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.

Estimate the hidden deployment cost of predictive models to improve patient care

Nature Medicine volume 26, pages 18–19 (2020)Cite this article

Subjects

An Author Correction to this article was published on 14 April 2020

This article has been updated

Although examples of algorithms designed to improve healthcare delivery abound, for many, clinical integration will not be achieved. The deployment cost of machine learning models is an underappreciated barrier to success. Experts propose three criteria that, assessed early, could help estimate the deployment cost.

Despite the impressive performance of healthcare-oriented machine learning models in research settings, examples of models that have been successfully deployed in a clinical setting are few1,2,3,4, to the detriment of machine learning researchers, healthcare organizations and, most importantly, patients. This is in part because the criteria typically used to evaluate model performance—such as sensitivity, specificity or area under the receiver operating characteristic (AUROC) curve—ignore downstream realities of deploying the model into clinical operations5,6. One such pressing reality is the cost of implementation.

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

Relevant articles

Open Access articles citing this article.

Access options

Rent or buy this article

Prices vary by article type

from$1.95

to$39.95

Learn more

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

Change history

References

  1. Yu, K.-H., Beam, A. L. & Kohane, I. S. Nat. Biomed. Eng. 2, 719–731 (2018).

    Article  Google Scholar 

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

    Article  CAS  Google Scholar 

  3. Esteva, A. et al. Nat. Med. 25, 25–29 (2019).

    Article  Google Scholar 

  4. Amarasingham, R., Patzer, R. E., Huesch, M., Nguyen, N. Q. & Xie, B. Health Aff. 33, 1148–1154 (2014).

    Article  Google Scholar 

  5. Wiens, J. et al. Nat. Med. https://doi.org/10.1038/s41591-019-0548-6 (2019).

  6. Shah, N. H., Milstein, A. & Bagley, S. C. JAMA https://doi.org/10.1001/jama.2019.10306 (2019).

  7. Cresswell, K. & Sheikh, A. Int. J. Med. Inform. 82, e73–e86 (2013).

    Article  Google Scholar 

  8. Larson, D. B. et al. Radiology 287, 313–322 (2018).

    Article  Google Scholar 

  9. Halabi, S. S. Validation of an artificial intelligence-based algorithm for skeletal age assessment. ClinicalTrials.gov https://clinicaltrials.gov/ct2/show/NCT03530098 (last updated October 2019).

  10. Escobar, G. J. et al. J. Hosp. Med. 11 (Suppl. 1), S18–S24 (2016).

    Article  Google Scholar 

  11. Dummett, B. A. et al. J. Hosp. Med. 11 (Suppl. 1), S25–S31 (2016).

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Department of Pediatrics, Stanford University School of Medicine, Stanford, CA, USA

    Keith E. Morse

  2. Department of Clinical Informatics, Lucile Packard Children’s Hospital, Palo Alto, CA, USA

    Keith E. Morse

  3. Center for Biomedical Informatics Research, Stanford University, Stanford, CA, USA

    Steven C. Bagley & Nigam H. Shah

Authors
  1. Keith E. Morse
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Steven C. Bagley
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Nigam H. Shah
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Keith E. Morse.

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

Morse, K.E., Bagley, S.C. & Shah, N.H. Estimate the hidden deployment cost of predictive models to improve patient care. Nat Med 26, 18–19 (2020). https://doi.org/10.1038/s41591-019-0651-8

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41591-019-0651-8

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