Skip to content. | Skip to navigation


Personal tools

This is SunRain Plone Theme
You are here: Home / Training / 2016 / Exploring the epileptogenic network through EEG/MEG source localization

Exploring the epileptogenic network through EEG/MEG source localization

Dr. Christophe Grova, Department of Physics, Concordia University

Exploring the epileptogenic network through EEG/MEG source localization

Dr. C. Grova

  • CREATE-MIA Event
  • Seminar
When Apr 08, 2016
from 02:45 PM to 03:45 PM
Where McConnell Engineering MC437
Attendees All CREATE-MIA Trainees
Add event to calendar vCal


During this talk we will introduce how to use measurement of ongoing electrophysiology from EEG or MEG data to localize the generator of epileptic discharges. After a brief introduction of source localization procedures in general, I will present the framework of the Maximum Entropy on the Mean (MEM) as a source localization technique particularly apropriate to reconstruct the generators of transient epileptic activity recorded from either EEG and MEG data : (i) MEM can recover the underlying spatial extent of those generators along the cortical surface and (ii) MEM can localize epileptic discharges in specific frequency bands. Detection of spontaneous epileptic discharges in EEG/MEG from background activity requires synchronized neuronal activity over a minimum area of cortex to be detected from scalp recordings. A minimum area of 6-10cm2 has been suggested in EEG and of 4cm2 in MEG. To assess the ability to recover such spatial extent,  MEM results were compared with other source localizations using realistic simulations, when applied to clinical EEG/MEG data and when compared quantitatively with intracranial EEG data. In a second step, I will introduce the wavelet-based MEM (wMEM), as a method particularly well-adapted to localize transient oscillatory patterns, by extended MEM formalism using the discrete wavelet framework. We recently demonstrated the relevance of wMEM to localize the generators of epilepsy specific High Frequency Oscillations (40-160Hz) when detected from MEG, as well as the localization of the seizure onset zone when applied to EEG/MEG acquired during seizures.


Dr Grova received in 1998 an Engineering degree and a Master of Science degree in biomedical engineering from the University of Technology of Compiègne, France, with a specialization in image and signal processing. In 2002, he  completed a PhD at University of Rennes 1 (France) for a project involving registration of nuclear medicine image data (SPECT). From 2003 to 2008, he worked in the team of Dr, Jean Gotman at the Montreal Neurological Institute as a postdoctoral fellow and became, in 2008, assistant Professor in both Biomedical Engineering and Neurology and Neurosurgery Departments at McGill University, where he created the Multimodal Functional Imaging Lab. Since July 2014, Dr Grova joined as an assistant Professor, the department of Physics and PERFORM center at Concordia University, where he is currently chairing the PERFORM applied bio-imaging committee (ABC). His research projects are now incorporating high-density EEG/fMRI, EEG/NIRS and EEG/MEG acquisitions proposing a multimodal framework to analyze resting state functional connectivity in healthy and pathological conditions, with a specific focus in epilepsy.

« December 2021 »
Funded by NSERC

Funding provided by NSERC