Meeting Banner
Abstract #3051

Gaussian Map Descriptors for Alzheimer Detection Using T1-weighted Magnetic Resonance Imaging

Inas A. Yassine1, Nourhan Zayed2, and Shereen E. Morsy1

1Biomedical Engineering and Systems, Cairo University, Cairo, Egypt, 2Computer and Systems, Electronics Research Institute, Cairo, Egypt

Recently, Alzheimer’s Disease (AD) is one of the most emerging elderly diseases. In this study, we propose employing Gaussian map descriptors to discriminate between AD, Mild Cognitive Impairment (MCI) and Normal subjects using T1-weighted Magnetic resonance images (MRI) downloaded from Alzheimer's disease Neuroimaging Initiative (ADNI) website. Extracted Gaussian map descriptors, calculated for the hippocampus, such as Gaussian curvature and mean curvature, were then fed to the support vector machine (SVM) for classification purposes. The Gaussian curvature outperformed mean curvature in case of normal to abnormal, and AD to MCI discrimination with accuracies of 69.5%, and 98.3% respectively.

This abstract and the presentation materials are available to members only; a login is required.

Join Here