Multiscale Node-Edge Interactions in Complex Networks: A Case Study Using Brain Network.

Area of Science:

• Complex network analysis
• Graph theory
• Neuroscience

Background:

• Research on relationships between nodes and edges in complex networks is ongoing.
• Multiscale node-edge interaction (MNEI) information has not been previously explored in complex networks.

Purpose of the Study:

• To introduce and demonstrate a novel method for analyzing multiscale node-edge interaction (MNEI) information within complex networks.
• To apply this method to brain networks for classifying individuals with Alzheimer's disease, mild cognitive impairment, and healthy controls.

Main Methods:

• Decomposition of node and edge attributes into nine MNEI components in the graph frequency domain.
• Transformation of MNEI components back to the vertex domain for analysis.
• Utilizing MNEI features from both domains for subject classification.

Main Results:

• A majority of MNEI features showed significant differences across the studied groups.
• The classification performance using MNEI features surpassed traditional brain network analysis methods.

Conclusions:

• This study is the first to capture MNEI in complex networks.
• The research presents a novel information fusion technique for integrating node and edge attributes under network topological constraints.
• The approach offers a new method for multi-scale complex network decomposition with potential for broad applications.