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: | Brian McCrindlle |
Institution: | McMaster University |
Department: | Electrical and Computer Engineering |
Country: | |
Proposed Analysis: | Much of the recent work done in the field of Deep Learning has involved improving the accuracy of predictions through the development of various network architectures or algorithms. Though the accuracy of deep networks is crucial to performance, the addition of uncertainty quantification in classification tasks can lead to improved predictions and overall algorithm trust, especially in medicine where errors are often a matter of life or death. This type of metric has not been widely employed in Deep Learning architectures today and must be further researched in the context of real-world problems. Therefore, the primary question at the forefront of our proposed research is to determine if the recent advances in Deep Learning uncertainty quantification can be applied to a set of Magnetic Resonance (MR) Images of Alzheimer’s disease patients. In this application, the goal of high disease classification accuracy and interpretable uncertainty metrics are paramount. A Variational Bayesian Deep Learning architecture will be employed to quantify the uncertainties on the large set of diverse set of MR images that can be provided through ADNI. |
Additional Investigators | |
Investigator's Name: | Michael Noseworthy |
Proposed Analysis: | Academic Supervisor on the project. Co-Director of the School of Biomedical Engineering. Professor in the Department of Electrical and Computer Engineering. Director of Medical Imaging Physics and Engineering, Imaging Research Centre, St. Joseph’s Healthcare, Hamilton. |