Predicting Brain Age Based on Spatial and Temporal Features of Human Brain Functional Networks.

Area of Science:

• Neuroscience
• Medical Imaging
• Computational Biology

Background:

• Human brain functional networks change with age.
• Functional MRI (fMRI) is used to study brain activity and connectivity.
• Previous research indicates age-related alterations in functional connectivity.

Purpose of the Study:

• To develop predictive models for estimating brain age using resting-state fMRI data.
• To identify age-related components within human functional brain networks.
• To explore feature extraction and reduction methods for brain network analysis.

Main Methods:

• Acquired resting-state fMRI data from subjects.
• Constructed functional network models using functional connectivity (FC) matrices.
• Applied Principal Component Analysis (PCA) to identify age-correlated components.
• Extracted predictor variables including component coefficients, edge features, and fALFF-based temporal features.
• Trained three regression models to predict brain age.

Main Results:

• Identified intrinsic, age-related, and other components shaping functional network architecture.
• Developed predictive models capable of accurately estimating brain age.
• Demonstrated that extracted features contain sufficient information for age prediction.

Conclusions:

• Functional network organization is influenced by intrinsic and age-related factors.
• Predictive modeling using fMRI-derived network features can accurately estimate brain age.
• The study highlights the potential of network neuroscience approaches for understanding brain aging.