Activation network improves spatiotemporal modelling of human brain communication processes.

Area of Science:

• Neuroscience
• Computational Biology
• Network Science

Background:

• Dynamic functional networks (DFN) model brain communication using time series correlations.
• Current DFN methods have limited sensitivity to dynamic changes due to non-dynamic dependencies.
• This limitation hinders the understanding of real-time brain communication dynamics.

Purpose of the Study:

• To introduce an activation network framework based on the activity of functional connectivity (AFC).
• To extract novel connectivity patterns reflecting time-specific brain communication fluctuations.
• To overcome the limitations of DFN in capturing dynamic brain processes.

Main Methods:

• Developed the AFC framework to eliminate non-dynamic dependencies in statistical correlations.
• Validated the AFC method using simulated data, showing high correlation with ground truth.
• Applied AFC to autism spectrum disorder (ASD) and COVID-19 datasets.

Main Results:

• AFC extracted richer topological reorganization information compared to DFN.
• AFC identified significant inter-regional connections and efficient temporal reconfiguration.
• AFC detected decreased brain information processing in patients, distinguishing them from controls, unlike DFN.
• Combined AFC and DFN achieved successful classification of ASD and COVID-19 patients.

Conclusions:

• The AFC framework offers a more sensitive approach to analyzing brain communication dynamics.
• AFC reveals neural mechanisms of brain dynamics previously obscured by DFN.
• This method holds potential for clinical applications in neurological and infectious diseases.