Predicting Progression to Septic Shock in the Emergency Department Using an Externally Generalizable Machine-Learning Algorithm.
Journal
Annals of emergency medicine
ISSN: 1097-6760
Titre abrégé: Ann Emerg Med
Pays: United States
ID NLM: 8002646
Informations de publication
Date de publication:
04 2021
04 2021
Historique:
received:
30
04
2020
revised:
30
10
2020
accepted:
10
11
2020
pubmed:
19
1
2021
medline:
13
4
2021
entrez:
18
1
2021
Statut:
ppublish
Résumé
Machine-learning algorithms allow improved prediction of sepsis syndromes in the emergency department (ED), using data from electronic medical records. Transfer learning, a new subfield of machine learning, allows generalizability of an algorithm across clinical sites. We aim to validate the Artificial Intelligence Sepsis Expert for the prediction of delayed septic shock in a cohort of patients treated in the ED and demonstrate the feasibility of transfer learning to improve external validity at a second site. This was an observational cohort study using data from greater than 180,000 patients from 2 academic medical centers between 2014 and 2019, using multiple definitions of sepsis. The Artificial Intelligence Sepsis Expert algorithm was trained with 40 input variables at the development site to predict delayed septic shock (occurring greater than 4 hours after ED triage) at various prediction windows. We then validated the algorithm at a second site, using transfer learning to demonstrate generalizability of the algorithm. We identified 9,354 patients with severe sepsis, of whom 723 developed septic shock at least 4 hours after triage. The Artificial Intelligence Sepsis Expert algorithm demonstrated excellent area under the receiver operating characteristic curve (>0.8) at 8 and 12 hours for the prediction of delayed septic shock. Transfer learning significantly improved the test characteristics of the Artificial Intelligence Sepsis Expert algorithm and yielded comparable performance at the validation site. The Artificial Intelligence Sepsis Expert algorithm accurately predicted the development of delayed septic shock. The use of transfer learning allowed significantly improved external validity and generalizability at a second site. Future prospective studies are indicated to evaluate the clinical utility of this model.
Identifiants
pubmed: 33455840
pii: S0196-0644(20)31386-X
doi: 10.1016/j.annemergmed.2020.11.007
pmc: PMC8554871
mid: NIHMS1748754
pii:
doi:
Types de publication
Journal Article
Multicenter Study
Observational Study
Research Support, N.I.H., Extramural
Research Support, Non-U.S. Gov't
Research Support, U.S. Gov't, Non-P.H.S.
Validation Study
Langues
eng
Sous-ensembles de citation
IM
Pagination
395-406Subventions
Organisme : NIEHS NIH HHS
ID : K01 ES025445
Pays : United States
Organisme : NCATS NIH HHS
ID : KL2 TR002381
Pays : United States
Organisme : NCATS NIH HHS
ID : UL1 TR002378
Pays : United States
Commentaires et corrections
Type : UpdateOf
Informations de copyright
Copyright © 2020 American College of Emergency Physicians. Published by Elsevier Inc. All rights reserved.
Références
JAMA. 2016 Dec 13;316(22):2402-2410
pubmed: 27898976
Resuscitation. 2013 Apr;84(4):465-70
pubmed: 23295778
JAMA. 2016 Feb 23;315(8):801-10
pubmed: 26903338
J Am Med Inform Assoc. 2014 Jul-Aug;21(4):699-706
pubmed: 24481703
Int J Emerg Med. 2016 Dec;9(1):10
pubmed: 26908009
Acad Emerg Med. 2013 May;20(5):433-40
pubmed: 23672356
Nat Med. 2019 Jan;25(1):37-43
pubmed: 30617331
Crit Care Explor. 2019 Sep 13;1(9):e0046
pubmed: 32166288
Acad Emerg Med. 2010 Apr;17(4):383-90
pubmed: 20370777
Acad Emerg Med. 2016 Mar;23(3):269-78
pubmed: 26679719
Science. 2006 Jul 28;313(5786):504-7
pubmed: 16873662
NPJ Digit Med. 2018 May 8;1:18
pubmed: 31304302
Crit Care Med. 2007 Jun;35(6):1477-83
pubmed: 17440421
Circulation. 2015 Nov 17;132(20):1920-30
pubmed: 26572668
JAMA. 2020 Jan 28;323(4):305-306
pubmed: 31904799
Ann Emerg Med. 2019 Apr;73(4):334-344
pubmed: 30661855
Nat Med. 2020 Sep;26(9):1351-1363
pubmed: 32908284
Crit Care Med. 2003 Mar;31(3):670-5
pubmed: 12626967
Am J Respir Crit Care Med. 2017 Apr 1;195(7):906-911
pubmed: 27649072
Ann R Coll Surg Engl. 2006 Oct;88(6):571-5
pubmed: 17059720
J Am Med Inform Assoc. 2018 Oct 1;25(10):1419-1428
pubmed: 29893864
Shock. 2016 May;45(5):518-24
pubmed: 26717110
Resuscitation. 2010 Aug;81(8):932-7
pubmed: 20637974
Crit Care Med. 2015 May;43(5):983-8
pubmed: 25668750
JAMA. 2017 Oct 3;318(13):1241-1249
pubmed: 28903154
J Am Med Inform Assoc. 2017 Jan;24(1):198-208
pubmed: 27189013
JMIR Med Inform. 2016 Sep 30;4(3):e28
pubmed: 27694098
Shock. 2007 Dec;28(6):636-643
pubmed: 18092378
Crit Care Med. 2018 Apr;46(4):547-553
pubmed: 29286945
Biomed Inform Insights. 2017 Jun 12;9:1178222617712994
pubmed: 28638239
Crit Care Med. 2020 Feb;48(2):210-217
pubmed: 31939789
Infect Control Hosp Epidemiol. 2018 Aug;39(8):994-996
pubmed: 29932042
Intensive Care Med. 2020 Mar;46(3):383-400
pubmed: 31965266
BMJ Open. 2018 Jan 26;8(1):e017833
pubmed: 29374661
Ann Emerg Med. 2014 Jul;64(1):48-54
pubmed: 24680548
J Crit Care. 2011 Dec;26(6):620-5
pubmed: 21596517
Balkan Med J. 2018 Jan 20;35(1):8-17
pubmed: 28903886
Cell. 2018 Feb 22;172(5):1122-1131.e9
pubmed: 29474911
J Intensive Care. 2017 Jul 12;5:43
pubmed: 28717513
Crit Care Med. 2017 Mar;45(3):486-552
pubmed: 28098591
BMJ Open Respir Res. 2017 Nov 9;4(1):e000234
pubmed: 29435343
J Hosp Med. 2016 Nov;11(11):757-762
pubmed: 27352032