CMStatistics 2016: Start Registration
View Submission - CMStatistics
B0674
Title: A latent variable regression model for assessing mixed type biomarkers Authors:  Zheyu Wang - Johns Hopkins University (United States) [presenting]
Krisztian Sebestyen - Johns Hopkins University (United States)
Sarah Monsell - University of Washington (United States)
Abstract: The purpose is to present a model-based clustering method via finite mixture modeling framework for mixed typed manifest variables with possible differential covariates. It is motivated by research questions in Alzheimer's disease (AD) with aims: to evaluate the accuracy of imaging biomarkers in AD prognosis, and to integrate biomarker information and standard clinical test results into the diagnoses. One challenge in such biomarker studies is that it is often desired or necessary to conduct the evaluation without relying on clinical diagnoses or some other standard references. This is because 1) biomarkers may provide prognostic information long before any standard reference can be acquired; 2) these references are often based on or provide unfair advantage to standard tests. Therefore, they can mask the prognostic value of a useful biomarker, especially when the biomarker is much more accurate than the standard tests. In addition, the biomarkers and existing tests may be of mixed type and vastly different distributions. We present a model-based clustering method to evaluate the prognostic value of biomarkers in addition to standard tests without relying on potentially inaccurate reference diagnoses. Maximum likelihood parameter estimation is carried out via the EM algorithm. Accuracy measures and the ROC curves of the biomarkers are derived subsequently. Finally, we illustrate the method with a real example in AD.