Individual patterns of functional connectivity in neonates as revealed by surface-based Bayesian modeling

Area of Science:

• Neuroscience
• Developmental Neuroscience
• Neuroimaging

Background:

• Resting-state functional connectivity (rsFC) is crucial for understanding brain development.
• Estimating rsFC in infants is challenging due to data acquisition complexities.
• Previous methods struggled to capture individual variability in infant brain networks.

Purpose of the Study:

• To characterize individual variability in brain network organization in a large cohort of term-born infants.
• To develop and validate a novel data-driven Bayesian framework for infant fMRI analysis.
• To investigate the relationship between age and functional brain connectivity in early development.

Main Methods:

• Utilized resting-state fMRI data from 289 infants from the developing Human Connectome Project (dHCP) database.
• Employed a novel data-driven Bayesian framework for network estimation.
• Conducted analysis on the cortical surface using surface-based registration with neonatal atlases for enhanced alignment.

Main Results:

• Successfully estimated subject-level maps for eight distinct brain networks.
• Generated individual functional parcellation maps revealing significant inter-subject differences.
• Identified a significant positive relationship between infant age and mean connectivity strength across all brain regions, including higher-order networks.

Conclusions:

• The study demonstrates the efficacy of surface-based methods and Bayesian approaches for analyzing infant fMRI data.
• The findings provide novel insights into individual variability and age-related changes in early brain network organization.
• This framework advances the study of neurodevelopmental trajectories in very young populations.