Publication

Abstract : Alzheimer's disease (AD) is a degenerative neurological disorder that causes brain deterioration.   Alzheimer's disease can be effectively treated and the degradation of brain tissue can be halted with early detection.   Memory loss is one of the symptoms of AD. Even though AD has no known cure and severely impairs the lives of those who suffer from it, early identification may be helpful in starting the appropriate course of treatment to stop further brain deterioration.  The sources of the data are the ADNI datasets.  The images are preprocessed in order to lower noise in the resulting dataset.  During the pre-processing stage, a nonlocal means filter is used. The Histogram of Gradients (HOG) Model is used to extract the feature.   Using magnetic resonance imaging (MRI) and a proposed deep learning model known as Grey Wolf Optimization based Elephant Herding Optimization (GWO-EHO) architecture, the work focuses on early AD detection.    The four types of AD are normal controls, early mild cognitive impairment, and late mild cognitive impairment.   Data from the AD  neuroimaging study were used to assess this method's ability to identify the illness early.  The proposed GWO-EHO approach has attained maximum precision, sensitivity, recall, and accuracy, according to the test findings.