Simultaneous recordings for characterizing the links between modalities: the case of epileptic networks

It is now widely accepted that epilepsy is a disease of brain networks. During presurgical evaluation, a combination of brain mapping modalities can be used such as EEG, MEG, PET, funtional MRI in order to map these networks non-invasively. In a second phase, electrodes can be implanted within the brain, which can record directly within brain structures and confirm/ infirm clinical hypotheses - this is Stereotaxic EEG (SEEG). Each modality has advantages and disadvantages; in order to optimise the use of brain mapping tools, it is therefore crucial to understand how to make best use of these complementary methods. In this context, two questions arise: what are the best markers of epileptogenic tissues, as visible on intracerebral signals, and how can we retrieve them non-invasively.  In this talk, I  will present recent results on network measures and characterization of high frequency activities. I will also discuss strategies based on simultaneous recordings in for characterizing the links between modalities, in animal models and in patients. In particular, I will show the feasibility of simultaneous EEG-MEG-SEEG recordings thanks to a visual stimulation paradigm and discuss their application to the development of signal processing methods, modelling and diagnosis.