Cross-correlation analysis of epileptiform propagation using wavelets

We have analyzed cortical and subcortical field recordings from spatially distinct neural circuits in order to support the hypothesis that spatially distinct brain locations display correlated ictal activity epileptic seizures. Field recordings have been obtained from cortex (CTX), anterior thalamic nuclei (AN), posterior thalamus (PT) and hippocampus (HPC) during pentylenetetrazol (PTZ) seizures in anesthetized animals. We use Wavelet Transform Cross-Correlation (WTCC) method in order to quantify the common activity between two recordings at particular bands of interest. In contrary to Fourier Transform Coherence (FTC), we show that WTCC provides a more reliable estimate of band-specific common activity or cross-coherence between two epileptic sources. Although most of the signal power is located at higher frequencies (15-30Hz), results from WTCC reveal significant mean cross-correlation estimates (∼0.7-0.8) at primarily the lower regions of the spectrum (0-10Hz). The behavior observed in the brain recordings analyzed in this paper lets us differentiate between local and global behavior, where the global behavior is assumed to be due to a pacemaker function which is a quasi-periodic train of impulse functions that differentially excites various areas of the brain.