<![CDATA[Alzheimer’s disease is a progressive neurodegenerative disorder characterized by cognitive decline, particularly in memory and reasoning abilities [1]. Early detection of disease severity plays a critical role in improving clinical decision-making [4]. This study aims to classify Alzheimer’s dementia levels using a public MRI dataset from Kaggle consisting of four classes: Non-Demented, Very Mild Demented, Mild Demented, and Moderate Demented [11]. This research compares conventional machine learning methods, namely Decision Tree and Random Forest, with a deep learning approach using Convolutional Neural Network (CNN) [4][8]. The experimental stages include preprocessing, data splitting, model training, and evaluation using accuracy, precision, recall, F1-score, and confusion matrix [4][10]. The results show that Random Forest outperforms Decision Tree with an accuracy of 89%, while CNN achieves the highest performance at 92% [2][3][4]. These findings indicate that CNN is more effective in extracting spatial features from MRI images compared to traditional machine learning methods [2][3][9].]]>
Copyrights © 2026
