Hybrid Deep Learning Architecture for Glaucoma Detection: Integrating a Multi-Network CNN Ensemble with ANFIS

Glaucoma is a leading cause of irreversible blindness worldwide, making early and accurate diag¬nosis critical for preventing severe vision loss [1]. While deep learning has advanced computer-aided diagnostics, traditional single-stream Convolutional Neural Networks (CNNs) [2] often strug¬gle with overfitting on small datasets and managing diagnostic uncertainties in overlapping disease patterns. To address these limitations, this thesis proposes a novel hybrid architecture that inte¬grates a multi-network CNN ensemble with an Adaptive Neuro-Fuzzy Inference System (ANFIS), as implemented and evaluated in the accompanying Python programmatic framework. Specifi¬cally, the Python pipeline pre-processes retinal fundus images via Region of Interest (ROI) ex¬traction and enhancement, feeding them into a parallel feature extraction module utilizing pre-trained ResNet, DenseNet, and MobileNet backbones. These fused, high-level structural features are subsequently passed into the ANFIS module, which applies adaptive fuzzy reasoning to ef¬fectively manage diagnostic uncertainty and subtle structural variations. Evaluated on aggregated benchmark datasets—including RIM-ONE, DRISHTI-GS1, and ACRIMA—the custom MultiNet-ANFIS Python program demonstrates superior diagnostic performance. When directly compared to standalone baseline models (such as standard ResNet18, DenseNet121, and MobileNet) within the script’s testing loop, the proposed CNN-ANFIS framework achieves significantly higher Ac¬curacy, Precision, and Area Under the Curve (AUC). By synergizing the robust feature extraction capabilities of ensemble deep learning with the interpretable decision-making of fuzzy logic, this programmatically validated model successfully mitigates classification errors and offers a highly efficient, scalable solution for automated glaucoma screening.

Deep Learning, Convolutional Neural Networks, Tele-Ophthalmology

1. L. Li et al., “A large-scale database and a cnn model for attention-based glaucoma detec-tion,” IEEE Transactions on Medical Imaging, vol. 39, no. 2, pp. 413–424, 2019. [Google Scholar] [Crossref]

2. M. Govindan, V. Dhakshnamurthy, K. Sreerangan, M. Nagarajan, and S. Rajamanickam, “A framework for early detection of glaucoma in retinal fundus images using deep learning,” Engineering Proceedings, vol. 62, no. 1, p. 3, 2024. [Google Scholar] [Crossref]

3. S. Kotagiri, S. K. Krishnamoorthy, V. A. Mahadevan, and S. Nagarajan, “Shuffle fuzzy attention network for glaucoma detection from fundus images,” Computers and Electrical Engineering, vol. 133, p. 111 024, 2026, ISSN: 0045-7906. [Google Scholar] [Crossref]

4. E. Tonti et al., “Artificial intelligence and advanced technology in glaucoma: A review,” Journal of Personalized Medicine, vol. 14, no. 10, p. 1062, 2024. [Google Scholar] [Crossref]

5. R. Rajalakshmi, R. Subashini, R. M. Anjana, and V. Mohan, “Automated diabetic retinopa¬thy detection in smartphone-based fundus photography using artificial intelligence,” Eye, vol. 32, no. 6, pp. 1138–1144, 2018. [Google Scholar] [Crossref]

6. J. Yuen, S. Pike, S. Khachikyan, and S. Nallasamy, “Telehealth in ophthalmology: Digital horizons in remote eye care,” in Digital Health, Exon Publications, 2022. [Google Scholar] [Crossref]

7. M. Mohammadpour, Z. Heidari, M. Mirghorbani, and H. Hashemi, “Smartphones, tele-ophthalmology, and vision 2020,” International Journal of Ophthalmology, vol. 10, no. 12, pp. 1909–1918, 2017. [Google Scholar] [Crossref]

8. C. Hanson, M. T. S. Tennant, and C. J. Rudnisky, “Optometric referrals to retina specialists: Evaluation and triage via teleophthalmology,” Telemedicine and e-Health, vol. 14, no. 5, pp. 441–445, 2008. [Google Scholar] [Crossref]

9. A. S. Mousavi, S. F. M. Baigi, F. Dahmardeh, M. R. Mehneh, and R. Darrudi, “Tele-ophthalmology: A systematic review of randomized controlled trials,” Frontiers in Health Informatics, vol. 12, p. 140, 2023. [Google Scholar] [Crossref]

10. V. Lakshminarayanan, J. Zelek, and A. McBride, “Smartphone science in eye care and medicine,” Optics & Photonics News, 2015. [Google Scholar] [Crossref]

11. P. Li et al., “Usability testing of a smartphone-based retinal camera among first-time users in the primary care setting,” BMJ Innovations, vol. 5, no. 4, pp. 120–126, 2019. [Google Scholar] [Crossref]

12. N. A. Aldossary, “Artificial intelligence in optometry: Potential benefits and key challenges: A narrative review,” International Journal of Advanced Computer Science and Applications, vol. 16, no. 8, 2025. [Google Scholar] [Crossref]

13. L. F. F. M. Santos, M. A´ . Sa´nchez-Tena, C. Alvarez-Peregrina, J.-M. Sa´nchez-Gonza´lez, and C. Martinez-Perez, “The role of artificial intelligence in optometric diagnostics and research: Deep learning and time-series forecasting applications,” Technologies, vol. 13, no. 2, p. 77, 2025. [Google Scholar] [Crossref]

14. D. B. Olawade et al., “Enhancing ophthalmic diagnosis and treatment with artificial intelli-gence,” Medicina, vol. 61, no. 3, p. 433, 2025. [Google Scholar] [Crossref]

15. L. Gonza´lez-Vides, J. L. Herna´ndez-Verdejo, and P. Can˜adas-Sua´rez, “Eye tracking in op-tometry: A systematic review,” Journal of Eye Movement Research, vol. 16, no. 3, pp. 1–55, 2023. [Google Scholar] [Crossref]

16. M. Anwar et al., “E-glaunet: A cnn-based ensemble deep learning model for glaucoma de¬tection and staging using retinal fundus images,” Computers, Materials & Continua, vol. 84, no. 2, pp. 3477–3502, 2025. [Google Scholar] [Crossref]

17. R. Kashyap, R. Nair, S. Gangadharan, M. Botto-Tobar, S. Farooq, and A. Rizwan, “Glau-coma detection and classification using improved u-net deep learning model,” Healthcare, vol. 10, no. 12, p. 2497, 2022. [Google Scholar] [Crossref]

18. S. Chen, D. Wei, C. Hong, L. Li, X. Qiu, and H. Jia, “Glauseg-net: Retinal fundus medical image automatic segmentation with multi-task learning for glaucoma early screening,” IEEE Access, 2024. [Google Scholar] [Crossref]

• What the Desert Fathers Teach Data Scientists: Ancient Ascetic Principles for Ethical Machine-Learning Practice
• Comparative Analysis of Some Machine Learning Algorithms for the Classification of Ransomware
• Comparative Performance Analysis of Some Priority Queue Variants in Dijkstra’s Algorithm
• Transfer Learning in Detecting E-Assessment Malpractice from a Proctored Video Recordings.
• Dual-Modal Detection of Parkinson’s Disease: A Clinical Framework and Deep Learning Approach Using NeuroParkNet