Sistem Deteksi Level Diabetic Retinopathy Melalui Citra Fundus Mata dengan Menggunakan Metode CNN (Convolutional Neural Network)

Mochamad Andy Ardyansyah (Universitas Sangga Buana, Indonesia)
Gunawansyah (Universitas Sangga Buana, Indonesia)

G-Tech : Jurnal Teknologi Terapan
Vol 7 No 4 (2023): G-Tech, Vol. 7 No. 4 Oktober 2023

Diabetes is one disease that has a serious impact on eye health, especially in a condition known as Diabetic Retinopathy (DR). DR can cause retinal damage and potentially lead to vision loss. Therefore, early detection and regular monitoring are essential. This study aims to develop a system for detecting Diabetic Retinopathy levels on fundus images of the eye using the Convolutional Neural Network (CNN) method. CNN is one technique in the field of Deep Learning that has proven effective in complex image analysis such as medical images. The dataset used is an image of the fundus of the eye sourced from kaggle and has been labeled in each class. The system is made using matlab software that can classify Diabteic Retinopathy into five classes. The test results obtained the best results with an accuracy rate as high as 85%