Early Identification of Neurodevelopmental Disorders with in utero MRI

Oregon Health & Science University, Portland OR

Investigators

Abstract

Project Summary There is a lack of longitudinal research examining neurodevelopmental trajectories spanning the fetal to postnatal period. The ability to characterize normative within-individual brain growth across this dynamic epoch is imperative to establishing the degree to which gestational measures can be used to predict neurodevelopmental abnormalities in individuals prior to later symptom onset. The goal of this proposal is to characterize the relationship between gestational and infant measures of neurodevelopment and determine the degree to which these measures are associated both structurally and functionally across the beginning of an individual's lifespan. The proposal leverages a dataset comprised of 229 serially imaged pre- and postnatal brain scans. Individuals received longitudinal structural T2-weighted scans beginning at mid gestation (postconceptional day 85) continuing through birth and into late infancy/early adolescence (postnatal 11 months) with the inclusion of postnatal diffusion-weighted imaging. Additionally, postnatal behavioral testing was conducted at one month of age. Aim 1 will characterize multiple cortical developmental indices (cortical folding, surface area, and volumetric expansion) across early development. Here, I test the hypothesis that in utero measures are associated with postnatal measures within-individuals using mixed effect modeling. Aim 2 will expand upon the previous aim to identify the degree to which gestational cortical macroscopic measures are associated with postnatal subcortical white matter maturation (fractional anisotropy and normalized T2-intensity), with the prediction that measures of gestational curvature will be associated with initial postnatal white matter organizational development. Finally, Aim 3 seeks to determine whether the developmental indices and relationships generated from the previous two aims explain individual variability in behavioral sensorimotor development. We hypothesize that gestational measures can alone predict composite measures of behavioral sensorimotor task performance and will be assessed using a Generalized Mixed Modelling approach. Together, these aims will provide a wide-ranging characterization of neurodevelopment, linking pre- and postnatal development measures longitudinally, which represent an under-studied and exceedingly important area in developmental research. The results generated from the proposed experiments will have implications for identifying the gestational measures best suited for early identification of neurodevelopmental abnormalities.