|
Abstract:
|
In the study of Alzheimer's disease (AD), researchers often collect repeated measurements of clinical variables, event history, and functional data (high dimension magnetic resonance imaging), to better understand the diseases. An accurate prediction of the time to dementia based on the information collected is particularly helpful for monitoring, screening and management of such neurodegenerative disease. We propose a series of novel functional joint models (FJM) to account for functional data as either predictor or response in longitudinal and survival sub-models in the joint modeling framework. We develop a Bayesian approach for statistical inference and a personalized dynamic prediction framework for predicting the patients' future health outcome and risk of event. The proposed model is applied to the motivating Alzheimer's Disease Neuroimaging Initiative (ADNI) study, suggesting that incorporating the imaging data as functional predictors into the model could improve the predictive ability.
|
