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: | Youho Myong |
Institution: | Seoul National University Hospital |
Department: | Department of Rehabilitation Medicine |
Country: | |
Proposed Analysis: | Dear whom this may concern, My name is Dr. Youho Myong. I am a board-certified physician currently practicing at the Department of Rehabilitation Medicine, Seoul National University Hospital, Seoul, Korea. I am also affiliated with the Department of Biomedical Engineering at Seoul National University College of Medicine. My proposal aims to enhance early detection performance of Alzheimer's disease by augmenting neuroimaging biomarkers via iterative imaging refinement. Recently, there have been attempts to augment ADNI, AIBL, and DOD ADNI's 1.5T brain MR to synthezed 3.5T MR (Zhou et al, 2021, Alzheimer's Research and Therapy). Zhou's report showed promising results in enhancing CNN based classification of Alzheimer's; nevertheless, there remains room for improvement in terms of enhancement magnitude. Medical image data augmentation has been relying primarily on generative adversarial networks (GAN). My team will engage a different approach: Google's SR3 (https://iterative-refinement.github.io/). Using SR3, and using normal 7T brain MRIs as reference, we aim to improve the resolution of the existing brain MRs to 7T level. After enhancement, we will evaluate whether such augmentation results in better early detection of Alzheimer's than using only the original 1.5T or 3T MRIs. As of now, we plan to use CNN-based classifiers for evaluations, but we are actively researching for better classification tools, so this plan may change later. Seeing from Zhou's research, we strongly believe that it will. Not only will this allow for better performance of models on AIBL and ADNI's data, we expect such methods will have immediate impact for many clinical settings where they have limited access to high-resolution MRI. Thank you very much for your time and consideration. Sincerely yours, Youho |
Additional Investigators |