Current challenges and future directions for brain age prediction in children and adolescents

Abstract

Abstract Advancements in computational techniques have enhanced our understanding of human brain development, particularly through high-dimensional data from magnetic resonance imaging (MRI). One notable approach is the brain-age prediction framework, which predicts biological age from neuroimaging data and calculates the brain age gap (BAG), a marker of deviation from chronological age. Most commonly applied to adult samples, this approach is now increasingly used in children and adolescents. However, several considerations must be taken into account when applying brain-age prediction in youth. In this Perspective, we outline important challenges and provide recommendations for researchers as well as future directions for the field.