Cyst and Tumor Lesion Segmentation on Dental Panoramic Images using Active Contour Models

Abstract

Active contours, or snakes, are computer-generated curves that move within images to find object boundaries. They are often used in computer vision and image analysis to detect and locate objects, and to describe their shape. Thus active contour can be used for object segmentation, especially the lesion in medical images. This paper presents the application of active contour models (Snakes) for the segmentation of lesions in dental panoramic image. The aim is to assist the clinical expert in locating potentially cyst or tumor cases for further analysis (e.g. classification of cyst or tumor lesion). In order to apply the snake formulation, color images were converted into gray images. Then, with correct parameters, we can create a snake that is attracted to edges or termination. Initializing contour, choosing parameter value and number of iteration affect the behaviour of the snake in a particular way. Using Receiver Pperating Characteristic (ROC), an average accuracy rate of 99.67 % is obtained. Examples of Snake segmentation results of lesions are presented.