Classification of Alzheimer′s Disease Using Nonlinear Independent Component Analysis

The goal of this study is to classify magnetic resonance imaging (MRI) of brains of patients with Alzheimer′s Disease (AD) and those without AD and then to identify changes in brain MRI in early stages of AD. A novel approach based on the diffusion map framework, which is considered to be one of the leading manifold learning methods, is used for this classification. Diffusion mapping provides dimensionality reduction of the data as well as pattern recognition that can be used to distinguish brains of patients with AD from brains of patients without AD. A new algorithm, which is an extension of diffusion maps, constructs coordinates that generate efficient geometric representations of the complex structures in the MRI. In addition, this method is adapted to the MRI and accounts for the variability in calibration of the MRI of different patients. The algorithm is tested on MRI data from patients who developed AD and those who did not.