Retina Hastalıklarının Gerçek Zamanlı Tespiti için Yapay Zeka Destekli Yeni Bir Hesaplama Sisteminin Geliştirilmesi

Abstract

Dünya çapında görme bozukluğu ve körlüğün önemli bir nedeni olan retina hastalıkları, geri dönüşü olmayan görme kaybını önlemek için genellikle erken ve doğru tanı gerektirir. Optik Koherens Tomografi (OCT), retina katmanlarının ayrıntılı olarak görüntülenmesini sağlayan gelişmiş bir görüntüleme tekniğidir ve koroidal neovaskülarizasyon (CNV), diyabetik maküler ödem (DMÖ) ve drusen gibi retina bozukluklarının teşhisinde yaygın olarak kullanılmaktadır. Bu çalışmada, OCT görüntüleri kullanılarak retina hastalıklarının gerçek zamanlı tespiti için yeni bir yapay zeka (AI) destekli bilgisayar destekli tanı sistemi geliştirilmiştir. GoogleNet, ResNet, EfficientNet ve DenseNet dahil olmak üzere derin öğrenme modelleri uygulanmış ve karşılaştırmalı olarak değerlendirilmiştir. DenseNet-201, %94,42 doğruluk ve 1,00 AUC ile üstün performans göstererek bu çalışma için birincil model olmuştur. Sistem görüntü doğrulama, veri fazlalığını önlemek için hashing ve klinik kullanım için kullanıcı dostu bir arayüzü entegre etmektedir. Önerilen yaklaşım sadece tanısal doğruluğu artırmakla kalmayıp aynı zamanda klinisyenler üzerindeki zaman yükünü de azaltmaktadır. Gelecekteki çalışmalar, modelin farklı klinik veri kümeleriyle genelleştirilmesini geliştirmeye ve sistemi mevcut sağlık altyapılarına entegre etmeye odaklanacaktır.

Keywords

• Retina Hastalıkları
• Optik Koherens Tomografi
• Yapay Zeka
• Derin Öğrenme
• DenseNet
• Karar Destek Sistemi.

Development of a New Computational System Supported by Artificial Intelligence for Detection of Real-Time Retinal Diseases

Abstract

Retinal diseases such as choroidal neovascularization (CNV), diabetic macular edema (DME), and drusen are among the leading causes of vision loss worldwide, requiring early and accurate diagnosis to prevent irreversible damage. Optical Coherence Tomography (OCT) provides high-resolution imaging of retinal structures, making it a valuable tool in ophthalmological diagnosis. This study presents a novel artificial intelligence (AI)-supported computer-aided diagnostic system for the real-time classification of retinal diseases using OCT images. The proposed system integrates a DenseNet-201 deep learning model with a hash-based data integrity mechanism and a user-friendly interface for clinical deployment. The DenseNet-201 model achieved superior performance with an accuracy of 94.42%, an F1- score of 0.9442, and an AUC of 1.00, outperforming other widely used models such as GoogleNet, ResNet50, and EfficientNetB0. Unlike existing systems, our approach includes automatic image validation, eliminates data redundancy through hashing, and is optimized for practical use via the Gradio interface. These features address major limitations in prior studies, such as a lack of real-time capability, data inconsistency, and insufficient clinical integration. The system not only improves diagnostic accuracy but also reduces clinician workload, ensuring faster and more reliable decision-making in the detection of retinal diseases. This work demonstrates the feasibility of deploying AI-powered diagnostic tools in real-world ophthalmic settings and lays the groundwork for future development of integrated, scalable healthcare solutions.

Keywords

• Retinal Diseases
• Optical Coherence Tomography
• Artificial Intelligence
• Deep Learning
• DenseNet
• Decision Support System

References

1. [1] World Health Organization. (2019). World report on vision. Geneva: WHO.
2. [2] Resnikoff, S., Lansingh, V. C., Washburn, L., Felch, W. C., & Gauthier, T. M. (2020). Vision loss and its impact on quality of life. Ophthalmic Epidemiology, 27(2), 85–90.
3. [3] Lamoureux, E. L., & Fenwick, E. K. (2016). Health-related quality of life and visual impairment. Current Opinion in Ophthalmology, 27(3), 238–243.
4. [4] Forrester, J. V., Dick, A. D., McMenamin, P. G., Roberts, F., & Pearlman, E. (2015). The Eye: Basic Sciences in Practice. Elsevier Health Sciences.
5. [5] Berger, John, Ways of Seeing, Penguin Books, UK 2008, p.7-33.
6. [6] Kolb, H. (2005). Simple Anatomy of the Retina. Webvision.
7. [7] Curcio, C.A., Sloan, K.R., Kalina, R.E.,Hendrickson, A.E. (1990) Human photoreceptor topography. The Journal of Comparative Neurology, 292 (4), 497-523.
8. [8] Dandekar SS, Jenkins SA, Peto T, et al.: Autofluoresence imaging of choroidal neovascularization due to age-related macular degeneration. Arch Ophthalmol. 2005;123:1507-1513.

9. [9] Bhende, M, Shetty, S, Parthasarathy, M. K., & Ramya, S. (2018). Optical coherence tomography: A guide to interpretation of common macular diseases. Indian journal of ophthalmology, 66(1), 20.
10. [10] Erdoğan, Alper, Epiretinal Membran Cerrahisinde Prognozu Etkileyen Faktörler, Uzmanlık tezi, s.8.
11. [11] Deutman AF, Hansen LMAA: Dominantly inherited drusen of Bruch's membrane. Br J Ophthalmol 1970; 34:373-382.
12. [12] Aydın,Ali,Bilge,A.Hamdi, Optik Koherens Tomografinin Glokomda Yeri,s.78.
13. [13] Li, F., Chen, H., Liu, Z., Zhang, X.-D., Jiang, M.-S., Wu, Z.-Z., and Zhou, K.-Q. (2019). Deep learning-based automated detection of retinal diseases using optical coherence tomography images. Biomedical Optics Express, 10(12), 6204–6226. https://doi.org/10.1364/BOE.10.006204
14. [14] Serener, A., and Serte, S. (2019). Dry and wet age-related macular degeneration classification using OCT images and deep learning. 2019 Scientific Meeting on Electrical-Electronics & Biomedical Engineering and Computer Science (EBBT), 1–5. https://ieeexplore.ieee.org/document/8741768
15. [15] Motozawa, N., An, G., Takagi, S., Kitahata, S., Mandai, M., Hirami, Y., Yokota, H., Akiba, M., Tsujikawa, A., Takahashi, M., & Kurimoto, Y. (2019). Optical coherence tomography- based deep-learning models for classifying normal and age- related macular degeneration and exudative and non-exudative age-related macular degeneration changes. Ophthalmology and Therapy, 8(4), 527–539. https://doi.org/10.1007/s40123-019- 00207-y
16. [16] Sun, W., Zhang, H., & Yao, Z. (2020). Automatic diagnosis of macular diseases from OCT volume based on its two- dimensional feature map and convolutional neural network with attention mechanism. Journal of Biomedical Optics, 25(9), 096004. https://doi.org/10.1117/1.JBO.25.9.096004
17. [17] Acar, E., Türk, Ö., Ertuğrul, Ö. F., and Aldemir, E. (2021). Employing deep learning architectures for image-based automatic cataract diagnosis. Turkish Journal of Electrical Engineering & Computer Sciences, 29(5), 2649–2662. https://doi.org/10.3906/elk-2103-77
18. [18] Shi, X., Keenan, T. D. L., Chen, Q., De Silva, T., Thavikulwat, A. T., Broadhead, G., Bhandari, S., Cukras, C., Chew, E. Y., and Lu, Z. (2021). Improving interpretability in machine diagnosis: Detection of geographic atrophy in OCT scans. Ophthalmology Science, 1(3), 100038. https://doi.org/10.1016/j.xops.2021.100038
19. [19] Rajagopalan, N., Venkateswaran, N., Josephraj, A. N., and Srithaladevi, E. (2021). Diagnosis of retinal disorders from Optical Coherence Tomography images using CNN. PLoS ONE, 16(7), e0254180. https://doi.org/10.1371/journal.pone.0254180
20. [20] Şahin, M. E. (2022). A deep learning-based technique for diagnosing retinal disease by using optical coherence tomography (OCT) images. Turkish Journal of Science & Technology, 17(2), 417–426. https://doi.org/10.55525/tjst.1128395
21. [21] Elsharkawy, M., Sharafeldeen, A., Soliman, A., Khalifa, F., Ghazal, M., El-Daydamony, E., Atwan, A., Sandhu, H. S., & El-Baz, A. (2022). A novel computer-aided diagnostic system for early detection of diabetic retinopathy using 3D-OCT higher-order spatial appearance model. Diagnostics, 12(2), 532. https://doi.org/10.3390/diagnostics12020532
22. [22] He, T., Zhou, Q., and Zou, Y. (2022). Automatic detection of age-related macular degeneration based on deep learning and local outlier factor algorithm. Diagnostics, 12(2), 532. https://doi.org/10.3390/diagnostics12020532
23. [23] Aykat, Ş., & Şenan, S. (2023). Advanced detection of retinal diseases via novel hybrid deep learning approach. Traitement du Signal, 40(6), 2367–2382. https://doi.org/10.18280/ts.400604
24. [24] Baharlouei, Z., Rabbani, H., & Plonka, G. (2023). Wavelet scattering transform application in classification of retinal abnormalities using OCT images. Scientific Reports, 13, 19013. https://doi.org/10.1038/s41598-023-46200-1
25. [25] Kulyabin, M., Zhdanov, A., Nikiforova, A., Stepichev, A., Kuznetsova, A., Ronkin, M., Borisov, V., Bogachev, A., Korotkich, S., Constable, P. A., & Maier, A. (2024). OCTDL: Optical coherence tomography dataset for image-based deep learning methods. Scientific Data, 11(365). https://doi.org/10.1038/s41597-024-03182-7
26. [26] Gencer, G., & Gencer, K. (2025). Advanced retinal disease detection from OCT images using a hybrid squeeze and excitation enhanced model. PLOS ONE, 20(2), e0318657. https://doi.org/10.1371/journal.pone.0318657
27. [27] https://www.kaggle.com/datasets/paultimothymooney/kermany2018
28. [28] C. Cortes, X. Gonzalvo, V. Kuznetsov, M. Mohri, and S. Yang. Adanet: Adaptive structural learning of artificial neural networks. arXiv preprint arXiv:1607.01097, 2016.
29. [29] K. He, X. Zhang, S. Ren, and J. Sun. Deep residual learning for image recognition. In CVPR, 2016
30. [30] Urmamen H E. “Optik Koherans Tomografi Görüntüleri İle Retinal Hastalıkların Evrişimsel Sinir Ağı Kullanılarak Teşhis Edilmesi“. Yüksek Lisans Tez Çalışması, 2023, s.22
31. [31] Long, E. Shelhamer, and T. Darrell. Fully convolutional networks for semantic segmentation. In CVPR, 2015
32. [32] Özdemir E, Türkoğlu İ. Yazılım Güvenlik Açıklarının Evrişimsel Sinir Ağları (CNN) ile Sınıflandırılması, 2022
33. [33] Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., & Rabinovich, A. (2015). Going deeper with convolutions. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 1–9.
34. [34] 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 (CVPR), 770–778.
35. [35] Tan, M., & Le, Q. V. (2019). EfficientNet: Rethinking model scaling for convolutional neural networks. In Proceedings of the 36th International Conference on Machine Learning (ICML), PMLR, 6105–6114.
36. [36] Huang, G., Liu, Z., Van Der Maaten, L., & Weinberger, K. Q. (2017). Densely Connected Convolutional Networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 4700-4708.
37. [37] Gu, J., Wang, Z., Kuen, J., Ma, L., Shahroudy, A., Shuai, B., & Chen, T. (2018). Recent Advances in Convolutional Neural Networks. Pattern Recognition, 77, 354-377