Multi-Disease Retinal Classification Using EfficientNet-B3 and Targeted Albumentations: A Benchmark on Kaggle Retinal Fundus Images Dataset

Authors

  • Kurniawan Aji Saputra Information Engineering, Faculty of Computer Science, Universitas Dian Nuswantoro, Semarang, Indonesia
  • Farrikh Alzami Information Engineering, Faculty of Computer Science, Universitas Dian Nuswantoro, Semarang, Indonesia
  • Defri Kurniawan Information Engineering, Faculty of Computer Science, Universitas Dian Nuswantoro, Semarang, Indonesia
  • Muhammad Naufal Information Engineering, Faculty of Computer Science, Universitas Dian Nuswantoro, Semarang, Indonesia
  • Muslih Information Engineering, Faculty of Computer Science, Universitas Dian Nuswantoro, Semarang, Indonesia
  • Rama Aria Megantara Information Engineering, Faculty of Computer Science, Universitas Dian Nuswantoro, Semarang, Indonesia
  • Ricardus Anggi Pramunendar Information Engineering, Faculty of Computer Science, Universitas Dian Nuswantoro, Semarang, Indonesia

DOI:

10.33395/sinkron.v10i1.15530

Keywords:

albumentations, cataract, diabetic retinopathy, EfficientNet-B3, fundus images, glaucoma, retinal disease classification

Abstract

Retinal diseases remain one of the leading causes of blindness worldwide. This study develops a deep learning pipeline for multiclass retinal disease classification using EfficientNet-B3 combined with Albumentations to improve generalization. We target four classes: cataract, diabetic retinopathy, glaucoma, and normal. We use the Kaggle Retinal Disease dataset (4,217 fundus images) divided into 70% training, 10% validation, and 20% testing. Images are resized to 224×224 and augmented with horizontal flip, random brightness contrast, CLAHE, shiftscale rotate, crop, gamma correction, and elastic transformation. The EfficientNet-B3 backbone is refined after head training with warm-up and learning rate regularization (batch normalization, dropout). After 50 epochs, the best validation performance reaches 0.9526, and on the hold-out test set, the model achieves 95.38% overall accuracy. The F1 scores per class were 1.0000 (diabetic retinopathy), 0.9685 (cataract), 0.9255 (normal), and 0.9184 (glaucoma). Confusion analysis showed that most errors involved glaucoma being misclassified as normal, likely due to optic disc similarities. These results demonstrate that EfficientNet-B3 with targeted augmentation provides accurate and reliable multi-disease screening of fundus images, with the potential to support faster and more consistent triage in clinical workflows. Future research should expand clinical validation and explore attention mechanisms or multimodal input to reduce glaucoma-normal ambiguity.

GS Cited Analysis

Downloads

Download data is not yet available.

References

Aldrees, A., Min, H., Dutta, A. K., Daradkeh, Y. I., & Anjum, M. (2025). Improving Fundus Detection Precision in Diabetic Retinopathy Using Derivative-Based Deep Neural Networks. CMES - Computer Modeling in Engineering and Sciences, 142(3), 2487–2511. https://doi.org/10.32604/cmes.2025.061103

Alzami, F., Abdussalam, Megantara, R. A., Fanani, A. Z., & Purwanto. (2019). Diabetic Retinopathy Grade Classification based on Fractal Analysis and Random Forest. 2019 International Seminar on Application for Technology of Information and Communication (ISemantic), 272–276. https://doi.org/10.1109/ISEMANTIC.2019.8884217

Anwar, M., Farhan, S., Ul Haq, Y., Azeem, W., Ilyas, M., Voicu, R. C., & Tanveer, M. H. (2025). E-GlauNet: A CNN-Based Ensemble Deep Learning Model for Glaucoma Detection and Staging Using Retinal Fundus Images. Computers, Materials and Continua, 84(2), 3477–3502. https://doi.org/10.32604/cmc.2025.065141

Arshad Hussain, M. S., Babu, S., Sri Satya Sai, M. K., Siddartha, K., & Naik, K. B. (2024). Retinal Disease Classification using EfficientNet-B3. Proceedings - 2024 4th International Conference on Soft Computing for Security Applications, ICSCSA 2024, 337–344. https://doi.org/10.1109/ICSCSA64454.2024.00060

Basit, S. A., Al-Absi, H. R. H., Musleh, S., & Alam, T. (2025). Cardiometabolic biomarker prediction based on retinal fundus image. Engineering Applications of Artificial Intelligence, 160. https://doi.org/10.1016/j.engappai.2025.111734

Bilal, H., Keles, A., & Bendechache, M. (2025). Advances in disease detection through retinal imaging: A systematic review. In Computers in Biology and Medicine (Vol. 194). Elsevier Ltd. https://doi.org/10.1016/j.compbiomed.2025.110412

Cahya, F. N., Hardi, N., Riana, D., & Hadianti, S. (2021). Klasifikasi Penyakit Mata Menggunakan Convolutional Neural Network ( CNN). SISTEMASI: Jurnal Sistem Informasi, 10(3),618-626. https://doi.org/https://doi.org/10.32520/stmsi.v10i3.1248

Di Giammarco, M., Santone, A., Cesarelli, M., Martinelli, F., & Mercaldo, F. (2025). Explainable retinal disease classification and localization through Convolutional Neural Networks. Image and Vision Computing, 162. https://doi.org/10.1016/j.imavis.2025.105667

Hasan, M. N., Pial, M. E. R., Das, S., Siddique, N., & Wang, H. (2025). DIA-VXNET: A framework for automated diabetic eye disease detection using transfer learning with feature fusion network. Biomedical Signal Processing and Control, 100. https://doi.org/10.1016/j.bspc.2024.106907

Jatmoko, C., & Lestiawan, H. (2024). PREDIKSI PENYAKIT MATA MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORK. (SEMNAS RISTEK) Seminar Nasional Riset Dan Inovasi Teknologi. https://doi.org/https://doi.org/10.30998/semnasristek.v8i01.7129

Jatmoko, C., Lestiawan, H., Agustina, F., & Erawan, L. (2024). Comparative Study of Classification of Eye Disease Types Using DenseNet and EfficientNetB3. Kinetik: Game Technology, Information System, Computer Network, Computing, Electronics, and Control. https://doi.org/10.22219/kinetik.v9i3.1931

Meedeniya, D., Shyamalee, T., Lim, G., & Yogarajah, P. (2025). Glaucoma identification with retinal fundus images using deep learning: Systematic review. In Informatics in Medicine Unlocked (Vol. 56). Elsevier Ltd. https://doi.org/10.1016/j.imu.2025.101644

Mellor, J., Jiang, W., Fleming, A., McGurnaghan, S. J., Blackbourn, L., Styles, C., Storkey, A. J., McKeigue, P. M., & Colhoun, H. M. (2023). Can deep learning on retinal images augment known risk factors for cardiovascular disease prediction in diabetes? A prospective cohort study from the national screening programme in Scotland. International Journal of Medical Informatics, 175. https://doi.org/10.1016/j.ijmedinf.2023.105072

Mohan, R., Kadry, S., Yassine, S., & Rajinikanth, V. (2025). Healthy/Glaucoma Fundus Retinal Image Classification using Butterfly Algorithm Optimized ResNet-Features. Procedia Computer Science, 258, 1804–1812. https://doi.org/10.1016/j.procs.2025.04.432

Mouhafid, M., Zhou, Y., Shan, C., & Xiao, Z. (2025). A Robust Approach to Early Glaucoma Identification from Retinal Fundus Images using Dirichlet-based Weighted Average Ensemble and Bayesian Optimization. Current Medical Imaging Formerly Current Medical Imaging Reviews, 21. https://doi.org/10.2174/0115734056335762250128095107

Nur, M., Muhlashin, I., Stefanie, A., Universitas, S., Karawang, J. H., Ronggo, W., & Karawang, I. (2023). KLASIFIKASI PENYAKIT MATA BERDASARKAN CITRA FUNDUS MENGGUNAKAN YOLO V8. In Jurnal Mahasiswa Teknik Informatika (Vol. 7, Issue 2).

Putri, C. A., & Rakasiwi, S. (2025). Diagnosis Dini Penyakit Mata: Klasifikasi Citra Fundus Retina dengan Convolutional Neural Network VGG-16. Edumatic: Jurnal Pendidikan Informatika, 9(1), 208–216. https://doi.org/10.29408/edumatic.v9i1.29571

Qulub, M. S., & Agustin, S. (2024). INDENTIFIKASI PENYAKIT MATA DENGAN KLASIFIKASI CITRA FOTO FUNDUS MENGUNAKAN CONVOLUTIONAL NEURAL NETWORK (CNN). In Jurnal Mahasiswa Teknik Informatika (Vol. 8, Issue 5).

Tiwari, S., Sahu, K., Vijh, S., & Awasthi, C. (2025). Deep Learning-Based Classification of Ocular Toxoplasmosis Fundus Images: A Comparative Study of CNN and SVM Models. Procedia Computer Science, 259, 1189–1197. https://doi.org/10.1016/j.procs.2025.04.074

Xu, X., Li, J., Guan, Y., Zhao, L., Zhao, Q., Zhang, L., & Li, L. (2021). GLA-Net: A global-local attention network for automatic cataract classification. Journal of Biomedical Informatics, 124. https://doi.org/10.1016/j.jbi.2021.103939

Downloads


Crossmark Updates

How to Cite

Saputra, K. A., Alzami, F., Kurniawan, D., Naufal, M., Muslih, M., Megantara, R. A. ., & Pramunendar, R. A. . (2026). Multi-Disease Retinal Classification Using EfficientNet-B3 and Targeted Albumentations: A Benchmark on Kaggle Retinal Fundus Images Dataset. Sinkron : Jurnal Dan Penelitian Teknik Informatika, 10(1), 232-241. https://doi.org/10.33395/sinkron.v10i1.15530

Issue

Vol. 10 No. 1 (2026): Article Research January 2026

Section

Articles

License

Copyright (c) 2025 Kurniawan Aji Saputra, Farrikh Alzami, Defri Kurniawan, Muhammad Naufal, Muslih Muslih, Rama Aria Megantara, Ricardus Anggi Pramunendar

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

Crossref
Scopus
Google Scholar
Europe PMC

Most read articles by the same author(s)