According to Alzheimer’s Disease International, 5.2% of the world’s population over 60 years old is estimated to be living with dementia, with the vast majority (~70-80%) the result of Alzheimer’s disease (AD). Considering the prevalence of AD and the projected population aging, by 2030, over 100 million people are expected to be living with AD. After several decades of focused clinical and genetic research and despite numerous advancements in the understanding of the disease, the pathophysiological mechanisms of AD remain unknown. In order to address this problem, and inspired by advances in the characterization of the brain’s connectome, availability of fMRI data and enhanced AI research capabilities enabled by HiperGator 3.0/AI, in this catalyst project, we will combine recently developed advanced functional magnetic resonance imaging (fMRI) techniques with novel network theory and image analysis methods, to identify quantitative biomarkers that can help the early diagnosis and guide treatment of AD.

Specifically, we aim to use machine learning to juxtapose connectomic analysis with brain imaging analysis to model disease progression and identify biomarkers in AD. Understanding connectome characteristics can lead to identification of critical neural nodes and connections underlying AD while imaging analysis can provide unfiltered diagnostic criteria. These findings can be combined to pinpoint AD biomarkers and used to model disease progression through targeted network attacks.