K-means clustering and classification of medical images based on regions-of-interest

摘要

Cirrhosis of the liver is characterized by the presence of widespread nodules and fibrosis in the liver, resulting in characteristic texture patterns. Computerized analysis of hepatic texture patterns is usually based on regions-of-interest (ROIs). However, not all ROIs are typical representatives of the disease stage of the liver from which the ROIs originated. The uncertainties in the labels of ROI feature vectors present a problem in supervised classifiers which represent the majority of the commonly used classifiers. This paper investigates the use of an unsupervised classifier, the k-means clustering, in classifying ROI based data.