Online Program Home
  My Program

Abstract Details

Activity Number: 285 - Enhance Risk Assessment Through Novel Statistical Measures and Methods for Complex Time-to-Event Data
Type: Topic Contributed
Date/Time: Tuesday, August 1, 2017 : 8:30 AM to 10:20 AM
Sponsor: Section on Risk Analysis
Abstract #322979 View Presentation
Title: Progression Risk Prediction with Copula Model in Age-Related Macular Degeneration (AMD) Patients
Author(s): Ying Ding* and Yi Liu and Wei Chen
Companies: University of Pittsburgh and and University of Pittsburgh
Keywords: Bivariate ; Copula ; Progression ; Risk prediction
Abstract:

Age-related Macular Degeneration (AMD) is a polygenic and progressive neurodegenerative disease, which is a leading cause of blindness in developed countries. Some patients with AMD maintain good vision for a long time with little disease progression over time, while others quickly advance to vision-threatening late AMD. The progressions of two eyes within the same subject are often correlated. In this research, we first develop a computationally efficient copula-based score test, of which the dependence between bivariate progression times is explicitly modeled, to identify susceptible demographical/environmental and genetic risk factors associated with AMD progression. Then, using a large randomized trial data, Age-related Eye Disease Study (AREDS), we establish copula-based prediction models to predict the joint progression-free probability of two eyes within a subject. Finally, we evaluate and validate the prediction models using another independent large randomized trial AREDS2.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2017 program

 
 
Copyright © American Statistical Association