A baseline for the multivariate comparison of resting-state networks

Area of Science:

• Neuroimaging
• Systems Neuroscience
• Biostatistics

Background:

• Large functional and structural MRI datasets require standardized analysis for diagnostic relevance.
• Establishing robust processing and statistical methods is crucial for reproducible neuroimaging research.

Purpose of the Study:

• To introduce a multivariate mega-analytic approach for optimizing sensitivity and reducing testing in neuroimaging.
• To identify the effects of age and gender on resting-state networks (RSNs) in healthy individuals.

Main Methods:

• Utilized a mega-analytic approach on resting-state fMRI data from 603 healthy adolescents and adults.
• Data preprocessing via an automated pipeline (SPM); analysis using group independent component analysis.
• Evaluated RSNs using time course spectral power, spatial map intensity, and functional network connectivity.

Main Results:

• Demonstrated significant effects of age on RSNs, primarily showing decreased network coherence and connectivity with aging.
• Observed smaller but distinct gender effects: females exhibited stronger intra-network connectivity, males stronger inter-network connectivity.
• Sensorimotor networks showed particular gender-related connectivity differences.

Conclusions:

• The proposed multivariate analytic approach provides a robust framework for neuroimaging studies.
• Age-related decreases in network coherence and connectivity are significant in healthy individuals.
• Identified sex-based differences in functional brain network organization offer a baseline for future research.