Multi-input Deep Convolutional Neural Network Based on Transfer Learning for Assisted Diagnosis of Alzheimer’s Disease

Alzheimer’s Disease is the most common form of dementia which initially impairs the memory and finally progresses to death. There is no effective treatment for this irreversible disease. The latest innovations in multimodal neuroimaging data and artificial intelligence technology made it possible to detect this disease in the early stage, which has become a major research area in neuroscience. We proposed a deep learning algorithm using pre-train Restnet50 that takes both gray matter and white matter into account which would have a potential improvement to the existing CAD methods of AD diagnosis is utilized for the classification of brain images among Cognitively Normal (CN), Early Mild Cognitive Impairment (EMCL), Late Mild Cognitive Impairment (LMCI), Alzheimer’s Disease (AD), ensuring very precise and accurate diagnosis.