Topological state-space estimation of functional human brain networks

Area of Science:

• Neuroscience
• Network Science
• Computational Biology

Background:

• Functional human brain networks exhibit dynamic changes over time.
• Understanding these dynamics is crucial for characterizing brain states.
• Existing methods often struggle to capture temporal complexity.

Purpose of the Study:

• To introduce a novel topological data analysis (TDA) technique for estimating brain network state spaces.
• To compare the efficacy of this new method against traditional clustering approaches like k-means.
• To explore the genetic heritability of dynamic brain network states.

Main Methods:

• Utilized topological data analysis (TDA) with Wasserstein distance for network comparison.
• Developed a data-driven approach to cluster brain networks into distinct topological states.
• Employed a twin study design to investigate the heritability of identified brain network states.

Main Results:

• The TDA-based method effectively identified distinct topological states in resting-state brain networks.
• This technique demonstrated superior performance over k-means clustering by incorporating temporal dynamics.
• Significant heritability was observed for the dynamic changes in brain network topology.

Conclusions:

• The proposed TDA method provides a powerful tool for analyzing dynamic functional brain networks.
• Dynamic brain network states possess substantial heritable components.
• These findings suggest that brain network topology and its dynamics may harbor significant genetic information.