Modeling the Progression of Alzheimer's Disease (AD) Using the Alzheimer's Disease Neuroimaging Initiative (ADNI) Cohort

• Activity Number: 177
• Type: Contributed
• Date/Time: Monday, July 30, 2012 : 10:30 AM to 12:20 PM
• Sponsor: Biometrics Section
• Author(s): Bruno Michel Jedynak*+ and Andrew Lang and Bo Liu and Brian Scott Caffo and Jerry Prince and Brad Wyman and Yanwei Zhang
• Companies: The Johns Hopkins University and The Johns Hopkins University and The Johns Hopkins University and The Johns Hopkins University and The Johns Hopkins University and Pfizer Inc. and Pfizer Inc.
• Keywords: generalized bilinear model ; alternating optimization ; Alzheimer's disease ; ADNI

Abstract

The Alzheimer's Disease Neuroimaging Initiative (ADNI) has been collecting data from over 800 subjects in the last several years. Using this data, we have developed a model of AD progression. Seven biomarkers were selected by experts for their relevance. The following principles were then used: All subjects follow a common disease progression but differ in their age of onset and rate of progression; as the disease progresses, each biomarker changes continuously and monotonically following a sigmoid shaped curve; and in the longitudinal period over which biomarkers are observed the rate of progression of a given subject is constant. The resulting statistical model is a generalized bilinear model. We fitted the parameters using least squares and alternating optimization. We were able to prove the identifiability of the model and to estimate the variance of the estimated parameters using bootstrap. This method is possibly the first to report a multiple biomarker, data-driven approach to assess time-dependent changes of biomarkers in Alzheimer's disease and to localize subjects on a scale of disease progression over the entire range of progression represented in the ADNI cohort.