Glaucoma Detection Based on Cup-to-Disc Ratio in Retinal Fundus Image Using Support Vector Machine

Glaucoma is damage to the optic nerve due to increased pressure on the eyeball which can damage the optic nerve. Efforts that can be made to prevent glaucoma are early detection through retinal screening. To evaluate glaucoma, the most widely used indicator is the cup-to-disc ratio. Currently, cup-to-disc ratio evaluation is performed manually by a trained ophthalmologist or with expensive equipment such as Heidelberg Retinal Tomography (HRT) and Optical Coherence Tomography (OCT). This study aims to develop a system that can assist ophthalmologists in conducting analysis based on the value of the cup-to-disc ratio to determine the severity of glaucoma by using image processing. This system has 4 main stages, namely fundus image preprocessing to improve image quality, segmentation to detect and separate optical discs and optical cups from the background, feature extraction based on cup-to-disc ratio, area, perimeter, and circularity values. Classification based on the severity of glaucoma into early, moderate, and advanced glaucoma. From the system test which was carried out on 40 data, the accuracy level was 95%, precision was 93.91%, the recall was 91.37% and specificity was 95.86%.