Eur. Phys. J. Special Topics 223, 2933–2941 (2014)
https://doi.org/10.1140/epjst/e2014-02306-8

Regular Article

Brain network dynamics characterization in epileptic seizures

Joint directed graph and pairwise synchronization measures

¹ Bioinformatics Interunits Graduate Program, University of São Paulo, São Paulo, Brazil
² Hospital Israelita Albert Einstein, São Paulo, Brazil
³ Department of Neurology, School of Medicine, University of São Paulo, São Paulo, Brazil
⁴ Department of Neurosciences and Behavior, Ribeirão Preto School of Medicine, University of São Paulo, Ribeirão Preto, Brazil
⁵ Institute of Mathematic and Statistics, University of São Paulo, São Paulo, Brazil
⁶ Department of Telecommunications and Control Engineering, Escola Politécnica, University of São Paulo, São Paulo, Brazil
⁷ Department of Radiology and Oncology, School of Medicine, University of São Paulo, São Paulo, Brazil

^a e-mail: abner@usp.br

Received: 27 May 2014
Revised: 17 October 2014
Published online: 10 December 2014

Abstract

Here we propose and evaluate a new approach to analyse multichannel mesial temporal lobe epilepsy EEG data from eight patients through complex network and synchronization theories. The method employs a Granger causality test to infer the directed connectivity graphs and a wavelet transform based phase synchronization measure whose characteristics allow studying dynamical transitions during epileptic seizures. We present a new combined graph measure that quantifies the level of network hub formation, called network hub out-degree, which closely reflects the level of synchronization observed during the ictus.