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.

  • Correspondence
  • Published:

Developing specific reporting guidelines for diagnostic accuracy studies assessing AI interventions: The STARD-AI Steering Group

Nature Medicine volume 26, pages 807–808 (2020)Cite this article

Subjects

Access through your institution
Buy or subscribe

To the Editor — Artificial intelligence (AI)-based technologies dominate medical headlines and are routinely touted as the panacea for a number of longstanding deficiencies across health systems globally. Stakeholders from healthcare, government, computer science and industry backgrounds are confident that AI can be positioned to tackle (1) the high rate of avoidable medical errors, (2) workflow inefficiencies and (3) delivery inefficiencies associated with contemporary healthcare provision1. Despite these lofty ambitions, the integration of AI into everyday practice within the health sector has been limited thus far.

So far, the majority of AI interventions that are close to translation are predominantly in the field of medical diagnostics. In the current paradigm, diagnostic investigations require timely interpretation from an expert clinician in order to generate a diagnosis and to direct subsequent episodes of care. The recurring issue with this system is that diagnostic services are inundated with large volumes of work, which can often supersede workforce capacity. In order to address this, diagnostic AI algorithms, with notable examples targeted toward breast cancer, lung cancer and diabetic retinopathy, have positioned themselves as medical devices2 that may achieve diagnostic accuracy comparable to that of an expert clinician while concurrently alleviating health-resource use. However, although this paradigm shift may seem imminent, it is crucial to note that much of the evidence supporting diagnostic algorithms has been disseminated in the absence of AI-specific reporting guidelines. Without this guidance, and in a relatively nascent areas, key stakeholders are poorly placed to appraise quality and compare diagnostic accuracy between studies.

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

Relevant articles

Open Access articles citing this article.

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. Topol, E. J. Nat. Med. 25, 44–56 (2019).

    Article  CAS  Google Scholar 

  2. US Food and Drug Administration. https://www.fda.gov/medical-devices/digital-health/software-medical-device-samd (2018).

  3. Bossuyt, P. M. et al. Br. Med. J. 351, h5527 (2015).

    Article  Google Scholar 

  4. Bluemke, D. A. et al. Radiology https://doi.org/10.1148/radiol.2019192515 (2019).

    Article  PubMed  Google Scholar 

  5. The EQUATOR Network. https://www.equator-network.org/library/reporting-guidelines-under-development/reporting-guidelines-under-development-for-other-study-designs/#STARDAI (2019).

  6. Liu, X. et al. Nat. Med. 25, 1467–1468 (2019).

    Article  CAS  Google Scholar 

  7. Collins, G. S. & Moons, K. G. M. Lancet 393, 1577–1579 (2019).

    Article  Google Scholar 

Download references

Acknowledgements

Infrastructure support was provided by the NIHR Imperial Biomedical Research Centre.

Author information

Authors and Affiliations

  1. Institute of Global Health Innovation, Imperial College London, London, UK

    Viknesh Sounderajah, Hutan Ashrafian, Ravi Aggarwal, Alan Karthikesalingam, Dominic King & Ara Darzi

  2. Department of Surgery and Cancer, Imperial College London, London, UK

    Viknesh Sounderajah, Hutan Ashrafian, Ravi Aggarwal, Alan Karthikesalingam, Dominic King, Sheraz R. Markar & Ara Darzi

  3. DeepMind, London, UK

    Jeffrey De Fauw

  4. Institute of Inflammation and Ageing, College of Medical and Dental Sciences, University of Birmingham, Birmingham, UK

    Alastair K. Denniston & Xiaoxuan Liu

  5. University Hospitals Birmingham NHS Foundation Trust, Birmingham, UK

    Alastair K. Denniston & Xiaoxuan Liu

  6. Health Data Research UK, London, UK

    Alastair K. Denniston & Xiaoxuan Liu

  7. Public Health England, London, UK

    Felix Greaves

  8. Department of Radiology, University of Ottawa, Ottawa, Canada

    Matthew D. F. McInnes

  9. Clinical Epidemiology Program, Ottawa Hospital Research Institute, Ottawa, Canada

    Matthew D. F. McInnes

  10. Division of Health Policy and Management, Harvard T.H. Chan School of Public Health, Boston, MA, USA

    Trishan Panch

  11. School of Public Health, Imperial College London, London, UK

    Jonathan Pearson-Stuttard

  12. Singapore Eye Research Institute, Singapore National Eye Center, Singapore, Singapore

    Daniel S. W. Ting

  13. Journal of the American Medical Association, Chicago, IL, USA

    Robert M. Golub

  14. Ottawa Hospital Research Institute, Ottawa, Canada

    David Moher

  15. Department of Clinical Epidemiology, Biostatistics and Bioinformatics, University of Amsterdam, Amsterdam, the Netherlands

    Patrick M. Bossuyt

Authors
  1. Viknesh Sounderajah
    View author publications

    Search author on:PubMed Google Scholar

  2. Hutan Ashrafian
    View author publications

    Search author on:PubMed Google Scholar

  3. Ravi Aggarwal
    View author publications

    Search author on:PubMed Google Scholar

  4. Jeffrey De Fauw
    View author publications

    Search author on:PubMed Google Scholar

  5. Alastair K. Denniston
    View author publications

    Search author on:PubMed Google Scholar

  6. Felix Greaves
    View author publications

    Search author on:PubMed Google Scholar

  7. Alan Karthikesalingam
    View author publications

    Search author on:PubMed Google Scholar

  8. Dominic King
    View author publications

    Search author on:PubMed Google Scholar

  9. Xiaoxuan Liu
    View author publications

    Search author on:PubMed Google Scholar

  10. Sheraz R. Markar
    View author publications

    Search author on:PubMed Google Scholar

  11. Matthew D. F. McInnes
    View author publications

    Search author on:PubMed Google Scholar

  12. Trishan Panch
    View author publications

    Search author on:PubMed Google Scholar

  13. Jonathan Pearson-Stuttard
    View author publications

    Search author on:PubMed Google Scholar

  14. Daniel S. W. Ting
    View author publications

    Search author on:PubMed Google Scholar

  15. Robert M. Golub
    View author publications

    Search author on:PubMed Google Scholar

  16. David Moher
    View author publications

    Search author on:PubMed Google Scholar

  17. Patrick M. Bossuyt
    View author publications

    Search author on:PubMed Google Scholar

  18. Ara Darzi
    View author publications

    Search author on:PubMed Google Scholar

Corresponding author

Correspondence to Hutan Ashrafian.

Ethics declarations

Competing interests

D.K., J.D.F. and A.K. are employees of Google Health. A.D. is an adviser at Google DeepMind/Health.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Sounderajah, V., Ashrafian, H., Aggarwal, R. et al. Developing specific reporting guidelines for diagnostic accuracy studies assessing AI interventions: The STARD-AI Steering Group. Nat Med 26, 807–808 (2020). https://doi.org/10.1038/s41591-020-0941-1

Download citation

  • Published:

  • Issue date:

  • DOI: https://doi.org/10.1038/s41591-020-0941-1

This article is cited by

Search

Advanced search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing