| Marie-Therese Horstmann 1,2,3, Christian E. Elger 1, Klaus Lehnertz 1,2,3 1Department of Epileptology, Univ. Bonn, 2Helmholtz-Institute for Radiation and Nuclear Physics, Univ. Bonn, 3Interdisciplinary Center for Complex Systems, Univ. Bonn We investigate whether graph theoretical approaches are suitable to characterize the dynamics of epileptic brain networks on time scales of days. We analyzed invasive multi-channel recordings of brain electrical activity that were acquired from patients with focal epilepsies during pre-surgical evaluation. Using a moving-window approach we estimated the strength of interactions (via mean phase coherence) between all pairs of sampled brain regions to define functional network links and estimated the global network characteristics average shortest path length L and clustering coefficient C. Both, C and L exhibit a large variability possibly reflecting the influence of various physiological and pathophysiological states. |
|