SEGMENTASI PADA CITRA PANORAMIK GIGI DENGAN METODE TWO-STAGE SOM DAN T-CLUSTER
Abstrak: Segmentation on
medical image requires good quality due to affect the interpretation and
diagnosis of medical experts. On medical image segmentation, there is merging
phase to increase the quality of the segmentation result. However, stopping
criteria on merging phase was determined manually by medical experts. It
implied the subjectivity of segmentation result. To increase the objectivity of
segmentation result, a method to automate merging phase on medical image
segmentation is required. Therefore, we propose a novel method on medical image
segmentation which combine two-stage SOM and T-cluster method. Experiments were
performed on dental panoramic as medical image sample and evaluated by using
segmentation quality formula. Experiments show that the proposed method can
perform segmentation on dental panoramic image automatically and objectively
with the best average of segmentation quality value is 4,40.
Index Terms: dental panoramic image, image segmentation, medical image,
Self-Organizing Map, T-cluster
Penulis: Khoirul Umam, Fidi
Wincoko Putro, Gulpi Qorik Oktagalu Pratamasunu
Kode Jurnal: jptkomputerdd140327