A computational biomarker of juvenile myoclonic epilepsy from resting-state MEG

Abstract

Objective Functional networks derived from resting-state scalp EEG from people with idiopathic (genetic) generalized epilepsy (IGE) have been shown to have an inherent higher propensity to generate seizures than those from healthy controls when assessed using the concept of brain network ictogenicity (BNI). Herein we test whether the BNI framework is applicable to resting-state MEG and whether it may achieve higher classification accuracy relative to previous studies using EEG. Methods The BNI framework consists in deriving a functional network from apparently normal brain activity, placing a mathematical model of ictogenicity into the network and then computing how often such network generates seizures in silico. We consider data from 26 people with juvenile myoclonic epilepsy (JME) and 26 healthy controls. Results We find that resting-state MEG functional networks from people with JME are characterized by a higher propensity to generate seizures (i.e. BNI) than those from healthy controls. We found a classification accuracy of 73%. Conclusions The BNI framework is applicable to MEG and capable of differentiating people with epilepsy from healthy controls. The observed classification accuracy is similar to previously achieved in scalp EEG. Significance The BNI framework may be applied to resting-state MEG to aid in epilepsy diagnosis.