Spatio-temporal modeling of connectome-scale brain network interactions via time-evolving graphs.

Area of Science:

• Neuroscience
• Computational Neuroscience
• Network Science

Background:

• Functional brain networks show dynamic patterns in fMRI data.
• Spatial overlap and temporal evolution of connectome-scale network interactions remain underexplored.

Purpose of the Study:

• To develop a novel spatio-temporal modeling framework for connectome-scale functional brain network interactions.
• To investigate how functional brain networks spatially overlap and how these interactions evolve over time.

Main Methods:

• Group-wise dictionary learning to create consistent brain network templates for a common reference space.
• Weighted time-evolving graphs and a dynamic behavioral mixed-membership model (DBMM) to analyze temporal dynamics of network interactions.

Main Results:

• The framework successfully identified meaningful and diverse behavioral patterns of connectome-scale network interactions.
• Identified network behaviors were distinct across various Human Connectome Project (HCP) tasks (motor, working memory, language, social).
• The temporal dynamics of network interactions corresponded well with specific task designs.

Conclusions:

• The proposed framework offers a novel approach to characterizing human brain function.
• Provides a quantitative description of the temporal evolution of spatial overlaps/interactions of connectome-scale brain networks.
• Enables analysis within a standardized reference space for cross-individual and cross-task comparisons.