Investigating WSD-Based Metrics for Brain Network Analysis

The connectome, a comprehensive map of neural connections in the brain, has been extensively studied at the macroscale level to understand complex brain behavior. The macroscale connectome can be modeled as a network, thereby allowing the application of network analysis techniques to explore its characteristics. In spectral graph theory, the weighted spectral distribution (WSD) has been proposed to analyze networks (e.g., communication networks) effectively. In this paper, we introduce new metrics based on the WSD for brain network analysis. Using actual connectome data from fMRI, we evaluate the effectiveness of our WSD-based metrics for brain network analysis. Our findings show that these metrics more accurately reflect the subject’s age and gender compared to other metrics calculated from the connectome data.