There are many active research projects accessing and applying shared ADNI data. Use the search above to find specific research focuses on the active ADNI investigations. This information is requested annually as a requirement for data access.
| Principal Investigator | |
| Principal Investigator's Name: | Peter Millar |
| Institution: | Washington University in St. Louis |
| Department: | Neurology |
| Country: | |
| Proposed Analysis: | Modeling brain-predicted age in preclinical and early symptomatic Alzheimer disease - Machine learning techniques allow for accurate predictions of individuals’ age using structural or functional MRI data (Cole & Franke, 2017). These “brain-predicted age” estimates may reflect a phenotypic manifestation of underlying age-related biological change. Interestingly, brain-predicted age is elevated in many disease states, including Alzheimer disease (AD) (Franke & Gaser, 2019), but has not been evaluated in relation to preclinical AD. In the proposed study, we will estimate brain-predicted age from multimodal imaging data (including structural/volumetric MRI, functional connectivity, and diffusion tensor imaging) in control participants across the adult lifespan. Importantly, we will test if these estimates are elevated as a function of preclinical AD in separate samples. Hypotheses: We predict that using a multi-modal approach will allow for more accurate age prediction than using structural or functional imaging data alone. We predict that brain-predicted age estimates will be elevated in association with preclinical AD in separate test samples. Justification: The machine learning approach requires a large training sample to build a model that can accurately predict age based on structural and/or functional imaging data. Thus, we seek to build a large sample of structural and functional imaging data from healthy control participants spanning the adult lifespan by combining across publicly available datasets in order to train this model. |
| Additional Investigators |
