Inferring group-wise consistent multimodal brain networks via multi-view spectral clustering

Area of Science:

• Neuroimaging
• Computational Neuroscience
• Network Science

Background:

• Quantitative analysis of structural and functional brain networks using diffusion tensor imaging (DTI) and functional magnetic resonance imaging (fMRI) is a growing field.
• The consistency of these brain networks across different neuroimaging modalities and individuals remains largely unexplored.

Purpose of the Study:

• To develop a novel approach for inferring group-wise consistent brain subnetworks from multimodal DTI and resting-state fMRI data.
• To assess the regularity and consistency of structural and functional brain networks across individuals and imaging modalities.

Main Methods:

• Utilized multi-view spectral clustering on cortical networks derived from dense individualized and common connectivity-based cortical landmarks (DICCCOL).
• Applied the algorithms to DTI data from 100 healthy young females and 50 healthy young males.
• Constructed multimodal brain networks using DICCCOL landmarks.

Main Results:

• Achieved consistent multimodal brain networks within and across different groups (gender, imaging modality).
• Demonstrated substantially improved inter-modality and inter-subject consistency in derived brain networks.
• Examined the functional roles of the identified consistent brain networks.

Conclusions:

• The novel approach successfully infers group-wise consistent brain subnetworks from multimodal neuroimaging data.
• The findings highlight improved inter-modality and inter-subject consistency, advancing brain network analysis.
• This method provides a robust framework for understanding brain network regularity.