Liver CT image analysis using TVSeg algorithm and DWPT-based rough set classifier

• Main Authors: Hsun-Ting Kuo, 郭訓廷
• Other Authors: none
• Format: Others
• Language: en_US
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/29567570348670018023

碩士 === 國立雲林科技大學 === 資訊管理系碩士班 === 100 === Recently, high prevalence and low age of liver cancer, more and more researchers concern about detecting diagnosing hepatic tumors or even lesions in abdominal computed tomography (CT) images. CT imaging is a robust, straightforward technique, and it is used for the inspection of liver diseases the most (such as liver lesion, liver tumor, and hepatitis). Abdominal CT is a well-established tool and widely applied to related detection work. Furthermore, CT images segmentation is one of important tools in many clinical applications of computer aided diagnosis and therapy planning. The research on automated methods involves different organs among which the liver is the most emphasized. However, the great amount of CT scans of each patient still is a challenge for specialists. Hence, several literatures have proposed methods for automatic or semi-automatic diagnosis by computer-aided detection (CAD) to assist manual inspection. This study proposed a method combining semi-automatic segmentation of region of interest (ROI) using TVSeg algorithm, feature filtering using discrete wavelet packet transform, and supervised rough set (RS) theory to classify the two datasets. The proposed method can make sure that input CT image need to be detected out the actual information of tumors. In experimental verification, two dataset is implemented; 2/3 for training and the remained 1/3 for testing. Through the classification of rough sets theory, based on LEM2 algorithm, the datasets can be classified. The results display a higher accuracy and verify that proposed method can improve the efficiency.