Development of a Machine Learning Model for Diagnosis of Alzheimer's Disease Stages Using Brain Image Patterns

The goal of this project was to increase the diagnostic accuracy of AI modelings in the detection of Alzheimer's disease. Alzheimer's disease is a complex condition that requires early diagnosis for effective treatment. However, current diagnostic methods can have limitations. In this study, we developed a more advanced research model to detect AD. By utilizing both linear regression models and neural network, our model achieved excellent results with a test accuracy of 81.34% for the neural network model and 93.05% for the linear model Unlike earlier models that focused on binary classification of Alzheimer's disease, our model goes beyond that by classifying different stages and severities of Alzheimer's. We utilized a large dataset found on Kaggle called the "Alzheimer's Dataset" [1] consisting of 5,121 images of four classes (Mild Demented, Non Demented, Very Mild Demented, and Moderate Demented). This research redemonstrates the potential of machine learning to improve diagnostic accuracy for Alzheimer's disease.