Alzheimer Disease Classification Using Deep Neural Network

Abstract

Alzheimer's disease is a complex neurodegenerative disorder with profound implications for individuals and healthcare systems. Early and accurate diagnosis is critical for effective intervention and treatment. This study explores the potential of deep learning algorithms to predict Alzheimer's disease stages using MRI segmentation data. Four distinct algorithms, namely MobileNet, CNN, DenseNet, and Inception V3, were evaluated for their performance in classifying AD stages. MobileNet emerged as the top-performing algorithm, followed by CNN, DenseNet, and Inception V3. The study's findings highlight the promise of utilizing deep learning techniques for early Alzheimer's disease detection. However, the study's shortcomings, such as the quantity of the dataset and the assessment limits, are acknowledged. Despite these limitations, the findings are encouraging, and show the promise of using cutting-edge technologies to improve the early diagnosis and care of Alzheimer's More research is needed to have been validated to improve these systems for better patient care for better treatment of Alzheimer's disease.