A Comprehensive Review and Current Methods for Classifying Alzheimer's Disease Using Feature Extraction and Machine Learning Techniques

Abstract

AbstractThe automatic detection of Alzheimer's disease (AD), particularly in its early stages, is crucial for maintaining human health. Alzheimer's disease appears to have a protracted incubation time due to the fact that it is a neurodegenerative ailment. Consequently, it is crucial to examine Alzheimer's symptoms at various stages. The use of a fusion feature set to classify AD from MRIs is reviewed in this work along with other feature extraction and Machine Learning (ML) techniques. To extract texture-based information, techniques including the Gray Level Coherence Matrix (GLCM), Gray Level Run Length Matrix (GLRLM), and Gray Level Difference Method (GLDM) are used. It is anticipated that the contributions of the extracted spatial domain from these methods would result in a more effective classification-based fusion feature set. According to the study under review, classification frameworks developed using these extracted variables show promise for the customized diagnosis and clinical progression prediction. Finally, this study addressed several future research possibilities and gave a thorough overview of the difficulties in classifying AD.KeywordsAlzheimer’s diseaseFeature extractionGray levelSupervised classification