Large scale brain networks in epilepsy Hitten P. Zaveri a* , Steven M. Pincus b , Irina I. Goncharova a , Robert B. Duckrow a,c , Susan S. Spencer a,c a Departments of Neurology, c Neurosurgery, Yale University, New Haven, CT 06520, USA; b Guilford, CT 06437, USA ABSTRACT Two studies of brain networks, performed on interictal intracranial EEGs recorded during the presurgical evaluation of patients with epilepsy, are presented in this report. In the first we examine pairwise relationships between pre-defined brain regions in 12 patients, 6 with medial temporal onset of seizures and 6 with frontal and parietal onset of seizures. We demonstrate that differences, in pairwise relationships between brain regions, allow a distinction of these two groups of patients. In the second study we evaluate short, mid, and long-distance brain connectivity as a function of distance to the seizure onset area in another 2 patients. We demonstrate that the measures of brain connectivity distinguish between brain areas which are close to and far from the seizure onset area. The results of the two studies may help both define large scale brain networks involved in the generation of seizures, and localize the area of seizure onset. Keywords: Localization related epilepsy, seizure onset area, brain networks, brain connectivity, coherence, Teager energy. 1. INTRODUCTION The pre-surgical evaluation of patients with epilepsy, in whom seizures cannot be controlled by medications, involves a series of increasingly more invasive investigations, and may require, as a final step, continuous monitoring of brain electrical activity directly from the cortex and subcortical structures (intracranial EEG, icEEG) over days and weeks to record seizures and locate their origin. The icEEG has helped establish the concept of a seizure onset area where seizures in localization related epilepsy are first expressed, and from where they may spread. This definition has been central to research on seizure generation and the control of epilepsy. Investigations of seizure generation have often focused on brain excitation and inhibition, and argued for a lack of balance between the two at the seizure onset area, such that out- of-control excitation sets up a cascade of events which culminates in seizure. The success of surgical control of epilepsy is predicated on detecting this area of the brain and resecting or disconnecting it. The success of new approaches such as stimulation of the brain with a brain implantable device or local delivery of drugs is predicated as well on an accurate definition of the seizure onset area. A considerable amount of evidence, from animal models of epilepsy as well as clinical, icEEG and imaging studies in humans, suggests the involvement of cortical and subcortical networks in the generation of partial and generalized onset seizures [1]. The underlying thesis of these studies is that seizures are an expression of network rather than focal brain abnormality. The specific brain networks which may be involved in the generation of seizures remain poorly defined and poorly understood. Intracranial EEG studies of brain networks can be conducted with pairwise measures of relationship (bivariate measures). Pairwise measures of relationship have been used to study seizures and the peri-seizure interval, typically in studies of seizure spread, seizure onset, and seizure prediction, and on interictal or background icEEG (ongoing activity between seizures). In a recent study of the background icEEG we demonstrated increased long-range theta coherence, between amygdala and the frontal lobe, accompanies the process of kindling in a rat model of epilepsy [2]. Further, Towle and coworkers, used the coherence of the background icEEG to define the functional borders of brain regions [3]. Schevon and coworkers sought to demonstrate that local hypersynchrony is a marker of epileptogenic cortex * hitten.zaveri@yale.edu Advanced Signal Processing Algorithms, Architectures, and Implementations XVIII, edited by Franklin T. Luk, Proc. of SPIE Vol. 7074, 70740T, (2008) · 0277-786X/08/$18 · doi: 10.1117/12.801365 Proc. of SPIE Vol. 7074 70740T-1 2008 SPIE Digital Library -- Subscriber Archive Copy