Classification of Liver Disease Based on US Images

Liver disease is progressive, asymptotic and potentially fatal Diseases. In this study, an automatic hierarchical procedure to classify and stage liver disease using ultrasound images is described. The database for this work is the ultra sonographic images of liver disease along with the healthy conditions. Initially the contrast enhancement is applied to the input image that helps to identify the object, after that discrete wavelet transform is applied which helps to remove the speckle noise, then the approximate component is subjected to K-mean clustering which segments the image with respect to the minimum Euclidian distance. The classification strategy is performed using the classifier such as Neural Network. It is used to analyze the Liver disease which will be useful to doctors for the second opinion.