Real-time multimodal imaging of functional brain networks using simultaneous EEG and fMRI acquisitions

Dr. Frédéric Grouiller - FNS encouragement de projets

The combination of EEG and functional MRI is very compelling to study the functioning of the human brain thanks to the high complementarity of electric and hemodynamic signals. Despite the technical challenge of recording sensitive signal of EEG into the strong magnetic field environment of the MRI, this non-invasive multimodal technique is now widely used in the neuroscience community.

The aim of this proposal is to achieve simultaneous and bimodal EEG and fMRI neurofeedback in the context of emotion regulation. Although attempts to correct strong MR-related artefacts in the EEG in real-time have been performed recently, artefact removal algorithms are not yet optimal and too limited to process online the EEG during the simultaneous fMRI acquisition. Up to now, very rare studies take advantage of both modalities in real-time and most of them only processed specific EEG electrodes by extracting signal in particular frequency bands.

 

In order to achieve our goal of bimodal neurofeedback, this proposal aims at filling this gap by implementing robust and reliable algorithms for the real-time corrections of MR-related artefacts in EEG and by integrating in real-time electrophysiological and hemodynamic brain signals. These algorithms will be based on previous work of the applicant especially at Ultra-High Field in order to obtain EEG with an excellent signal-to-noise ratio and without major artefacts residuals in real-time. One of the originality of this project is to exploit the full EEG scalp topography in real-time in order to propose advanced multimodal integration approaches. These pioneering developments will be a major cornerstone for several applications in both fundamental and clinical neuroscience, opening new opportunities for real-time neurofeedback aiming at regulating brain activity in health and disease. These efficient artefact corrections and multimodal integrations would also open the exciting opportunities to perform online electrical source imaging or online mapping of EEG microstates. The potential applications are not limited to the study of emotion regulation but might also impact considerably epilepsy by non-invasive online mapping of epileptic networks and neurological or psychiatric rehabilitation particularly in stroke or in depression.

 

As proof-of-concept for emotion regulation, we will evaluate the feasibility of simultaneous and bimodal connectivity-based neurofeedback through two applications: attention and emotion regulation. Bimodal neurofeedback is expected to outperform the performance of unimodal neurofeedback given the different nature of EEG and fMRI signals and thanks to their complementarity in terms of spatial and temporal resolutions.

 

The applicant and his collaborators have a strong expertise in multimodal imaging and particularly in the technical development of EEG-fMRI. He occupies a permanent position as MRI Laboratory Head at the Brain and Behaviour Laboratory of the University of Geneva which offers him the possibility to conduct independent research (50% of his work load). This project is an excellent opportunity for him to develop his own independent research that could benefit other users of the platform in addition to the neuroimaging community.

Team