Volume 10 • Issue 1 • PP: 01–07 • 2025
Machine Learning Algorithm Comparison for Four-Class Retinal Disease Classification Using Digital Fundus Images
Open Access & Copyright
© 2025 The Author(s). Published by ASPG. This article is licensed under the Creative Commons Attribution 4.0 International License (CC BY 4.0).
View PDF Abstract
Retinal diseases are a major cause of vision loss, making timely and accurate diagnosis important for effective treatment. This study compares convolutional neural networks (CNN), k-nearest neighbors (K-NN), and support vector machines (SVM) for classifying digital fundus images into four categories: normal retina, cataract, glaucoma, and other retinal diseases. The models are evaluated using accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and F-score. CNN achieves the highest reported accuracy (94.05%) and sensitivity (0.9977), with an F-score of 0.9693. The results indicate that CNN can be useful for automated retinal screening, while its low specificity (0.0178) highlights the need to reduce false-positive predictions. The comparison illustrates both the potential and the limitations of machine-learning-based retinal disease classification.
Keywords
References
[1] B. Hassan, H. Raja, T. Hassan, M. U. Akram, H. Raja,A. A. Abd-alrazaq, S. Yousefi, and N. Werghi, “A com-prehensive review of artificial intelligence models for screening major retinal diseases,” Artificial Intelligence Review, vol. 57, no. 5, p. 111, 2024.
[2] A. Grzybowski, K. Jin, J. Zhou, X. Pan, M. Wang, J. Ye, and T. Y. Wong, “Retina fundus photograph-based arti-ficial intelligence algorithms in medicine: A systematic review,” Ophthalmology and Therapy, vol. 13, no. 8, pp. 2125–2149, 2024.
[3] N. U. Haq, T. Waheed, K. Ishaq, M. A. Hassan, N. Safie, N. F. Elias, and M. Shoaib, “Computationally efficient deep learning models for diabetic retinopathy detection: A systematic literature review,” Artificial Intelligence Review, vol. 57, p. 309, 2024.
[4] J. I. Lim, A. V. Rachitskaya, J. A. Hallak, S. Gholami, and M. N. Alam, “Artificial intelligence for retinal dis-eases,” Asia-Pacific Journal of Ophthalmology, vol. 13, no. 4, p. 100096, 2024.
[5] M. Guo, D. Gong, and W. Yang, “In-depth analysis of research hotspots and emerging trends in ai for retinal diseases over the past decade,” Frontiers in Medicine, vol. 11, p. 1489139, 2024.
[6] Y. Jin, L. Liang, J. Li, K. Xu, W. Zhou, and Y. Li, “Artificial intelligence and glaucoma: A lucid and com-prehensive review,” Frontiers in Medicine, vol. 11, p. 1423813, 2024.
[7] S. Islam, R. C. Deo, P. D. Barua, J. Soar, P. Yu, and U. R. Acharya, “Retinal health screening using artificial intelligence with digital fundus images: A review of the last decade (2012–2023),” IEEE Access, vol. 12, pp. 176 630–176 685, 2024.
[8] J. Yao, J. Lim, G. Y. S. Lim, J. C. L. Ong, Y. Ke, T. F. Tan, T.-E. Tan, S. Vujosevic et al., “Novel artificial intel-ligence algorithms for diabetic retinopathy and diabetic macular edema,” Eye and Vision, vol. 11, p. 23, 2024.
[9] N. Nasr El-Den, M. Elsharkawy, I. Saleh, M. Ghazal, A. Khalil et al., “Ai-based methods for detecting and classifying age-related macular degeneration: A comprehensive review,” Artificial Intelligence Review, vol. 57, p. 237, 2024.
[10] Y. Gao, F. Xiong, J. Xiong, Z. Chen, Y. Lin, X. Xia, Y. Yang, G. Li, and Y. Hu, “Recent advances in the ap-plication of artificial intelligence in age-related macular degeneration,” BMJ Open Ophthalmology, vol. 9, no. 1, p. e001903, 2024.
[13] M. H. Akpinar, M. Akpinar, A. Sengur, O. Faust, L. Tong, F. Molinari, and U. R. Acharya, “Artificial intelligence in retinal screening using oct images: A re-view of the last decade (2013–2023),” Computer Meth-ods and Programs in Biomedicine, vol. 254, p. 108253, 2024.
[14] T. D. Nguyen, D.-T. Le, J. Bum, S.-H. Kim, S. J. Song, and H. Choo, “Retinal disease diagnosis using deep learning on ultra-wide-field fundus images,” Diagnos-tics, vol. 14, no. 1, p. 105, 2024.
[15] S. Benbakreti, S. Benbakreti, and U. Ozkaya, “The clas-sification of eye diseases from fundus images based on cnn and pretrained models,” Acta Polytechnica, vol. 64, no. 1, pp. 1–11, 2024.
[16] A. Mihalache, R. S. Huang, M. Cruz-Pimentel, N. S. Patil, M. M. Popovic, B. U. Pandya, R. Shor, A. Pereira, and R. H. Muni, “Artificial intelligence chatbot interpre-tation of ophthalmic multimodal imaging cases,” Eye, vol. 38, no. 13, pp. 2491–2493, 2024.
[17] L. La Franca, C. Rutigliani, L. Checchin, R. Lattanzio, F. Bandello, and M. V. Cicinelli, “Rate and predictors of misclassification of active diabetic macular edema as detected by an automated retinal image analysis system,” Ophthalmology and Therapy, vol. 13, no. 6, pp. 1553–1567, 2024.
[18] P. Bansal, N. Harjai, M. Saif, S. H. Mugloo, and P. Kaur, “Utilization of big data classification models in digitally enhanced optical coherence tomography for medical di-agnostics,” Neural Computing and Applications, vol. 36, no. 1, pp. 225–239, 2024.
[19] S. Wang, W. Shen, Z. Gao, X. Jiang, Y. Wang, Y. Li, X. Ma, W. Wang, S. Xin, W. Ren, K. Jin, and J. Ye, “Enhancing the ophthalmic ai assessment with a fundus image quality classifier using local and global atten-tion mechanisms,” Frontiers in Medicine, vol. 11, p. 1418048, 2024.
[20] R. Alizadeh Eghtedar, A. Vard, M. Malekahmadi, and A. Peyman, “A new computer-aided diagnosis tool based on deep learning methods for automatic detec-tion of retinal disorders from oct images,” International Ophthalmology, vol. 44, no. 1, p. 110, 2024.
[21] A. J. Prabha, C. Venkatesan, M. S. Fathimal, K. K. Nithiyanantham, and S. P. A. Kirubha, “Rd-oct net: Hy-brid learning system for automated diagnosis of macular diseases from oct retinal images,” Biomedical Physics and Engineering Express, vol. 10, no. 2, p. 025033, 2024.
[22] X. Xu, M. Zhang, S. Huang, X. Li, X. Kui, and J. Liu, “The application of artificial intelligence in diabetic retinopathy: Progress and prospects,” Frontiers in Cell and Developmental Biology, vol. 12, p. 1473176, 2024.
[23] G. S. Reiter, J. Mai, S. Riedl, K. Birner, S. Frank, H. Bo-gunovic, and U. Schmidt-Erfurth, “Ai in the clinical management of ga: A novel therapeutic universe re-quires novel tools,” Progress in Retinal and Eye Re-search, vol. 103, p. 101305, 2024.
[24] E. R. Martin et al., “Ocular biomarkers: Useful inciden-tal findings by deep learning algorithms in fundus pho-tographs,” Eye, vol. 38, no. 13, pp. 2581–2588, 2024.
[25] S. Liu, W. Wang, L. Deng, and H. Xu, “Cnn-trans model: A parallel dual-branch network for fundus image clas-sification,” Biomedical Signal Processing and Control, vol. 96, p. 106621, 2024.
[26] G. Corradetti, A. Verma, J. Tojjar, L. Almidani, D. On-cel, M. Emamverdi, A. Bradley, S. Lindenberg, M. G. Nittala, and S. R. Sadda, “Retinal imaging findings in inherited retinal diseases,” Journal of Clinical Medicine, vol. 13, no. 7, p. 2079, 2024.
[27] X. Xia, Y. Li, G. Xiao, K. Zhan, J. Yan, C. Cai, Y. Fang, and G. Huang, “Benchmarking deep models on reti-nal fundus disease diagnosis and a large-scale dataset,” Signal Processing: Image Communication, vol. 127, p. 117151, 2024.
[28] H. Kang, H. Yang, M. Kim, and H. Choo, “Data effi-cient image classification for retinal disease diagnosis,” Journal of Internet Computing and Services, vol. 25, no. 3, pp. 19–25, 2024.
[29] J. Ong and J. Chhablani, “Advances in imaging-based machine learning and therapeutic technology in the man-agement of retinal diseases,” Medicina, vol. 60, no. 11, p. 1794, 2024.
[30] S. Ejaz, R. Baig, Z. Ashraf, M. M. Alnfiai, M. M. Alna-hari, and R. M. Alotaibi, “A deep learning framework for the early detection of multi-retinal diseases,” PLOS ONE, vol. 19, no. 7, p. e0307317, 2024.
[31] S. Touma, B. A. Hammou, F. Antaki, M. C. Boucher, and R. Duval, “Comparing code-free deep learning models to expert-designed models for detecting retinal diseases from optical coherence tomography,” Interna-tional Journal of Retina and Vitreous, vol. 10, no. 1, p. 36, 2024.
[32] Y. Chen, X. Zhang, J. Yang, G. Han, H. Zhang, M. Lai, and J. Zhao, “Hdb-net: Hierarchical dual-branch net-work for retinal layer segmentation in diseased oct im-ages,” Biomedical Optics Express, vol. 15, no. 9, pp. 5359–5383, 2024.
[33] T. F. Tan, S. Y.-H. Chang, and D. S. W. Ting, “Deep learning for precision medicine: Guiding laser therapy in ischemic retinal diseases,” Cell Reports Medicine, vol. 4, no. 7, p. 101239, 2023.
[34] P. K. Karn and W. H. Abdulla, “On machine learning in clinical interpretation of retinal diseases using oct images,” Bioengineering, vol. 10, no. 4, p. 407, 2023.
[35] M. Subramanian et al., “Diagnosis of retinal diseases based on bayesian optimization deep learning network using optical coherence tomography images,” Computa-tional Intelligence and Neuroscience, p. 8014979, 2022.
[36] P. Burlina, W. Paul, T. Y. A. Liu, and N. M. Bressler, “Detecting anomalies in retinal diseases using genera-tive, discriminative, and self-supervised deep learning,” JAMA Ophthalmology, vol. 140, no. 2, p. 185, 2022.
[37] K. M. Kim et al., “Development of a fundus image-based deep learning diagnostic tool for various retinal diseases,” Journal of Personalized Medicine, vol. 11, no. 5, p. 321, 2021.
[38] T. K. Yoo, J. Y. Choi, and H. K. Kim, “Feasibility study to improve deep learning in oct diagnosis of rare retinal diseases with few-shot classification,” Medical and Bio-logical Engineering and Computing, vol. 59, no. 2, pp. 401–415, 2021.
[39] Q. Zhang, Z. Liu, J. Li, and G. Liu, “Identifying dia-betic macular edema and other retinal diseases by opti-cal coherence tomography image and multiscale deep learning,” Diabetes, Metabolic Syndrome and Obesity, vol. 13, pp. 4787–4800, 2020.
Cite This Article
Choose your preferred format
Publisher's Note
The statements, opinions, and data presented in this article are solely those of the author(s) and do not necessarily represent those of ASPG, the journal, or its editors. ASPG and the editors disclaim responsibility for any harm arising from the use of any ideas, methods, instructions, or products described in this article, to the fullest extent permitted by applicable law.