Investigating the Progression of Alzheimer’s Disease Using Digital Volume Correlation Algorithm and Strain As a Metric

Abstract

In the United States, Alzheimer’s disease (AD) affects one in ten people ages 65 and older. In most patients, the first indication of AD is the inability to remember new information, and symptoms grow to include behavior changes and increasing confusion and suspicions surrounding loved ones and daily events. As the disease progresses, the cortex and hippocampus regions of the brain decrease in size, allowing the fluid-filled ventricles within the brain to increase. New and innovative therapies to delay the onset of the disease and progression of the symptoms are being discovered. For example, the antibody solanezumab is undergoing clinical trials to determine its ability to reduce the levels of beta-amyloid in the brain, a known risk factor of AD. Consequently, the ability to identify patients who could benefit from the therapies will be invaluable. The purpose of this study is to determine if the digital volume correlation (DVC) algorithm can detect and track the onset and progression of AD using magnetic resonance imaging (MRI) scans of the head. DVC measures the deformation and strain of the volumetric MRI dataset by tracking the changes in its grey value pattern. A collection of MRI datasets of a patient’s head, which include scans from a baseline visit and visits at 6 months, 12 months, and every 12 months thereafter, is used in our analysis. A strain is applied to each set of MRI scans prior to implementation of the digital volume correlation algorithm. The DVC algorithm is then applied to the dataset and the resulting error between the expected and calculated strain is computed. A decrease in the contrast of the MRI dataset will correlate to additional error by the algorithm. As a result, an increase in the calculated strain error is anticipated to correlate with an increase in the ventricles in the brain, or progression of the disease, over the time period of interest.