Alzheimer's disease diagnosis and classification using deep learning techniques.

Abstract

Alzheimer's disease is an incurable neurodegenerative disease that affects brain memory mainly in aged people. Alzheimer's disease occurs worldwide and mainly affects people aged older than 65 years. Early diagnosis for accurate detection is needed for this disease. Manual diagnosis by health specialists is error prone and time consuming due to the large number of patients presenting with the disease. Various techniques have been applied to the diagnosis and classification of Alzheimer's disease but there is a need for more accuracy in early diagnosis solutions. The model proposed in this research suggests a deep learning-based solution using DenseNet-169 and ResNet-50 CNN architectures for the diagnosis and classification of Alzheimer's disease. The proposed model classifies Alzheimer's disease into Non-Dementia, Very Mild Dementia, Mild Dementia, and Moderate Dementia. The DenseNet-169 architecture outperformed in the training and testing phases. The training and testing accuracy values for DenseNet-169 are 0.977 and 0.8382, while the accuracy values for ResNet-50 were 0.8870 and 0.8192. The proposed model is usable for real-time analysis and classification of Alzheimer's disease.