Classification of Ovarian Cysts on Ultrasound Images Using Watershed Segmentation and Contour Analysis

Abstract

Ovarian cyst is a disease that occurs in the uterus of a woman, the method of detection and analysis is carried out by experts by looking at and observing the size of the cyst and the characteristics of the cyst on an ultrasound device. The accuracy of manual ovarian cyst measurement analysis on ultrasound examination results often results in errors, therefore a tool is needed to calculate the size of the cyst and detect the characteristics of the cyst based on the papillary growth in the cyst. Ultrasound image from the hospital as input from the system, then a preprocessing process is carried out to remove noise in the image, the next step is the segmentation process using the watershed method, the segmentation results will be used for feature extraction by detecting cysts and papillary and their sizes using contour analysis with the bounding box method. The extraction feature will be used for cyst classification. This system has an accuracy rate of 97.8%.