SVM Based Liver Tumor Classification from Computerized Tomography Images
Priya V.1, Biju V. G.2
1Priya V., M.Tech Scholar, Department of Computer Science and Engineering, College of Engineering Munnar, Idukki, Kerala, India.
2Biju V. G., Associate Professor, Department of Electronic and Communication Engineering, College of Engineering Munnar, Idukki, Kerala, India.
Manuscript received on May 08, 2015. | Revised Manuscript received on May 24, 2015. | Manuscript published on May 31, 2015. | PP: 31-36 | Volume-2 Issue-6, May 2015. | Retrieval Number: F0303052615
Open Access | Ethics and Policies | Cite
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Accurate liver segmentation and tumor detection on computerized tomography (CT) images is a crucial task in the cases where surrounding tissues have intensities similar to that of the liver and lesions reside at the liver edges. In this paper, an automated method to segment liver portion, followed by tumor area from abdominal CT image is proposed. For this, the CT images are pre-processed by median filter to remove noise from the image and liver is segmented using localized region based active contouring algorithm. Tumor is detected from segmented liver using seed region growing algorithm. Using Grey Level Co-occurrence Matrix (GLCM), the texture features of the tumor are extracted. Support Vector machine (SVM) is used to classify the tumor as either benign or malignant based on these texture features. The performances of liver segmentation and tumor detection are evaluated by using Segmentation Matching Factor (SMF), Dice coefficient (DICE COEFF), Root Mean Square Error (RMSE) and Peak Signal to Noise Ratio (PSNR). Experimental results show that the proposed method has a lower error in segmenting the liver and is able to detect and classify all tumors from the liver accurately.
Keywords: CT image, Localized region based active contours, Region Growing Algorithm, GLCM, SVM.