Deep Learning-Based Alzheimer’s Classification Using Optimized VGG16

Early detection and prevention of Alzheimer’s disease (AD) is a vital and complex undertaking. The primary challenge lies in establishing a definitive and precise diagnosis of Alzheimer’s disease during its initial phases. Consequently, numerous investigations were conducted to facilitate the early detection of Alzheimer’s disease. However, the current machine learning model exhibits lower accuracy. This study introduced a deep learning model that utilizes an optimized VGG16 architecture to accurately detect Alzheimer’s disease (AD) and tackle these issues. The suggested approach consists of a preprocessor and a classification model. The purpose of the preprocessor is to transform the MRI picture into a format that optimizes the performance of the classification model. In addition, this study utilized GradCAM to identify and emphasize Regions Of Interest in MRI scans of the Demented Brain compared to the Normal Brain. The proposed model achieves an accuracy of 99.12% on the testing dataset. The assessment of this model has yielded exceptional evaluation metrics and outperformed the majority of existing methodologies.