Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal EDUMATIC: Jurnal Pendidikan Informatika
Putri, Chana Amelinda
Unknown Affiliation
Computer Science & IT Education

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Diagnosis Dini Penyakit Mata: Klasifikasi Citra Fundus Retina dengan Convolutional Neural Network VGG-16 Putri, Chana Amelinda; Rakasiwi, Sindhu
Jurnal Pendidikan Informatika (EDUMATIC) Vol 9 No 1 (2025): Edumatic: Jurnal Pendidikan Informatika
Publisher : Universitas Hamzanwadi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29408/edumatic.v9i1.29571

Abstract

Retinal fundus image-based eye disease classification is important to support early diagnosis of vision disorders such as cataracts, glaucoma, and diabetic retinopathy. This study aims to diagnose early eye diseases with retinal fundus image classification using Convolutional Neural Network VGG-16. The model was developed to detect cataract, glaucoma, and diabetic retinopathy to support early diagnosis. The dataset used comes from Kaggle, including 4,217 retinal fundus images consisting of 1,038 cataract, 1,007 glaucoma, 1,098 diabetic retinopathy, and 1,074 normal images. The images were processed through normalization, augmentation, and resizing to 224×224 pixels, with the dataset divided in a ratio of 80:10:10 for training, validation, and testing. Results showed that the VGG-16 model with transfer learning achieved 88% accuracy, a 10% increase from the previous 75% in the CNN model without transfer learning. This model has the potential to be integrated in clinical decision support systems or mobile applications to improve the speed and accuracy of diagnosis. Limitations of the study include the limited dataset size and potential data bias that may affect the accuracy of the model in detecting eye diseases early, so future research is recommended to use larger and more diverse datasets, as well as explore other deep learning architectures to improve classification performance.
Co-Authors Sindhu Rakasiwi