Abstract / Summary This proposal aims to perform novel, secondary analyses on large archives of publicly-available fMRI studies in order to quantify the functional characteristics of white matter (WM) and their changes during normal aging and in the progression to Alzheimer’s Disease (AD). Blood oxygenation level dependent (BOLD) effects have been used to detect neural activity in grey matter (GM) for many years, but BOLD signals in WM have traditionally been ignored so they have not been considered in previous analyses. However, WM BOLD signals have been evaluated in relatively small numbers of healthy brains, and these studies have shown that WM shows robust, tract-specific BOLD changes in response to stimuli, WM exhibits inter-regional correlations in a resting state similar to those used to infer functional connectivity in cortex, and signal fluctuations within WM tracts in a resting state show strong correlations to specific GM cortical volumes engaged together in functional networks. We therefore propose to adapt the tools developed for analyzing GM connectivity and for diffusion imaging of WM to analyze the functional changes in WM with age in >7,900 imaging studies available publicly. In Aim 1, we will detect and characterize changes in WM functional networks with normal aging by analyzing subjects from the Baltimore Longitudinal Study of Aging (BLSA), the Open Access Series of Imaging Studies (OASIS-3) and the Alzheimer’s Disease Neuroimaging Initiative (ADNI). Neuroimaging studies of AD suggest that WM abnormalities exist at a preclinical stage of the disease, so detecting and quantifying altered WM function may be an important metric of functional changes in this disorder. In Aim 2 therefore we will measure alterations in WM functional networks in subjects enrolled in the ADNI, BLSA, and OASIS-3 databases. In both Aims we will also measure co-variations of WM connectivities with behavioral, clinical and genetic assessments to establish how WM functional metrics reflect behavior and cognition and change with increasing cognitive impairment. In Aim 3 we will extend the creation of atlases and machine learning from current studies of diffusion MRI to quantifying WM functional MRI by creating age-adjusted atlases of WM functional MRI properties to enable normative comparisons and provide canonical templates for both functional and structural connectivity network analyses. We will also apply data-driven deep learning to identify individual signatures of impairment on a whole-brain basis. A failure of white matter functional integrity is clearly implicated in aging and neurodegeneration. This proposal will develop new understandings of the functional changes in WM across the lifespan, identify pathological changes in WM function with AD, and create new data-driven tools for interpretation of WM fMRI.