EcoSta 2022: Start Registration
View Submission - EcoSta2022
A0608
Title: Joint curve registration for longitudinal and survival data with application to Alzheimer's disease onset prediction Authors:  Tianhao Wang - Rush University Medical Center (United States) [presenting]
Abstract: In studies of Alzheimer's disease (AD), there is great interest in understanding the progression of cognitive markers and developing a prognostic model for AD onset using the longitudinal cognitive markers. The conventional joint modeling approach for longitudinal and survival data requires a predetermined time scale, typically the time since baseline, in which every subject is assumed to have a comparable risk profile. However, in many observational AD studies, the participants entered the studies with heterogeneous cognition status at baseline, which leads to heterogeneous and incomparable risk profiles in time since baseline. We introduce a novel joint modeling approach based on the functional curve registration method. It assumes the longitudinal trajectories follow a flexible common shape function with a person-specific disease progression pattern characterized by a random curve registration function, which is further used to create a homogenous and comparable time scale for survival analysis. We propose a personalized dynamic prediction framework that can be updated as new observations are collected to reflect the patient's latest cognition status. Simulation studies and application to data from the Rush Religious Orders Study and Memory and Aging Project demonstrate the effectiveness of this new approach.