Time-frequency models for single event magneto-electro encephalography
analysis

                           Maureen Clerc

                  Odyssie Laboratory (INRIA-ENPC-ENS) 
                            Sophia Antipolis
                                France

Signals measured in EEG and MEG, such as event-related potentials and
fields, oscillations or epileptic spikes, often present significant
variability across different realizations.  This variability may
convey useful information, but is lost in the classical procedure of
signal averaging. We present a general method for modeling and
tracking M/EEG events, applicable for both low and high frequencies
and taking into account the spatial structure (topography) of the
events. In this study, we investigate the extraction of spatial and
time-frequency patterns from the data in order to build
topography-time-frequency templates.  The topography is handled by a
Principal Components Analysis, and the time-frequency templates are
initialized by considering the average of the energy in the time
frequency plane, over all realizations. Then the parameters of the
templates are adjusted to each realization by a gradient descent
method, with a penalization on their dispersion.  Results will be
shown on the recovery of synthetic data, and of epileptic spikes, with
different signal to noise ratios.