Automatic segmentation and classification of liver tumor in CT images using adaptive hybrid technique and Contourlet based ELM classifier

The liver is necessary for survival and is also prone to many diseases. CT examinations can be used to plan and properly administer radiation treatments for tumors and to guide biopsies and other minimally invasive procedure. Manual segmentation and classification of CT image is a tedious task and time consuming process which is impractical for large amount of data. Fully automatic and unsupervised methods eliminate the need for manual interaction. In this paper, evaluation of potential role of the adaptive hybrid segmentation algorithm, Contourlet transform and the Extreme Learning Machine in the differential diagnosis of liver tumors in CT images are proposed. The liver is segmented from CT images using adaptive threshold method and morphological processing. Extraction of tumor is done by means of Fuzzy C Means (FCM) clustering from the segmented liver region. The statistical and textural information are obtained from the extracted tumor using Contourlet Transform. The features like mean, standard deviation and entropy of the obtained sub bands are calculated and stored in a feature vector. The extracted features are fed as input to Extreme Learning Machine classifier to classify the diseases such as hepatoma, hemangioma and cholangiocarcinoma. The segmentation results are compared with the experts results and analyzed. The classifier differentiates the tumor with relatively high accuracy and provides a second opinion to the radiologist.