An Effective Diagnosis of Alzheimer’s Disease with the Use of Deep Learning based CNN Model

In recent years, Alzheimer’s disease has become a major health concern. Over 45 million individuals worldwide suffer from this illness. Alzheimer’s disease is a neurodegenerative illness of unidentified etiology and process that causes cognitive deterioration, and it mostly affects the elderly. Dementia, the gradual and irreversible loss of brain cells, is the leading cause of Alzheimer’s disease. Those with this illness lost the ability to think, read, and do other basic tasks. A machine learning system may aid in this situation by making accurate predictions about the onset of illness. Dementia diagnosis in a broad patient population is the major objective. This research presents the findings and analysis of many Machine Learning models for identifying dementia. To alleviate this obstacle and help in the diagnosis of AD will create a deep-learning architecture that uses stacked auto-encoders and a softmax output layer. The proposed technique can analyze numerous classes in a single setting, unlike earlier procedures, and only needs a small number of labeled training samples and basic domain expertise to get started. In the present studies, a substantial improvement in performance when it came to classifying all diagnostic subgroups. Using the proposed technique can run these time series through a Convolutional Neural Network (CNN) and ResNet50 model to extract the distinguishing patterns across stages. With an F1-score of 0.99 and an accuracy of 99.91 percent, the CNN-based technique outperformed the standard feature-based classifiers by a significant margin.