Classification of Retinal Diseases in Fundus Images Using Hybrid Deep Learning Based on Multiscale Feature Fusion

Abstract

Diabetic retinopathy (DR) and glaucoma affect more than 400 million people worldwide and are among the leading causes of preventable blindness. Early diagnosis of these diseases is critical to preventing vision loss. However, in regions with limited access to healthcare, the shortage of qualified ophthalmologists makes timely diagnosis of these diseases difficult. In this study, a hybrid deep learning architecture combining the powerful feature extraction capabilities of EfficientNet-B3 and ResNet50 architectures was developed for the automated and accurate diagnosis of diabetic retinopathy and glaucoma from fundus images. The proposed HybridNet architecture adopts a multi-scale feature fusion strategy, simultaneously analyzing both micro-scale lesions (microaneurysms, hard exudate, retinal hemorrhages) and macro-scale structural tissue changes (optic disc geometry, retinal vascular network morphology) in fundus images. Additionally, integrated Grad-CAM heat maps provide clinical interpretability, demonstrating the model's focus on medically critical areas such as pathological hemorrhages, the optic disc region, and vascular structures during the decision-making phase, offering physicians reliable guidance in the diagnostic process. Experimental findings show that the developed system achieved high-performance metrics such as a 98.0% F1-score and 0.99 AUC in the DR class, a 0.972 F1-score in the Glaucoma class, and a 0.978 F1-score in the Healthy class. These results demonstrate that the proposed hybrid architecture can create a reliable screening infrastructure even in regions with limited access to healthcare and can offer transparent and high-performance diagnostic capabilities in clinical decision support systems. In conclusion, this study makes a significant contribution to the field of AI-powered medical image analysis, highlighting the vital role that deep learning technology plays in the diagnosis of retinal diseases and the democratization of healthcare.

Keywords

• Fundus imaging
• Artificial intelligence
• Hybrid deep learning
• Diabetic retinopathy
• Glaucoma

Ethical Statement

Ethics committee approval was not required for this study because of there was no study on animals or humans.

Thanks

This study was not financially supported by any institution or organization.

References

1. Abbas, Q., Qureshi, I., Yan, J., & Shaheed, K. (2022). Machine learning methods for diagnosis of eye-related diseases: A systematic review study based on ophthalmic imaging modalities. Archives of Computational Methods in Engineering, 29(6), 3861–3918. https://doi.org/10.1007/s11831-022-09720-z
2. Abràmoff, M. D., Lavin, P. T., Birch, M., Shah, N., & Folk, J. C. (2018). Pivotal trial of an autonomous AI-based diagnostic system for detection of diabetic retinopathy in primary care offices. npj Digital Medicine, 1(1), Article 39. https://doi.org/10.1038/s41746-018-0040-6
3. Al-Shabrawey, M., Zhang, W., & McDonald, D. (2015). Diabetic retinopathy: Mechanism, diagnosis, prevention, and treatment. BioMed Research International, 2015, Article 854593. https://doi.org/10.1155/2015/854593
4. Arun, N., Gaw, N., Singh, P., Kanter, P., Krishnan, R., Ennab, M., Hoebel, K., Zaim, S., & Kalpathy-Cramer, J. (2021). Assessing the trustworthiness of saliency maps for localizing abnormalities in medical imaging. Radiology: Artificial Intelligence, 3(6), Article e200267. https://doi.org/10.1148/ryai.2021200267
5. Aykat, Ş., & Senan, S. (2023). Derin öğrenme kullanılarak fundus görüntülerinden katarakt ve diyabetik retinopati tespiti [Detection of cataract and diabetic retinopathy from fundus images using deep learning]. Mühendislik Bilimleri ve Araştırmaları Dergisi, 5(2), 312–324. https://doi.org/10.46387/bjesr.1332567
6. Chankhachon, S., Kansomkeat, S., Bhurayanontachai, P., & Intajag, S. (2025). Deep learning network with illuminant augmentation for diabetic retinopathy segmentation using comprehensive anatomical context integration. Diagnostics, 15(21), Article 2762. https://doi.org/10.3390/diagnostics15212762
7. Chattopadhay, A., Sarkar, A., Howlader, P., & Balasubramanian, V. N. (2018). Grad-CAM++: Generalized gradient-weighted class activation mapping for visual explanations of deep convolutional networks. In 2018 IEEE Winter Conference on Applications of Computer Vision (WACV) (pp. 839–847). IEEE. https://doi.org/10.1109/WACV.2018.00097
8. Grauslund, J. (2022). Diabetic retinopathy screening in the era of artificial intelligence. Diabetologia, 65(9), 1415–1423. https://doi.org/10.1007/s00125-022-05727-0

9. Gulshan, V., Peng, L., Coram, M., Stumpe, M. C., Wu, D., Narayanaswamy, A., Seneviratne, S., Gadepalli, R., Johnston, P., & Webster, D. R. (2016). Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs. JAMA, 316(22), 2402–2410. https://doi.org/10.1001/jama.2016.17216
10. Gulum, M. A., Trombley, C. M., & Kantardzic, M. (2021). A review of explainable deep learning cancer detection models in medical imaging. Applied Sciences, 11(10), Article 4573. https://doi.org/10.3390/app11104573
11. He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 770–778). IEEE. https://doi.org/10.1109/CVPR.2016.90
12. Kılıç, Ş. (2025). HybridVisionNet: An advanced hybrid deep learning framework for automated multi-class ocular disease diagnosis using fundus imaging. Ain Shams Engineering Journal, 16(5), Article 103351. https://doi.org/10.1016/j.asej.2025.103594
13. Kocak, B., Klontzas, M. E., Stanzione, A., Meddeb, A., Demircioğlu, A., Bluethgen, C., & Cuocolo, R. (2025). Evaluation metrics in medical imaging AI: Fundamentals, pitfalls, misapplications, and recommendations. European Journal of Radiology Open: Artificial Intelligence, 3, 100030.
14. LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521(7553), 436–444. https://doi.org/10.1038/nature14539
15. Nørgaard, M. F., & Grauslund, J. (2018). Automated screening for diabetic retinopathy: A systematic review. Ophthalmic Research, 60(1), 9–17. https://doi.org/10.1159/000486284
16. Pan, Y., Liu, J., Cai, Y., Yang, X., Zhang, Z., Long, H., & Tan, Z. (2023). Fundus image classification using Inception V3 and ResNet-50 for the early diagnostics of fundus diseases. Frontiers in Physiology, 14, Article 1126780. https://doi.org/10.3389/fphys.2023.1126780
17. Polater, S. N., & Işık, A. H. (2024). Diyabetik retinopati tespiti için derin öğrenmeye dayalı sınıflandırma [Deep learning-based classification for diabetic retinopathy detection]. Uluborlu Mesleki Bilimler Dergisi, 7(2), 13–19.
18. Sayres, R., Taly, A., Rahimy, E., Blumer, K., Coz, D., Hammel, N., Seneviratne, S., & Webster, D. R. (2019). Using a deep learning algorithm and integrated gradients explanation to assist grading for diabetic retinopathy. Ophthalmology, 126(4), 552–564. https://doi.org/10.1016/j.ophtha.2018.11.016
19. Selvaraju, R. R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., & Batra, D. (2017). Grad-CAM: Visual explanations from deep networks via gradient-based localization. In Proceedings of the IEEE International Conference on Computer Vision (ICCV) (pp. 618–626). IEEE. https://doi.org/10.1109/ICCV.2017.74
20. Sharmin, S., Rashid, M. R., Khatun, T., Hasan, M. Z., Uddin, M. S., & Marzia. (2024). A dataset of color fundus images for the detection and classification of eye diseases. Data in Brief, 57, Article 110979. https://doi.org/10.1016/j.dib.2024.110979
21. Sounderajah, V., Ashrafian, H., & Darzi, A. (2025). The STARD-AI reporting guideline for diagnostic accuracy studies using artificial intelligence. Nature Medicine, 31(1), 1–10. https://doi.org/10.1038/s41591-025-03953-8
22. Sun, H., Saeedi, P., Karuranga, S., Pinkepank, M., Ogurtsova, K., Harrison, S., & Magliano, D. J. (2022). IDF Diabetes Atlas: Global, regional and country-level diabetes prevalence estimates for 2021 and projections for 2045. Diabetes Research and Clinical Practice, 183, Article 109119. https://doi.org/10.1016/j.diabres.2021.109119
23. Sutradhar, I., Gayen, P., & Hasan, M. (2019). Eye diseases: The neglected health condition among urban slum population of Dhaka, Bangladesh. BMC Ophthalmology, 19, Article 38. https://doi.org/10.1186/s12886-019-1043-z
24. Tamim, N., Elshrkawey, M., & Nassar, H. (2021). Accurate diagnosis of diabetic retinopathy and glaucoma using retinal fundus images based on hybrid features and genetic algorithm. Applied Sciences, 11(13), Article 6178. https://doi.org/10.3390/app11136178
25. Tan, M., & Le, Q. (2019). EfficientNet: Rethinking model scaling for convolutional neural networks. In International Conference on Machine Learning (pp. 6105–6114). PMLR.
26. Teo, Z. L., Kwee, A., Lim, J. C., Lam, C. S., Ho, D., Maurer-Stroh, S., & Ting, D. S. (2023). Artificial intelligence innovation in healthcare: Relevance of reporting guidelines for clinical translation from bench to bedside. Annals of the Academy of Medicine, Singapore, 52(4), 199–212. https://doi.org/10.47102/annals-acadmedsg.2022452
27. Teo, Z. L., Tham, Y. C., Yu, M., Cheng, C. Y., Wong, T. Y., & Sabanayagam, C. (2020). Do we have enough ophthalmologists to manage vision-threatening diabetic retinopathy? A global perspective. Eye, 34(7), 1255–1261. https://doi.org/10.1038/s41433-020-0776-5
28. Tham, Y. C., Li, X., Wong, T. Y., Quigley, H. A., Aung, T., & Cheng, C. Y. (2014). Global prevalence of glaucoma and projections of glaucoma burden through 2040: A systematic review and meta-analysis. Ophthalmology, 121(11), 2081–2090. https://doi.org/10.1016/j.ophtha.2014.05.013
29. Uçar, M. (2021). Glokom hastalığının evrişimli sinir ağı mimarileri ile tespiti [Detection of glaucoma disease with convolutional neural network architectures]. Dokuz Eylül Üniversitesi Mühendislik Fakültesi Fen ve Mühendislik Dergisi, 23(68), 521–529.
30. World Health Organization. (2019). World report on vision. World Health Organization.
31. Yıldırım, Ö., & Özbay, F. A. (2022). Fundus görüntülerinden derin öğrenme teknikleri ile glokom hastalığının tespiti [Detection of glaucoma disease from fundus images with deep learning techniques]. Avrupa Bilim ve Teknoloji Dergisi, (44), 1–6. https://doi.org/10.31590/ejosat.1216404
32. Zhou, Y., Chia, M. A., Wagner, S. K., Ayhan, M. S., Williamson, D. J., Struyven, R. R., & Keane, P. A. (2023). A foundation model for generalizable disease detection from retinal images. Nature, 622(7981), 156–163. https://doi.org/10.1038/s41586-023-06555-x