Estimation of brain age delta from brain imaging

• Published in: NeuroImage (Orlando, Fla.) Vol. 200; pp. 528 - 539
• Main Authors: Smith, Stephen M., Vidaurre, Diego, Alfaro-Almagro, Fidel, Nichols, Thomas E., Miller, Karla L.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 15.10.2019; Elsevier Limited; Academic Press
• Subjects: Age; Age determination; Aged; Aging; Aging - physiology; Bias; Brain - diagnostic imaging; Brain - physiology; Brain aging; Brain imaging; Cognitive ability; Computer Simulation; Databases, Factual; Health Status; Humans; Magnetic Resonance Imaging; Medical imaging; Models, Statistical; Neuroimaging; Neuroimaging - statistics & numerical data; Neuropsychological Tests - statistics & numerical data; Phenotype; Phenotypes; UK biobank; United Kingdom; United Kingdom; United Kingdom > UK; UK biobank; Brain aging; Brain imaging
• ISSN: 1053-8119; 1095-9572; 1095-9572
• DOI: 10.1016/j.neuroimage.2019.06.017

It is of increasing interest to study “brain age” - the apparent age of a subject, as inferred from brain imaging data. The difference between brain age and actual age (the “delta”) is typically computed, reflecting deviation from the population norm. This therefore may reflect accelerated aging (positive delta) or resilience (negative delta) and has been found to be a useful correlate with factors such as disease and cognitive decline. However, although there has been a range of methods proposed for estimating brain age, there has been little study of the optimal ways of computing the delta. In this technical note we describe problems with the most common current approach, and present potential improvements. We evaluate different estimation methods on simulated and real data. We also find the strongest correlations of corrected brain age delta with 5,792 non-imaging variables (non-brain physical measures, life-factor measures, cognitive test scores, etc.), and also with 2,641 multimodal brain imaging-derived phenotypes, with data from 19,000 participants in UK Biobank. •It is of interest to study "brain age'', as inferred from brain imaging data.•The delta between brain age and actual age is typically computed.•We describe problems with the most common current approach.•We present potential improvements.•We evaluate methods on simulated data and data from UK Biobank.