The main objective is to develop a computational imaging approach to assess, non-invasively, the microstructure of cerebral vasculature in aging subjects. Our approach consists of interpreting images by invoking the physics underlying the imaging process and employing high performance computing to extract intrinsic microvascular parameters in the normal aging brain. No other techniques are available to provide this critical information about microcirculation in the brain tissue of healthy subjects in vivo. When developed and validated as we propose, the new technique will usher in a continuum of research examining the variation of the metabolic support of brain neurons and glial cells in aging or disease.