Study of Alzheimer Disease Prediction Using Various Classifiers

Abstract

Alzheimer's disease (AD) is a chronic, lifelong neurological condition that causes memory and thought ability loss. The disorder has no known cure and is almost always fatal. The disease's progression takes a specific path for each person, making prediction extremely difficult. Much of the time, the signs are misdiagnosed as conditions associated with old age. Early Alzheimer's disease diagnosis is critical because it allows dementia patients and their families to better plan for the disease's progression. It also allows patients to take advantage of available medications that can alleviate some of the side effects of dementia and improve their quality of life by making the disease easier to handle. There are several Neuroimaging instruments and examinations available to examine the brain. Magnetic resonance imaging (MRI), functional magnetic resonance imaging (fMRI), and positron emission tomography (PET) scans are some of the most commonly used brain imaging instruments. Using Neuroimaging info, machine learning methods assisted computer scientists in classifying and distinguishing various stages of Alzheimer's disease. However, distinguishing between Alzheimer's phases is difficult. The majority of previous works categorized Neuroimaging data into binary or three groups. For classification, various machine learning approaches have been used, including traditional methods such as the general linear model (GLM) and multi-voxel methods such as support vector machines (SVMs).