Age-related changes in human brain functional connectivity using graph theory and machine learning techniques in...

Area of Science:

• Neuroscience
• Network Science
• Gerontology

Background:

• Normal brain aging involves cognitive and motor function decline.
• Understanding age-related brain changes is crucial due to global population aging.
• Functional magnetic resonance imaging (fMRI) and graph theory offer tools to model brain networks.

Purpose of the Study:

• To investigate age-related alterations in brain network functional connectivity.
• To model changes in brain network communication across different lifespan age groups.
• To identify specific brain regions most impacted by normal aging.

Main Methods:

• Utilized functional magnetic resonance imaging (fMRI) data from the Human Connectome Project (HCP).
• Calculated Pearson correlation-based connectivity networks and applied graph theory measures.
• Employed statistical tests and machine learning (SVM, KNN, decision tree) for feature selection and age group classification.

Main Results:

• Observed decreased global efficiency and increased transitivity in resting-state brain networks with age.
• Identified amygdala, putamen, hippocampus, precuneus, and temporal gyri as significantly affected brain regions.
• Achieved 82.2% classification accuracy for age groups using Fisher score and a decision tree classifier.

Conclusions:

• Graph theory analysis of functional connectivity effectively reveals normal age-related brain changes.
• Identified brain regions and network alterations can serve as biomarkers for monitoring brain health.
• This approach provides insights into the aging brain and potential for health monitoring.