GLAUCOMA DETECTION BASED ON QUANTIFICATION OF OPTIC DISC AND OPTIC CUP CHARACTERISTICS ON RETINAL IMAGES

• Main Author: Nafis Al Mustofa, Anas
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/58304
• Institution: Institut Teknologi Bandung

Glaucoma is eye disease which is caused by increase of intraocular pressure. The pressure is damaging optic nerve head and could lead to partially or even entirely loss of eyesight if there is no appropriate treatment. In Indonesia, lot of glaucoma patient were visiting hospital after they had severe stage of glaucoma that is when they had loss much of their eyesight. In this stage, the medication is difficult to do, and the patient have a big chance to continue to suffer irreversible blindness. Therefore, early glaucoma detection is crucial so that glaucoma can be treated as fast as possible. The glaucoma detection system provides robust glaucoma detection in a matter of second just by inputting patient’s digital retinal image. Currently, there are many developments in this field that have been carried out using various types of approaches. One of the approaches is by quantification of optic disc and cup’s characteristics e.g. cup to disc ratio. In this approach, the segmentation of optic disc and cup play an important role in glaucoma detection. However, there’s still space of improvements that can be conducted so that highest accuracy detection on many possible images is achieved. Moreover, an effective feature and classification method is still unknown. In this research, the author had developed an automation of glaucoma detection method based on the quantification of optical disc and cup characteristic. Proposed method successfully localizes optic disc with 99.99 ± 0,12% proportion within region of interest. Optic disc segmentation achieves fscore 0.935 ± 0.031 in Drishti-GS dataset and fscore 0.950 ± 0.028 in Refuge dataset. Furthermore, optic cup segmentation achieves fscore 0,832 ± 0,071 in Drishti-GS dataset and fscore 0,871 ± 0,061 in Refuge dataset. Glaucoma classification achieves 0,760 accuracy, 0,781 sensitivity, and 0,722 specificity in Drishti-GS, and 0,945 accuracy, 0,700 sensitivity, and 0,972 in Refuge datasets.