Graph coarse-graining reveals differences in the module-level structure of functional brain networks.

Area of Science:

• Neuroimaging
• Network Science
• Computational Neuroscience

Background:

• Functional magnetic resonance imaging (fMRI) network analysis simplifies complex brain data.
• Understanding network structures and differences between networks remains challenging.
• Current modular partitioning methods struggle with comparing distinct networks.

Purpose of the Study:

• To present a novel coarse-graining framework for analyzing and comparing fMRI brain networks.
• To enable transparent and statistically verifiable comparisons of module-level connectivity.
• To overcome limitations of independent network partitioning.

Main Methods:

• Developed a coarse-graining framework using a single set of data-driven modules.
• Applied the framework to fMRI data from 13 healthy subjects during rest and movie viewing.
• Compared module-level connectivity differences between rest and movie states.

Main Results:

• Independent partitioning of rest and movie networks yielded limited insights.
• The coarse-graining framework successfully identified module-level structural differences.
• Observed increased intra-module connectivity in the visual cortex during movie viewing.

Conclusions:

• The coarse-graining framework provides a transparent method for comparing brain network structures.
• This approach can quantify differences in brain connectivity due to stimuli or clinical conditions.
• The method enhances understanding of brain network dynamics and functional organization.