Academic Journal
View in EDS
Linked Full Text

A hybrid dual-stream CNN framework with dynamic data augmentation and improved Manta Ray Foraging Optimization for robust glaucoma detection.

Bibliographic Details
Title: A hybrid dual-stream CNN framework with dynamic data augmentation and improved Manta Ray Foraging Optimization for robust glaucoma detection.
Authors: Atia A; Department of Information Systems, Faculty of Computers and Information, Kafrelsheikh University, Kafrelsheikh, Egypt. azzamohamed@fci.kfs.edu.eg., Abdel-Kader H; Department of Information Systems, Faculty of Computers and Information, Menoufia University, Shibin El-Kom, Al Minufiyah, Egypt., Abo-Seida OM; Department of Computer Science, Faculty of Computers and Information, Kafrelsheikh University, Kafrelsheikh, Egypt., Abdelatey A; Department of Information Systems, Faculty of Computers and Information, Menoufia University, Shibin El-Kom, Al Minufiyah, Egypt.; Faculty of Computers and Artificial Intelligence, Menoufia National University, Tukh Tambisha, Al Minufiyah, Egypt.
Source: Scientific reports [Sci Rep] 2026 Apr 17; Vol. 16 (1). Date of Electronic Publication: 2026 Apr 17.
Publication Type: Journal Article
Language: English
Journal Info: Publisher: Nature Publishing Group Country of Publication: England NLM ID: 101563288 Publication Model: Electronic Cited Medium: Internet ISSN: 2045-2322 (Electronic) Linking ISSN: 20452322 NLM ISO Abbreviation: Sci Rep Subsets: MEDLINE
Imprint Name(s): Original Publication: London : Nature Publishing Group, copyright 2011-
MeSH Terms: Glaucoma*/diagnosis , Glaucoma*/diagnostic imaging , Image Processing, Computer-Assisted*/methods , Image Interpretation, Computer-Assisted*/methods, Convolutional Neural Networks ; Humans ; Algorithms ; Deep Learning
Abstract: A progressive neurological condition, glaucoma is one of the main causes of irreversible blindness in the globe. Early detection is crucial to preventing irreversible vision loss however conventional diagnostic methods are often time-consuming, and heavily reliant on clinical expertise. This study presents an innovative deep learning framework designed to automate glaucoma detection while addressing key challenges such as data imbalance, image quality inconsistencies, and the need for accurate feature extraction. The proposed framework begins with a dedicated preprocessing pipeline that enhances image clarity, and isolates clinically relevant regions of interest. To handle class imbalance, a novel hybrid data augmentation strategy is introduced, combining geometric transformations with adaptive Gaussian noise injection that adjusts intensity according to image characteristics, thereby simulating realistic clinical variability. At its core, the framework leverages a dual-stream CNN architecture that integrates DenseNet121 for structural feature extraction and ResNet50 for texture representation. A lightweight channel-wise attention mechanism is then introduced to selectively emphasize clinically significant channels while suppressing redundant features, thereby balancing efficiency and interpretability. To optimize model performance while reducing computational overhead, an Improved Manta Ray Foraging Optimization (IMRFO) algorithm is employed. IMRFO enhances the standard MRFO with Partial Centroid Opposition-Based Learning (PCOBL) to dynamically fine-tune hyperparameters, including augmentation settings and transfer learning configurations. Experimental validation was conducted on four public benchmark datasets, ACRIMA, Drishti-Gs, ORIGA, and RIM-ONE-DL, demonstrating the framework's superior performance across all evaluation metrics. The model achieved 100.00% accuracy, precision, recall, and AUC on both ACRIMA and Drishti-Gs, with losses of 0.003 and 0.001, respectively. On ORIGA, it reached 99.70% accuracy, 99.80% precision, 99.30% recall, and 99.50% AUC (loss = 0.01). On RIM-ONE-DL, the model scored 99.90% accuracy, 99.70% precision, 99.50% recall, and 99.70% AUC (loss = 0.006). These findings confirm the framework's robustness and clinical applicability for effective glaucoma screening.
(© 2026. The Author(s).)
Competing Interests: Declarations. Competing interests: The authors declare no competing interests.
References: Moreno, M. V., Houriet, C. & Grounauer, P. A. Ocular phantom-based feasibility study of an early diagnosis device for glaucoma. Sensors 21(2), 579 (2021). (PMID: 33467457783066910.3390/s21020579)
Tam, Y. C. et al. Global prevalence of glaucoma and projections of glaucoma burden through 2040: A systematic review and meta-analysis. Ophthalmology 121(11), 2081–2090 (2014). (PMID: 10.1016/j.ophtha.2014.05.013)
Kolář, R. Detection of glaucomatous eye via color fundus images using fractal dimensions. Radioengineering 17(3), 109–114 (2018).
Michelson, G., Warntges, S., Hornegger, J. & Lausen, B. The papilla as a screening parameter for early diagnosis of glaucoma. Dtsch. Arztebl. Int. 105(34–35), 583–589 (2018).
Pannu, R. et al. Enhanced glaucoma classification through advanced segmentation by integrating cup-to-disc ratio and neuro-retinal rim features. Comput. Med. Imaging Graph. https://doi.org/10.1016/j.compmedimag.2025.102559 (2025). (PMID: 10.1016/j.compmedimag.2025.10255940315660)
Zubair, M., Yamin, A. & Khan, S. A. Automated detection of optic disc for the analysis of retina using color fundus image. 2013 IEEE International Conference on Imaging Systems and Techniques (IST). IEEE, 2013.‏.
Kumar, B. N., Chauhan, R. P. & Dahiya, N. Detection of glaucoma using image processing techniques: A review. In Proceedings of the 2016 International Conference on Microelectronics, Computing, and Communications (MicroCom), Durgapur, India, pp. 1–9, 2016.
Qiu, K. et al. Application of the ISNT rules on retinal nerve fiber layer thickness and neuroretinal rim area in healthy myopic eyes. Acta Ophthalmol. 96(2), 161–167 (2018). (PMID: 2919715710.1111/aos.13586)
Shoukat, A. & Akbar, S. Artificial intelligence techniques for glaucoma detection through retinal images. In Artificial Intelligence and Internet of Things, Springer, pp. 1–20, 2021.
Reguant, R., Brunak, S. & Saha, S. Understanding inherent image features in CNN-based assessment of diabetic retinopathy. Sci. Rep. 11(1), 1–12 (2021). (PMID: 10.1038/s41598-021-89225-0)
Vaghjiani, D., et al. Visualizing and understanding inherent image features in CNN-based glaucoma detection. In 2020 Digital Image Computing: Techniques and Applications (DICTA), IEEE, pp. 1–3, 2020.
Owais, M., et al. A comprehensive review tracing the evolution of volumetric medical imaging analysis from classic CNNs to emerging AI-agents. Authorea Preprints (2025).‏.
Zubair, M. et al. An interpretable framework for gastric cancer classification using multi-channel attention mechanisms and transfer learning approach on histopathology images. Sci. Rep. 15(1), 13087 (2025). (PMID: 402404571200378710.1038/s41598-025-97256-0)
Guo, J., Azzopardi, G., Shi, C., Jansonius, N. M. & Petkov, N. Automatic determination of vertical cup-to-disc ratio in retinal fundus images for glaucoma screening. IEEE Access 7, 8527–8541 (2019). (PMID: 10.1109/ACCESS.2018.2890544)
Shoba, S. G. & Therese, A. B. Detection of glaucoma disease in fundus images based on morphological operation and finite element method. Biomed. Signal Process. Control. 62, 101986 (2020). (PMID: 10.1016/j.bspc.2020.101986)
Pruthi, J., Khanna, K. & Arora, S. Optic cup segmentation from retinal fundus images using glowworm swarm optimization for glaucoma detection. Biomed. Signal Process. Control 60, 102004 (2020). (PMID: 10.1016/j.bspc.2020.102004)
Qureshi, I., Khan, M. A., Sharif, M., Saba, T. & Ma, J. Detection of glaucoma based on cup-to-disc ratio using fundus images. Int. J. Intell. Syst. Technol. Appl. 19, 1–16 (2020).
Kirar, B. S., Reddy, G. R. S. & Agrawal, D. K. Glaucoma detection using SS-QB-VMD-Based Fine Sub-Band Images From Fundus Images. IETE J. Res. 1–12 (2021).
Martins, J., Cardoso, J. & Soares, F. Offline computer-aided diagnosis for glaucoma detection using fundus images targeted at mobile devices. Comput. Methods Programs Biomed. 192, 105341 (2020). (PMID: 3215553410.1016/j.cmpb.2020.105341)
Shinde, R. Glaucoma detection in retinal fundus images using U-Net and supervised machine learning algorithms. Intell. Med. 5, 100038 (2021).
Song, W. T., Lai, I.-C. & Su, Y.-Z. A statistical robust glaucoma detection framework combining Retinex, CNN, and DOE using fundus images. IEEE Access 9, 103772–103783 (2021). (PMID: 10.1109/ACCESS.2021.3098032)
Nazir, T. et al. Retinal image analysis for diabetes-based eye disease detection using deep learning. Appl. Sci. 10, 6185 (2020). (PMID: 10.3390/app10186185)
Nazir, T., Irtaza, A. & Starovoitov, V. Optic disc and optic cup segmentation for glaucoma detection from blur retinal images using improved Mask-RCNN. Int. J. Opt. 2021, 6641980 (2021). (PMID: 10.1155/2021/6641980)
Serte, S. & Serener, A. Graph-based saliency and ensembles of convolutional neural networks for glaucoma detection. IET Image Process. 15, 797–804 (2020). (PMID: 10.1049/ipr2.12063)
Nayak, D. R., Das, D., Majhi, B., Bhandary, S. V. & Acharya, U. R. ECNet: An evolutionary convolutional network for automated glaucoma detection using fundus images. Biomed. Signal Process. Control 67, 102559 (2021). (PMID: 10.1016/j.bspc.2021.102559)
Sreng, S. et al. Deep learning for optic disc segmentation and glaucoma diagnosis on retinal images. Appl. Sci. 10(14), 4916 (2020). (PMID: 10.3390/app10144916)
Elangovan, P. & Nath, M. K. Glaucoma assessment from color fundus images using convolutional neural network. Int. J. Imaging Syst. Technol. 31(2), 955–971 (2021). (PMID: 10.1002/ima.22494)
Chaudhary, P. K. & Pachori, R. B. Automatic diagnosis of glaucoma using two-dimensional Fourier-Bessel series expansion based empirical wavelet transform. Biomed. Signal Process. Control 64, 102237 (2021). (PMID: 10.1016/j.bspc.2020.102237)
Gheisari, S. et al. A combined convolutional and recurrent neural network for enhanced glaucoma detection. Sci. Rep. 11(1), 1945 (2021). (PMID: 33479405782023710.1038/s41598-021-81554-4)
de Sales Carvalho, N. R. et al. Automatic method for glaucoma diagnosis using a three-dimensional convoluted neural network. Neurocomputing 438, 72–83 (2021). (PMID: 10.1016/j.neucom.2020.07.146)
Lin, M. et al. Automated diagnosing primary open-angle glaucoma from fundus image by simulating human’s grading with deep learning. Sci. Rep. 12(1), 14080 (2022). (PMID: 35982106938853610.1038/s41598-022-17753-4)
Kashyap, R., et al. Glaucoma detection and classification using improved U-Net Deep Learning Model. Healthcare. 10(12). MDPI (2022).‏.
Nawaz, M. et al. An efficient deep learning approach to automatic glaucoma detection using optic disc and optic cup localization. Sensors 22(2), 434 (2022). (PMID: 35062405878079810.3390/s22020434)
Saha, S., Vignarajan, J. & Frost, S. A fast and fully automated system for glaucoma detection using color fundus photographs. Sci. Rep. 13(1), 18408 (2023). (PMID: 378912381061181310.1038/s41598-023-44473-0)
Fan, R., et al. Detecting glaucoma from fundus photographs using deep learning without convolutions: Transformer for improved generalization. Ophthalmol. Sci. 3(1), 100233 (2023).‏.
Shoukat, A. et al. Automatic diagnosis of glaucoma from retinal images using deep learning approach. Diagnostics 13(10), 1738 (2023). (PMID: 372382221021771110.3390/diagnostics13101738)
Velpula, V. K. & Sharma, L. D. Multi-stage glaucoma classification using pre-trained convolutional neural networks and voting-based classifier fusion. Front. Physiol. 14, 1175881 (2023). (PMID: 373831461029361710.3389/fphys.2023.1175881)
Muduli, D. et al. Retinal imaging based glaucoma detection using modified pelican optimization based extreme learning machine. Sci. Rep. 14(1), 29660 (2024). (PMID: 396137991160695710.1038/s41598-024-79710-7)
Subha, K. J., Rajavel, R. & Paulchamy, B. An improved ensemble deep learning framework for glaucoma detection. Multimedia Tools Appl. https://doi.org/10.1007/s11042-024-20088-z (2024). (PMID: 10.1007/s11042-024-20088-z)
Sharma, S. K. et al. An evolutionary supply chain management service model based on deep learning features for automated glaucoma detection using fundus images. Eng. Appl. Artif. Intell. 128, 107449 (2024). (PMID: 10.1016/j.engappai.2023.107449)
Sujithra, B. S. & Albert Jerome, S. Identification of glaucoma in fundus images utilizing gray wolf optimization with deep convolutional neural network-based resnet50 model. Multimedia Tools Appl. 83(16), 49301–49319 (2024). (PMID: 10.1007/s11042-023-17506-z)
Rangaiah, P. K. B. & Augustine, R. Enhanced glaucoma detection using U-Net and U-Net+ architectures using deep learning techniques. Photodiagn. Photodyn. Ther. https://doi.org/10.1016/j.pdpdt.2025.104621 (2025). (PMID: 10.1016/j.pdpdt.2025.104621)
Sivakumar, R. & Penkova, A. Enhancing glaucoma detection through multi-modal integration of retinal images and clinical biomarkers. Eng. Appl. Artif. Intell. 143, 110010 (2025). (PMID: 10.1016/j.engappai.2025.110010)
Wang, A. X. et al. Addressing imbalance in health data: Synthetic minority oversampling using deep learning. Comput. Biol. Med. 188, 109830 (2025). (PMID: 3998336110.1016/j.compbiomed.2025.109830)
Wang, Z., et al. A comprehensive survey on data augmentation. IEEE Trans. Knowl. Data Eng. (2025).‏.
Zhao, W., Zhang, Z. & Wang, L. Manta ray foraging optimization: An effective bio-inspired optimizer for engineering applications. Eng. Appl. Artif. Intell. 87, 103300 (2020). (PMID: 10.1016/j.engappai.2019.103300)
Abdullahi, M. et al. Manta ray foraging optimization algorithm: Modifications and applications. IEEE Access 11, 53315–53343 (2023). (PMID: 10.1109/ACCESS.2023.3276264)
Ewees, A. A., Elaziz, M. A. & Oliva, D. A new multi-objective optimization algorithm combined with opposition-based learning. Expert Syst. Appl. 165, 113844 (2021). (PMID: 10.1016/j.eswa.2020.113844)
Li, J. et al. A dual opposition-based learning for differential evolution with protective mechanism for engineering optimization problems. Appl. Soft Comput. 113, 107942 (2021). (PMID: 10.1016/j.asoc.2021.107942)
Mahdavi, S., Rahnamayan, S. & Deb, K. Partial opposition-based learning using current best candidate solution. In 2016 IEEE Symposium Series on Computational Intelligence (SSCI). IEEE, 2016.‏.
Si, T. & Dutta, R. Partial opposition-based particle swarm optimizer in artificial neural network training for medical data classification. Int. J. Inf. Technol. Decis. Mak. 18(05), 1717–1750 (2019). (PMID: 10.1142/S0219622019500329)
Si, T. et al. Pcobl: A novel opposition-based learning strategy to improve metaheuristics exploration and exploitation for solving global optimization problems. IEEE Access 11, 46413–46440 (2023). (PMID: 10.1109/ACCESS.2023.3273298)
Diaz-Pinto, A., et al. CNNs for automatic glaucoma assessment using fundus images: an extensive validation. Biomed. Eng. Online 18, 1–19 (2019).‏.
Zhang, Z., et al. Origa-light: An online retinal fundus image database for glaucoma analysis and research. In 2010 Annual International Conference of the IEEE Engineering in Medicine and Biology. IEEE, 2010.‏.
Batista, F. J. F. et al. Rim-one dl: A unified retinal image database for assessing glaucoma using deep learning. Image Anal. Stereol. 39(3), 161–167 (2020). (PMID: 10.5566/ias.2346)
Sivaswamy, J., et al. Drishti-gs: Retinal image dataset for optic nerve head (onh) segmentation. 2014 IEEE 11th International Symposium on Biomedical Imaging (ISBI). IEEE, 2014.‏.
Ahmed, M., Seraj, R. & Islam, S. M. S. The k-means algorithm: A comprehensive survey and performance evaluation. Electronics 9(8), 1295 (2020). (PMID: 10.3390/electronics9081295)
Lakshmi, K. S. & Sargunam, B. Exploration of AI-powered DenseNet121 for effective diabetic retinopathy detection. Int. Ophthalmol. 44(1), 90 (2024). (PMID: 3836709810.1007/s10792-024-03027-7)
Lin, C.-L. & Wu, K.-C. Development of revised ResNet-50 for diabetic retinopathy detection. BMC Bioinf. 24(1), 157 (2023). (PMID: 10.1186/s12859-023-05293-1)
Jin, X. et al. Delving deep into spatial pooling for squeeze-and-excitation networks. Pattern Recognit. 121, 108159 (2022). (PMID: 10.1016/j.patcog.2021.108159)
Han, J., Pei, J., & Tong, H. Data mining: concepts and techniques. Morgan kaufmann (2022).‏.
Powers, D. M. W. Evaluation: from precision, recall and F-measure to ROC, informedness, markedness and correlation. arXiv preprint arXiv:2010.16061  (2020).‏.
Owais, M. et al. Unified synergistic deep learning framework for multimodal 2-d and 3-d radiographic data analysis: Model development and validation. IEEE Access https://doi.org/10.1109/ACCESS.2024.3487575 (2024). (PMID: 10.1109/ACCESS.2024.3487575)
Zubair, M. et al. A comprehensive review of techniques, algorithms, advancements, challenges, and clinical applications of multi-modal medical image fusion for improved diagnosis. Comput. Methods Programs Biomed. https://doi.org/10.1016/j.cmpb.2025.109014 (2025). (PMID: 10.1016/j.cmpb.2025.10901440946521)
Contributed Indexing: Keywords: Dual-stream convolutional neural network (CNN); Dynamic Gaussian noise; Glaucoma detection; Lightweight channel-wise attention mechanism
Entry Date(s): Date Created: 20260417 Date Completed: 20260715 Latest Revision: 20260715
Update Code: 20260715
PubMed Central ID: PMC13090355
DOI: 10.1038/s41598-026-45384-6
PMID: 41998000
Database: MEDLINE
Description
ISSN:2045-2322
DOI:10.1038/s41598-026-45384-6