Abstract
This study uses a deep learning algorithm to analyze optic disc photographs (ODPs) and classify eyes as glaucomatous or healthy based on optic nerve appearance. ODPs from three databases were independently graded by two glaucoma specialists. Images were preprocessed using the open-source language R and RimNet, a deep learning model for optic disc segmentation, to prepare inputs for training. The model was developed with Python based on Google’s Vision Transformer (ViT). After assessing the ODPs, the model provided an output between 0 and 1 to predict glaucoma likelihood (≥ 0.5 signified glaucoma). Model performance was evaluated using the area under the receiver operating curve (AUC), where 1 indicates perfect classification. A total of 1,432 glaucomatous eyes with a mean MD of − 2.09 dB were analyzed using the model. The model achieved AUCs of 1.00, 0.98, and 1.00 in the training, validation, and test phases respectively. Overall accuracy in test images was 0.987, sensitivity was 0.994, and specificity was 0.969 with grader labels as the ground truth. A later assessment using 956 advanced glaucomatous eyes (MD < − 6 dB) with a mean MD of − 11.71 reached 99.9% accuracy. The model demonstrated high accuracy in detecting glaucomatous optic nerve damage from ODPs. The model’s strong potential for early disease detection suggests that deep learning can be a valuable, cost-effective tool in glaucoma screening, especially in resource-limited regions. Our study demonstrates the potential of deep learning in providing accessible, early-stage glaucoma detection, supporting global efforts to prevent vision loss.
Similar content being viewed by others
Data availability
The data underlying this article are de-identified patient records and are available upon request to Dr. Joseph Caprioli, Ophthalmology, Jules Stein Eye Institute, Los Angeles, CA 90095, USA; Caprioli@jsei.ucla.edu.
References
Steinmetz, J. D. et al. Causes of blindness and vision impairment in 2020 and trends over 30 years, and prevalence of avoidable blindness in relation to VISION 2020: The Right to Sight: An analysis for the Global Burden of Disease Study. Lancet Glob. Health. 9(2), e144–e160. https://doi.org/10.1016/S2214-109X(20)30489-7 (2021).
Vision impairment and blindness. Accessed 25 July 2023. https://www.who.int/news-room/fact-sheets/detail/blindness-and-visual-impairment.
Burr, J. M. et al. The clinical effectiveness and cost-effectiveness of screening for open angle glaucoma: A systematic review and economic evaluation. Health Technol. Assess. 11(41), iii–iv. https://doi.org/10.3310/hta11410 (2007).
Moyer, V. A. Screening for glaucoma: U.S. Preventive Services Task Force recommendation statement. Ann. Intern. Med. 159(7), 484–489. https://doi.org/10.7326/0003-4819-159-6-201309170-00686 (2013).
Leske, M. C. The epidemiology of open-angle glaucoma: A review. Am. J. Epidemiol. 118(2), 166–191. https://doi.org/10.1093/oxfordjournals.aje.a113626 (1983).
Sommer, A. et al. Relationship between intraocular pressure and primary open angle glaucoma among white and black Americans. The Baltimore Eye Survey. Arch. Ophthalmol. Chic Ill 1960 109(8), 1090–1095. https://doi.org/10.1001/archopht.1991.01080080050026 (1991).
Hollows, F. C. & Graham, P. A. Intra-ocular pressure, glaucoma, and glaucoma suspects in a defined population. Br. J. Ophthalmol. 50(10), 570–586 (1966).
Kass, M. A. et al. The ocular hypertension treatment study: A randomized trial determines that topical ocular hypotensive medication delays or prevents the onset of primary open-angle glaucoma. Arch. Ophthalmol. Chic Ill. 1960 120(6), 701–713. https://doi.org/10.1001/archopht.120.6.701 (2002).
Balazsi, A. G., Drance, S. M., Schulzer, M. & Douglas, G. R. Neuroretinal rim area in suspected glaucoma and early chronic open-angle glaucoma. Correlation with parameters of visual function. Arch. Ophthalmol. Chic. Ill. 1960 102(7), 1011–1014. https://doi.org/10.1001/archopht.1984.01040030813022 (1984).
Quigley, H. A., Addicks, E. M. & Green, W. R. Optic nerve damage in human glaucoma. III. Quantitative correlation of nerve fiber loss and visual field defect in glaucoma, ischemic neuropathy, papilledema, and toxic neuropathy. Arch. Ophthalmol. Chic. Ill. 1960 100(1), 135–146. https://doi.org/10.1001/archopht.1982.01030030137016 (1982).
Varma, R., Steinmann, W. C. & Scott, I. U. Expert agreement in evaluating the optic disc for glaucoma. Ophthalmology 99(2), 215–221. https://doi.org/10.1016/s0161-6420(92)31990-6 (1992).
Tielsch, J. M., Katz, J., Quigley, H. A., Miller, N. R. & Sommer, A. Intraobserver and interobserver agreement in measurement of optic disc characteristics. Ophthalmology 95(3), 350–356. https://doi.org/10.1016/s0161-6420(88)33177-5 (1988).
Zangwill, L., Shakiba, S., Caprioli, J. & Weinreb, R. N. Agreement between clinicians and a confocal scanning laser ophthalmoscope in estimating cup/disk ratios. Am. J. Ophthalmol. 119(4), 415–421. https://doi.org/10.1016/s0002-9394(14)71226-7 (1995).
Resnikoff, S. et al. Estimated number of ophthalmologists worldwide (International Council of Ophthalmology update): Will we meet the needs?. Br. J. Ophthalmol. 104(4), 588–592. https://doi.org/10.1136/bjophthalmol-2019-314336 (2020).
Johnson, G. J. & Foster, A. Training in community ophthalmology. Int. Ophthalmol. 14(3), 221–226. https://doi.org/10.1007/BF00158322 (1990).
Ahn, J. M. et al. A deep learning model for the detection of both advanced and early glaucoma using fundus photography. PLoS ONE 13(11), e0207982. https://doi.org/10.1371/journal.pone.0207982 (2018).
Christopher, M. et al. Performance of deep learning architectures and transfer learning for detecting glaucomatous optic neuropathy in fundus photographs. Sci. Rep. 8, 16685. https://doi.org/10.1038/s41598-018-35044-9 (2018).
Nouri-Mahdavi, K. et al. Macular ganglion cell/inner plexiform layer measurements by Spectral Domain Optical Coherence Tomography for detection of early glaucoma and comparison to retinal nerve fiber layer measurements. Am. J. Ophthalmol. 156(6), 1297-1307.e2. https://doi.org/10.1016/j.ajo.2013.08.001 (2013).
Bouris, E. et al. A neural network for automated image quality assessment of optic disc photographs. J. Clin. Med. 12(3), 1217. https://doi.org/10.3390/jcm12031217 (2023).
Rasheed, H. A. et al. RimNet: A deep neural network pipeline for automated identification of the optic disc rim. Ophthalmology Science 3(1), 100244. https://doi.org/10.1016/j.xops.2022.100244 (2023).
Alexey, D. An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv: 2010.11929. (2020).
Fu, Z. (2022). Vision Transformer: Vit and its Derivatives. arXiv preprint arXiv:2205.11239.
Tan, M., & Le, Q. Efficientnetv2: Smaller models and faster training. In International Conference on Machine Learning, 10096–10106. PMLR (2021).
Phan, S. et al. Evaluation of deep convolutional neural networks for glaucoma detection. Jpn. J. Ophthalmol. 63(3), 276–283. https://doi.org/10.1007/s10384-019-00659-6 (2019).
Zulfira, F. Z., Suyanto, S. & Septiarini, A. Segmentation technique and dynamic ensemble selection to enhance glaucoma severity detection. Comput. Biol. Med. 139, 104951. https://doi.org/10.1016/j.compbiomed.2021.104951 (2021).
Caprioli, J. Clinical evaluation of the optic nerve in glaucoma. Trans. Am. Ophthalmol. Soc. 92, 589–641 (1994).
Damms, T. & Dannheim, F. Sensitivity and specificity of optic disc parameters in chronic glaucoma. Invest. Ophthalmol. Vis. Sci. 34(7), 2246–2250 (1993).
Jonas, J. B., Gusek, G. C. & Naumann, G. O. Optic disc, cup and neuroretinal rim size, configuration and correlations in normal eyes. Invest. Ophthalmol. Vis. Sci. 29(7), 1151–1158 (1988).
Bengtsson, B. The variation and covariation of cup and disc diameters. Acta Ophthalmol. (Copenh) 54(6), 804–818. https://doi.org/10.1111/j.1755-3768.1976.tb01801.x (1976).
How, A. C. S., Tan, G. S. W. & Chan, Y. H. Population prevalence of tilted and torted optic discs among an adult Chinese population in Singapore: The Tanjong Pagar Study. Arch. Ophthalmol. 127(7), 894–899. https://doi.org/10.1001/archophthalmol.2009.134 (2009).
Grassi, L. et al. Phenotypic expressions of the optic disc in primary open-angle glaucoma. Eye https://doi.org/10.1038/s41433-023-02627-4 (2023).
Pan, C. W., Dirani, M., Cheng, C. Y., Wong, T. Y. & Saw, S. M. The age-specific prevalence of myopia in Asia: A meta-analysis. Optom. Vis. Sci. 92(3), 258–266. https://doi.org/10.1097/OPX.0000000000000516 (2015).
Girkin, C. A. Differences in optic nerve structure between individuals of predominantly African and European ancestry: Implications for disease detection and pathogenesis. Clin. Ophthalmol. (Auckl. N.Z.) 2(1), 65–69 (2008).
Zangwill, L. M. et al. Racial differences in optic disc topography: baseline results from the confocal scanning laser ophthalmoscopy ancillary study to the ocular hypertension treatment study. Arch. Ophthalmol. Chic. Ill. 1960 122(1), 22–28. https://doi.org/10.1001/archopht.122.1.22 (2004).
Tielsch, J. M. et al. Racial variations in the prevalence of primary open-angle glaucoma. The Baltimore Eye Survey. JAMA 266(3), 369–374 (1991).
Varma, R. et al. Race-, age-, gender-, and refractive error-related differences in the normal optic disc. Arch. Ophthalmol. Chic. Ill. 1960 112(8), 1068–1076. https://doi.org/10.1001/archopht.1994.01090200074026 (1994).
Li, Z. et al. Efficacy of a deep learning system for detecting glaucomatous optic neuropathy based on color fundus photographs. Ophthalmology 125(8), 1199–1206. https://doi.org/10.1016/j.ophtha.2018.01.023 (2018).
Cho, J., Lee, K., Shin, E., Choy, G., Do, S. How much data is needed to train a medical image deep learning system to achieve necessary high accuracy? Published online 19 Nov 2015. Accessed 2 Oct 2023. http://arxiv.org/abs/1511.06348.
Dai, L. et al. A deep learning system for detecting diabetic retinopathy across the disease spectrum. Nat. Commun. 12(1), 3242. https://doi.org/10.1038/s41467-021-23458-5 (2021).
Ting, D. S. W. et al. Development and validation of a deep learning system for diabetic retinopathy and related eye diseases using retinal images from multiethnic populations with diabetes. JAMA 318(22), 2211–2223. https://doi.org/10.1001/jama.2017.18152 (2017).
Gulshan, V. et al. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs. JAMA 316(22), 2402–2410. https://doi.org/10.1001/jama.2016.17216 (2016).
Funding
Dr. Caprioli received grants from the Simms/Mann Family Foundation, the Payden Memorial Foundation, and Research to Prevent Blindness. The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Author information
Authors and Affiliations
Contributions
E.B. conceived the study, developed the model architecture, conducted the analysis, and wrote the main manuscript text, B.K.L. drove the project forward, conducted data analysis, and edited the manuscript, E.M. , J.L., and Z.F. developed the model and analyzed the data, J.L. contributed to the computer science methodology, O.P.O. and S.W.J. provided clinical expertise and image grading, Z.F. performed statistical analysis, E.M. and O.A. assisted with data processing, and J.C. supervised the study and provided funding. All authors reviewed the manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Bouris, E., Leyva, B.K., Odugbo, O.P. et al. A vision transformer model for the detection of glaucoma from optic disc photographs. Sci Rep (2026). https://doi.org/10.1038/s41598-026-44662-7
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-026-44662-7
