Neonatal brain-age models in full- and preterm infants

Area of Science:

• Neuroscience
• Medical Imaging
• Developmental Biology

Background:

• Prematurity poses risks to infant brain development and neurodevelopmental outcomes.
• Current biomarkers for identifying at-risk preterm infants are limited.
• Accurate assessment of brain maturation in neonates is crucial.

Purpose of the Study:

• To develop and validate a white matter neonatal brain-age model using diffusion MRI.
• To assess the model's accuracy in characterizing preterm infant brains.
• To determine if brain-age prediction can inform infant health beyond clinical measures.

Main Methods:

• Constructed neonatal brain-age prediction models using diffusion magnetic resonance imaging (dMRI) white matter features.
• Utilized two datasets: the developing Human Connectome Project (dHCP) and a clinical sample (LPCH).
• Evaluated model performance and the association of brain-age gap with health complications.

Main Results:

• White matter features accurately predicted brain-age in both healthy (dHCP) and preterm (LPCH) infants.
• Prediction accuracy was within 3.9 days for dHCP and 6.6 days for LPCH.
• The brain-age gap showed no significant association with a composite score of prematurity complications.

Conclusions:

• Tractometry-derived brain-age models effectively characterize neonatal brain maturation.
• These models demonstrate accuracy with both research and clinical dMRI data.
• The brain-age gap's limited sensitivity to clinical complications suggests a need for multimodal biomarkers.