Precision mapping of functional brain network trajectories during early development

Preterm birth is a known risk factor for neurodevelopmental disabilities, but early cognitive assessments often fail to predict long-term outcomes. This limitation underscores the need for alternative biomarkers that reflect early brain organization. Resting-state functional connectivity is a powerful tool to study functional brain organization during the perinatal period. However, most fMRI studies in infant populations use group-level analyses that average subject-specific data across several weeks of development, reducing sensitivity to subtle, time-sensitive deviations from typical brain trajectories. Using a novel precision functional mapping approach, we estimated individual resting-state networks (RSNs) in a large cohort of neonates (N = 352, gestational age at birth: 25.6–42.3 weeks) from the developing Human Connectome Project. RSN connectivity strength increased linearly with age at scan, especially in higher-order networks. In particular, the default mode network (DMN) exhibited marked changes in topography and connectivity strength, evolving from an immature organization in preterm infants to a more adult-like pattern in term-born infants. Longitudinal data from a subset of preterm infants (N = 15) confirmed ongoing network development shortly after birth. Despite this maturation, preterm infants did not reach the connectivity levels of term-born infants by term-equivalent age. These findings highlight the potential of individualized RSN mapping as an early marker of neurodevelopmental trajectories.