High-Order Graphical Topology Analysis of Brain Functional Connectivity Networks Using fMRI.

Area of Science:

• Neuroscience
• Graph Theory
• Medical Imaging

Background:

• Brain connectivity networks are often analyzed using classical graph theory, which may overlook crucial high-order topological properties.
• A comprehensive understanding of brain topology requires examining metrics beyond low-order indicators.

Purpose of the Study:

• To develop and validate a graphical topology analysis framework for brain connectivity networks using functional Magnetic Resonance Imaging (fMRI).
• To investigate high-order topological features, phase transitions, and propose a novel metric ('redundant energy') for brain chaos.
• To assess the framework's reproducibility, generalizability, and application in distinguishing healthy controls from Parkinson's disease patients.

Main Methods:

• Utilized functional Magnetic Resonance Imaging (fMRI) data to construct brain connectivity networks.
• Applied a graphical topology analysis framework to examine high-order topological metrics across varying graph sparsity levels.
• Investigated topological phase transitions and introduced a 'redundant energy' indicator to quantify brain chaos.

Main Results:

• High-order metrics revealed topological phase transitions indicative of brain criticality.
• Classical graph indicators showed sharp changes near critical points, driven by brain regions with high node curvatures.
• Significant alterations in high-order topological features were identified in Parkinson's disease patients, correlating with disease severity.

Conclusions:

• The proposed framework effectively captures high-order topological features of brain networks.
• The study provides a novel perspective on brain topology, enhancing comprehension of brain function in health and disease.
• High-order topological analysis holds potential for understanding neurological disorders like Parkinson's disease.