Model for cascading failures in functional networks: application to epileptic patients with generalized tonic-clonic seizures.

The dynamic process of epilepsy is modeled as a cascading failure model in functional networks derived from graph theory. The aim is to test whether cascading failure identified from functional magnetic resonance imaging data could simulate epileptic discharges in 18 subjects with generalized tonic-clonic seizure and 17 demographically matched healthy controls. A cascading failure model was used to simulate the neural networks underlying generalized tonic-clonic seizure and healthy controls by stimulation of the node with the greatest number of connections. Results showed that the efficiency of generalized tonic-clonic seizure dropped significantly when compared to controls. Particular nodes whose efficiency altered significantly showed a correlation with the symptoms of generalized tonic-clonic seizure. Results also indicated that the left middle frontal lobe may be a potential focal area in the initiation of generalized tonic-clonic seizure.