EEG-fMRI method. Therefore, component-based EEG-fMRI can be considered as a reliable predictor of SOZ in patients with focal epilepsy and included as part of clinical evaluation for patients with medically resistant epilepsy.