A Comparison of Static and Dynamic Functional Connectivities for Identifying Subjects and Biological Sex Using...

Area of Science:

• Neuroscience
• Cognitive Science
• Brain Imaging

Background:

• Resting-state functional connectivity (FC) reveals correlated brain region activities without tasks.
• Initial studies assumed stationary FC, but recent research explores dynamic network models.
• Dynamic models offer better brain activity representation but are computationally intensive.

Purpose of the Study:

• To quantitatively compare static and dynamic FC models for identifying intrinsic individual connectivity patterns.
• To assess the efficacy of FC patterns as unique identifiers for individuals.
• To investigate the use of FC patterns for biological sex identification.

Main Methods:

• Utilized data from the Human Connectome Project.
• Compared static and dynamic functional connectivity (FC) models.
• Employed partial correlation for FC analysis.
• Applied edge consistency, edge variability, and differential power measures.

Main Results:

• Intrinsic individual brain connectivity patterns act as a 'fingerprint' for subject identification.
• Static FC, especially with partial correlation, more accurately captures individual brain connectivity.
• Individual brain connectivity patterns remained invariant over several months.
• Successful identification of biological sex using individual connectivity patterns and group FC matrices.

Conclusions:

• Static FC models are more effective for capturing stable, individual-specific brain connectivity patterns.
• Intrinsic brain connectivity patterns can reliably distinguish individuals and identify biological sex.
• FC patterns offer a robust, long-term identifier of individual brain networks.