Copyright © 2018 Authors. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. International Journal of Engineering & Technology, 7 (2.21) (2018) 140-143 International Journal of Engineering & Technology Website: www.sciencepubco.com/index.php/IJET Research paper Automatic segmentation of chondroblastoma from X-ray images using active contour and levelset method P.Y. Muhammed Anshad 1* , Dr. S.S. Kumar 2 1 Research Scholar, Department of Electronics & Communication, Noorul Islam University, India 2 Associate Professor, Electronics and Instrumentation, Noorul Islam University, India *Corresponding author E-mail: anshadpy@gmail.com Abstract Chondroblastoma is a benign but locally aggressive bone tumor found usually in the age below 25 years. Chondroblastoma is a destructive type of lesion with a thin radio dense border which is normally seen in the epiphysis of long bones. The benign tumors have similarities in pathology and could be related with histogenic similarity. This tumor reduces the strength of affected bone and may leads to death if not treated early. Chondroblastoma can be diagnosed from X-ray/CT/MRI images and the treatment is its removal by surgical methods. Diagnosis of Chondroblastoma is difficult due to the similarities with other benign tumors like chondromyxoid fibroma. To reduce diagnostic errors, computer aided methods can adopt. This work focuses on automatic segmentation of Chondroblastoma using active contour and level set method which gives better segmentation results and a mild stone to CAD design. Keywords: Chondroblastoma, computer aided methods, segmentation, active contour, levelset method, CAD, 1. Introduction Chondroblastoma is a benign bone tumor most frequently arise in the epiphyses of long bones, with 75% occurring in the humerus area. They tumor size commonly seen is 3-4 cm. Figure 1 shows the anatomy of normal knee and Chondroblastoma affected knee. It clearly shows some lesion around the cartilage area. Chondroblastoma is diagnosed from X-ray/CT images and the tumor can be removed by surgical method. In the proposed method, the system can identify tumor area using automatic seed point selection method, can identify the region of interest (ROI) using segmentation method and can calculate the volume of ROI which assists the doctors for the successful removal of tumor. Figure 1: a). Normal knee anatomy b). Chondroblastoma affected knee In most of theimage based diagnostic operations need individual objects to be divided into different from the image. Image segmentation is a fundamental task which is responsible for the separation of different parts using homogeneity. The job of segmentation is to divide an image into its basic and disjoint sub- regions that are identical based on their properties like color, quality, intensity etc. Segmentation algorithms are commonly based on either discontinuity or similarity with sub regions. The purpose of segmentation is to divide an image into different sub- regions. The role of classification is to identify the divided sub- regions. Thus, segmentation and classification are general functions as individual and sequential process. Active contours work on the basis of relaxation [1] methods, and this valid to different image segmentation issues. Active contours also can be used for image segmentation and boundary tracking in snakes by Kass et al. [2]. The basic idea is to start with initial boundary shapes which is represented in a type of enclosed curves. The contours are iteratively change them by applying expansion or shrink operations based on the constraints of images. These are called contour evolution and made by the minimization of some energy function like fixed region based segmentation methods [3]. 2. Background In the area of medicine, diagnosis is one of the biggest and fundamental task to start treatment. There is a well-developed system already available in the field of diagnosis like X-Ray, MRI, CT etc. These machineries only provide the images of the affected area for diagnosis. The decision is still manual in most of the cases. This may increase the possibility of misdiagnosis. Here comes the importance of CAD system. Now a day’s more researches going on in the field of CAD. Segmentation is one of the difficult and important area in CAD systems. There is a variety of segmentation methods available. This is possible to use semiautomatic or fully automatic segmentation methods which depend on the area or method of segmentation. Bone tumor images varies in size, shapes, position and its appearance. It’s based on reducing an object function by updating membership function to cluster centers. The object function used here is the weighted sum of distances from cluster centers. The weighted mean of data is the cluster center anditeration is continued till the operation between iterations exceeds a threshold value. This method is more time consuming and to overcome this problem. Atkins proposed such a model for brain tumor segmentation. Here