Alzheimer’s Disease Detection Using Transfer Learning: Performance Analysis of InceptionResNetV2 and ResNet50 Models

Alzheimer’s disease (AD), a well-recognized form of dementia, is characterized by cognitive decline and memory impairment, significantly impacting an individual’s daily functioning. Recent investigations have explored the potential of innovative imaging techniques, such as machine learning and deep learning algorithms in conjunction with magnetic resonance imaging (MRI), for diagnosing AD. This study utilizes convolutional neural network (CNN) models to automatically classify brain MRI images, capturing relevant features associated with Alzheimer’s disease. The effectiveness of CNN models in distinguishing between the four classes of AD (Mild Demented, Very Mild Demented, Non-Demented, and Moderate Demented) is compared against traditional techniques. The CNN architectures employed include InceptionResNetV2 and ResNet50. The performance of each CNN model is rigorously evaluated, leading to the selection of the model with the highest accuracy. This research contributes to advancing diagnostic methodologies for Alzheimer’s disease, holding promise for improving early detection and intervention strategies.