An Intelligent Detection of Alzheimer Disease Using Advanced Deep Learning Approaches

Alzheimer's disease is becoming more common daily, especially in elderly persons over the age of 55 and eventually becomes the cause of their demise. It is the most common kind of dementia in the neurological system. It causes a continuous decline in cognitive function, ultimately eradicating a person's ability to think and understand since early diagnosis is one of the most essential components for improving medical and therapeutic results. Compared to conventional machine learning methods, deep learning models outperform them in diagnosing AD since they do not require manually created feature extraction. The significant idea of this study is to describe how to classify MRIs of AD using transfer learning and convolutional neural networks (CNNs) and to create a deep learning-based pipeline for accurately identifying and categorizing the various phases of AD. The recommended analysis makes use of learning models that have already been trained and evaluated, such as VGG19, Inception V3, and Convolutional Neural Network (CNN), to reach test accuracies of 95.5%, 91.5%, and 92.16%, respectively. They are recognizing that VGG19 affords the highest degree of accuracy.