Making machine learning matter to clinicians: model actionability in medical decision-making.


Journal

NPJ digital medicine
ISSN: 2398-6352
Titre abrégé: NPJ Digit Med
Pays: England
ID NLM: 101731738

Informations de publication

Date de publication:
24 Jan 2023
Historique:
received: 01 04 2022
accepted: 13 01 2023
entrez: 23 1 2023
pubmed: 24 1 2023
medline: 24 1 2023
Statut: epublish

Résumé

Machine learning (ML) has the potential to transform patient care and outcomes. However, there are important differences between measuring the performance of ML models in silico and usefulness at the point of care. One lens to use to evaluate models during early development is actionability, which is currently undervalued. We propose a metric for actionability intended to be used before the evaluation of calibration and ultimately decision curve analysis and calculation of net benefit. Our metric should be viewed as part of an overarching effort to increase the number of pragmatic tools that identify a model's possible clinical impacts.

Identifiants

pubmed: 36690689
doi: 10.1038/s41746-023-00753-7
pii: 10.1038/s41746-023-00753-7
pmc: PMC9871014
doi:

Types de publication

Journal Article Review

Langues

eng

Pagination

7

Informations de copyright

© 2023. The Author(s).

Références

JAMA. 2019 Oct 8;322(14):1351-1352
pubmed: 31393527
Stud Health Technol Inform. 2019 Jul 30;263:23-34
pubmed: 31411150
Circulation. 2015 Nov 17;132(20):1920-30
pubmed: 26572668
Ann Am Thorac Soc. 2018 Mar;15(3):371-379
pubmed: 29283699
BMC Med. 2019 Oct 29;17(1):195
pubmed: 31665002
Intensive Care Med. 2001 Nov;27(11):1750-5
pubmed: 11810118
J Gen Intern Med. 2018 Jan;33(1):103-115
pubmed: 28936618
Pediatr Crit Care Med. 2020 Dec;21(12):1051-1058
pubmed: 32740190
AMIA Annu Symp Proc. 2018 Dec 05;2018:887-896
pubmed: 30815131
AMIA Jt Summits Transl Sci Proc. 2020 May 30;2020:181-190
pubmed: 32477637
J Crit Care. 2017 Jun;39:6-10
pubmed: 28088009
Aust Health Rev. 2018 Aug;42(4):395-402
pubmed: 28578757
Crit Care. 2017 Dec 28;21(Suppl 3):314
pubmed: 29297399
N Engl J Med. 1989 Jun 1;320(22):1489-91
pubmed: 2497349
Nat Med. 2018 Nov;24(11):1716-1720
pubmed: 30349085
J Gen Intern Med. 2007 May;22(5):632-8
pubmed: 17443371
Diagnosis (Berl). 2015 Jun;2(2):97-103
pubmed: 26955512
Arch Intern Med. 2008 Apr 14;168(7):741-8
pubmed: 18413557
NPJ Digit Med. 2021 Feb 19;4(1):32
pubmed: 33608661
Qual Saf Health Care. 2008 Apr;17(2):122-6
pubmed: 18385406
BMJ. 2016 Jan 25;352:i6
pubmed: 26810254
JAMA. 2019 Jun 18;321(23):2281-2282
pubmed: 31107500

Auteurs

Daniel E Ehrmann (DE)

Department of Critical Care Medicine and Labatt Family Heart Centre, The Hospital for Sick Children, Toronto, ON, Canada. Dehrmann@umich.edu.
Congenital Heart Center at Mott Children's Hospital and the University of Michigan Medical School, Ann Arbor, MI, USA. Dehrmann@umich.edu.

Shalmali Joshi (S)

Center for Research on Computation on Society, Harvard University, Cambridge, MA, USA.

Sebastian D Goodfellow (SD)

Department of Critical Care Medicine and Labatt Family Heart Centre, The Hospital for Sick Children, Toronto, ON, Canada.
Faculty of Applied Science and Engineering, University of Toronto, Toronto, ON, Canada.

Mjaye L Mazwi (ML)

Department of Critical Care Medicine and Labatt Family Heart Centre, The Hospital for Sick Children, Toronto, ON, Canada.
Department of Paediatrics, University of Toronto, Toronto, ON, Canada.

Danny Eytan (D)

Department of Critical Care Medicine and Labatt Family Heart Centre, The Hospital for Sick Children, Toronto, ON, Canada.
Department of Medicine, Technion, Haifa, Israel.

Classifications MeSH