Segmentation of Breast Using Ultrasound Image for Detection Breast Cancer

Breast cancer condition when cancer cells form in breast tissue. Cancer can also form in fat or connective tissue in the breast. Is a type of cancer that is deadly and difficult to detect at an early age but along with the development of technology in the health sector many methods are used for early detection of cancer, one of which is by using USG (Ultrasonography). Ultrasound is an imaging procedure using high-frequency sound wave technology to produce images of the inner body, such as organs and soft tissues. The ultrasound test results have very low quality, giving rise to many different perceptions in diagnosing the disease. From these problems sparked an idea to detect breast cancer automatically. The watershed transform algorithm is used in the segmentation process to produce the location of the cancer and can distinguish objects based on background. The result of segmentation using watershed is the second segmentation process using thresholding binaries to separate the image of the cancer as the object being observed. The final step is the process of calculating the area of cancer. The result of this paper is a comparison of the calculation of the area of cancer between Hospital data and trial results, from the results of these trials the system accuracy reached 88.65% of all data tested and an error of 11.35%. From these tests it can be concluded that the method used is the watershed transform algorithm capable of segmenting the image of the breast ultrasound image.