Commute Time as a Method to Explore Brain Functional Connectomes.

Area of Science:

• Neuroscience
• Graph Theory
• Network Science

Background:

• Graph theory metrics like shortest path length are common in neuroscience for analyzing brain networks.
• These metrics assume information flow is limited to shortest paths, a potentially restrictive assumption.
• Resting-state functional magnetic resonance imaging (rs-fMRI) is frequently used to assess brain networks.

Purpose of the Study:

• To investigate the utility of commute time as a measure of regional centrality in functional connectomes.
• To compare commute time with shortest path-based metrics in brain network analysis.
• To evaluate the impact of head motion artifacts on both metrics.

Main Methods:

• Utilized commute time as a novel tool for regional centrality analysis on functional connectomes.
• Employed the ADHD-200 database for empirical validation.
• Compared commute time with shortest path-based metrics at both group and subject levels.

Main Results:

• At the group level, commute time and shortest path lengths were highly correlated.
• At the subject level, commute time demonstrated significantly less susceptibility to head motion artifacts compared to shortest path metrics.
• Despite group-level similarity, subject-level analysis revealed distinct advantages for commute time.

Conclusions:

• Commute time provides a valuable alternative to shortest path metrics for analyzing functional brain networks.
• Commute time is particularly advantageous in connectomic studies involving populations prone to motion artifacts, such as children or clinical groups.
• This metric offers a more robust approach to understanding brain network organization when dealing with real-world data limitations.