Comparison of phase synchrony information flow in human EEG through wavelet phase synchronization analysis

The nonlinear interdependencies between fluctuations of three simultaneously brain recordings have been intensively studied in pairs, and the synchrony flow differences have been compared. Chavez et al. used phase synchronization analysis and nonlinear correlation coefficients in the specific frequency band of 10-25 Hz in two patients and identified the instantaneous synchrony drop in the focal area several minutes before seizures. Here we focus on the possibility of tracking distinguishable pre-seizure patterns observed before seizure onset using simultaneous synchrony information in 4 specific frequency bands in three adjacent channels of multichannel iEEG recording. The present study on 21 patients and 77 seizures compares the capability of the wavelet phase synchronization in different pairs of channels in predicting seizures in the sense of correct predictions (C.P.R), false positives per hour (FP/h), and average length of prediction.