Machine learning for the real-time assessment of left ventricular ejection fraction in critically ill patients: a bedside evaluation by novices and experts in echocardiography.
Artificial intelligence
Echocardiography
Left ventricular ejection fraction
Machine learning
Novice
Point of care ultrasound
Journal
Critical care (London, England)
ISSN: 1466-609X
Titre abrégé: Crit Care
Pays: England
ID NLM: 9801902
Informations de publication
Date de publication:
14 12 2022
14 12 2022
Historique:
received:
28
08
2022
accepted:
07
10
2022
entrez:
14
12
2022
pubmed:
15
12
2022
medline:
17
12
2022
Statut:
epublish
Résumé
Machine learning algorithms have recently been developed to enable the automatic and real-time echocardiographic assessment of left ventricular ejection fraction (LVEF) and have not been evaluated in critically ill patients. Real-time LVEF was prospectively measured in 95 ICU patients with a machine learning algorithm installed on a cart-based ultrasound system. Real-time measurements taken by novices (LVEF LVEF Machine learning-enabled real-time measurements of LVEF were strongly correlated with manual measurements obtained by experts. The accuracy of real-time LVEF measurements was excellent, and the precision was fair. The reproducibility of LVEF measurements was better with the machine learning system. The specificity to detect left ventricular dysfunction was excellent both for experts and for novices, whereas the sensitivity could be improved. NCT05336448. Retrospectively registered on April 19, 2022.
Sections du résumé
BACKGROUND
Machine learning algorithms have recently been developed to enable the automatic and real-time echocardiographic assessment of left ventricular ejection fraction (LVEF) and have not been evaluated in critically ill patients.
METHODS
Real-time LVEF was prospectively measured in 95 ICU patients with a machine learning algorithm installed on a cart-based ultrasound system. Real-time measurements taken by novices (LVEF
RESULTS
LVEF
CONCLUSION
Machine learning-enabled real-time measurements of LVEF were strongly correlated with manual measurements obtained by experts. The accuracy of real-time LVEF measurements was excellent, and the precision was fair. The reproducibility of LVEF measurements was better with the machine learning system. The specificity to detect left ventricular dysfunction was excellent both for experts and for novices, whereas the sensitivity could be improved.
TRIAL REGISTRATION
NCT05336448. Retrospectively registered on April 19, 2022.
Identifiants
pubmed: 36517906
doi: 10.1186/s13054-022-04269-6
pii: 10.1186/s13054-022-04269-6
pmc: PMC9749290
doi:
Banques de données
ClinicalTrials.gov
['NCT05336448']
Types de publication
Clinical Study
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
386Informations de copyright
© 2022. The Author(s).
Références
Echocardiography. 2021 Jun;38(6):982-992
pubmed: 33982820
JAMA Cardiol. 2021 Jun 1;6(6):624-632
pubmed: 33599681
Circ Cardiovasc Imaging. 2021 Jun;14(6):e012293
pubmed: 34126754
J Am Coll Cardiol. 2019 Mar 26;73(11):1317-1335
pubmed: 30898208
Crit Care. 2017 Nov 17;21(1):279
pubmed: 29149863
Circ Cardiovasc Imaging. 2019 Sep;12(9):e009303
pubmed: 31522550
Br J Anaesth. 2022 Nov;129(5):e116-e119
pubmed: 36031414
Int J Cardiovasc Imaging. 2021 Feb;37(2):577-586
pubmed: 33029699
Intensive Care Med. 2019 Jun;45(6):770-788
pubmed: 30911808
Chest. 2020 Nov;158(5):2107-2118
pubmed: 32707179