Professor Danielle Jenine Harvey, UC Davis
Friday April 28, 2023
111:10 am - 11:45 am
FMonitoring machine learning (ML)-based risk prediction algorithms in the presence of confounding medical interventions
Abstract:
Alzheimer’s disease (AD) is the leading cause of dementia and currently affects more than 5 million older adults in the US and over 50 million people worldwide. Clinical trials have largely failed, leaving limited treatment options for those suffering from the disease. However, clinical trials are still ongoing at various stages of the disease process, but still face numerous challenges, especially at early stages of the disease when we might have the best chance to intervene. The Alzheimer’s Disease Neuroimaging Initiative (ADNI) is an ongoing multi-site longitudinal study that began in 2004 and is designed to develop clinical, imaging, genetic, and biochemical biomarkers for the early detection and tracking of AD. In this talk, I will provide a brief overview of Alzheimer’s disease and the challenges faced in clinical trials. I will then describe ADNI and discuss prediction models that have been built using ADNI data that could support clinical trials. I will end with a discussion of some current limitations of ADNI and how the next phase of ADNI is hoping to overcome them.