|
Activity Number:
|
64
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, July 29, 2007 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #308957 |
|
Title:
|
Comparing Approaches for Predicting Prostate Cancer from Longitudinal Data
|
|
Author(s):
|
Christopher Morrell*+ and Larry J. Brant and Shan L. Sheng
|
|
Companies:
|
Loyola College in Maryland and National Institute on Aging and National Institute on Aging
|
|
Address:
|
Mathematical Sciences Department, Baltimore, MD, 21210,
|
|
Keywords:
|
Classification ; Linear Mixed-Effects Models ; Sensitivity and Specificity
|
|
Abstract:
|
Classification approaches are compared using longitudinal data to predict the onset of disease. The data are modeled using linear mixed-effects models. Posterior probabilities are computed of group membership starting with the first observation and adding observations until the subject is classified as developing the disease or until the last measurement is used. From the longitudinal analysis we first use the marginal distributions of the mixed-effects models. Next, conditional on group-specific random effects, the conditional distribution is used to compute the posterior probabilities. The third approach uses the distributions of the random effects. Finally, the subjects' data is summarized by the most recent value and rate of change which are used in a logistic regression model to obtain formulae that can be applied at each visit to obtain probabilities of group membership.
|
