Estimation of gender-specific connectional brain templates using joint multi-view cortical morphological network...

Area of Science:

• Neuroimaging
• Brain Network Analysis
• Computational Neuroscience

Background:

• Estimating population-level connectional brain templates (CBTs) that capture shared and differential patterns across individuals is crucial for understanding brain organization.
• Gender fingerprinting, identifying sex-specific brain characteristics, lacks robust methods for integrating population-level network data.
• Existing approaches have not explored gender-specific CBTs derived from multi-view cortical morphological networks (CMNs).

Purpose of the Study:

• To develop and validate a novel method for estimating gender-specific connectional brain templates (CBTs).
• To identify gender-discriminative regions of interest (ROIs) using multi-view cortical morphological networks (CMNs).
• To advance the field of gender fingerprinting through advanced neuroimaging analysis.

Main Methods:

• Proposed the Multi-View Clustering and Fusion Network (MVCF-Net), a novel method for fusing multi-view network data.
• Estimated CMNs from T1-weighted MRI, deriving views from specific cortical attributes (e.g., thickness).
• Employed manifold optimization for multi-view network clustering, followed by linear and non-linear fusion techniques to generate CBTs.

Main Results:

• MVCF-Net successfully generated representative and well-centered CBTs for male and female populations independently.
• The method identified highly discriminative ROIs that mark significant gender differences.
• Key gender-discriminative ROIs included the lateral occipital cortex, pars opercularis (left hemisphere), and middle temporal gyrus, lingual gyrus (right hemisphere).

Conclusions:

• The proposed MVCF-Net is effective in estimating gender-specific CBTs and identifying neuroanatomical markers of gender differences.
• This study provides the first evidence for gender-specific CBTs derived from multi-view CMNs.
• The identified ROIs offer potential biomarkers for future gender-based brain research.