SEGMENTASI CITRA CT SCAN LUNG MENGGUNAKAN DETEKSI TEPI SOBEL DAN METODE DISTANCE REGULARIZED LEVEL SET EVOLUTION (DRLSE)

The lung is one of the organs in the respiratory system that serves as a place to exchange oxygen with carbon dioxide in the blood. Disturbance the lung causes the patient difficult breathing, difficult to activity, lack of oxygen in fact if not quickly detected can cause death. To determine the symptoms of a patient's illness is generally carried out laboratory tests, where these tests are quite expensive and sometimes cause injuries and the result was sometimes long to be known x-ray images, this causes many people who indicated suffering from lung disease. The previous method capable of performing image segmentation ct scan lung difficult or not clarified to determination of edge detection and it still takes a long time for processing on ct scan lung image. Therefore, in this study to implement segmentation on ct scan lung image by using edge detection sobel and Distance Regularized Level Set Evolution Method to accelerate the segmentation of medical images. This method is capable of performing image segmentation ct scan lung with average an accuracy of 95.08% and average the Area Under the Curve (AUC) on relaive operating characteristic curve (ROC) amounted to 90.66%