Image Retrieval Framework for Anomalies Detection in X-Ray Radiographical Images

Abstract

Medical imaging is well-known since ages and has become very popular now-a-days. Medical imaging solely based on datasets. The medical dataset consists of highly sensitive and valuable information. Since last two decades, lots of images are added to the domain, hence an efficient image retrieval is on the radars of many researchers. The Content Based Medical Image Retrieval System (CBMIRS) retrieve an image of interest from large set of medical images. The heart of CBMIRS is image retrieval strategy. There are numerous strategies developed, but does not address all classes of images. Different modalities and orientation of images makes the retrieval system difficult for searching and retrieving. This paper uses the histogram of image-based strategy for image retrieval, which is insensitive for any orientation and modality of the image. The validation is carried out with large set of X-rays radiographical images. The results suggest efficiency of the used strategy.