Applying Deep Learning Models to Classify Magnetic Resonance Imaging (MRI) to Support Early Detection of Alzheimer’s Disease
Molloy Faculty Mentor
Helen Dang
Presenter Major
Computer Science
Presentation Type
Oral
Location
H339, 3rd floor, Barbara H. Hagan Center for Nursing
Start Date
28-4-2025 7:22 PM
End Date
28-4-2025 7:29 PM
Description (Abstract)
Alzheimer's Disease (AD) is a degenerative brain syndrome that causes memory loss and prohibits the patient from performing the simplest of tasks. It affects over 50 million people worldwide, usually those over the age of 65. Before getting to Alzheimer’s however, there are other stages along the way. The first stage is “mild cognitive impairment non-convertible” (MCInc), which results naturally from human aging. Then, there is “mild cognitive impairment convertible” (MCIc) which means the person will develop AD in a few years. This condition is difficult to detect using traditional medical techniques. In this study, a Deep Learning model will be applied to classify medical images (magnetic resonance imaging (MRI)) to support early detection of AD. Experiments and model evaluations performed on available datasets promise positive results.
Data availability: The datasets used in this study were obtained from
the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database.
This was an open-access dataset available from the link (http://adni.loni.usc.edu)
- The data here are open access datasets via link. So no license is required
Keywords
Machine Learning
Related Pillar(s)
Study
Applying Deep Learning Models to Classify Magnetic Resonance Imaging (MRI) to Support Early Detection of Alzheimer’s Disease
H339, 3rd floor, Barbara H. Hagan Center for Nursing
Alzheimer's Disease (AD) is a degenerative brain syndrome that causes memory loss and prohibits the patient from performing the simplest of tasks. It affects over 50 million people worldwide, usually those over the age of 65. Before getting to Alzheimer’s however, there are other stages along the way. The first stage is “mild cognitive impairment non-convertible” (MCInc), which results naturally from human aging. Then, there is “mild cognitive impairment convertible” (MCIc) which means the person will develop AD in a few years. This condition is difficult to detect using traditional medical techniques. In this study, a Deep Learning model will be applied to classify medical images (magnetic resonance imaging (MRI)) to support early detection of AD. Experiments and model evaluations performed on available datasets promise positive results.
Data availability: The datasets used in this study were obtained from
the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database.
This was an open-access dataset available from the link (http://adni.loni.usc.edu)
- The data here are open access datasets via link. So no license is required
