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:

Steps to avoid overuse and misuse of machine learning in clinical research

Nature Medicine volume 28, pages 1996–1999 (2022)Cite this article

Subjects

Machine learning algorithms are a powerful tool in healthcare, but sometimes perform no better than traditional statistical techniques. Steps should be taken to ensure that algorithms are not overused or misused, in order to provide genuine benefit for patients.

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 SpringerLink
  • Instant access to the full article PDF.

USD 39.95

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

Fig. 1: Model fitting.

References

  1. Christodoulou, E. et al. J. Clin. Epidemiol. 110, 12–22 (2019).

    Article  Google Scholar 

  2. Gravesteijn, B. Y. et al. J. Clin. Epidemiol. 122, 95–107 (2020).

    Article  Google Scholar 

  3. Zou, J. et al. Nature 559, 324–326 (2018).

    Article  CAS  Google Scholar 

  4. Topol, E. J. Nat. Med. 26, 1318–1320 (2020).

    Article  CAS  Google Scholar 

  5. Vasey, B. et al. Nat. Med. 28, 924–933 (2022).

    Article  CAS  Google Scholar 

  6. DeGrave, A. J. et al. Nat. Mach. Intell. 3, 610–619 (2021).

    Article  Google Scholar 

  7. Wynants, L. et al. Br. Med. J. 369, m1328 (2020).

    Article  Google Scholar 

  8. Abràmoff, M. D. et al. npj Digit. Med. 1, 39 (2018).

    Article  Google Scholar 

  9. Shin, S. et al. ESC Heart Fail. 8, 106–115 (2021).

    Article  Google Scholar 

  10. Cho, S. M. et al. Can. J. Cardiol. 37, 1207–1214 (2021).

    Article  Google Scholar 

  11. Uddin, S. et al. BMC Med. Inform. Decis. Mak. 19, 281 (2019).

    Article  Google Scholar 

  12. Volovici, V. et al. J. Neurotrauma 36, 3183–3189 (2019).

    Article  Google Scholar 

  13. Shashikumar, S. P. et al. npj Digit. Med. 4, 134 (2021).

    Article  Google Scholar 

  14. Cancer Genome Atlas Research Network. et al. Nat. Genet. 45, 1113–1120 (2013).

    Article  Google Scholar 

  15. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders 5th edn (APA, 2013).

  16. Futoma, J. et al. Lancet Digit. Health 2, e489–e492 (2020).

    Article  Google Scholar 

  17. Pisică, D. et al. World Neurosurg. 161, 230–239.e236 (2022).

    Article  Google Scholar 

  18. Marek, S. et al. Nature 603, 654–660 (2022).

    Article  CAS  Google Scholar 

  19. Rudin, C. Nat. Mach. Intell. 1, 206–215 (2019).

    Article  Google Scholar 

  20. Richens, J. G. et al. Nat. Commun. 11, 3923 (2020).

    Article  CAS  Google Scholar 

  21. Andaur Navarro, C. L. et al. Br. Med. J. 375, n2281 (2021).

    Google Scholar 

  22. Stulberg, J. J. et al. JAMA Surg. 153, 586–587 (2018).

    Article  Google Scholar 

  23. Twinanda, A. P. et al. IEEE Trans. Med. Imaging 36, 86–97 (2017).

    Article  Google Scholar 

Download references

Acknowledgements

We would like to thank M. van Bilsen for the figure and F. Liu for her valuable advice. V.V. wishes to thank D. Volovici for opening up the world of probability, statistics and machine learning.

Author information

Authors and Affiliations

  1. Department of Neurosurgery, Erasmus MC University Medical Center, Rotterdam, the Netherlands

    Victor Volovici

  2. Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Singapore

    Nicholas L. Syn & Joseph J. Zhao

  3. Department of Surgery, National University Hospital, National University Health System, Singapore, Singapore

    Nicholas L. Syn

  4. Cambridge Centre for AI in Medicine, University of Cambridge, Cambridge, UK

    Ari Ercole

  5. Programme in Health Services and Systems Research, Duke–NUS Medical School, Singapore, Singapore

    Nan Liu

Authors
  1. Victor Volovici
    View author publications

    Search author on:PubMed Google Scholar

  2. Nicholas L. Syn
    View author publications

    Search author on:PubMed Google Scholar

  3. Ari Ercole
    View author publications

    Search author on:PubMed Google Scholar

  4. Joseph J. Zhao
    View author publications

    Search author on:PubMed Google Scholar

  5. Nan Liu
    View author publications

    Search author on:PubMed Google Scholar

Contributions

V.V. conceived the idea, drafted the first manuscript, conceptualized the figure and supervised the work; N. S. substantially revised the manuscript and critically read all versions of the manuscript. A. E., J. J. Z. and N. L. made substantial revisions and approved the final manuscript.

Corresponding author

Correspondence to Victor Volovici.

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

Volovici, V., Syn, N.L., Ercole, A. et al. Steps to avoid overuse and misuse of machine learning in clinical research. Nat Med 28, 1996–1999 (2022). https://doi.org/10.1038/s41591-022-01961-6

Download citation

  • Published:

  • Version of record:

  • Issue date:

  • DOI: https://doi.org/10.1038/s41591-022-01961-6

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