Automated Echocardiographic Quantification of Left Ventricular Ejection Fraction Without Volume Measurements Using a Machine Learning Algorithm Mimicking a Human Expert.
echocardiography
endocardium
left ventricular function
machine learning
observer variation
Journal
Circulation. Cardiovascular imaging
ISSN: 1942-0080
Titre abrégé: Circ Cardiovasc Imaging
Pays: United States
ID NLM: 101479935
Informations de publication
Date de publication:
09 2019
09 2019
Historique:
entrez:
17
9
2019
pubmed:
17
9
2019
medline:
9
6
2020
Statut:
ppublish
Résumé
Echocardiographic quantification of left ventricular (LV) ejection fraction (EF) relies on either manual or automated identification of endocardial boundaries followed by model-based calculation of end-systolic and end-diastolic LV volumes. Recent developments in artificial intelligence resulted in computer algorithms that allow near automated detection of endocardial boundaries and measurement of LV volumes and function. However, boundary identification is still prone to errors limiting accuracy in certain patients. We hypothesized that a fully automated machine learning algorithm could circumvent border detection and instead would estimate the degree of ventricular contraction, similar to a human expert trained on tens of thousands of images. Machine learning algorithm was developed and trained to automatically estimate LVEF on a database of >50 000 echocardiographic studies, including multiple apical 2- and 4-chamber views (AutoEF, BayLabs). Testing was performed on an independent group of 99 patients, whose automated EF values were compared with reference values obtained by averaging measurements by 3 experts using conventional volume-based technique. Inter-technique agreement was assessed using linear regression and Bland-Altman analysis. Consistency was assessed by mean absolute deviation among automated estimates from different combinations of apical views. Finally, sensitivity and specificity of detecting of EF ≤35% were calculated. These metrics were compared side-by-side against the same reference standard to those obtained from conventional EF measurements by clinical readers. Automated estimation of LVEF was feasible in all 99 patients. AutoEF values showed high consistency (mean absolute deviation =2.9%) and excellent agreement with the reference values: Machine learning algorithm for volume-independent LVEF estimation is highly feasible and similar in accuracy to conventional volume-based measurements, when compared with reference values provided by an expert panel.
Sections du résumé
BACKGROUND
Echocardiographic quantification of left ventricular (LV) ejection fraction (EF) relies on either manual or automated identification of endocardial boundaries followed by model-based calculation of end-systolic and end-diastolic LV volumes. Recent developments in artificial intelligence resulted in computer algorithms that allow near automated detection of endocardial boundaries and measurement of LV volumes and function. However, boundary identification is still prone to errors limiting accuracy in certain patients. We hypothesized that a fully automated machine learning algorithm could circumvent border detection and instead would estimate the degree of ventricular contraction, similar to a human expert trained on tens of thousands of images.
METHODS
Machine learning algorithm was developed and trained to automatically estimate LVEF on a database of >50 000 echocardiographic studies, including multiple apical 2- and 4-chamber views (AutoEF, BayLabs). Testing was performed on an independent group of 99 patients, whose automated EF values were compared with reference values obtained by averaging measurements by 3 experts using conventional volume-based technique. Inter-technique agreement was assessed using linear regression and Bland-Altman analysis. Consistency was assessed by mean absolute deviation among automated estimates from different combinations of apical views. Finally, sensitivity and specificity of detecting of EF ≤35% were calculated. These metrics were compared side-by-side against the same reference standard to those obtained from conventional EF measurements by clinical readers.
RESULTS
Automated estimation of LVEF was feasible in all 99 patients. AutoEF values showed high consistency (mean absolute deviation =2.9%) and excellent agreement with the reference values:
CONCLUSIONS
Machine learning algorithm for volume-independent LVEF estimation is highly feasible and similar in accuracy to conventional volume-based measurements, when compared with reference values provided by an expert panel.
Identifiants
pubmed: 31522550
doi: 10.1161/CIRCIMAGING.119.009303
pmc: PMC7099856
doi:
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
e009303Commentaires et corrections
Type : CommentIn
Références
Am Heart J. 1982 Sep;104(3):603-6
pubmed: 7113901
J Am Soc Echocardiogr. 2013 Nov;26(11):1267-73
pubmed: 23993695
JACC Cardiovasc Imaging. 2016 Jul;9(7):769-782
pubmed: 27318718
Radiol Phys Technol. 2017 Mar;10(1):23-32
pubmed: 28211015
J Cardiol. 2018 Jul;72(1):74-80
pubmed: 29472129
Data (Basel). 2017 Mar;2(1):
pubmed: 28243594
Med Image Anal. 2012 Jul;16(5):933-51
pubmed: 22465077
Eur J Echocardiogr. 2003 Mar;4(1):59-67
pubmed: 12565064
Circulation. 2018 Oct 16;138(16):1623-1635
pubmed: 30354459
Arch Cardiovasc Dis. 2017 Nov;110(11):580-589
pubmed: 28566200
J Cardiovasc Magn Reson. 2008 Jul 07;10:35
pubmed: 18605997
J Am Soc Echocardiogr. 2003 Aug;16(8):824-31
pubmed: 12878991
J Am Soc Echocardiogr. 2015 Jan;28(1):1-39.e14
pubmed: 25559473
Eur Heart J Cardiovasc Imaging. 2018 Jan 1;19(1):47-58
pubmed: 28159984
J Am Soc Echocardiogr. 2017 Nov;30(11):1049-1058
pubmed: 28916243
J Am Soc Echocardiogr. 2017 Sep;30(9):879-885
pubmed: 28688857
Cardiovasc Ultrasound. 2018 Jan 23;16(1):2
pubmed: 29357888
JACC Cardiovasc Imaging. 2008 Jul;1(4):413-23
pubmed: 19356461