ALGORITHMS FOR DETECTION OF EYE DISEASES USING ARTIFICIAL INTELLIGENCE AND FRACTAL DIMENSION CALCULATION METHODS
Received: 2026-05-27
Published: 2026-06-06
Abstract
The rapid global increase in the prevalence of diabetic retinopathy, glaucoma, and other retinal vascular diseases requires development of automated, scalable, and highly accurate early-detection systems. This paper presents a novel integrated algorithm that combines deep-learning-based retinal image analysis with fractal dimension computation to enhance ophthalmologic diagnostics. Our methodology employs a U-Net encoder–decoder architecture for precise retinal blood vessel segmentation, followed by the box-counting algorithm to compute the fractal dimension (D) of the vessel network. An EfficientNet-B4 convolutional neural network classifier, trained under transfer-learning principles, then integrates fractal-derived features with deep visual features to classify retinal pathologies including diabetic retinopathy, glaucoma, and hypertensive retinopathy. The system was evaluated on widely validated public datasets: EyePACS (n=88,702), Messidor-2, DRIVE, STARE, RIM-ONE v3, and HRF. The proposed approach achieved accuracy 94.6%, sensitivity 93.8%, specificity 95.2%, F1-score 94.5%, and AUC 0.968 for diabetic retinopathy detection, surpassing all single-modality baselines. Scientific novelty lies in the systematic integration of fractal geometry as a mathematically grounded, interpretable biomarker alongside deep neural network classification, simultaneously improving accuracy and clinical explainability of vessel-morphology changes.
Keywords
List of references
-
Gulshan V., Peng L., Coram M., et al. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs. JAMA. 2016;316(22):2402–2410.
-
Ting D.S.W., Cheung C.Y.L., Lim G., 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. 2017;318(22):2211–2223.
-
Ronneberger O., Fischer P., Brox T. U-Net: Convolutional networks for biomedical image segmentation. MICCAI 2015. Lecture Notes in Computer Science, vol 9351. Springer; 2015:234–241.
-
He K., Zhang X., Ren S., Sun J. Deep residual learning for image recognition. CVPR 2016:770–778.
-
Abràmoff M.D., Lavin P.T., Birch M., Shah N., Folk J.C. Pivotal trial of an autonomous AI-based diagnostic system for detection of diabetic retinopathy in primary care offices. npj Digital Medicine. 2018;1:39.
-
Tan M., Le Q.V. EfficientNet: Rethinking model scaling for convolutional neural networks. ICML 2019:6105–6114.
-
Cheung N., Donaghue K.C., Liew G., et al. Quantitative assessment of early diabetic retinopathy using fractal analysis. Diabetes Care. 2009;32(1):106–110.
-
Mandelbrot B.B. The Fractal Geometry of Nature. W. H. Freeman and Company; 1982.
-
Orlando J.I., Fu H., Breda J.B., et al. REFUGE challenge: A unified framework for evaluating automated methods for glaucoma assessment from fundus photographs. Medical Image Analysis. 2020;59:101570.
-
Goodfellow I., Bengio Y., Courville A. Deep Learning. MIT Press; 2016.
-
Masters B.R. Fractal analysis of the vascular tree in the human retina. Annual Review of Biomedical Engineering. 2004;6:427–452.
-
Staal J., Abràmoff M.D., Niemeijer M., Viergever M.A., van Ginneken B. Ridge-based vessel segmentation in color images of the retina. IEEE Trans. Med. Imaging. 2004;23(4):501–509.
-
Zhu C., Zou B., Zhao R., et al. Retinal vessel segmentation in fundus images: A comprehensive review of deep learning approaches. IEEE Access. 2021;9:113474–113498.
-
Li T., Bo W., Hu C., et al. Applications of deep learning in fundus images: A review. Medical Image Analysis. 2021;69:101971.
-
Simonyan K., Zisserman A. Very deep convolutional networks for large-scale image recognition. ICLR 2015.
-
Chen X., Xu Y., Yan S., Wong D.W.K., Wong T.Y., Liu J. Automatic feature learning for glaucoma detection based on deep learning. MICCAI 2015;8:669–677.
-
Jiang Y., Deng Z., Yang F., et al. Joint retinal vessel segmentation and optic disc detection with dual branch network. Knowledge-Based Systems. 2022;257:109810.
-
Tham Y.C., Li X., Wong T.Y., Quigley H.A., Aung T., Cheng C.Y. Global prevalence of glaucoma and projections of glaucoma burden through 2040. Ophthalmology. 2014;121(11):2081–2090.
-
Teo Z.L., Tham Y.C., Yu M., et al. Global prevalence of diabetic retinopathy and projection of burden through 2045. Ophthalmology. 2021;128(11):1580–1591.
-
World Health Organization. World report on vision. WHO; 2019.
-
Pizer S.M., Amburn E.P., Austin J.D., et al. Adaptive histogram equalization and its variations. Computer Vision, Graphics and Image Processing. 1987;39(3):355–368.
-
Zhou Z., Siddiquee M.M.R., Tajbakhsh N., Liang J. UNet++: A nested U-Net architecture for medical image segmentation. Deep Learning in Medical Image Analysis. Springer; 2018:3–11.
-
Lim G., Yanagihara R.T., Lee M.L., Goh J.H.L., Lee A.Y. Artificial intelligence solutions for retinal image analysis in ophthalmologic diseases. Current Ophthalmology Reports. 2020;8:68–82.
-
Sarhan A., Hudson N., Peto T., et al. Retinal vessel tortuosity and its relation to hypertension: a review. Ophthalmic Epidemiology. 2023;30(1):1–12.
-
Huang G., Liu Z., Van der Maaten L., Weinberger K.Q. Densely connected convolutional networks. CVPR 2017:4700–4708.
About the Authors
License
This work is licensed under a Creative Commons Attribution 4.0 International License.