Individual variability in functional organization of the neonatal brain

• Published in: NeuroImage (Orlando, Fla.) Vol. 253; p. 119101
• Main Authors: Molloy, M. Fiona, Saygin, Zeynep M.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.06.2022; Elsevier Limited; Elsevier
• Subjects: Adult; Adults; Attention; Brain; Brain - diagnostic imaging; Brain architecture; Cognition; Connectivity; Connectome; Development; fMRI; Functional morphology; Gene expression; Humans; Individuality; Infant; Infant, Newborn; Information processing; Magnetic Resonance Imaging; Medical imaging; Neonates; Nerve Net - diagnostic imaging; Neural networks; Neurosciences; Newborn babies; Premature birth; Quality control; Resting state; Sensorimotor system; Time series; Visual perception; Development; Neonates; fMRI; Connectivity; Resting state
• ISSN: 1053-8119; 1095-9572
• DOI: 10.1016/j.neuroimage.2022.119101

The adult brain is organized into distinct functional networks, forming the basis of information processing and determining individual differences in behavior. Is this network organization genetically determined and present at birth? And what is the individual variability in this organization in neonates? Here, we use unsupervised learning to uncover intrinsic functional brain organization using resting-state connectivity from a large cohort of neonates (Developing Human Connectome Project). We identified a set of symmetric, hierarchical, and replicable networks: sensorimotor, visual, default mode, ventral attention, and high-level vision. We quantified individual variability across neonates, and found the most individual variability in the ventral attention networks. Crucially, the variability of these networks was not driven by SNR differences or differences from adult networks (Yeo et al., 2011). Finally, differential gene expression provided a potential explanation for the emergence of these distinct networks and identified potential genes of interest for future developmental and individual variability research. Overall, we found neonatal connectomes (even at the voxel-level) can reveal broad individual-specific information processing units. The presence of individual differences in neonates and the framework for personalized parcellations demonstrated here has the potential to improve prediction of behavior and future outcomes from neonatal and infant brain data.