Hybrid Deep Learning and Morphological Processing for High-Accuracy Diabetic Retinopathy Detection and Grading

  • Authors

    • Dr. Ambuj Kumar Agarwal Kuala Lumpur University of Science & Technology (KLUST), (formerly known as Infrastructure University Kuala Lumpur (IUKL)), Unipark Suria, Jalan Ikram-Uniten, Kajang, Selangor, Malaysia and Department of Computer Science & Engineering, Sharda School of Engineering & Technology, Sharda University, Greater Noida, India
    • Prof. Dr. Abu Bakar Bin Abdul Hamid Kuala Lumpur University of Science & Technology (KLUST), (formerly known as Infrastructure University Kuala Lumpur (IUKL)), Unipark Suria, Jalan Ikram-Uniten, Kajang, Selangor, Malaysia
    • Danish Ather Amity University in Tashkent, Uzbekistan
    • Indrajit De Department of CSE (AIML) and CSBS, IEM-IIT Joint Center of Research, IEM-UEM Kolkata
    • Dr. Naveen Tewari Associate Professor, School of Computing, Graphic Era Hill University, Bhimtal Campus, Uttarakhand, India
    • Dr. Kunchanapaalli Rama Krishna Professor, Department of CSIT, K L Deemed to be University Vaddeswaram 522502, Andhra Pradesh, India
    https://doi.org/10.14419/pg20rf29

    Received date: August 12, 2025

    Accepted date: September 17, 2025

    Published date: September 22, 2025

  • APTOS 2019; Deep Residual Attention; Diabetic Retinopathy; Hybrid Segmentation; ResNet-DenseNet Fusion

  • Abstract

    Diabetic retinopathy (DR) is a microvascular complication of diabetes and the leading cause of preventable blindness among working‑age adults. The rapid increase in global diabetes prevalence has motivated the development of automated screening systems that can support overburdened healthcare providers in resource‑constrained settings. In this paper we propose a comprehensive computer‑aided diagnosis (CAD) framework for DR that integrates deep learning with traditional image processing to deliver accurate and interpretable screening. The pipeline first uses a Retinal_Denoiser network to suppress noise and enhance vasculature; the denoiser incorporates a residual attention mechanism to preserve fine lesions. A subsequent Hy_Retinal_Segmentation module combines a U‑Net backbone with morphological opening, closing and top‑hat transforms to isolate vessels and lesions. Finally, a hybrid classification network (Retinal_Net101) fuses high‑level features from ResNet‑101 and DenseNet‑201 branches to grade DR severity. The proposed framework is modular and can be trained end‑to‑end. We evaluate the system on publicly available EyePACS, APTOS 2019 and DDR datasets as well as a synthetic dataset built from the IDRiD segmentation ground truth. Comparisons against conventional preprocessing, segmentation and classification methods demonstrate significant improvements in peak signal‑to‑noise ratio (PSNR), Dice score and grading accuracy. The ablation study shows that each module contributes independently to performance; combining denoising, segmentation and hybrid classification yields a classification accuracy of 98.5 % compared with 94.0 % for a baseline pipeline. Our method outperforms state‑of‑the‑art approaches and offers interpretable intermediate outputs that could facilitate clinician trust. We conclude by discussing limitations and promising directions for future research.

     

  • References

    1. S. V. Chilukoti, A. S. Maida, and X. Hei, "Diabetic retinopathy detection using transfer learning from pre-trained convolutional neural network models," IEEE J. Biomed. Health Inf., vol. 20, pp. 1–10, 2022.
    2. M. Al-Smadi, M. Hammad, Q. B. Baker, and A. A. Sa’ad, "A transfer learning with deep neural network approach for diabetic retinopathy classifi-cation," Int. J. Electr. Comput. Eng., vol. 11, no. 4, p. 3492, 2021.
    3. Elsharkawy, M., Sharafeldeen, A., Soliman, A., Khalifa, F., Ghazal, M., El-Daydamony,E. & El-Baz, A. A novel computer-aided diagnostic system for early detection of diabetic retinopathy using 3D-OCT higher-order spatial appearance model. Diagnostics 12(2), 461 (2022).
    4. S. Akhtar and S. Aftab, "A framework for diabetic retinopathy detection using transfer learning and data fusion," Int. J. Inform. Technol. Comput. Sci. (IJITCS), vol. 16, no. 6, pp. 61–73, 2024.
    5. R. Sebti, S. Zroug, L. Kahloul, and S. Benharzallah, "A deep learning approach for the diabetic retinopathy detection," in Proc. Int. Conf. Smart City Applications, pp. 459–469, Springer, 2021.
    6. A. K. Gangwar and V. Ravi, "Diabetic retinopathy detection using transfer learning and deep learning," in Evolution in Computational Intelligence: Frontiers in Intelligent Computing: Theory and Applications (FICTA 2020), vol. 1, pp. 679–689, Springer, 2021.
    7. M. K. Jabbar, J. Yan, H. Xu, Z. Ur Rehman, and A. Jabbar, "Transfer learning-based model for diabetic retinopathy diagnosis using retinal images," Brain Sci., vol. 12, no. 5, p. 535, 2022.
    8. S. Ghosh and A. Chatterjee, "Transfer-ensemble learning based deep convolutional neural networks for diabetic retinopathy classification," arXiv preprint arXiv:2308.00525, 2023.
    9. A. Bilal, X. Liu, M. Shafiq, Z. Ahmed, and H. Long, "NIMEQ-SACNet: A novel self-attention precision medicine model for vision-threatening diabetic retinopathy using image data," Comput. Biol. Med., vol. 171, p. 108099, 2024.
    10. S. Akhtar, S. Aftab, M. Ahmad, and B. Ihnaini, "A classification framework for diabetic retinopathy detection using transfer learning," in Proc. 2024 2nd Int. Conf. Cyber Resilience (ICCR), pp. 1–5, IEEE, 2024.
    11. M. Ahmad, M. Alfayad, S. Aftab, M. A. Khan, A. Fatima, B. Shoaib, and N. S. Elmitwal, "Data and machine learning fusion architecture for car-diovascular disease prediction," Comput. Mater. Contin., vol. 69, no. 2, 2021.
    12. E. AbdelMaksoud, S. Barakat, and M. Elmogy, "A computer-aided diagnosis system for detecting various diabetic retinopathy grades based on a hybrid deep learning technique," Med. Biol. Eng. Comput., vol. 60, no. 7, pp. 2015–2038, 2022.
    13. A. Bora et al., "Predicting the risk of developing diabetic retinopathy using deep learning," Lancet Digit. Health, vol. 3, no. 1, pp. e10–e19, 2021.
    14. F. Li et al., "Deep learning-based automated detection for diabetic retinopathy and diabetic macular oedema in retinal fundus photographs," Eye, vol. 36, no. 7, pp. 1433–1441, 2022.
    15. S. Akhtar, S. Aftab, M. Ahmad, and A. Akhtar, "Diabetic retinopathy severity grading using transfer learning techniques," Int. J. Eng. Manuf. (IJEM), vol. 14, no. 6, pp. 41–53, 2024.
    16. L. K. Singh, H. Garg, and M. Khanna, "An artificial intelligence-based smart system for early glaucoma recognition using OCT images," in Re-search Anthology on Improving Medical Imaging Techniques for Analysis and Intervention, pp. 1424–1454, IGI Global, 2023.
    17. L. K. Singh, M. Khanna, and S. Thawkar, "A novel hybrid robust architecture for automatic screening of glaucoma using fundus photos, built on feature selection and machine learning-nature driven computing," Expert Syst., vol. 39, no. 10, p. e13069, 2022.
    18. S. Abbas et al., "Data and ensemble machine learning fusion based intelligent software defect prediction system," Comput. Mater. Contin., vol. 75, no. 3, 2023.
    19. M. Khanna, L. K. Singh, and H. Garg, "A novel approach for human diseases prediction using nature inspired computing & machine learning ap-proach," Multimed. Tools Appl., vol. 83, no. 6, pp. 17773–17809, 2024.
    20. L. K. Singh, H. Garg, and Pooja, "Automated glaucoma type identification using machine learning or deep learning techniques," in Adva. Mach. Intell. Interact. Med. Image Anal., pp. 241–263, 2020.
    21. L. K. Singh, M. Khanna, S. Thawkar, and R. Singh, "Nature-inspired computing and machine learning based classification approach for glaucoma in retinal fundus images," Multimed. Tools Appl., vol. 82, no. 27, pp. 42851–42899, 2023.
    22. S. H. Kassani et al., "Diabetic retinopathy classification using a modified xception architecture," in Proc. 2019 IEEE Int. Symp. Signal Process. Inf. Technol. (ISSPIT), pp. 1–6, IEEE, 2019.
    23. [23] A. Bilal, L. Zhu, A. Deng, H. Lu, and N. Wu, "AI-based automatic detection and classification of diabetic retinopathy using U-Net and deep learning," Symmetry, vol. 14, no. 7, p. 1427, 2022.
    24. P. M. Ebin and P. Ranjana, "A variant binary classification model for no-DR mild-DR detection using CLAHE images with transfer learning," in Proc. 2022 Int. Conf. Comput., Commun., Security Intell. Syst. (IC3SIS), pp. 1–5, IEEE, 2022.
    25. C. Bhardwaj, S. Jain, and M. Sood, "Transfer learning based robust automatic detection system for diabetic retinopathy grading," Neural Comput. Appl., vol. 33, no. 20, pp. 13999–14019, 2021.
    26. W. L. Alyoubi, M. F. Abulkhair, and W. M. Shalash, "Diabetic retinopathy fundus image classification and lesions localization system using deep learning," Sensors, vol. 21, no. 11, p. 3704, 2021.
    27. G. Çinarer, K. Kilic, and T. Parlar, "A deep transfer learning framework for the staging of diabetic retinopathy," J. Sci. Rep.-A., no. 051, pp. 106–119, 2022.
    28. B. Kotiyal and H. Pathak, "Diabetic retinopathy binary image classification using Pyspark," Int. J. Math. Eng. Manag. Sci., vol. 7, no. 5, p. 624, 2022.
    29. A. Skouta, A. Elmoufidi, S. Jai-Andaloussi, and O. Ochetto, "Automated binary classification of diabetic retinopathy by convolutional neural net-works," in Advances on Smart and Soft Computing: Proc. ICACIn 2020, pp. 177–187, Springer, 2021.
    30. P. Uppamma and S. Bhattacharya, "Diabetic retinopathy detection: A blockchain and African vulture optimization algorithm-based deep learning framework," Electronics, vol. 12, no. 3, p. 742, 2023.
    31. N. Sikder et al., "Severity classification of diabetic retinopathy using an ensemble learning algorithm through analyzing retinal images," Symmetry, vol. 13, no. 4, p. 670, 2021.
    32. M. Ghazal, S. S. Ali, A. H. Mahmoud, A. M. Shalaby, and A. El-Baz, "Accurate detection of non-proliferative diabetic retinopathy in optical co-herence tomography images using convolutional neural networks," IEEE Access, vol. 8, pp. 34387–34397, 2020.
    33. H. Liu et al., "Hybrid model structure for diabetic retinopathy classification," J. Healthc. Eng., vol. 2020, 2020.
    34. A. Jabbar et al., "A lesion-based diabetic retinopathy detection through hybrid deep learning model," IEEE Access, 2024.
    35. A. M. Moustari, Y. Brik, B. Attallah, and R. Bouaouina, "Two-stage deep learning classification for diabetic retinopathy using gradient weighted class activation mapping," Automatika, vol. 65, no. 3, pp. 1284–1299, 2024.
    36. H. Shakibania, S. Raoufi, B. Pourafkham, H. Khotanlou, and M. Mansoorizadeh, "Dual branch deep learning network for detection and stage grad-ing of diabetic retinopathy," Biomed. Signal Process. Control., vol. 93, p. 106168, 2024.
    37. L. Dai et al., "A deep learning system for predicting time to progression of diabetic retinopathy," Nat. Med., vol. 30, no. 2, pp. 584–594, 2024.
    38. Y. Jin et al., "Deep learning-driven automated quality assessment of ultra-widefield optical coherence tomography angiography images for diabetic retinopathy," Vis. Comput., pp. 1–11, 2024.

  • Downloads

  • How to Cite

    Agarwal , D. A. K. ., Abdul Hamid , P. D. A. B. B. ., Ather, D. . ., De , I. ., Tewari , D. N. ., & Krishna , D. K. R. . (2025). Hybrid Deep Learning and Morphological Processing for High-Accuracy Diabetic Retinopathy Detection and Grading. International Journal of Basic and Applied Sciences, 14(5), 805-812. https://doi.org/10.14419/pg20rf29