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: | Peiyan ZHANG |
Institution: | Hong Kong University of Science and Technology |
Department: | Computer Science and Engineering |
Country: | |
Proposed Analysis: | We study the subtype of Alzheimer’s Disease (AD) and their genetic basis, aiming to pinpoint novel genetic markers at a finer granularity than typical studies dichotomizing the cohort into case (AD) group and control group. We trained a small-changes-invariant robust deep learning model over 3D fMRI images classifying AD patients and control group with 70% prediction accuracy over a held-out cohort. We then leverage model interpretation techniques to ask the model: for every fMRI sample, what perturbations we need to introduce to flip the model’s prediction. Although some of these perturbations with smaller magnitude appear to be noises, others tend to congregate at certain regions of the brain. We further cluster these patterns, and consider each cluster as a subtype of Alzheimer’s disease. Finally, we use standard GWAS tools to identify the SNPs that are associated with these subtypes. |
Additional Investigators |