Signal processing for characterizing dynamic networks with application to epilepsy presurgical mapping

Orateur : Christian BENAR (Inserm researcher, Institut de Neurosciences des Systèmes, Marseille)

A major interest of neuroscience is to define the spatio-temporal dynamics of brain networks. Noninvasive electrophysiological tools such as EEG and MEG are useful tools in that respect, as they allow measuring brain activity with millisecond temporal precision. However, these techniques remain surface recordings that require solving a difficult inverse problem. It is thus important to understand in which condition brain activity can be retrieved from the surface. In this talk, I will present simultaneous recordings of depth and surface signals in patients as a strategy to design signal processing strategies for recovering information on network dynamics from the surface. I will also present developments in time-frequency analysis for high frequency activities in epilepsy, as well as the AnyWave software designed in the team that helps transferring methodological developments towards clinicians.

Télécharger dans mon agenda