Segmentation-Driven Glaucoma Classification from Retinal Fundus Images Using Optic Disk Localization, Linear and Nonlinear Multiple Instance Learning Frameworks

Mahbub Hasan

doi:doi:10.11648/j.mcs.20251005.11

Research Article |

| Peer-Reviewed

Segmentation-Driven Glaucoma Classification from Retinal Fundus Images Using Optic Disk Localization, Linear and Nonlinear Multiple Instance Learning Frameworks

Mahbub Hasan^*

Published in Mathematics and Computer Science (Volume 10, Issue 5)

Received: 24 August 2025 Accepted: 4 September 2025 Published: 25 September 2025

Views: Downloads:

Download PDF

Share This Article

Twitter
Linked In
Facebook

Abstract

In recent years, glaucoma has been one of the leading causes of vision loss, and its early detection is essential for preventing irreparable damage. This study proposes an efficient classification approach using segmentation on retinal fundus images from the MESSIDOR dataset, focusing on optic disk segmentation to enhance disease-specific feature extraction while minimizing irrelevant background information. The pipeline includes preprocessing steps such as image resizing and green channel extraction, followed by segmentation of both the optic cup and optic disk using traditional image processing techniques, including pixel clustering with superpixel methods, global thresholding, and the Circular Hough Transform (CHT). Performance is evaluated using metrics such as testing correctness (TC), sensitivity, specificity, F1-score, and computational time under 5-fold, 10-fold, and Leave-One-Out (LOO) cross-validation (CV). Results show that preprocessing significantly improves accuracy across all models, with mi-SPSVM under 10-fold CV achieving the best results, obtaining 90.00% sensitivity and a 67.09% F1-score. Meanwhile, mi-KSPSVM achieved the lowest computational time, processing a single image in just 0.09s. These findings demonstrate that incorporating segmentation into the classification task can enhance both diagnostic accuracy and computational efficiency, making the approach suitable for real-time applications and resource-limited environments. Overall, this study presents a promising and scalable solution for automated OD screening, with future work aimed at integrating deep learning-based segmentation to improve precision further.

Published in	Mathematics and Computer Science (Volume 10, Issue 5)
DOI	10.11648/j.mcs.20251005.11
Page(s)	83-91
Creative Commons	This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.
Copyright	Copyright © The Author(s), 2025. Published by Science Publishing Group

Keywords

Optic Disk, Optic Cup, Segmentation, Classification, Multiple Instance Learning, Machine Learning, Glaucoma

References

[1]	Almazroa, A., Burman, R., Raahemifar, K., & Lakshminarayanan, V. (2015). Optic Disc and Optic Cup Segmentation Methodologies for Glaucoma Image Detection: A Survey. Journal of ophthalmology, 2015, 180972. https://doi.org/10.1155/2015/180972
[2]	Abdullah, M., Fraz, M. M., & Barman, S. A. (2016). Localization and segmentation of optic disc in retinal images using circular Hough transform and grow-cut algorithm. PeerJ, 4, e2003. https://doi.org/10.7717/peerj.2003
[3]	Rehman, Z. U., Naqvi, S. S., Khan, T. M., Arsalan, M., Khan, M. A., & Khalil, M. A. (2019). Multi-parametric optic disc segmentation using superpixel based feature classification. Expert Systems with Applications, 120, 461-473. https://doi.org/10.1016/j.eswa.2018.12.008
[4]	Ünver, H. M., Kökver, Y., Duman, E., & Erdem, O. A. (2019). Statistical Edge Detection and Circular Hough Transform for Optic Disk Localization. Applied Sciences, 9(2), 350. https://doi.org/10.3390/app9020350
[5]	Ramani, R. G., & Shanthamalar, J. J. (2020). Improved image processing techniques for optic disc segmentation in retinal fundus images. Biomedical Signal Processing and Control, 58, 101832. https://doi.org/10.1016/j.bspc.2019.101832
[6]	Dietter, J., Haq, W., Ivanov, I. V., Norrenberg, L. A., Völker, M., Dynowski, M., ... & Ueffing, M. (2019). Optic disc detection in the presence of strong technical artifacts. Biomedical Signal Processing and Control, 53, 101535. https://doi.org/10.1016/j.bspc.2019.04.012
[7]	Zhao, X., Wang, H., & Peng, Z. (2019, December). Fine segmentation of fundus optic disc based on SLIC superpixel. In AOPC 2019: AI in Optics and Photonics (Vol. 11342, pp. 155-163). SPIE. https://doi.org/10.1117/12.2547992
[8]	Fu, H., Li, F., Xu, Y., Liao, J., Xiong, J., Shen, J., ... & iChallenge-GON study group. (2020). A retrospective comparison of deep learning to manual annotations for optic disc and optic cup segmentation in fundus photographs. Translational vision science & technology, 9(2), 33-33. https://doi.org/10.1167/tvst.9.2.33
[9]	Liu, B., Pan, D. & Song, H. Joint optic disc and cup segmentation based on densely connected depthwise separable convolution deep network. BMC Med Imaging 21, 14 (2021). https://doi.org/10.1186/s12880-020-00528-6
[10]	Mahmood, M. T., & Lee, I. H. (2022). Optic disc localization in fundus images through accumulated directional and radial blur analysis. Computerized Medical Imaging and Graphics, 98, 102058. https://doi.org/10.1016/j.compmedimag.2022.102058
[11]	Toptas¸, B., Toptas¸, M. & Hanbay, D. Detection of Optic Disc Localization from Retinal Fundus Image Using Optimized Color Space. J Digit Imaging 35, 302-319 (2022). https://doi.org/10.1007/s10278-021-00566-8
[12]	Ali A, Wan Zaki WMD, Hussain A et al. Vessel masking and Hough transform for optic disc localisation from retinal images [version 1; peer review: 1 approved with reservations, 1 not approved]. F1000Research 2022, 11: 181 https://doi.org/10.12688/f1000research.73390.1
[13]	Ávila, F. J., Bueno, J. M., & Remón, L. (2022). Superpixel-Based Optic Nerve Head Segmentation Method of Fundus Images for Glaucoma Assessment. Diagnostics, 12(12), 3210. https://doi.org/10.3390/diagnostics12123210
[14]	Septiarini, A., Hamdani, H., Setyaningsih, E., Junirianto, E., & Utaminingrum, F. (2023). Automatic method for optic disc segmentation using deep learning on retinal fundus images. Healthcare Informatics Research, 29(2), 145-151. https://doi.org/10.4258/hir.2023.29.2.145
[15]	Jana, S., & Sarkar, P. (2024). Optic Disc and Vessel Segmentation from Fundus Image using Miniunet Architecture. medRxiv, 2024-12. https://doi.org/10.1101/2024.12.28.24319728
[16]	Zhou, L., Zhao, Y., Yang, J., Yu, Q., & Xu, X. (2018). Deep multiple instance learning for automatic detection of diabetic retinopathy in retinal images. IET Image Processing, 12(4), 563-571. https://doi.org/10.1049/iet-ipr.2017.0636
[17]	Papadopoulos, A., Topouzis, F., & Delopoulos, A. (2021). An interpretable multiple-instance approach for the detection of referable diabetic retinopathy in fundus images. Scientific Reports, 11(1), 14326. https://doi.org/10.1038/s41598-021-93632-8
[18]	Bi, Q. et al. (2021). Local-Global Dual Perception Based Deep Multiple Instance Learning for Retinal Disease Classification. In: de Bruijne, M., et al. Medical Image Computing and Computer Assisted Intervention - MICCAI 2021. MICCAI 2021. Lecture Notes in Computer Science (), vol 12908. Springer, Cham. https://doi.org/10.1007/978-3-030-87237-3_6
[19]	Cao, P., Hou, Q., Song, R., Wang, H., & Zaiane, O. (2022). Collaborative learning of weakly-supervised domain adaptation for diabetic retinopathy grading on retinal images. Computers in Biology and Medicine, 144, 105341. https://doi.org/10.1016/j.compbiomed.2022.105341
[20]	Mueller, S., Wintergerst, M. W., Falahat, P., Holz, F. G., Schaefer, C., Schahab, N., ... & Schultz, T. (2022). Multiple instance learning detects peripheral arterial disease from high-resolution color fundus photography. Scientific Reports, 12(1), 1389. https://doi.org/10.1038/s41598-022-05169-z
[21]	Elmoufidi, A., Skouta, A., Jai-Andaloussi, S., & Ouchetto, O. (2022). Deep multiple instance learning for automatic glaucoma prevention and auto-annotation using color fundus photography. Progress in Artificial Intelligence, 11(4), 397-409. https://doi.org/10.1007/s13748-022-00292-4
[22]	Zhu, W., Qiu, P., Lepore, N., Dumitrascu, O. M., & Wang, Y. (2023, March). Self-supervised equivariant regularization reconciles multiple-instance learning: Joint referable diabetic retinopathy classification and lesion segmentation. In 18th International Symposium on Medical Information Processing and Analysis (Vol. 12567, pp. 100-107). SPIE. https://doi.org/10.1117/12.2669772
[23]	Bi, Q., Sun, X., Yu, S., Ma, K., Bian, C., Ning, M., ... & Zheng, Y. (2023). MIL-ViT: A multiple instance vision transformer for fundus image classification. Journal of Visual Communication and Image Representation, 97, 103956. https://doi.org/10.1016/j.jvcir.2023.103956
[24]	Matten, P., Scherer, J., Schlegl, T., Nienhaus, J., Stino, H., Niederleithner, M., ... & Schmoll, T. (2023). Multiple instance learning based classification of diabetic retinopathy in weakly-labeled widefield OCTA en face images. Scientific Reports, 13(1), 8713. https://doi.org/10.1038/s41598-023-35713-4
[25]	Avolio, M., Fuduli, A., Vocaturo, E., & Zumpano, E. (2023). Multiple Instance Learning for Diabetic Retinopathy Detection. In SEBD (pp. 22-31). https://www.semanticscholar.org/paper/Multiple-Instance-Learning-for-Diabetic-Retinopathy-Avolio-Fuduli/239d08976fe53b4b795a3f81b5fc2163074d722b
[26]	Li, R., Gu, Y., Wang, X., & Pan, J. (2023). A Cross-Domain Weakly Supervised Diabetic Retinopathy Lesion Identification Method Based on Multiple Instance Learning and Domain Adaptation. Bioengineering, 10(9), 1100. https://doi.org/10.3390/bioengineering10091100
[27]	Fan, J., Yang, T., Wang, H., Zhang, H., Zhang, W., Ji, M., & Miao, J. (2025). A self-supervised equivariant refinement classification network for diabetic retinopathy classification. Journal of Imaging Informatics in Medicine, 38(3), 1796-1811. https://doi.org/10.1007/s10278-024-01270-z
[28]	Wei, X., Liu, Y., Zhang, F., Geng, L., Shan, C., Cao, X., & Xiao, Z. (2025). MSTNet: Multi-scale spatial-aware transformer with multi-instance learning for diabetic retinopathy classification. Medical Image Analysis, 102, 103511. https://doi.org/10.1016/j.media.2025.103511
[29]	Lai, Q., Vong, C. M., Yan, T., Wong, P. K., & Liang, X. (2025). Hybrid multiple instance learning network for weakly supervised medical image classification and localization. Expert Systems with Applications, 260, 125362. https://doi.org/10.1016/j.eswa.2024.125362
[30]	Astorino, A., Fuduli, A. and Gaudioso, M., 2019. A Lagrangian relaxation approach for binary multiple instance classification. IEEE transactions on neural networks and learning systems, 30(9), pp.2662-2671. https://doi.org/10.1109/TNNLS.2018.2885852
[31]	Avolio, M. and Fuduli, A., 2020. A semiproximal support vector machine approach for binary multiple instance learning. IEEE transactions on neural networks and learning systems, 32(8), pp.3566-3577. https://doi.org/10.1109/TNNLS.2020.3015442
[32]	Avolio, M., Fuduli, A., Vocaturo, E. and Zumpano, E., 2023, May. On detection of diabetic retinopathy via multiple instance learning. In Proceedings of the 27th International Database Engineered Applications Symposium (pp. 170-176). https://doi.org/10.1145/3589462.3589490
[33]	Avolio, M. and Fuduli, A., 2024. The semiproximal SVM approach for multiple instance learning: a kernel-based computational study. Optimization Letters, 18(2), pp.635-649. https://doi.org/10.1007/s11590-023-02022-8
[34]	Decencière, Etienne and Zhang, Xiwei and Cazuguel, Guy and Lay, Bruno and Cochener, Béatrice and Trone, Caroline and Gain, Philippe and Ordó˜ nez-Varela, John-Richard and Massin, Pascale and Erginay, Ali, 2014. Feedback on a publicly distributed image database: the Messidor database. Image Analysis & Stereology, pp.231-234. https://doi.org/10.5566/ias.1155

Cite This Article

Plain Text BibTeX RIS

APA Style

Hasan, M. (2025). Segmentation-Driven Glaucoma Classification from Retinal Fundus Images Using Optic Disk Localization, Linear and Nonlinear Multiple Instance Learning Frameworks. Mathematics and Computer Science, 10(5), 83-91. https://doi.org/10.11648/j.mcs.20251005.11

Copy | Download

ACS Style

Hasan, M. Segmentation-Driven Glaucoma Classification from Retinal Fundus Images Using Optic Disk Localization, Linear and Nonlinear Multiple Instance Learning Frameworks. Math. Comput. Sci. 2025, 10(5), 83-91. doi: 10.11648/j.mcs.20251005.11

Copy | Download

AMA Style

Hasan M. Segmentation-Driven Glaucoma Classification from Retinal Fundus Images Using Optic Disk Localization, Linear and Nonlinear Multiple Instance Learning Frameworks. Math Comput Sci. 2025;10(5):83-91. doi: 10.11648/j.mcs.20251005.11

Copy | Download

@article{10.11648/j.mcs.20251005.11,
  author = {Mahbub Hasan},
  title = {Segmentation-Driven Glaucoma Classification from Retinal Fundus Images Using Optic Disk Localization, Linear and Nonlinear Multiple Instance Learning Frameworks
},
  journal = {Mathematics and Computer Science},
  volume = {10},
  number = {5},
  pages = {83-91},
  doi = {10.11648/j.mcs.20251005.11},
  url = {https://doi.org/10.11648/j.mcs.20251005.11},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.mcs.20251005.11},
  abstract = {In recent years, glaucoma has been one of the leading causes of vision loss, and its early detection is essential for preventing irreparable damage. This study proposes an efficient classification approach using segmentation on retinal fundus images from the MESSIDOR dataset, focusing on optic disk segmentation to enhance disease-specific feature extraction while minimizing irrelevant background information. The pipeline includes preprocessing steps such as image resizing and green channel extraction, followed by segmentation of both the optic cup and optic disk using traditional image processing techniques, including pixel clustering with superpixel methods, global thresholding, and the Circular Hough Transform (CHT). Performance is evaluated using metrics such as testing correctness (TC), sensitivity, specificity, F1-score, and computational time under 5-fold, 10-fold, and Leave-One-Out (LOO) cross-validation (CV). Results show that preprocessing significantly improves accuracy across all models, with mi-SPSVM under 10-fold CV achieving the best results, obtaining 90.00% sensitivity and a 67.09% F1-score. Meanwhile, mi-KSPSVM achieved the lowest computational time, processing a single image in just 0.09s. These findings demonstrate that incorporating segmentation into the classification task can enhance both diagnostic accuracy and computational efficiency, making the approach suitable for real-time applications and resource-limited environments. Overall, this study presents a promising and scalable solution for automated OD screening, with future work aimed at integrating deep learning-based segmentation to improve precision further.
},
 year = {2025}
}

Copy | Download

TY  - JOUR
T1  - Segmentation-Driven Glaucoma Classification from Retinal Fundus Images Using Optic Disk Localization, Linear and Nonlinear Multiple Instance Learning Frameworks

AU  - Mahbub Hasan
Y1  - 2025/09/25
PY  - 2025
N1  - https://doi.org/10.11648/j.mcs.20251005.11
DO  - 10.11648/j.mcs.20251005.11
T2  - Mathematics and Computer Science
JF  - Mathematics and Computer Science
JO  - Mathematics and Computer Science
SP  - 83
EP  - 91
PB  - Science Publishing Group
SN  - 2575-6028
UR  - https://doi.org/10.11648/j.mcs.20251005.11
AB  - In recent years, glaucoma has been one of the leading causes of vision loss, and its early detection is essential for preventing irreparable damage. This study proposes an efficient classification approach using segmentation on retinal fundus images from the MESSIDOR dataset, focusing on optic disk segmentation to enhance disease-specific feature extraction while minimizing irrelevant background information. The pipeline includes preprocessing steps such as image resizing and green channel extraction, followed by segmentation of both the optic cup and optic disk using traditional image processing techniques, including pixel clustering with superpixel methods, global thresholding, and the Circular Hough Transform (CHT). Performance is evaluated using metrics such as testing correctness (TC), sensitivity, specificity, F1-score, and computational time under 5-fold, 10-fold, and Leave-One-Out (LOO) cross-validation (CV). Results show that preprocessing significantly improves accuracy across all models, with mi-SPSVM under 10-fold CV achieving the best results, obtaining 90.00% sensitivity and a 67.09% F1-score. Meanwhile, mi-KSPSVM achieved the lowest computational time, processing a single image in just 0.09s. These findings demonstrate that incorporating segmentation into the classification task can enhance both diagnostic accuracy and computational efficiency, making the approach suitable for real-time applications and resource-limited environments. Overall, this study presents a promising and scalable solution for automated OD screening, with future work aimed at integrating deep learning-based segmentation to improve precision further.

VL  - 10
IS  - 5
ER  -

Copy | Download

Author Information

Mahbub Hasan

Department of Computer Science and Engineering, Southeast University, Dhaka, Bangladesh

Contact Email

http://orcid.org/0009-0000-7906-1823

Download PDF

Submit an Article

Sections

Plain Text BibTeX RIS

APA Style

Hasan, M. (2025). Segmentation-Driven Glaucoma Classification from Retinal Fundus Images Using Optic Disk Localization, Linear and Nonlinear Multiple Instance Learning Frameworks. Mathematics and Computer Science, 10(5), 83-91. https://doi.org/10.11648/j.mcs.20251005.11

Copy | Download

ACS Style

Hasan, M. Segmentation-Driven Glaucoma Classification from Retinal Fundus Images Using Optic Disk Localization, Linear and Nonlinear Multiple Instance Learning Frameworks. Math. Comput. Sci. 2025, 10(5), 83-91. doi: 10.11648/j.mcs.20251005.11

Copy | Download

AMA Style

Hasan M. Segmentation-Driven Glaucoma Classification from Retinal Fundus Images Using Optic Disk Localization, Linear and Nonlinear Multiple Instance Learning Frameworks. Math Comput Sci. 2025;10(5):83-91. doi: 10.11648/j.mcs.20251005.11

Copy | Download

@article{10.11648/j.mcs.20251005.11,
  author = {Mahbub Hasan},
  title = {Segmentation-Driven Glaucoma Classification from Retinal Fundus Images Using Optic Disk Localization, Linear and Nonlinear Multiple Instance Learning Frameworks
},
  journal = {Mathematics and Computer Science},
  volume = {10},
  number = {5},
  pages = {83-91},
  doi = {10.11648/j.mcs.20251005.11},
  url = {https://doi.org/10.11648/j.mcs.20251005.11},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.mcs.20251005.11},
  abstract = {In recent years, glaucoma has been one of the leading causes of vision loss, and its early detection is essential for preventing irreparable damage. This study proposes an efficient classification approach using segmentation on retinal fundus images from the MESSIDOR dataset, focusing on optic disk segmentation to enhance disease-specific feature extraction while minimizing irrelevant background information. The pipeline includes preprocessing steps such as image resizing and green channel extraction, followed by segmentation of both the optic cup and optic disk using traditional image processing techniques, including pixel clustering with superpixel methods, global thresholding, and the Circular Hough Transform (CHT). Performance is evaluated using metrics such as testing correctness (TC), sensitivity, specificity, F1-score, and computational time under 5-fold, 10-fold, and Leave-One-Out (LOO) cross-validation (CV). Results show that preprocessing significantly improves accuracy across all models, with mi-SPSVM under 10-fold CV achieving the best results, obtaining 90.00% sensitivity and a 67.09% F1-score. Meanwhile, mi-KSPSVM achieved the lowest computational time, processing a single image in just 0.09s. These findings demonstrate that incorporating segmentation into the classification task can enhance both diagnostic accuracy and computational efficiency, making the approach suitable for real-time applications and resource-limited environments. Overall, this study presents a promising and scalable solution for automated OD screening, with future work aimed at integrating deep learning-based segmentation to improve precision further.
},
 year = {2025}
}

Copy | Download

TY  - JOUR
T1  - Segmentation-Driven Glaucoma Classification from Retinal Fundus Images Using Optic Disk Localization, Linear and Nonlinear Multiple Instance Learning Frameworks

AU  - Mahbub Hasan
Y1  - 2025/09/25
PY  - 2025
N1  - https://doi.org/10.11648/j.mcs.20251005.11
DO  - 10.11648/j.mcs.20251005.11
T2  - Mathematics and Computer Science
JF  - Mathematics and Computer Science
JO  - Mathematics and Computer Science
SP  - 83
EP  - 91
PB  - Science Publishing Group
SN  - 2575-6028
UR  - https://doi.org/10.11648/j.mcs.20251005.11
AB  - In recent years, glaucoma has been one of the leading causes of vision loss, and its early detection is essential for preventing irreparable damage. This study proposes an efficient classification approach using segmentation on retinal fundus images from the MESSIDOR dataset, focusing on optic disk segmentation to enhance disease-specific feature extraction while minimizing irrelevant background information. The pipeline includes preprocessing steps such as image resizing and green channel extraction, followed by segmentation of both the optic cup and optic disk using traditional image processing techniques, including pixel clustering with superpixel methods, global thresholding, and the Circular Hough Transform (CHT). Performance is evaluated using metrics such as testing correctness (TC), sensitivity, specificity, F1-score, and computational time under 5-fold, 10-fold, and Leave-One-Out (LOO) cross-validation (CV). Results show that preprocessing significantly improves accuracy across all models, with mi-SPSVM under 10-fold CV achieving the best results, obtaining 90.00% sensitivity and a 67.09% F1-score. Meanwhile, mi-KSPSVM achieved the lowest computational time, processing a single image in just 0.09s. These findings demonstrate that incorporating segmentation into the classification task can enhance both diagnostic accuracy and computational efficiency, making the approach suitable for real-time applications and resource-limited environments. Overall, this study presents a promising and scalable solution for automated OD screening, with future work aimed at integrating deep learning-based segmentation to improve precision further.

VL  - 10
IS  - 5
ER  -

Copy | Download