Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Article
  • Published:

Artificial intelligence to distinguish retinal vein occlusion patients using color fundus photographs

Eye volume 37, pages 2026–2032 (2023)Cite this article

Subjects

Abstract

Purpose

Our aim is to establish an AI model for distinguishing color fundus photographs (CFP) of RVO patients from normal individuals.

Methods

The training dataset included 2013 CFP from fellow eyes of RVO patients and 8536 age- and gender-matched normal CFP. Model performance was assessed in two independent testing datasets. We evaluated the performance of the AI model using the area under the receiver operating characteristic curve (AUC), accuracy, precision, specificity, sensitivity, and confusion matrices. We further explained the probable clinical relevance of the AI by extracting and comparing features of the retinal images.

Results

Our model achieved an average AUC was 0.9866 (95% CI: 0.9805–0.9918), accuracy was 0.9534 (95% CI: 0.9421–0.9639), precision was 0.9123 (95% CI: 0.8784–9453), specificity was 0.9810 (95% CI: 0.9729–0.9884), and sensitivity was 0.8367 (95% CI: 0.7953–0.8756) for identifying fundus images of RVO patients in training dataset. In independent external datasets 1, the AUC of the RVO group was 0.8102 (95% CI: 0.7979–0.8226), the accuracy of 0.7752 (95% CI: 0.7633–0.7875), the precision of 0.7041 (95% CI: 0.6873–0.7211), specificity of 0.6499 (95% CI: 0.6305–0.6679) and sensitivity of 0.9124 (95% CI: 0.9004–0.9241) for RVO group. There were significant differences in retinal arteriovenous ratio, optic cup to optic disc ratio, and optic disc tilt angle (p = 0.001, p = 0.0001, and p = 0.0001, respectively) between the two groups in training dataset.

Conclusion

We trained an AI model to classify color fundus photographs of RVO patients with stable performance both in internal and external datasets. This may be of great importance for risk prediction in patients with retinal venous occlusion.

This is a preview of subscription content, access via your institution

Access options

Buy this article

  • Purchase on Springer Link
  • Instant access to full article PDF
Buy now

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: Workflow of our experimental design.
Fig. 2: The diagnostic efficiency of our AI model.
Fig. 3: Representative images of fundus photograph and their heat maps in training dataset and independent external testing datasets.

Similar content being viewed by others

Data availability

The datasets analysed during the current study are available from the corresponding author on reasonable request.

References

  1. Ip M, Hendrick A. Retinal vein occlusion review. Asia Pac J Ophthalmol (Philos). 2018;7:40–5.

    Google Scholar 

  2. Rogers S, McIntosh R, Cheung N, Lim L, Wang J, Mitchell P, et al. The prevalence of retinal vein occlusion: pooled data from population studies from the United States, Europe, Asia, and Australia. Ophthalmology. 2010;117:313–9.e1.

    Article  PubMed  Google Scholar 

  3. Di Capua M, Coppola A, Albisinni R, Tufano A, Guida A, Di Minno MN, et al. Cardiovascular risk factors and outcome in patients with retinal vein occlusion. J Thromb Thrombolysis. 2010;30:16–22.

    Article  PubMed  Google Scholar 

  4. Hayreh SS, Zimmerman B, McCarthy MJ, Podhajsky P. Systemic diseases associated with various types of retinal vein occlusion. Am J Ophthalmol. 2001;131:61–77.

    Article  CAS  PubMed  Google Scholar 

  5. Hirano Y, Suzuki N, Tomiyasu T, Kurobe R, Yasuda Y, Esaki Y, et al. Multimodal imaging of microvascular abnormalities in retinal vein occlusion. J. Clin. Med. 2021;10:405.

  6. Kim MJ, Woo SJ, Park KH, Kim TW. Retinal nerve fiber layer thickness is decreased in the fellow eyes of patients with unilateral retinal vein occlusion. Ophthalmology. 2011;118:706–10.

    Article  PubMed  Google Scholar 

  7. Maltsev DS, Kulikov AN, Burnasheva MA, Chhablani J. Prevalence of resolved paracentral acute middle maculopathy lesions in fellow eyes of patients with unilateral retinal vein occlusion. Acta Ophthalmol. 2020;98:e22–e28.

    Article  PubMed  Google Scholar 

  8. Arslan GD, Guven D, Demir M, Alkan AA, Ozcan D. Microvascular and functional changes according to the fundus location of the affected arteriovenous crossing in patients with branch retinal vein occlusion. Indian J Ophthalmol. 2021;69:1189–96.

    Article  PubMed  PubMed Central  Google Scholar 

  9. Pinhas A, Dubow M, Shah N, Cheang E, Liu CL, Razeen M, et al. FELLOW EYE CHANGES IN PATIENTS WITH NONISCHEMIC CENTRAL RETINAL VEIN OCCLUSION: Assessment of Perfused Foveal Microvascular Density and Identification of Nonperfused Capillaries. Retina. 2015;35:2028–36.

    Article  PubMed  PubMed Central  Google Scholar 

  10. Shin YI, Nam KY, Lee SE, Lim HB, Lee MW, Jo YJ, et al. Changes in peripapillary microvasculature and retinal thickness in the fellow eyes of patients with unilateral retinal vein occlusion: An OCTA study. Invest Ophthalmol Vis Sci. 2019;60:823–9.

    Article  PubMed  Google Scholar 

  11. Lim SH, Kim M, Chang W, Sagong M. Comparison of the lamina cribrosa thickness of patients with unilateral branch retinal vein occlusion and healthy subjects. Retina. 2017;37:515–21.

    Article  PubMed  Google Scholar 

  12. Chan MMH, Thomas AS, Yoon SP, Leitner D, Fekrat S. Clinical characteristics of patients with CRVO in one eye with subsequent RVO in the fellow eye: a retrospective observational study. Ophthalmic Surg, lasers Imaging Retin. 2019;50:444–9.

    Article  Google Scholar 

  13. McIntosh RL, Rogers SL, Lim L, Cheung N, Wang JJ, Mitchell P, et al. Natural history of central retinal vein occlusion: an evidence-based systematic review. Ophthalmology. 2010;117:1113–1123 e15.

    Article  PubMed  Google Scholar 

  14. Sakaue H, Katsumi O, Hirose T. Electroretinographic findings in fellow eyes of patients with central retinal vein occlusion. Arch Ophthalmol. 1989;107:1459–62.

    Article  CAS  PubMed  Google Scholar 

  15. Ferraz DA, Tovar-Moll F, Belfort R Jr. Artificial intelligence: from the retina to the brain. Arq Bras Oftalmol. 2021;84:197–198.

    Article  PubMed  Google Scholar 

  16. Date RC, Jesudasen SJ, Weng CY. Applications of deep learning and artificial intelligence in retina. Int Ophthalmol Clin. 2019;59:39–57.

    Article  PubMed  Google Scholar 

  17. Schmidt-Erfurth U, Sadeghipour A, Gerendas BS, Waldstein SM, Bogunovic H. Artificial intelligence in retina. Prog Retin Eye Res. 2018;67:1–29.

    Article  PubMed  Google Scholar 

  18. Zhang Y, Shi J, Peng Y, Zhao Z, Zheng Q, Wang Z, et al. Artificial intelligenceenabled screening for diabetic retinopathy: a real-world, multicenter and prospective study. BMJ Open Diab Res Care 2020;8:e001596.

  19. Yim J, Chopra R, Spitz T, Winkens J, Obika A, Kelly C, et al. Predicting conversion to wet age-related macular degeneration using deep learning. Nat Med. 2020;26:892–9.

    Article  CAS  PubMed  Google Scholar 

  20. Prabhakar B, Singh RK, Yadav KS. Artificial intelligence (AI) impacting diagnosis of glaucoma and understanding the regulatory aspects of AI-based software as medical device. Comput Med Imaging Graph. 2020;87:101818.

    Article  PubMed  Google Scholar 

  21. Li JO, Liu H, Ting DSJ, Jeon S, Chan RVP, Kim JE, et al. Digital technology, tele-medicine and artificial intelligence in ophthalmology: A global perspective. Prog Retin Eye Res. 2021;82:100900.

  22. Goldhagen BE, Al-Khersan H. Diving deep into deep learning: an update on artificial intelligence in retina. Curr Ophthalmol Rep. 2020;8:121–8.

    Article  PubMed  PubMed Central  Google Scholar 

  23. Adhi M, Filho MA, Louzada RN, Kuehlewein L, de Carlo TE, Baumal CR, et al. Retinal capillary network and foveal avascular zone in eyes with vein occlusion and fellow eyes analyzed with optical coherence tomography angiography. Invest Ophthalmol Vis Sci. 2016;57:OCT486–94.

    Article  PubMed  Google Scholar 

  24. Costanzo E, Parravano M, Gilardi M, Cavalleri M, Sacconi R, Aragona E, et al. Microvascular retinal and choroidal changes in retinal vein occlusion analyzed by two different optical coherence tomography angiography devices, ophthalmologica. J Int d’ophtalmologie Int J Ophthalmol Z fur Augenheilkd. 2019;242:8–15.

    Article  Google Scholar 

  25. Chen L, Yuan M, Sun L, Wang Y, Chen Y. Evaluation of microvascular network with optical coherence tomography angiography (OCTA) in branch retinal vein occlusion (BRVO). BMC Ophthalmol. 2020;20:154.

    Article  PubMed  PubMed Central  Google Scholar 

  26. Vieira MJ, Campos A, do Carmo A, Arruda H, Martins J, Sousa JP. Thrombophilic risk factors for retinal vein occlusion. Sci Rep. 2019;9:18972.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  27. Stojakovic T, Scharnagl H, Marz W, Winkelmann BR, Boehm BO, Schmut O. Low density lipoprotein triglycerides and lipoprotein(a) are risk factors for retinal vascular occlusion. Clin Chim Acta. 2007;382:77–81.

    Article  CAS  PubMed  Google Scholar 

  28. Nagasato D, Tabuchi H, Ohsugi H, Masumoto H, Enno H, Ishitobi N, et al. Deep-learning classifier with ultrawide-field fundus ophthalmoscopy for detecting branch retinal vein occlusion. Int J Ophthalmol. 2019;12:94–99.

    PubMed  PubMed Central  Google Scholar 

  29. Nagasato D, Tabuchi H, Ohsugi H, Masumoto H, Enno H, Ishitobi N, et al. Deep neural network-based method for detecting central retinal vein occlusion using ultrawide-field fundus ophthalmoscopy. J Ophthalmol. 2018;2018:1875431.

    Article  PubMed  PubMed Central  Google Scholar 

  30. Nagasato D, Tabuchi H, Masumoto H, Enno H, Ishitobi N, Kameoka M, et al. Automated detection of a nonperfusion area caused by retinal vein occlusion in optical coherence tomography angiography images using deep learning. PLoS One. 2019;14:e0223965.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  31. Hayreh SS. Photocoagulation for retinal vein occlusion. Prog Retin Eye Res. 2021;85:100964.

  32. Karska-Basta I, Kubicka-Trzaska A, Romanowska-Dixon B, Undas A. Thrombophilia—a risk factor of retinal vein occlusion?. Klin Ocz. 2013;115:29–33.

    Google Scholar 

  33. Yildirim C, Yaylali V, Tatlipinar S, Kaptanoglu B, Akpinar S. Hyperhomocysteinemia: a risk factor for retinal vein occlusion. Ophthalmologica. 2004;218:102–6.

    Article  CAS  PubMed  Google Scholar 

  34. Undas A, Kubicka-Trzaska A. Thrombophilia as a risk factor for central retinal vein occlusion. Klin Ocz. 2003;105:221–4.

    Google Scholar 

  35. Chan EW, Wong TY, Liao J, Cheung CY, Zheng YF, Wang JJ, et al. Branch retinal vein occlusion and optic nerve head topographic parameters: the Singapore Indian eye study. Br J Ophthalmol. 2013;97:611–6.

    Article  PubMed  Google Scholar 

  36. Szigeti A, Schneider M, Ecsedy M, Nagy ZZ, Recsan Z. Optic disc morphology in unilateral branch retinal vein occlusion using spectral domain optical coherence tomography. BMC Ophthalmol. 2015;15:178.

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

We thank the Chengdu Ikangguobin Health Examination Center for provide the color fundus photographs.

Funding

This work was supported by The Project of National Key Research and Development (No. 2018YFC1106103) to MZ and Post-Doctor Research Project, West China Hospital, Sichuan University (2021HXBH030) to XR. And Natural Science Foundation of Sichuan Province (No. 2022NSFSC1285) to XR.

Author information

Author notes

  1. These authors contributed equally: Xiang Ren, Wei Feng.

Authors and Affiliations

  1. Department of Ophthalmology, Ophthalmic Laboratory, West China Hospital, Sichuan University, Chengdu, Sichuan, 610041, P. R. China

    Xiang Ren, Ruijin Ran, Yunxia Gao, Yu Lin, Xiangyu Fu, Yunhan Tao, Ting Wang & Ming Zhang

  2. Research Laboratory of Ophthalmology and Vision Sciences, State Key Laboratory of Biotherapy, West China Hospital, Sichuan University, Chengdu, Sichuan, 610041, P. R. China

    Xiang Ren, Yu Lin, Xiangyu Fu & Ting Wang

  3. Beijing Airdoc Technology Co Ltd, Beijing, China

    Wei Feng, Bin Wang, Lie Ju, Yuzhong Chen & Lanqing He

  4. Minda Hospital of Hubei Minzu University, Enshi, China

    Ruijin Ran

  5. ECSE, Faculty of Engineering, Monash University, Melbourne, VIC, Australia

    Lie Ju & Zongyuan Ge

  6. Chengdu Ikangguobin Health Examination Center Ltd, Chengdu, China

    Wu Xi & Xiaorong Liu

  7. eResearch Centre, Monash University, Melbourne, VIC, Australia

    Zongyuan Ge

Authors
  1. Xiang Ren
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Wei Feng
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Ruijin Ran
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Yunxia Gao
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Yu Lin
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Xiangyu Fu
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Yunhan Tao
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Ting Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Bin Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Lie Ju
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. Yuzhong Chen
    View author publications

    You can also search for this author in PubMed Google Scholar

  12. Lanqing He
    View author publications

    You can also search for this author in PubMed Google Scholar

  13. Wu Xi
    View author publications

    You can also search for this author in PubMed Google Scholar

  14. Xiaorong Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  15. Zongyuan Ge
    View author publications

    You can also search for this author in PubMed Google Scholar

  16. Ming Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

XR, BW, and MZ designed the study, directed the project, and interpreted the data. XR, YG, RR, YT, WF, LJ, LH, WX, XL, TW, and YC performed the experiments. GZ provided guidance for this project. XR wrote the paper, and YL, XF, GZ, and MZ contributed to editing. All authors contributed to the paper and approved the submitted version.

Corresponding author

Correspondence to Ming Zhang.

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

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Ren, X., Feng, W., Ran, R. et al. Artificial intelligence to distinguish retinal vein occlusion patients using color fundus photographs. Eye 37, 2026–2032 (2023). https://doi.org/10.1038/s41433-022-02239-4

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41433-022-02239-4

Search

Advanced search

Quick links