Detect Alzheimer’s Disease from Neuroimaging Using Artificial Intelligence Techniques

Alzheimer’s disease (AD) is the continuous deterioration of cognitive, behavioral, and social skills that affects a person’s ability to function independently. It is the most common cause of memory impairment and dementia. Early detection is considered one of the most important factors that helps make plans and prevent the development of symptoms of this disease. The purpose of this work is to compare machine learning (ML) and deep learning (DL) models to detect early warning signs of AD. In addition, a neural network is utilized to extract important features and distinguish between healthy brain imaging and that of those affected by mild cognitive impairment (MCI). A 3D convolutional neural network (3D-CNN) architecture is utilized to compute the features of neuroimaging data. After that, the feature vector is fed to the support vector machines (SVM) for final recognition. The system is assessed using a publicly available ADNI dataset for functional magnetic resonance imaging (fMRI). In the case of a 3D convolutional neural network, a recognition rate of 98.92% and 98.3% is attained on training and testing, respectively. In the case of the SVM model, an accuracy of 97.14% and 97.5% is achieved on training and testing, respectively.