A Novel Brain Network Construction Method for Exploring Age-Related Functional Reorganization.

Area of Science:

• Neuroscience
• Network Science
• Computational Biology

Background:

• Brain aging involves complex reorganization and changes in functional brain networks.
• Graph theory analysis of brain networks reveals age-related topological differences, but findings can be inconsistent.
• Current network construction methods may obscure crucial signals within noise, limiting the detection of subtle age-related changes.

Purpose of the Study:

• To introduce a novel subnetwork voting (SNV) method for constructing functional brain networks.
• To enhance the identification of statistically significant differences in brain network topology between young and elderly adults.
• To improve the accuracy of classifying age groups using brain network features.

Main Methods:

• A sliding window approach was employed to construct functional brain networks using the proposed subnetwork voting (SNV) method.
• Topological properties of brain networks were analyzed and compared between classic and SNV methods.
• Support vector machine (SVM) classification was performed using features derived from both network construction methods.

Main Results:

• The SNV method demonstrated consistency in identifying topological differences compared to classic methods.
• Statistical analysis revealed that the SNV method identified significantly more age-related differences between groups.
• SVM classification accuracy for distinguishing young from elderly adults reached 89.3% using the SNV method, outperforming the classic method.

Conclusions:

• The subnetwork voting (SNV) method offers improved consistency within groups and highlights inter-group differences in functional brain networks.
• This method holds potential for advancing the exploration and auxiliary diagnosis of aging and age-related neurological diseases.
• The SNV approach provides a more robust framework for analyzing age-related brain network alterations.