Brain maturation: predicting individual BrainAGE in children and adolescents using structural MRI.

BACKGROUND: Neural development during human childhood and adolescence involves highly coordinated and sequenced events, characterized by both progressive and regressive processes. Despite a multitude of results demonstrating the age-dependent development of gray matter, white matter, and total brain volume, a reference curve allowing prediction of structural brain maturation is still lacking but would be clinically valuable. For the first time, the present study provides a validated reference curve for structural brain maturation during childhood and adolescence, based on structural MRI data. METHODS AND
FINDINGS: By employing kernel regression methods, a novel but well-validated BrainAGE framework uses the complex multidimensional maturation pattern across the whole brain to estimate an individual's brain age. The BrainAGE framework was applied to a large human sample (n=394) of healthy children and adolescents, whose image data had been acquired during the NIH MRI study of normal brain development. Using this approach, we were able to predict individual brain maturation with a clinically meaningful accuracy: the correlation between predicted brain age and chronological age resulted in r=0.93. The mean absolute error was only 1.1 years. Moreover, the predicted brain age reliably differentiated between all age groups (i.e., preschool childhood, late childhood, early adolescence, middle adolescence, late adolescence). Applying the framework to preterm-born adolescents resulted in a significantly lower estimated brain age than chronological age in subjects who were born before the end of the 27th week of gestation, demonstrating the successful clinical application and future potential of this method.
CONCLUSIONS: Consequently, in the future this novel BrainAGE approach may prove clinically valuable in detecting both normal and abnormal brain maturation, providing important prognostic information.