Klasifikasi Citra X-ray Pada Penyakit Katarak, Retional Diabetik, dan Glukoma Menggunakan VGG 19

This study aims to improve the accuracy in classifying eye diseases, such as cataracts, diabetic retinopathy, and glaucoma, by utilizing the VGG19 deep learning model. Building on previous research that used retinal fundus image datasets, this study proposes a new approach to enhance model performance through more effective data augmentation and preprocessing techniques. The dataset used is sourced from public dataset providers such as Kaggle. The results show that applying the VGG19 model with a more balanced dataset significantly improves accuracy, with an accuracy of 90%, precision of 90%, recall of 90%, and an F1 score of 90%, which is higher than the previous study that achieved only 87,6% accuracy. This improvement can be attributed to the use of data augmentation and balancing techniques that addressed the dataset imbalance issues present in earlier studies, which had affected model performance. These findings demonstrate the great potential of the VGG19 model in supporting the early detection of eye diseases, which could enhance the efficiency of diagnosis and medical treatment in ophthalmology.