Klasifikasi Penyakit Retina melalui Citra OCT (Optical Coherence Tomography) Menggunakan K-Nearest Neighbor

Abstract

Retinal disease is a vision disorder that affects the macular health of the retina which is a serious problem that can affect normal vision. CNV (choroidal neovascularization), DME (diabetic macular edema), and drusen are types of retinal disease which are common causes of vision loss. Retinal disease can be diagnosed by using OCT (Optical Coherence Tomography) to produce an overview of the condition of the macular retina of the eye which indicates retinal disease. The purpose of using technology in this study is used as a medium for classifying retinal diseases. In this study, the method used is K-Nearest Neighbor to assist the classification process. The steps taken before carrying out the classification are the preprocessing stage (grayscaling, gaussian blur), and the canny edge detection stage. The results of this study indicate that this method is capable of classifying retinal diseases by obtaining an accuracy of 91.5%.