CAD System for Liver Diseases using Histological and Imaging features

Abstract

The current work for characterization of liver disease has been carried out using histological and imaging data. The BUPA liver disorders dataset created by University of California, Irvine has been considered as histological data. The ultrasound images of hepatocellular carcinoma (HCC) and Hemangioma (HEM) lesions were taken from ultrasoundcases.info. Laws ‘texture features were extracted from these images using Laws’ masks of length 3. In CAD system design 1, histological classification of liver diseases has been carried out using SVM classifier. In CAD system design 2, liver disease classification has been carried out using imaging features.