Changepoint Estimation in Shape or Other Markers for Alzheimer's Disease

• Activity Number: 110
• Type: Invited
• Date/Time: Monday, August 10, 2015 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistics in Imaging
• Author(s): Laurent Younes*
• Companies: The Johns Hopkins University
• Keywords: Medical Imaging ; Computational Anatomy ; Shape Statistics

Abstract

Diagnosis of Alzheimer's disease is made on the basis of cognitive tests which can reliably detect the disease once it has been progressing for a number of years already. In this talk, we will describe a shape analysis approach to estimate the average number of years before diagnosis, over which brain atrophy (hippocampus, amygdala, entorhinal cortex) has been progressing in disease population. We will also apply the approach to other markers (CSF, cognitive). This will be based on data from the BIOCARD dataset, a study in which initially healthy patients have been followed for almost 20 years, with some of them developing the disease over time.