AUTHOR=Wang Zhi Ji , Noh Byoung Ho , Kim Eun Seong , Yang Donghwa , Yang Shan , Kim Nam Young , Hur Yun Jung , Kim Heung Dong TITLE=Brain network analysis of interictal epileptiform discharges from ECoG to identify epileptogenic zone in pediatric patients with epilepsy and focal cortical dysplasia type II: A retrospective study JOURNAL=Frontiers in Neurology VOLUME=13 YEAR=2022 URL=https://www.frontiersin.org/journals/neurology/articles/10.3389/fneur.2022.901633 DOI=10.3389/fneur.2022.901633 ISSN=1664-2295 ABSTRACT=Objective

For patients with drug–resistant focal epilepsy, intracranial monitoring remains the gold standard for surgical intervention. Focal cortical dysplasia (FCD) is the most common cause of pharmacoresistant focal epilepsy in pediatric patients who usually develop seizures in early childhood. Timely removal of the epileptogenic zone (EZ) is necessary to achieve lasting seizure freedom and favorable developmental and cognitive outcomes to improve the quality of life. We applied brain network analysis to investigate potential biomarkers for the diagnosis of EZ that will aid in the resection for pediatric focal epilepsy patients with FCD type II.

Methods

Ten pediatric patients with focal epilepsy diagnosed as FCD type II and that had a follow–up after resection surgery (Engel class I [n = 9] and Engel class II [n = 1]) were retrospectively included. Time–frequency analysis of phase transfer entropy, graph theory analysis, and power spectrum compensation were combined to calculate brain network parameters based on interictal epileptiform discharges from ECoG.

Results

Clustering coefficient, local efficiency, node out–degree, and node out–strength with higher values are the most reliable biomarkers for the delineation of EZ, and the differences between EZ and margin zone (MZ), and EZ and normal zone (NZ) were significant (p < 0.05; Mann–Whitney U-test, two–tailed). In particular, the difference between MZ and NZ was significant for patients with frontal FCD (MZ > NZ; p < 0.05) but was not significant for patients with extra–frontal FCD.

Conclusions

Brain network analysis, based on the combination of time–frequency analysis of phase transfer entropy, graph theory analysis, and power spectrum compensation, can aid in the diagnosis of EZ for pediatric focal epilepsy patients with FCD type II.