Alzheimer’s disease (AD) is a progressive neurodegenerative disorder characterized by the neuropathological accumulation of amyloid beta (Ab) plaques and neurofibrillary tangles comprised of hyperphosphorylated tau. Recent data from multiple groups support that given enough time, Ab and tau spreads throughout the brain in a well-defined manner along brain networks, very similar to the way a prion protein travels throughout the brain. Notably, the olfactory system has been reported to be one of the first systems affected in AD. Our past work has focused on using Manganese Enhanced MRI (MEMRI) to assess axonal transport in the olfactory system in mouse models of AD. Indeed, we have reported that axonal transport deficits in the olfactory system are detectable prior to the development of learning and memory deficits and well before plaque formation. These data are consistent with the idea that AD pathology spreads throughout the brain, beginning with the olfactory system. Additionally, our prior work has also focused on the effects of reducing oxidative stress in mouse models of AD. When we reduced oxidative stress by overexpressing superoxide dismutase 2 (SOD-2) in mouse models of AD, we observed a complete recovery in learning and memory deficits, a complete recovery in axonal transport deficits in the olfactory system as well as an over 50% reduction in Ab plaque formation. Although oxidative stress has been identified as a significant player in the development of AD, efforts to reduce oxidative stress with antioxidants has met with limited success. Some of the reasons for this are thought to include the inability to target sufficient quantities of administered antioxidants to the appropriate regions within the brain. Additionally, clinically available antioxidants have poor solubility and do not readily enter cells. Thus, we have turned to nanotechnology and have been working with nano-antioxidants that are much more potent than clinically available antioxidants, are non-toxic and readily enter cells. We therefore hypothesize that protecting olfactory and adjoining structures with intranasally administered nano-antioxidants (PEG-HCCs) will slow down the progression of AD as assessed with behavioral assays, MEMRI, resting state fMRI (rs-fMRI) as well as 31P measurements and histology. We will also incorporate machine learning, specifically, a probabilistic graphical model, to determine the interactions of the readouts in Aims 1 and 2 in mouse models of AD and controls with and without treatment with PEG-HCCs. We also propose to incorporate machine learning to predict 1) the degree to which superoxide levels should be reduced to improve AD pathology and 2) which stages of AD (e.g. pre vs post plaque) are beyond rescue. Completion of this highly innovative project will have significant impact towards future AD therapeutic strategies.