An Automated View Classification Model for Pediatric Echocardiography Using Artificial Intelligence.
Artificial intelligence
Convolutional neural network
Echocardiography
Pediatric cardiology
View classification
Journal
Journal of the American Society of Echocardiography : official publication of the American Society of Echocardiography
ISSN: 1097-6795
Titre abrégé: J Am Soc Echocardiogr
Pays: United States
ID NLM: 8801388
Informations de publication
Date de publication:
12 2022
12 2022
Historique:
received:
13
03
2022
revised:
10
07
2022
accepted:
12
08
2022
pubmed:
2
9
2022
medline:
7
12
2022
entrez:
1
9
2022
Statut:
ppublish
Résumé
View classification is a key step toward building a fully automated system for interpretation of echocardiograms. However, compared with adult echocardiograms, creating a view classification model for pediatric echocardiograms poses additional challenges, such as greater variation in anatomy, structure size, and views. The aim of this study was to develop a computer vision model to autonomously perform view classification on pediatric echocardiographic images. Using a training set of 12,067 echocardiographic images from patients aged 0 to 19 years, a convolutional neural network model was trained to identify 27 preselected standard pediatric echocardiographic views which included anatomic sweeps, color Doppler, and Doppler tracings. A validation set of 6,197 images was used for parameter tuning and model selection. A test set of 9,684 images from 100 different patients was then used to evaluate model accuracy. The model was also evaluated on a per study basis using a second test set consisting of 524 echocardiograms from children with leukemia to identify six preselected views pertinent to cardiac dysfunction surveillance. The model identified the 27 preselected views with 90.3% accuracy. Accuracy was similar across age groups (89.3% for 0-4 years, 90.8% for 4-9 years, 90.0% for 9-14 years, and 91.2% for 14-19 years; P = .12). Examining the view subtypes, accuracy was 78.3% for the cine one location, 90.5% for sweeps with color Doppler, 82.2% for sweeps without color Doppler, and 91.1% for Doppler tracings. Among the leukemia cohort, the model identified the six preselected views on a per study basis with a positive predictive value of 98.7% to 99.2% and sensitivity of 76.9% to 94.8%. A convolutional neural network model was constructed for view classification of pediatric echocardiograms that was accurate across the spectrum of ages and view types. This work lays the foundation for automated quantitative analysis and diagnostic support to promote efficient, accurate, and scalable analysis of pediatric echocardiograms.
Sections du résumé
BACKGROUND
View classification is a key step toward building a fully automated system for interpretation of echocardiograms. However, compared with adult echocardiograms, creating a view classification model for pediatric echocardiograms poses additional challenges, such as greater variation in anatomy, structure size, and views. The aim of this study was to develop a computer vision model to autonomously perform view classification on pediatric echocardiographic images.
METHODS
Using a training set of 12,067 echocardiographic images from patients aged 0 to 19 years, a convolutional neural network model was trained to identify 27 preselected standard pediatric echocardiographic views which included anatomic sweeps, color Doppler, and Doppler tracings. A validation set of 6,197 images was used for parameter tuning and model selection. A test set of 9,684 images from 100 different patients was then used to evaluate model accuracy. The model was also evaluated on a per study basis using a second test set consisting of 524 echocardiograms from children with leukemia to identify six preselected views pertinent to cardiac dysfunction surveillance.
RESULTS
The model identified the 27 preselected views with 90.3% accuracy. Accuracy was similar across age groups (89.3% for 0-4 years, 90.8% for 4-9 years, 90.0% for 9-14 years, and 91.2% for 14-19 years; P = .12). Examining the view subtypes, accuracy was 78.3% for the cine one location, 90.5% for sweeps with color Doppler, 82.2% for sweeps without color Doppler, and 91.1% for Doppler tracings. Among the leukemia cohort, the model identified the six preselected views on a per study basis with a positive predictive value of 98.7% to 99.2% and sensitivity of 76.9% to 94.8%.
CONCLUSIONS
A convolutional neural network model was constructed for view classification of pediatric echocardiograms that was accurate across the spectrum of ages and view types. This work lays the foundation for automated quantitative analysis and diagnostic support to promote efficient, accurate, and scalable analysis of pediatric echocardiograms.
Identifiants
pubmed: 36049595
pii: S0894-7317(22)00428-X
doi: 10.1016/j.echo.2022.08.009
pmc: PMC9990955
mid: NIHMS1839606
pii:
doi:
Types de publication
Journal Article
Research Support, N.I.H., Extramural
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
1238-1246Subventions
Organisme : NHLBI NIH HHS
ID : R01 HL140731
Pays : United States
Organisme : NHLBI NIH HHS
ID : T32 HL007572
Pays : United States
Informations de copyright
Published by Elsevier Inc.
Références
J Am Med Inform Assoc. 2021 Aug 13;28(9):1834-1842
pubmed: 34279636
J Med Imaging (Bellingham). 2017 Jan;4(1):014502
pubmed: 28149925
Int J Cardiol. 2020 Oct 1;316:272-278
pubmed: 32507394
Circulation. 2007 Oct 23;116(17):1876-8
pubmed: 17965402
Am J Cardiol. 2009 Aug 1;104(3):419-28
pubmed: 19616678
J Magn Reson Imaging. 2019 Apr;49(4):939-954
pubmed: 30575178
Circulation. 2018 Oct 16;138(16):1623-1635
pubmed: 30354459
Eur Heart J Cardiovasc Imaging. 2019 Aug 1;20(8):925-931
pubmed: 30629127
J Am Soc Echocardiogr. 2006 Dec;19(12):1413-30
pubmed: 17138024
Acad Radiol. 2019 Jun;26(6):735-743
pubmed: 30076083
J Am Soc Echocardiogr. 2021 Apr;34(4):443-445
pubmed: 33276079
Nat Med. 2021 May;27(5):882-891
pubmed: 33990806
J Am Coll Cardiol. 2015 Sep 29;66(13):1456-66
pubmed: 26403342
Med Image Anal. 2021 Apr;69:101942
pubmed: 33418465
Nat Commun. 2021 May 11;12(1):2726
pubmed: 33976142
JAMA Cardiol. 2021 Jun 1;6(6):624-632
pubmed: 33599681
IEEE J Biomed Health Inform. 2020 Apr;24(4):994-1003
pubmed: 31831455
NPJ Digit Med. 2018;1:
pubmed: 30828647
NPJ Digit Med. 2018 Oct 18;1:59
pubmed: 31304338
Curr Opin Pediatr. 2015 Oct;27(5):587-96
pubmed: 26262579