Hub Patterns-Based Detection of Dynamic Functional Network Metastates in Resting State: A Test-Retest Analysis

Area of Science:

• Neuroscience
• Computational Neuroscience
• Network Science

Background:

• Resting-state functional networks capture brain dynamics relevant to physiology and pathology.
• Metastate analysis quantifies brain functional connectome dynamics but often overlooks topological structure.
• Node centrality, reflecting local network topology, has not been fully integrated into metastate analysis.

Purpose of the Study:

• To develop and validate a novel node centrality-based metastate analysis method for brain functional connectome dynamics.
• To assess the test-retest reliability and stability of the identified metastates.
• To explore the relationship between dynamic metastate hub regions and brain structural organization.

Main Methods:

• A node centrality-based clustering method was developed using time sequences of node centrality from resting-state fMRI data.
• A test-retest experiment with 23 healthy young volunteers (21-26 years) was conducted to evaluate metastate stability.
• Hub regions identified from metastates were compared with intrinsic sub-networks and structural connectivity data.

Main Results:

• The node centrality-based method identified repeatable dynamic metastate features across scans.
• High overlap was observed between metastate hub regions and intrinsic brain sub-networks.
• Identified metastate hub patterns showed significant overlap with structural hub regions.

Conclusions:

• The proposed node centrality-based metastate detection method reliably reveals meaningful dynamics of spontaneous brain activity.
• This approach provides insights into the underlying nature of brain dynamics.
• The findings suggest a strong link between dynamic functional network organization and the brain's structural connectome.