Seminar // Patricia Figueiredo - EEG signatures of fMRI dynamic functional connectivity states

Recurrent states of dynamic functional connectivity (dFC) have been
identified in BOLD-fMRI data, but their electrophysiological
underpinnings remain a matter of debate. In this talk, I will present
our recent findings regarding the investigation of potential EEG
correlates of fMRI-derived dFC states, based on simultaneous EEG-fMRI
recordings from a group of healthy volunteers and a group of epilepsy
patients. We employ a dictionary learning approach to identify a limited
number of dFC states on the BOLD-fMRI data and, for each dFC state
occurrence, we extract candidate EEG features based on spectral power
and microstates. By training a classifier of dFC states using these EEG
features, we find that the three EEG microstates with the greatest
global explained variance are the most predictive of the simultaneous
fMRI dFC states.