Correlation-based alignment of multichannel signals and application to paroxysmal events
Stiliyan Kalitzin, Wojciech B. Zbijewski, Juan de la Parra, D.N. Velis, L. Manshanden, F. Lopes da Silva · IEEE Transactions on Biomedical Engineering · 2002
We propose a new group-theoretical approach to the problem of alignment of time events in multichannel signal recordings. Such an alignment is an essential phase in the classification of transients in electroencephalogram/magnetoencephalogram (MEG) signals. A common reference frame is reconstructed applying a time translation transformation based on delayed mutual correlation functions of the individual events. The method is applied to MEG data sets recorded from epileptic patients showing paroxysmal interictal discharges.