A novel fuzzy blend scheme for image enhancement of CT scans bearing liver cancer

Abstract

ABSTRACT This paper deals with the design of an image enhancement algorithm which is capable of improving the quality of computed tomography (CT) images for liver cancer diagnosis. A novel soft computing technique namely fuzzy blend scheme (FBS) is proposed in this work. The FBS works in two stages. The first stage is the pre-processing stage in which the image is de-noised and sharpened by the application of a non-linear filter. In the second stage, the pre-processed output is blended with a novel fuzzy transformation function to give the contrast-enhanced image. The method is tested on 181 CT images from 15 patients of liver cancer named hepatocellular carcinoma. Quality metrics computed for the comparison of FBS with the existing methods include signal to noise ratio = 11.4270, enhancement measure = 12.30, structural similarity index = 91% and discrete entropy which is less than original image. The maximum computation time for FBS is 3.488 s. The proposed method can aid the radiologists and physicians in disease diagnosis. It will help in reducing the burden of ionic contrast reagents induced in patients for tumour visualisation.