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.

  • Comment
  • Published:

Bias in predictive models for vitreoretinal diseases: ethnic and socioeconomic disparities in artificial intelligence

Eye (2025)Cite this article

Subjects

Access through your institution
Buy or subscribe

The use of artificial intelligence (AI) in vitreoretinal diseases presents a promising avenue for achieving improved outcomes in vitreoretinal diseases. Predictive models are particularly promising in surgical pathology due to their potential to assist with patient selection, risk stratification, and prediction of surgical outcomes. However, as AI technologies have become increasingly incorporated into vitreoretinal surgery, it is crucial to address potential biases, particularly those related to ethnicity, socioeconomic status, and social background. Such biases can significantly impact the performance of AI models, potentially resulting in inequitable healthcare outcomes, particularly for minority or underrepresented groups.

AI systems are trained on large datasets, commonly from electronic health records, imaging data, and patient metadata [1]. When these datasets are unrepresentative or biased towards certain groups, the resulting models may perform poorly for individuals from different ethnic, socioeconomic, or geographical backgrounds.

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

Access options

Access through your institution

Buy this article

  • Purchase on SpringerLink
  • Instant access to full article PDF
Buy now

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

References

  1. Janett RS, Yeracaris PP. Electronic Medical Records in the American Health System: challenges and lessons learned. Cien Saude Colet. 2020;25:1293–304.

    Article  PubMed  Google Scholar 

  2. Gupta B, Neffendorf JE, Wong R, Laidlaw DAH, Williamson TH. Ethnic variation in vitreoretinal surgery: differences in clinical presentation and outcome. Eur J Ophthalmol. 2017;27:367–71.

    Article  PubMed  Google Scholar 

  3. Sivaprasad S, Gupta B, Gulliford MC, Dodhia H, Mohamed M, Nagi D, et al. Ethnic variations in the prevalence of diabetic retinopathy in people with diabetes attending screening in the United Kingdom (DRIVE UK). PLoS One. 2012;7:e32182.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  4. Mastropasqua R, Luo YH, Cheah YS, Egan C, Lewis JJ, da Cruz L. Black patients sustain vision loss while White and South Asian patients gain vision following delamination or segmentation surgery for tractional complications associated with proliferative diabetic retinopathy. Eye. 2017;31:1468–74.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  5. Chandra A, Banerjee P, Davis D, Charteris D. Ethnic variation in rhegmatogenous retinal detachments. Eye. 2015;29:803–7.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  6. Chen DK, Modi Y, Al-Aswad LA. Promoting transparency and standardization in ophthalmologic artificial intelligence: a call for artificial intelligence model card. Asia Pac J Ophthalmol. 2022;11:215–8.

    Article  Google Scholar 

  7. Jacoba CMP, Celi LA, Lorch AC, Fickweiler W, Sobrin L, Gichoya JW, et al. Bias and non-diversity of big data in artificial intelligence: focus on retinal diseases. Semin Ophthalmol. 2023;38:433–41.

    Article  PubMed  Google Scholar 

  8. Xu D, Uhr J, Patel SN, Pandit RR, Jenkins TL, Khan MA, et al. Sociodemographic factors influencing rhegmatogenous retinal detachment presentation and outcome. Ophthalmol Retin. 2021;5:337–41.

    Article  Google Scholar 

  9. Rehman Siddiqui MA, Abdelkader E, Hammam T, Murdoch JR, Lois N. Socioeconomic status and delayed presentation in rhegmatogenous retinal detachment. Acta Ophthalmol. 2010;88:e352–3.

    Article  PubMed  Google Scholar 

  10. Anguita R, Ting MYL, Makuloluwa A, Charteris DG. Causal factors for late presentation of retinal detachment. Eye. 2023;37:185–6.

    Article  PubMed  Google Scholar 

  11. Anguita R, Roth J, Makuloluwa A, Shahid S, Katta M, Khalid H, et al. Late presentation of retinal detachment: clinical features and surgical outcomes. Retina. 2021;41:1833–8.

    Article  PubMed  Google Scholar 

  12. Ferro Desideri L, Danilovska T, Bernardi E, Artemiev D, Paschon K, Hayoz M, et al. Artificial intelligence-enhanced OCT biomarkers analysis in macula-off rhegmatogenous retinal detachment patients. Transl Vis Sci Technol. 2024;13:21.

    Article  PubMed  PubMed Central  Google Scholar 

  13. Poh SSJ, Sia JT, Yip MYT, Tsai ASH, Lee SY, Tan GSW, et al. Artificial intelligence, digital imaging, and robotics technologies for surgical vitreoretinal diseases. Ophthalmol Retin. 2024;8:633–45.

    Article  Google Scholar 

  14. Mihan A, Pandey A, Van Spall HG. Mitigating the risk of artificial intelligence bias in cardiovascular care. Lancet Digit Health. 2024;6:e749–e54.

    Article  PubMed  CAS  Google Scholar 

  15. Sufian MA, Alsadder L, Hamzi W, Zaman S, Sagar A, Hamzi B. Mitigating algorithmic bias in AI-driven cardiovascular imaging for fairer diagnostics. Diagnostics. 2024;14:2675.

    Article  PubMed  PubMed Central  Google Scholar 

  16. Burlina P, Joshi N, Paul W, Pacheco KD, Bressler NM. Addressing artificial intelligence bias in retinal diagnostics. Transl Vis Sci Technol. 2021;10:13.

    Article  PubMed  PubMed Central  Google Scholar 

  17. DeCamp M, Lindvall C. Mitigating bias in AI at the point of care. Science. 2023;381:150–2.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  18. Mittermaier M, Raza MM, Kvedar JC. Bias in AI-based models for medical applications: challenges and mitigation strategies. NPJ Digit Med. 2023;6:113.

    Article  PubMed  PubMed Central  Google Scholar 

  19. Khan SM, Liu X, Nath S, Korot E, Faes L, Wagner SK, et al. A global review of publicly available datasets for ophthalmological imaging: barriers to access, usability, and generalisability. Lancet Digit Health. 2021;3:e51–e66.

    Article  PubMed  CAS  Google Scholar 

  20. Vagliano I, Byrne Salsas C, Wunn T, Schut MC. External validation and transportability of models to predict acute kidney injury in the intensive care unit. Stud Health Technol Inf. 2022;295:148–51.

    Google Scholar 

  21. Correa R, Pahwa K, Patel B, Vachon CM, Gichoya JW, Banerjee I. Efficient adversarial debiasing with concept activation vector—medical image case-studies. J Biomed Inf. 2024;149:104548.

    Article  Google Scholar 

  22. Zhang J, Wu J, Zhou XS, Shi F, Shen D. Recent advancements in artificial intelligence for breast cancer: Image augmentation, segmentation, diagnosis, and prognosis approaches. Semin Cancer Biol. 2023;96:11–25.

    Article  PubMed  CAS  Google Scholar 

  23. Yang J, Soltan AAS, Eyre DW, Yang Y, Clifton DA. An adversarial training framework for mitigating algorithmic biases in clinical machine learning. NPJ Digit Med. 2023;6:55.

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Department of Ophthalmology, Inselspital, University Hospital of Bern, Bern, Switzerland

    Lorenzo Ferro Desideri & Martin Zinkernagel

  2. Moorfields Eye Hospital NHS Foundation Trust, London, UK

    Ben Kirkpatrick & Rodrigo Anguita

Authors
  1. Lorenzo Ferro Desideri
    View author publications

    Search author on:PubMed Google Scholar

  2. Ben Kirkpatrick
    View author publications

    Search author on:PubMed Google Scholar

  3. Martin Zinkernagel
    View author publications

    Search author on:PubMed Google Scholar

  4. Rodrigo Anguita
    View author publications

    Search author on:PubMed Google Scholar

Contributions

RA conceived and designed the manuscript. LFD, BK, MZ and RA analysed and interpreted the literature. LFD, BK, MZ and RA drafted the manuscript and made critical revision of the manuscript.

Corresponding author

Correspondence to Rodrigo Anguita.

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.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Ferro Desideri, L., Kirkpatrick, B., Zinkernagel, M. et al. Bias in predictive models for vitreoretinal diseases: ethnic and socioeconomic disparities in artificial intelligence. Eye (2025). https://doi.org/10.1038/s41433-025-03990-0

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1038/s41433-025-03990-0

Search

Advanced search

Quick links