B0830
Title: Mediating role of neuroimaging data image-related cognitive decline
Authors: Shuo Chen - University of Maryland (United States) [presenting]
Hwiyoung Lee - University of Maryalnd (United States)
Abstract: Aging changes brain functions and structures in a downward trajectory and consequently leads to a decline in neurocognitive performance. Neuroimaging plays a central role in understanding the neurophysiology of brain aging. We consider aging as an independent variable while treating neuroimaging data and cognitive function as the multivariate mediators and outcome, respectively. We aim to investigate the mediation role of multivariate neuroimaging variables in age-related cognitive decline. We present a new multivariate mediation model that maximizes the mediation proportion to reflect that the brain is the primary organ determining cognitive function. Specifically, we consider the aggregating effect of selected neuroimaging variables as a functional brain age score, which can maximally explain age-related cognitive decline. We propose a new objective function that transforms mediation proportion maximization into a quadratic form with an $l_1$ penalty and $l_2$ constraint. We develop a computationally efficient algorithm to handle the nonconvexity of the objective function based on the alternating direction method of multipliers with semidefinite relaxation. We apply our method to 37,441 UK Biobank participants with whole-brain cortical thickness and white-matter integrity measures and cognitive performance scores. Our results show that the mediation effect of brain-imaging variables can explain more than 99\% of age-related cognitive decline.
