An Intelligent Deep Learning Framework for Ocular Disease Diagnostics: The RetinaVision Approach

Bonasi Poojitha, Dr.S.Usharani

Ocular Disease Detection; Deep Learning; EfficientNet, Fundus Imaging; Grad-CAM, Diabetic Retinopathy; Explainable AI.

Vision impairment and blindness caused by ocular diseases remain a major global health challenge, affecting over 2.2 billion people worldwide. Many conditions such as diabetic retinopathy, glaucoma, cataract, age-related macular degeneration, hypertension-related retinopathy, and myopia are progressive but preventable with early diagnosis. However, limited access to ophthalmologists, high consultation costs, and time-intensive manual diagnosis hinder timely detection, especially in developing regions. RetinaVision AI addresses these challenges through a deep learning-based automated system for retinal disease detection using fundus images. The system is built on the EfficientNet-B3 architecture, known for its high accuracy and computational efficiency. It employs two models: a multi-disease classifier that detects eight retinal conditions, and a specialized model for diabetic retinopathy severity grading across four levels.To enhance interpretability, the system integrates Grad-CAM, which generates heatmaps highlighting important regions in the retinal image, ensuring clinically relevant decision-making. RetinaVision AI is deployed as a user-friendly web application using Streamlit, enabling easy access through a browser. Overall, the system demonstrates the effective use of deep learning, transfer learning, and explainable AI for scalable and reliable ophthalmological diagnostics.

[1] M. Tan and Q. V. Le, “EfficientNet: Rethinking model scaling for convolutional neural networks,” Proc. ICML, vol. 97, pp. 6105–6114, 2019.
[2] 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, pp. 2402–2410, 2016.
[3] R. R. Selvaraju et al., “Grad-CAM: Visual explanations from deep networks via gradient-based localization,” Proc. ICCV, pp. 618–626, 2017.
[4] D. S. W. Ting et al., “Development and validation of a deep learning system for diabetic retinopathy and related eye diseases using retinal images from multiethnic populations with diabetes,” JAMA, vol. 318, no. 22, pp. 2211–2223, 2017.
[5] K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” Proc. CVPR, pp. 770–778, 2016.
[6] K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale image recognition,” arXiv preprint arXiv:1409.1556, 2014.
[7] J. Deng et al., “ImageNet: A large-scale hierarchical image database,” Proc. CVPR, pp. 248–255, 2009.
[8] A. Paszke et al., “PyTorch: An imperative style, high-performance deep learning library,” Adv. Neural Inf. Process. Syst., vol. 32, pp. 8026–8037, 2019.
[9] Aneeshkumar A. S., “Relevance of Data Mining for Presumption of Anarchy,” International Journal of Linguistics and computational Applications (IJLCA), IRDP Group of Journals, vol.3, issue 4, pp. 72-74, 2016.
[10] G. Bradski, “The OpenCV library,” Dr. Dobb’s J. Softw. Tools, vol. 25, no. 11, pp. 120–125, 2000.
[11] S. van der Walt, S. C. Colbert, and G. Varoquaux, “The NumPy array: A structure for efficient numerical computation,” Comput. Sci. Eng., vol. 13, no. 2, pp. 22–30, 2011.
[12] World Health Organization, World Report on Vision. Geneva, Switzerland: WHO Press, 2023.
[13] A. Esteva et al., “Dermatologist-level classification of skin cancer with deep neural networks,” Nature, vol. 542, pp. 115–118, 2017.
[14] Y. LeCun, Y. Bengio, and G. Hinton, “Deep learning,” Nature, vol. 521, pp. 436–444, 2015.
[15] Streamlit Inc., “Streamlit documentation,” 2024. [Online]. Available: https://docs.streamlit.io
[16] A. Dosovitskiy et al., “An image is worth 16×16 words: Transformers for image recognition at scale,” arXiv preprint arXiv:2010.11929, 2020.
[17] B. Zhou et al., “Learning deep features for discriminative localization,” Proc. CVPR, pp. 2921–2929, 2016.
[18] N. V. Chawla et al., “SMOTE: Synthetic minority over-sampling technique,” J. Artif. Intell. Res., vol. 16, pp. 321–357, 2002.
[19] Kaggle, “APTOS 2019 blindness detection dataset,” 2019. [Online]. Available: https://www.kaggle.com/c/aptos2019-blindness-detection
[20] T. Li et al., “Applications of deep learning in fundus images: A review,” Med. Image Anal., vol. 69, p. 101971, 2021.
[21] M. D. Abramoff et al., “Pivotal trial of an autonomous AI-based diagnostic system for detection of diabetic retinopathy in primary care offices,” npj Digital Medicine, vol. 1, p. 39, 2018.