Deep Learning-Based Classification of Diabetic Retinopathy Using ResNet Variants and Edge-Based Segmentation
Keywords:
Image Classification, Feature Extraction, ResNet-152, Diabetic RetinopathyAbstract
Diabetic Retinopathy is a leading cause of vision impairment and blindness among diabetic patients worldwide. Early detection and accurate classification of DR severity levels are essential for timely intervention and treatment. This study presents a deep learning-based framework for automated detection and classification of DR using high-resolution fundus images. The proposed approach leverages MobileNetV2 for lightweight feature extraction and employs three ResNet variants such as ResNet-18, ResNet-50, and ResNet-152 for classification of DR into five stages: No DR, Mild, Moderate, Severe, and Proliferative.The fundus images from the publicly available Kaggle DR dataset were pre-processed and segmented using Canny Edge Detection, enabling the model to focus on critical retinal structures such as blood vessels and lesions. Experimental results demonstrate that ResNet-152 achieves the best performance, with a validation accuracy of 90.15%, F1-score of 90.01%, and ROC-AUC of 0.94, outperforming baseline models including InceptionV3 (82.00%) and GoogleNet (87.23%).This work demonstrates the effectiveness of combining edge-based segmentation with transfer learning for DR detection, offering a robust and scalable solution for clinical decision support systems, especially in resource-constrained healthcare settings.
Downloads
References
Trucillo, P., & Di Maio, E. (2021). Classification and Production of Polymeric Foams among the Systems for Wound Treatment. Polymers, 13(10), 1608. https://doi.org/10.3390/polym13101608
Kropp M, Golubnitschaja O, Mazurakova A, Koklesova L, Sargheini N, Vo TKS, de Clerck E, Polivka J Jr, Potuznik P, Polivka J, Stetkarova I, Kubatka P, Thumann G. Diabetic retinopathy as the leading cause of blindness and early predictor of cascading complications-risks and mitigation. EPMA J. 2023 Feb 13;14(1):21-42. doi: 10.1007/s13167-023-00314-8. PMID: 36866156; PMCID: PMC9971534.
Tinajero MG, Malik VS. An update on the epidemiology of type 2 diabetes: a global perspective. Endocrinol Metab Clin North Am. 2021;50:337–355. doi: 10.1016/j.ecl.2021.05.013.
Lin X, Xu Y, Pan X, Xu J, Ding Y, Sun X, et al. Global, regional, and national burden and trend of diabetes in 195 countries and territories: an analysis from 1990. Sci Rep. 2020;10:14790. doi: 10.1038/s41598-020-71908-9.
Hussain N, Edraki M, Tahhan R, Sanalkumar N, Kenz S, Akasha NK, Mtemererwa B, Mohammed N. Telemedicine for diabetic retinopathy screening using an ultra-widefield fundus camera. Clin Ophthalmol. 2017 Aug 14;11:1477-1482. doi: 10.2147/OPTH.S135287. PMID: 28860696; PMCID: PMC5565372.
Shafi S, Parwani AV. Artificial intelligence in diagnostic pathology. DiagnPathol. 2023 Oct 3;18(1):109. doi: 10.1186/s13000-023-01375-z. PMID: 37784122; PMCID: PMC10546747.
Kaur K, Gurnani B, Nayak S, Deori N, Kaur S, Jethani J, Singh D, Agarkar S, Hussaindeen JR, Sukhija J, Mishra D. Digital Eye Strain- A Comprehensive Review. Ophthalmol Ther. 2022 Oct;11(5):1655-1680. doi: 10.1007/s40123-022-00540-9. Epub 2022 Jul 9. PMID: 35809192; PMCID: PMC9434525.
S. Kulkarni, N. Seneviratne, M. S. Baig, and A. H. A. Khan, “Artificial intelligence in medicine: where are we now?” Academic Radiology, vol. 27, no. 1, pp. 62–70, 2020.
Y. Mori, H. Neumann, M. Misawa, S. E. Kudo, and M. Bretthauer, “Artificial intelligence in colonoscopy: now on the market. What’s next?” Journal of Gastroenterology and Hepatology, vol. 36, 2021.
H. Endo, S. Kase, M. Saito et al., “Choroidal thickness in diabetic patients without diabetic retinopathy: a metametaanalysis,” American Journal of Ophthalmology, vol. 218, pp. 68–77, 2020.
S. Roy and D. Kim, “Retinal capillary basement membrane thickening: role in the pathogenesis of diabetic retinopathy,” Progress in Retinal and Eye Research, vol. 82, Article ID 100903, 2020.
L. Camara Neto, G. L. B. Ramalho, J. F. S. Rocha Neto, R. M. S. Veras, and F. N. S. Medeiros, “An unsupervised coarse-to-fine algorithm for blood vessel segmentation in fundus images,” Expert Systems with Applications, vol. 78, pp. 182–192, 2017.
R. Sundaram, R. KS, P. Jayaraman, and V. B, “Extraction of blood vessels in fundus images of retina through hybrid segmentation approach,” Mathematics, vol. 7, no. 2, p. 169, 2019.
Y. Zhao, L. Rada, K. Chen, S. P. Harding, and Y. Zheng, “Automated vessel segmentation using infinite perimeter active contour model with hybrid region information with application to retinal images,” IEEE Transactions on Medical Imaging, vol. 34, no. 9, pp. 1797–1807, 2015.
L. C. Rodrigues and M. Marengoni, “Segmentation of optic disc and blood vessels in retinal images using wavelets, mathematical morphology and Hessian-based multi-scale filtering,” Biomedical Signal Processing and Control, vol. 36, pp. 39–49, 2017.
Ç. Sazak, C. J. Nelson, and B. Obara, “+e multiscale bowlerhat transform for blood vessel enhancement in retinal images,” Pattern Recognition, vol. 88, pp. 739–750, 2019.
B. Savelli, A. Bria, A. Galdran et al., “Illumination correction by dehazing for retinal vessel segmentation,” in Proceedings of the 30th International Symposium on Computer-Based Medical Systems, pp. 219–224, +essaloniki, Greece, June 2017.
F. Girard, C. Kavalec, and F. Cheriet, “Joint segmentation and classification of retinal-arteries/veins from fundus images,” Artificial Intelligence in Medicine, vol. 94, pp. 96–109, 2019.
K. Hu, Z. Zhang, X. Niu et al., “Retinal vessel segmentation of color fundus images using multiscale convolutional neural network with an improved cross-entropy loss function,” Neurocomputing, vol. 309, pp. 179–191, 2018.
H. Fu, Y. Xu, S. Lin, D. W. Kee Wong, and J. Liu, “DeepVessel: retinal vessel segmentation via deep learning and conditional random field,” in Proceedings of the International Conference on Medical Image Computing and Computer-Assisted Intervention, pp. 132–139, Athens, Greece, October 2016.
T. A. Soomro, A. J. Afifi, J. Gao et al., “Boosting sensitivity of a retinal vessel segmentation algorithm with convolutional neural network,” in Proceedings of the International Conference on Digital Image Computing: Techniques and Applications, pp. 1–8, Sydney, Australia, November 2017.
Valizadeh, Amin, Jafarzadeh Ghoushchi, Saeid, Ranjbarzadeh, Ramin, Pourasad, Yaghoub, Presentation of a Segmentation Method for a Diabetic Retinopathy Patient’s Fundus Region Detection Using a Convolutional Neural Network, Computational Intelligence and Neuroscience, 2021, 7714351, 14 pages, 2021. https://doi.org/10.1155/2021/7714351
Chen, PN., Lee, CC., Liang, CM. et al. General deep learning model for detecting diabetic retinopathy. BMC Bioinformatics 22 (Suppl 5), 84 (2021). https://doi.org/10.1186/s12859-021-04005-x
Shi, B.; Zhang, X.;Wang, Z.; Song, J.; Han, J.; Zhang, Z.; Toe, T.T. GoogLeNet-based Diabetic-Retinopathy-Detection. In Proceedings of the 14th IEEE International Conference on Advanced Computational Intelligence (ICACI),Wuhan, China, 15–17 July 2022; pp. 246–249.
R., Y.; Sarobin, M.V.R.; Panjanathan, R.; Jasmine, S.G.; Anbarasi, L.J. Diabetic Retinopathy Classification Using CNN and Hybrid Deep Convolutional Neural Networks. Symmetry 2022, 14, 1932
Kurup, G.; Jothi, J.A.A.; Kanadath, A. Diabetic Retinopathy Detection and Classification using Pretrained Inception-v3. In Proceedings of the IEEE International Conference on Smart Generation Computing, Communication and Networking (SMART GENCON), Pune, India, 29–30 October 2021; pp. 1–6.
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
All papers should be submitted electronically. All submitted manuscripts must be original work that is not under submission at another journal or under consideration for publication in another form, such as a monograph or chapter of a book. Authors of submitted papers are obligated not to submit their paper for publication elsewhere until an editorial decision is rendered on their submission. Further, authors of accepted papers are prohibited from publishing the results in other publications that appear before the paper is published in the Journal unless they receive approval for doing so from the Editor-In-Chief.
IJISAE open access articles are licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. This license lets the audience to give appropriate credit, provide a link to the license, and indicate if changes were made and if they remix, transform, or build upon the material, they must distribute contributions under the same license as the original.
