Identifying and characterizing resting state networks in temporally dynamic functional connectomes

Area of Science:

• Neuroscience
• Functional Neuroimaging
• Brain Dynamics

Background:

• Resting state functional magnetic resonance imaging (fMRI) is crucial for identifying resting state networks (RSNs).
• Conventional RSN studies assume static networks, overlooking the brain's dynamic functional state changes.
• The functional connectome varies over time, even during rest.

Purpose of the Study:

• To characterize temporal brain dynamics in resting state using fMRI.
• To identify dynamic resting state networks (RSNs) and compare them with static RSNs.
• To investigate the stability and variability of RSNs during resting states.

Main Methods:

• Utilized temporally dynamic functional connectome patterns and a multi-view spectral clustering method.
• Extracted resting state clusters and RSNs based on the DICCCOL system.
• Compared dynamic clusters with static clusters derived from multi-subject functional connectomes.

Main Results:

• Some dynamic clusters mirrored static clusters, indicating stable RSNs like the visual and default mode networks.
• Two motor-related dynamic clusters corresponded to a single static cluster, suggesting high temporal variability in motor RSNs.
• Four dynamic clusters significantly differed from their static counterparts, highlighting their role in brain dynamics.

Conclusions:

• Resting state brain networks are not static and exhibit significant temporal variability.
• Specific RSNs, such as motor networks, display greater dynamic changes during rest.
• The identified dynamic networks offer novel insights into brain function and its dynamic interactions.