Graph Signal Processing for Identifying Structurefunction Coupling Using Multimodal Brain Imaging with fMRI and MEG

Résumé fourni par la source

Understanding how brain dynamics emerge from and are constrained by the underlying neuronal connectivity structure is a central question in neuroscience. Graph harmonic analysis has been used to quantify the coupling between the structural connectome and slow ($0.01-1 \text{Hz}$) brain activity fluctuations measured with functional magnetic resonance imaging (fMRI), revealing a macroscale gradient of the structure-function coupling that aligns with the unimodaltransmodal cortical organization. Magnetoencephalography (MEG) yields access to millisecond-resolution brain dynamics, enabling the spectral characterization of fast ($1-100 \text{Hz}$) neuronal oscillations. Neuronal oscillations are fundamental for healthy brain function, providing a temporal clocking mechanism for neuronal communication. Yet, how the rich spatiotemporal patterns of oscillation dynamics are related to the brain's structural connectome has remained poorly understood. To address this knowledge gap, we implemented the graph harmonic analysis to investigate how MEG oscillatory activities are constrained by the structural connectome and to evaluate their similarities with fMRI. Our results demonstrate that graph harmonic analysis can be used to quantify frequencydependent MEG structure-function relationships. Furthermore, we show a partial similarity between MEG and fMRI imaging modalities and their relation to the structural network, particularly in neural oscillations between$10-20 \text{hz}$. This work characterizes how neural oscillations flow on top of the structural network.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.