Evaluation of Current Dental Radiographs Segmentation Approaches in Computer- aided Applications Abdolvahab Ehsani Rad, Mohd Shafry Mohd Rahim, Amjad Rehman1, Ayman Altameem 2 and Tanzila Saba3 Faculty of Computing, Universiti Teknologi Malaysia, Malaysia, Kuala Lumpur, 'MIS Department, College of Business Administration, Salman bin Abdul Aziz University, Alkharj, KSA, 2College of Applied Studies and Community Services, King Saud University, Riyadh, KSA, 3 Department of Computer and Information Sciences, Prince Sultan University Riyadh KSA Abstract With a wide variety researches on Image segmentation techniques in biomedical and bioinformatics area, it is important to analyze the performance of these approaches in specific problems. Image segmentation is one of the most significant processes of dental X-ray image analysis. Therefore, to obtain the proper result, it is required to perform the accurate and efficient segmentation approach which proved itself in the aspect of X-ray image segmentation. The aim of this review paper is to understand the different image segmenta tion approaches which have been used for dental X-ray image analysis over the past studies. In this paper, different available approaches of dental X-ray image segmentation, reviewed and their advantages, disad vantages, and limitations are discussed. Keywords Dentalradiographs, Medical imagesegmentation, Segmentationmethods. 1. Introduction Dental X-ray imaging is one of the most common and low cost w ays to get the im age (information) of teeth. The X-ray images can be used in computer applications such as hum an identification system s or assisting in clinical aspects like dental diagnosis systems and dental treat- m ent system s. H ow ever, to analysis such an X-ray im age, w e need to use some process on the images to obtain the inform ation. M o st im p o rtant im ag e-p ro cessing procedure used for analyzing medical images and computer-aided m edical diagnosis system s is im age segm entation [1]. Segmentation of such medical images has more diffi- culty in process due to a vast variety in topologies, the intricacy of m edical structures, and poor im age qualities caused by some conditions such as noise, low contrast, and sim ilarity of body tissues, som e sort of artifacts, and lim itation of scanning m ethod s, w hich results in u nsuc- cessful segmentation. "Segmentation subdivides an im age into its constituent regions or object" [2]. In other definition, image segmentation procedure is defined as the process of extracting region of interest (ROI) from the image background. There are two basic properties which generally image segmentation approaches are based on, one of these tw o intensity values are: sim ilar- ity and discontinuity. The major approaches in the first m ethods are based on segm enting an im age into regions that are sim ilar according to a set of predefined criteria. The approach of second m ethods is to segm ent an im age based on sudden changes in intensity, such as edges in an im age [3]. W e can classify the segmentation methods based on the pixel values and their relationships into three areas; pixel-based, edge-based, and region-based. In pixel-based approaches, the classification is based on pixel gray-level values in images. Edge-based segmenta - tion approaches are based on abrupt changes of intensity values in image areas. And, region-based segmentation m ethods are based on differences in predefined values of neighboring pixels in the image. Figure 1 demonstrates the classification of segm entation approaches. M oreover, we can classify another group as "Hybrid" which is based on combination of other methods. Analysis of dental X-ray images has some difficulty in comparison with other medical images which makes segmentation a more challenging process. The dif- ficulties are like: sam ple of artifacts, im pacted teeth, variations of tooth, space between missing tooth, and also problems in imaging process. Figure 2 illustrates the difficulties w hich can appear on teeth im ages. Due to these problems, still finding the accurate and proper method in the segmentation of dental X-ray images is a challenging process. Nonetheless, many surveys on m ed ical im age p ro cessing have been p ublished in differ- ent journals, but none of them focused on dental X-ray images. To overcome this lack, in this review paper, the various approaches of image segmentation techniques that are w idely used in the area of computer vision w ith application to dental X-ray images are reviewed and 210 IETE TECHNICAL REVIEW | VOL 30 | ISSUE 3 | MAY-JUN 2013