Classification of alzheimer’s disease from structural magnetic resonance imaging using particle-bernstein polynomials algorithm

Abstract

Automated structural magnetic resonance imaging (MRI) classification has gained popularity for the early detection of mild cognitive impairment (MCI), the first stage of dementia condition with an increased risk of eventually developing Alzheimer’s disease (AD). In general, an MRI diagnosis system requires some fundamental activities: MRI processing, features selection, data classification. The aim of this paper is twofold: (i) first, a high-performance classification algorithm based on particle-Bernstein polynomials (PBPs), recently proposed for nonlinear regression of input–output data that combines low complexity and good accuracy, has been developed, (ii) second, an MRI-based computer-aided diagnosis (CAD) system for the classification of AD has been derived. Several experiments on a dataset from Alzheimer’s Disease Neuroimaging Initiative (ADNI) and comparisons with the state-of-the-art establish the performance of the method.