On Classification of BMD Images Using Machine Learning (ANN) Algorithm

Abstract

Osteoporosis is a disease which has affected major part of human body that is bones. It trends to reduce the mass of the bones and hence degrade the micro architecture of tissues of bones. There are multiple imaging technologies which have been used from the decades in diagnosis and the micro architecture of distressed as well as affected bones in order to find the bone density deficiency. Image processing methods are widely used like filtering, segmentation, classification, image enhancement as well as other pre-processing technique to diagnosis the affected bone structure easily. It will help to extract the vital information about deformed micro architecture pattern. In this paper we have collected the basic information of tissues such as osteoblast as well as osteoclast and also have presented a comparative study of various diagnosis technique based on the image processing for the osteoporosis. The main aim of this work is to assess the prevalence of osteoporosis and changes in bone mass with increasing age, compare bone health status of apparently healthy men, premenopausal and postmenopausal women. In this work, we have used ethical database of 260 subject 130 men, 80 women (premenopausal) and 50 women (postmenopausal). Bone mineral density (BMD) has been measured through dual energy X-ray absorptiometry at femoral neck.