Deep learning approaches to predicting Alzheimer's disease progression using both metabolic and amyloid PET imaging data
Youngho Seo, Professor
UC San Francisco
Closed. This professor is continuing with Fall 2024 apprentices on this project; no new apprentices needed for Spring 2025.
Keywords: deep learning, artificial intelligence, prediction, Alzheimer's disease, positron emission tomography, PET, image processing, FDG, glucose metabolism, amyloid
Alzheimer's disease (AD) is the most common type of dementia, and a debilitating condition that affects a large number of aging populations. When symptoms occur, there are a very limited set of management strategies. Most research focus on early detection of AD when the symptoms are not very apparent.
Our laboratory has pioneered the concept of using artificial intelligence methods to predict AD's progression. In our early work, we showed that AD progression can be predicted over 5 years early using metabolic PET imaging data. Since PET imaging of amyloid plaques is much more common for AD, we are interested in comparing the prediction performances using metabolic and amyloid imaging data, or combined.
Role: Training Opportunity
- Understand how AI/deep learning techniques can be used for disease progression prediction
- Work with real-world data
- Develop a computational model that can potentially help further understand AD progression
Qualifications: Qualifications: (Required) Some familiarity with programming languages, Python preferred; Some familiarity of convolutional neural network (Preferred) Some familiarity with medical imaging data
Hours: 6-8 hrs
Off-Campus Research Site: Programming work can be done remotely. The project mentor is available to meet online for meetings and discussions. This work is in collaboration with Prof. Jaewon Yang at UT Southwestern Medical Center.
Related website: http://www.radiology.ucsf.edu/physics
Related website: https://doi.org/10.1148/radiol.2018180958