A Comprehensive Review of Multimodal and Transformer-Based AI Models for Ophthalmic Disease Detection
DOI:
https://doi.org/10.64882/ijrt.v14.i3.1733Keywords:
Ophthalmology, Retinal Disease Diagnosis, Deep Learning, Vision Transformers (ViT), Multimodal Medical Imagin.Abstract
Impaired vision due to retinal and optic neuropathies continues to pose a major public health challenge. Early detection is critical to prevent irreversible visual impairment, especially for conditions such as diabetic retinopathy (DR), glaucoma, age-related macular degeneration (AMD), and cataracts. With the surge in artificial intelligence (AI) and deep learning (DL) capabilities, automated diagnostic frameworks are gaining traction as viable solutions to alleviate diagnostic workload, improve accessibility, and enhance screening accuracy. This review presents a comprehensive analysis of recent deep learning and hybrid machine learning models developed for the detection and classification of ophthalmic and neurodegenerative diseases. The synthesis spans fundus-based convolutional neural networks (CNNs), Vision Transformers (ViTs), multimodal architectures, and ensemble systems. Through comparative assessment, the paper highlights that ensemble CNNs and transformer-based methods consistently demonstrate superior accuracy, especially in multi-disease classification scenarios. However, challenges such as data imbalance, model interpretability, and deployment constraints persist. Despite these advancements, challenges remain, including dataset imbalance, limited model interpretability, and real-world deployment constraints. By consolidating recent trends, methodologies, and performance insights, this review aims to guide researchers and practitioners toward developing robust, interpretable, and scalable AI-driven solutions for ophthalmic disease screening and diagnosis.
References
T. Nazir, A. Irtaza, A. Javed, H. Malik, D. Hussain, and R. A. Naqvi, “Retinal image analysis for diabetes-based eye disease detection using deep learning,” Applied Sciences, vol. 10, no. 18, p. 6185, 2020, doi: 10.3390/app10186185.
M. Aamir, M. Irfan, T. Ali, G. Ali, A. Shaf, S. A. Sultan, A. Al Beshri, T. Alasbali, and M. H. Mahnashi, “An adoptive threshold-based multi-level deep convolutional neural network for glaucoma eye disease detection and classification,” Diagnostics, vol. 10, no. 8, p. 602, 2020, doi: 10.3390/diagnostics10080602.
P. Glaret Subin and P. Muthukannan, “Optimized convolution neural network based multiple eye disease detection,” Computers in Biology and Medicine, vol. 146, art. no. 105648, 2022, doi: 10.1016/j.compbiomed.2022.105648.
K. Prasad, P. S. Sajith, M. Neema, L. Madhu, and P. N. Priya, “Multiple eye disease detection using deep neural network,” in TENCON 2019—2019 IEEE Region 10 Conference, Kochi, India, 2019, pp. 2148–2153, doi: 10.1109/TENCON.2019.8929666.
H. Zhang et al., “Multi omics approaches to discover biomarkers of thyroid eye disease: A systematic review,” International Journal of Biological Sciences, vol. 20, no. 15, pp. 6038–6055, 2024, doi: 10.7150/ijbs.103977. Frontiersijbs.com+1
A. Bali and V. Mansotra, “Analysis of deep learning techniques for prediction of eye diseases: A systematic review,” Archives of Computational Methods in Engineering, vol. 31, no. 1, pp. 487–520, 2024, doi: 10.1007/s11831-023-09989-8. PMC+6ResearchGate+6MDPI+6
A. A. Ahmed, A. Aldhahab, and H. M. Al Abboodi, “Review of eye diseases detection and classification using deep learning techniques,” in BIO Web of Conferences, vol. 97, 2024, art. no. 00012, doi: [ADD DOI].
M. A. Iqbal, J. Kim, I. Han, and S. K. Kim, “Attention-driven feature fusion integrating Swin transformer and CNN models for improved ocular disease classification,” in 2024 International Conference on Engineering and Emerging Technologies (ICEET), Dubai, United Arab Emirates, 2024, pp. 1–6, doi: 10.1109/ICEET65156.2024.10913537.
J. Zhao et al., “Multi-label classification of retinal diseases based on fundus images using ResNet and Transformer,” Medical & Biological Engineering & Computing, vol. 62, pp. 3459–3469, 2024, doi: 10.1007/s11517-024-03144-6.
M. A. Jawad et al., “Towards improved fundus disease detection using Swin transformers,” Multimedia Tools and Applications, vol. 83, pp. 78125–78159, 2024, doi: 10.1007/s11042-024-18627-9.
H. Yu, “An eye disease image diagnosis method based on a hybrid machine learning and deep learning model,” in 2024 5th International Conference on Electronic Communication and Artificial Intelligence (ICECAI), Shenzhen, China, 2024, pp. 347–352, doi: 10.1109/ICECAI62591.2024.10675277.
R. C. Joshi, A. K. Sharma, and M. K. Dutta, “VisionDeep-AI: Deep learning-based retinal blood vessels segmentation and multi-class classification framework for eye diagnosis,” Biomedical Signal Processing and Control, vol. 94, art. no. 106273, 2024, doi: 10.1016/j.bspc.2024.106273.
M. N. Hasan, M. E. R. Pial, S. Das, N. Siddique, and H. Wang, “DIA-VXNET: A framework for automated diabetic eye disease detection using transfer learning with feature fusion network,” Biomedical Signal Processing and Control, vol. 100, pt. C, art. no. 106907, 2025, doi: 10.1016/j.bspc.2024.106907.
W. Hu et al., “FundusNet: A deep-learning approach for fast diagnosis of neurodegenerative and eye diseases using fundus images,” Bioengineering, vol. 12, p. 57, 2025, doi: 10.3390/bioengineering12010057.
J. Zhao, S. Li, Y. Hao, and C. Zhang, “Bridging the modality gap in multimodal eye disease screening: Learning modality shared-specific features via multi-level regularization,” IEEE Signal Processing Letters, vol. 32, pp. 586–590, 2025, doi: 10.1109/LSP.2025.3526094.
R. U. Karim et al., “Optimizing stroke recognition with MediaPipe and machine learning: An explainable AI approach for facial landmark analysis,” IEEE Access, vol. 13, pp. 32636–32660, 2025, doi: 10.1109/ACCESS.2025.3550577.
P. B. Singh et al., “HViTML: Hybrid vision transformer with machine learning-based classification model for glaucomatous eye,” Multimedia Tools and Applications, early access, 2025, doi: 10.1007/s11042-024-20544-w.
M. S. Hossain and M. Shorfuzzaman, “Early diabetic retinopathy cyber-physical detection system using attention-guided deep CNN fusion,” IEEE Transactions on Network Science and Engineering, vol. 12, no. 3, pp. 1898–1910, May–Jun. 2025, doi: 10.1109/TNSE.2025.3541138.
E. S. Cutur and N. G. Inan, “Multi-class classification of retinal eye diseases from ophthalmoscopy images using transfer learning-based vision transformers,” Journal of Digital Imaging, early access, 2025, doi: 10.1007/s10278-025-01416-7.
A. G. Tuwan et al., “Eye diseases multiclass classification using convolutional vision transformer,” TechRxiv, preprint, Feb. 2025, doi: 10.36227/techrxiv.173886337.74494015/v1.
S. El-Ateif and A. Idri, “Eye diseases diagnosis using deep learning and multimodal medical eye imaging,” Multimedia Tools and Applications, vol. 83, pp. 30773–30818, 2024, doi: 10.1007/s11042-023-16835-3.
A. Albelaihi and D. M. Ibrahim, “DeepDiabetic: An identification system of diabetic eye diseases using deep neural networks,” IEEE Access, vol. 12, pp. 10769–10789, 2024, doi: 10.1109/ACCESS.2024.3354854.
R. Roopashree, R. Ghosh, and M. M. Gaur, “Exploring the potential of deep learning in diagnosing eye diseases,” in 2024 15th International Conference on Computing Communication and Networking Technologies (ICCCNT), Kamand, India, 2024, pp. 1–6, doi: 10.1109/ICCCNT61001.2024.10724872.
A. Chaudhari, P. Shelke, P. Thombare, and S. Sandbhor, “Cost-effective real-time eye disease detection and classification using deep learning techniques,” in 2024 15th International Conference on Computing Communication and Networking Technologies (ICCCNT), Kamand, India, 2024, pp. 1–8, doi: 10.1109/ICCCNT61001.2024.10726189.
Downloads
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
