Certain improvements in alzheimer disease classification using novel fuzzy c means clustering for image segmentation

Abstract

newline newlineMedical images play a vital role in ensuring information on the anatomy of human body. Because of the invention of a number of digital image equipments including Computed Tomography (CT), Magnetic Resonance Imaging (MRI), and Positron Emission Tomography (PET/CT) are proving to be inevitable in the arena of diagnosis of diseased condition. Segmentation technique is used in surgical plans, anomaly identification, post-surgery appraisal and other medical applications. In spite of using automated and semi-automated image segmentation techniques, in almost all cases there is noise which is not relevant, intensity in homogeneity, bad contrast values and weak boundaries which are inherent in medical images. MRI and other types of image contains complex anatomical structures that need precise segmentation for medical diagnoses. Brain image segmentation as such is very complex and tough but accurate segmentations are needed to detect tumors at an early stage, edema and also brain cell death. Diagnosis is made possible through precise identification of tissues. Dementia is a chronic or progressive nature syndrome due to various brain illnesses affecting thinking, memory, behavior and ability to perform daily work. Dementia is a clinical syndrome of cognitive decline which interferes with occupational and social functioning. It is an anchor point of reference in revised Alzheimer s Disease (AD) diagnostic criteria. AD is a foremost public health problem in countries with long life expectancy. According to estimates about three million Americans and about two million Japanese suffer from AD. This work classifies brain images for dementia detection. newline