Abstract—Alzheimer’s Disease (AD) has been widely investigated to improve diagnosis and prognosis. A lot of studies depended on structural and texture properties have already been conducted to help for identifying AD patients. The Artificial Intelligence (AI) approach, including Machine Learning (ML) and Deep Learning (DL) models, has shown promise of extracting and merging the diverse data to develop predictive models for diagnosing AD. In this paper, an AD Classifier (ADC) model has been proposed for classifying AD severity using image processing methodology to extract textural information from the Magnetic Resonance Imaging (MRI), and classify the patients into four groups; Non Demented, Moderate Demented, Mild Demented and Very Mild Demented using longitudinal cross-sectional data from Open Access Series of Imaging Studies-3 (OASIS-3) sourced from Kaggle database. As shown in the proposed ADC model, four ML and two DL algorithms have been applied for classifying AD severity. The used ML algorithms are Decision Tree (DT), K-Nearest Neighbors (K-NN), Support Vector Machines (SVM), and Random Forest (RF), while the used DL algorithms are Residual Network with 50 Layers (ResNet-50) and Visual Geometry Group 19-layer Network (VGG-19). The feature extraction of the AD using MRI is done using Gray Level Co-occurrence Matrix (GLCM), Histogram of Oriented Gradient (HOG), and Local Binary Patterns (LBP) methods. A comparative study has been done among the used ML and DL algorithms to justify the proper algorithm for classifying AD severity. Looking at the results, it is found that the K-NN algorithm achieves the highest classification accuracy of 99.76% among the evaluated algorithms. Therefore, the K-NN algorithm could be favored for AD classifications because of its effective for early detection.
 
Keywords—Alzheimer disease, deep learning, texture analysis, machine learning
 
Copyright © 2026 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).