Healthy and unhealthy rat hippocampus cells classification: A neural based automated system for Alzheimer disease classification

Alzheimer's disease (AD) is a serious progressive neurodegenerative disease. It is characterized by a severe memory loss and weakening in cognitive function. The Hippocampal atrophy is associated with Alzheimer's disease, where the hippocampus cells died. The current work presented an effective algorithm to classify a set of hippocampus rat brain images into normal and abnormal. This attempt can be used to recognize the AD that appears in the patient's brain. This classification is performed based on the cell status (healthy or unhealthy) in a pool of 176 rat hippocampus brain images. The contribution of this paper is to perform such classification based on Haralick features as indicated for the AD diagnosis. During this study the default partitioning ratio of the image set for training, test, and validation phase are employed. This ratio achieved about 93.18% classification accuracy with the proposed system. While, it established classification accuracy of 97.16% value using Jack Knife protocols for the used features.