Study of the brain’s dynamic functional connectivity with simultaneous
fMRI-EEG: a network science approach

by Francisca Ayres Basto Soares Ribeiro

MSc thesis presentation and discussion.

Date: 2021-Jan-15   Time: 16:30
Room: Zoom

Abstract: The brain’s intrinsic organization into functional networks can be
assessed using several imaging techniques, mainly functional magnetic
resonance imaging (fMRI) and electroencephalography (EEG). While recent
studies have suggested a link between the dynamic functional connectivity
captured by these two modalities, the exact relationship between the fMRI and
EEG functional networks spatiotemporal organization is still unclear.
Furthermore, since these networks are spatially embedded, the question arises
whether the topological features captured can be explained exclusively by the
spatial constraints. We address these two questions by investigating
functional brain networks measured by fMRI and EEG data acquired
simultaneously during resting-state, using a community and motif analysis. For
this purpose, several established approaches were used, such as the Louvain
algorithm as well as its multiplex version, here applied to find partitions
combining fMRI and EEG for the first time, and also g-tries data structure to
efficiently count the occurrence of subgraphs. To explore the influence of
space in the topology captured, new approaches were applied, such as a
modified version of the Louvain algorithm, including a degree constrained
spatial null model in the modularity definition, but also a motif analysis
where the subgraphs are statistically tested against this spatial null model.
We show that even though fMRI and EEG functional connectomes are slightly
linked, the two modalities seem to capture different information, with most
topology, but not entirely, being explained by the spatial constraints.