Early Diagnosis and Detection of Vascular Dementia based on Mobile V Net Framework using DCNN

The Alzheimer disease is a long-term neurodegenerative illness. It is the leading causative agent of brain damage. Alzheimer's disease has an impact for memory, behaviour, and reasoning. It is possible to diagnose neurological illnesses including schizophrenia, Alzheimer's disease, and dementia with the exact segmentation of brain tissues in magnetic resonance imaging, which include the white matter, cerebrospinal fluid, and grey matter. Alzheimer's disease has been diagnosed by examining the patient's MRI scan. Throughout this analysis, our aim is to develop the Convolutional neural network design towards predicting Alzheimer illness. The Deep Convolutional Neural Network will be used in this study to classify Alzheimer's disease (CNN). To begin, we worked on the MRI images, splitting them into an appropriate format to prevent data leakage. In this paper we proposed a Mobile Net V2 Architecture model that is used in depth wise separable convolutions and has fewer parameters and numerous conceptual precisions. To minimise the amount of network parameters and maximize classification results, dense blocks, as proposed in Dense Nets, are incorporated Mobile Net. The Mobile Net architecture for Alzheimer's disease classification achieves an accuracy of 94%.