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:

Clinical AI tools must convey predictive uncertainty for each individual patient

Nature Medicine volume 29, pages 2996–2998 (2023)Cite this article

Subjects

Artificial intelligence tools usually aim to maximize predictive accuracy, but personalized measures of uncertainty, using new techniques such as conformal prediction, are needed for clinical artificial intelligence to realize its potential and improve human health.

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

Relevant articles

Open Access articles citing this article.

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: Conformal prediction in the clinic.

References

  1. Kelly, C. J., Karthikesalingam, A., Suleyman, M., Corrado, G. & King, D. BMC Med. 17, 1–9 (2019).

    Article  CAS  Google Scholar 

  2. Andaur Navarro, C. L. et al. BMJ 375, 2281 (2021).

    Article  Google Scholar 

  3. Gaube, S. et al. npj Digit. Med. 4, 1–8 (2021).

    Article  Google Scholar 

  4. Murdoch, B. BMC Med. Ethics 22, 1–5 (2021).

    Article  Google Scholar 

  5. Evans, W. E. & Relling, M. V. Science 286, 487–491 (1999).

    Article  CAS  PubMed  Google Scholar 

  6. Coventry, L. L., Finn, J. & Bremner, A. P. Heart Lung 40, 477–491 (2011).

    Article  PubMed  Google Scholar 

  7. Jackson, S. E. & Chester, J. D. Int. J. Cancer 137, 262–266 (2015).

    Article  CAS  PubMed  Google Scholar 

  8. der Kiureghian, A. & Ditlevsen, O. Struct. Saf. 31, 105–112 (2009).

    Article  Google Scholar 

  9. Finlayson, S. G. et al. N. Engl. J. Med. 385, 283–286 (2021).

    Article  PubMed  PubMed Central  Google Scholar 

  10. Langford, B. J. et al. Clin. Microbiol. Infect. 27, 520–531 (2021).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  11. Heesom, L. et al. J. Glob. Antimicrob. Resist. 22, 782–784 (2020).

    Article  PubMed  PubMed Central  Google Scholar 

  12. Buolamwini, J. & Gebru, T. in Proceedings of Machine Learning Research vol. 81 77–91 (PMLR, 2018).

  13. Mitra, R. et al. Nat. Mach. Intell. 5, 13–23 (2023).

    Article  Google Scholar 

  14. Barber, R. F., Candes, E. J., Ramdas, A. & Tibshirani, R. J. Ann. Statist. 51, 816–845 (2023).

    Article  Google Scholar 

  15. Vovk, V., Gammerman, A. & Shafer, G. Algorithmic Learning in a Random World (Springer, 2005).

Download references

Author information

Authors and Affiliations

  1. The Alan Turing Institute, London, UK

    Christopher R. S. Banerji, Tapabrata Chakraborti & Ben D. MacArthur

  2. University College London Hospitals, NHS Foundation Trust, London, UK

    Christopher R. S. Banerji

  3. UCL Cancer Institute, Faculty of Medical Sciences, University College London, London, UK

    Christopher R. S. Banerji & Tapabrata Chakraborti

  4. Roche Pharmaceuticals, Welwyn Garden City, UK

    Chris Harbron

  5. Faculty of Medicine, University of Southampton, Southampton, UK

    Ben D. MacArthur

  6. Mathematical Sciences, University of Southampton, Southampton, UK

    Ben D. MacArthur

Authors
  1. Christopher R. S. Banerji
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Tapabrata Chakraborti
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Chris Harbron
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Ben D. MacArthur
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding authors

Correspondence to Christopher R. S. Banerji or Ben D. MacArthur.

Ethics declarations

Competing interests

The authors declare no competing interests.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Banerji, C.R.S., Chakraborti, T., Harbron, C. et al. Clinical AI tools must convey predictive uncertainty for each individual patient. Nat Med 29, 2996–2998 (2023). https://doi.org/10.1038/s41591-023-02562-7

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41591-023-02562-7

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