Abstract
Timely diagnosis is paramount in ocular medicine to prevent irreversible vision impairment. Despite the promise of deep learning in automated diagnosis, many existing models are tailored for a singular disease, potentially overlooking coexistent pathologies. This research introduces a hybrid ResNet50-LSTM model, designed for the concurrent detection of multiple ocular conditions. The model achieved a 100% diagnostic accuracy on this dataset, outperforming several contemporary models. Central to the approach used in this research was the combination of two neural network architectures. The Convolutional Neural Network (CNN) adeptly extracts spatial features from retinal images. In tandem, the Long-Short Term Memory (LSTM) Recurrent Neural Network interprets these features sequentially, enhancing diagnostic precision. Given its robust performance and versatility, the model presents itself as a promising diagnostic tool, meriting consideration for clinical application.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Burlina, P., Pacheco, K., Joshi, N., Freund, D., Kong, J., Bressler, N.: Comparing humans and deep learning performance for grading AMD: a study in using universal deep features and transfer learning for automated AMD analysis. Comput. Biol. Med. 109, 79–86 (2019)
Esteva, A., et al.: Dermatologist-level classification of skin cancer with deep neural networks. Nature 542(7639), 115–118 (2017)
Fu, H., et al.: GANet: a deep learning framework for glaucoma diagnosis with gated attention mechanism. IEEE J. Biomed. Health Inf. 25(4), 1184–1194 (2021)
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 770–778 (2016)
Neha, K., Gour, D.: Automatic detection of diabetic retinopathy stages using deep convolutional neural network. In: 2019 IEEE International Conference on Electrical, Computer and Communication Technologies (ICECCT), pp. 1–5. IEEE (2019)
Saha, S., Srinivasan, S., Krishnan, S.M.: Ocular disease identification using deep learning. Expert Syst. Appl. 157, 113456 (2020)
Tavakoli, M., Rabbani, H.: Multi-task deep learning for the automated diagnosis of diabetic retinopathy, retinal vein occlusion and age-related macular degeneration. In: Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization, vol. 7, no. 2, pp. 177–186 (2019)
Ting, D.S.W., Cheung, C.Y.L.: Development and validation of a deep learning system for diabetic retinopathy and related eye diseases using retinal images from multiethnic populations with diabetes. In: JAMA, vol. 322, no. 17, pp. 1661–1670 (2019)
Wang, W., Wang, P.: Multi-scale densely connected convolutional features for retinal disease classification. IEEE Access 9, 81756–81766 (2021)
Zhang, L., Shen, C.: Deep learning for automated diabetic retinopathy diagnosis using a multi-scale residual network with attention mechanism. Pattern Recogn. 107, 107477 (2020)
George, Y., Antony, B.J., Ishikawa, H., Wollstein, G., Schuman, J.S., Garnavi, R.: Attention-guided 3D-CNN framework for glaucoma detection and structural-functional association using volumetric images. IEEE J. Biomed. Health Inf. 24(12), 3421–3430 (2020). https://doi.org/10.1109/JBHI.2020.3001019
Zhang, H., et al.: Automatic segmentation and visualization of choroid in oct with knowledge infused deep learning. IEEE J. Biomed. Health Inf. 24(12), 3408–3420 (2020). https://doi.org/10.1109/jbhi.2020.3023144
Azimi, B., Rashno, A., Fadaei, S.: Fully convolutional networks for fluid segmentation in retina images. In: 2020 International Conference on Machine Vision and Image Processing (MVIP), pp. 1–7 (2020). https://doi.org/10.1109/MVIP49855.2020.9116914
Zedan, M.J., Zulkifley, M.A., Ibrahim, A.A., Moubark, A.M., Kamari, N.A.M., Abdani, S.R.: Automated glaucoma screening and diagnosis based on retinal fundus images using deep learning approaches: a comprehensive review. Diagnostics 13(13), 2180 (2023)
Acknowledgements
International Development Research Centre (IDRC) and the Swedish International Development Cooperation Agency (SIDA) under the Artificial Intelligence for Development (AI4D) Africa Scholarship program with the Africa Center for Technology Studies (ACTS) for the funding provided.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Muchuchuti, S., Viriri, S. (2023). Retinal Disease Diagnosis with a Hybrid ResNet50-LSTM Deep Learning. In: Bebis, G., et al. Advances in Visual Computing. ISVC 2023. Lecture Notes in Computer Science, vol 14362. Springer, Cham. https://doi.org/10.1007/978-3-031-47966-3_28
Download citation
DOI: https://doi.org/10.1007/978-3-031-47966-3_28
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-47965-6
Online ISBN: 978-3-031-47966-3
eBook Packages: Computer ScienceComputer Science (R0)