Designing an Intelligent System for Predicting Alzheimer’s Disease

Alzheimer's disease (AD) is a degenerative progressive disorder that affects the brain's neurons and nerve cells, causing behavioral changes, memory loss, language skills, and thinking. It is a neurological condition with an exponentially increasing incidence rate, primarily affecting adults over 65. Contrary to popular belief, AD is not a normal aspect of aging and is the most prevalent type of dementia. In this work, CNN, Densenet169, and the Hybrid convolution recurrent neural network approach are used to detect Alzheimer's disease at an early stage. Data augmentation is utilized at preprocessing step to handle the small size of the dataset. The Hybrid CNN-RNN network design comprises convolution layers followed by a recurrent neural network (RNN). The combined model uses the RNN to extract relationships from MRI images and to account for temporal dependencies of the images during classification. Three algorithms are used for classifying AD and comparing their results. We have tested the model on MRI dataset. According to the results, the proposed CNN algorithm achieved higher accuracy than the Densenet169 and the hybrid Convolution-Recurrent Neural Network.