Detecting of endometriosis in ultrasound image using matlab image processing toolbox

Due to the advances in computer technology image processing method have developed rapidly and used in many areas. The goal of the thesis is to find out boundary of endometriosis which is a disease frequently seen in women in ultrasound image using image processing techniques. Image processing techniques is realized very difficult in with classical morphologic programming language. These difficulties are usually result of intensive mathematical operations. On the other hand the application program must be checked step by step. We used Matlab to overcome this problem in this thesisIn this study, to find the boundary of endometriosis cyst in a given ultrasound image active contour method is used based on techniques of curve evolution. In this study gray level ultrasound images are used. To evolve curve in images initial contour is tested different place and different dimension, algorithm run different iteration and results is monitorized. Difficulty is seen for detection of boundary in gray level ultrasound image. Also importance of initial contours?s place, dimension and iteration of algorithm is seen. After all real contour of endometriotic cyst in gray level ultrasound image is nearly obtained.Key Words : Endometriosis, Image Processing, Segmentation, Snake Model, Geometrical Deformation Model, Active Contour Model