The effects of machine learning algorithms in magnetic resonance imaging (MRI), and biomarkers on early detection of Alzheimer's disease

Abstract

Alzheimer's Disease (AD) is a disorder that worsens over time causing loss of memory and decline of cognitive functions. Current methods for diagnosis consist of neuroimaging scans, magnetic resonance imaging (MRI scans), positron emission tomography (PET scans), and identifying biomarkers in cerebrospinal fluid (CSF). New forms of advanced technology such as machine learning are rising to quickly diagnose AD. This work is a comprehensive review of the research that uses machine learning methods to classify AD cases early. It is a study to provide details for MRI scans and biomarkers used for the recognition of AD and evaluates the execution of both applications while using different classifiers. This paper will discuss and compare various machine learning methods that can be implemented for the classification of Alzheimer's disease. The applications of these algorithms (MRI and biomarkers) are also discussed ultimately proposing the best algorithm and application for classification.