Comparative Study of CNN Architectures for Ophthalmic Disease Detection

International Journal of Emerging Research in Science, Engineering, and Management
Vol. 2, Issue 3, pp. 18-24, March 2026.

https://doi.org/10.58482/ijersem.v2i3.3

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

Comparative Study of CNN Architectures for Ophthalmic Disease Detection

Full Text PDF

M Kumaraswamy

V Prakash

G Manasa

K Niranjan

N Lakshmi Charan Reddy

K Naveen Kumar

Department of CSE, Siddharth Institute of Engineering & Technology, Puttur, India.

Abstract: Common eye conditions, including diabetic retinopathy, glaucoma, cataracts, and age-related macular degeneration, pose a significant challenge to global health due to the potential for irreversible vision loss when these diseases are not detected and treated early enough. Currently, with the decrease in the price of retinal imaging, the increased availability of retinal image databases, and the advancement of machine learning through deep learning techniques, especially through the use of Convolutional Neural Networks (CNNs), automated detection of these common eye diseases is now feasible on a large scale. The objective of this research is to compare results from several different CNN models in order to determine which type of architecture is most effective at diagnosing ophthalmic conditions from fundus and OCT images. Each architecture is evaluated based on a series of performance metrics, including overall accuracy, sensitivity, specificity, computational complexity, training time, and robustness in the presence of image variation. Results from this comparative study showed large performance differences between CNN architectures, emphasizing the importance of selecting an appropriate CNN architecture for optimal diagnostic accuracy. This research presents important evidence regarding the strengths and weaknesses of popular convolutional neural network architectures and serves as a useful resource for developing improved and more reliable automated systems for assisting in the diagnosis of ophthalmologic disease.

Keywords: Ophthalmic Diseases, Deep Learning, Fundus and OCT Images, Comparative Study, Computer-Aided Diagnosis.

References: 

  1. V. Gulshan et al., “Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal Fundus photographs,” JAMA, vol. 316, no. 22, p. 2402, Nov. 2016, doi: 10.1001/jama.2016.17216.
  2. Y. LeCun, Y. Bengio, and G. Hinton, “Deep learning,” Nature, vol. 521, no. 7553, pp. 436–444, May 2015, doi: 10.1038/nature14539.
  3. K. Simonyan and A. Zisserman, “Very deep convolutional networks for Large-Scale image recognition,” arXiv.org, Sep. 04, 2014. https://arxiv.org/abs/1409.1556
  4. K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” arXiv.org, Dec. 10, 2015. https://arxiv.org/abs/1512.03385
  5. C. Szegedy et al., “Going deeper with convolutions,” 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston, MA, USA, 2015, pp. 1-9, doi: 10.1109/CVPR.2015.7298594.
  6. G. Huang, Z. Liu, L. Van Der Maaten and K. Q. Weinberger, “Densely Connected Convolutional Networks,” 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA, 2017, pp. 2261-2269, doi: 10.1109/CVPR.2017.243.
  7. M. Tan and Q. Le V., “EfficientNet: Rethinking model scaling for convolutional neural networks,” arXiv.org, May 28, 2019. https://arxiv.org/abs/1905.11946
  8. G. Litjens et al., “A survey on deep learning in medical image analysis,” Medical Image Analysis, vol. 42, pp. 60–88, Jul. 2017, doi: 10.1016/j.media.2017.07.005.
  9. H. Pratt, F. Coenen, D. M. Broadbent, S. P. Harding, and Y. Zheng, “Convolutional neural networks for diabetic retinopathy,” Procedia Computer Science, vol. 90, pp. 200–205, Jan. 2016, doi: 10.1016/j.procs.2016.07.014.
  10. D. S. W. Ting et al., “Artificial intelligence and deep learning in ophthalmology,” British Journal of Ophthalmology, Feb. 2019, doi: 10.1136/bjophthalmol-2018-313173.
  11. A. Krizhevsky, I. Sutskever, and G. E. Hinton, “ImageNet classification with deep convolutional neural networks,” Communications of the ACM, vol. 60, no. 6, pp. 84–90, May 2017, doi: 10.1145/3065386.
  12. A. Esteva et al., “Dermatologist-level classification of skin cancer with deep neural networks,” Nature, vol. 542, no. 7639, pp. 115–118, Jan. 2017, doi: 10.1038/nature21056.
  13. A. Krizhevsky, I. Sutskever, and G. E. Hinton, “ImageNet classification with deep convolutional neural networks,” Communications of the ACM, vol. 60, no. 6, pp. 84–90, May 2017, doi: 10.1145/3065386.
Download PDF

This article is part of Volume 2, Issue 3 (March 2026)