Deep Learning for Improved Precision and Reproducibility of Left Ventricular Strain in Echocardiography: A Test-Retest Study

Ivar M Salte; Andreas Østvik; Sindre H Olaisen; Sigve Karlsen; Thomas Dahlslett; Erik Smistad; Torfinn K Eriksen-Volnes; Harald Brunvand; Kristina H Haugaa; Thor Edvardsen; Håvard Dalen; Lasse Lovstakken; Bjørnar Grenne

doi:10.1016/j.echo.2023.02.017

Deep Learning for Improved Precision and Reproducibility of Left Ventricular Strain in Echocardiography: A Test-Retest Study

J Am Soc Echocardiogr. 2023 Jul;36(7):788-799. doi: 10.1016/j.echo.2023.02.017. Epub 2023 Mar 16.

Authors

Affiliations

¹ Department of Medicine, Hospital of Southern Norway, Kristiansand, Norway; Faculty of Medicine, University of Oslo, Oslo, Norway; ProCardio Center for Innovation, Department of Cardiology, Oslo University Hospital, Rikshospitalet, Oslo, Norway.
² Centre for Innovative Ultrasound Solutions and Department of Circulation and Medical Imaging, Norwegian University of Science and Technology, Trondheim, Norway; Medical Image Analysis, Health Research, SINTEF Digital, Trondheim, Norway.
³ Centre for Innovative Ultrasound Solutions and Department of Circulation and Medical Imaging, Norwegian University of Science and Technology, Trondheim, Norway.
⁴ Faculty of Medicine, University of Oslo, Oslo, Norway; Department of Medicine, Hospital of Southern Norway, Arendal, Norway.
⁵ Department of Medicine, Hospital of Southern Norway, Arendal, Norway.
⁶ Centre for Innovative Ultrasound Solutions and Department of Circulation and Medical Imaging, Norwegian University of Science and Technology, Trondheim, Norway; Clinic of Cardiology, St. Olavs University Hospital, Trondheim, Norway.
⁷ Faculty of Medicine, University of Oslo, Oslo, Norway; ProCardio Center for Innovation, Department of Cardiology, Oslo University Hospital, Rikshospitalet, Oslo, Norway; Faculty of Medicine, Karolinska Institutet and Cardiovascular Division, Karolinska University Hospital, Stockholm, Sweden.
⁸ Faculty of Medicine, University of Oslo, Oslo, Norway; ProCardio Center for Innovation, Department of Cardiology, Oslo University Hospital, Rikshospitalet, Oslo, Norway.
⁹ Centre for Innovative Ultrasound Solutions and Department of Circulation and Medical Imaging, Norwegian University of Science and Technology, Trondheim, Norway; Clinic of Cardiology, St. Olavs University Hospital, Trondheim, Norway; Levanger Hospital, Nord-Trøndelag Hospital Trust, Levanger, Norway.
¹⁰ Centre for Innovative Ultrasound Solutions and Department of Circulation and Medical Imaging, Norwegian University of Science and Technology, Trondheim, Norway; Clinic of Cardiology, St. Olavs University Hospital, Trondheim, Norway. Electronic address: bjornar.grenne@ntnu.no.

PMID: 36933849
DOI: 10.1016/j.echo.2023.02.017

Abstract

Aims: Assessment of left ventricular (LV) function by echocardiography is hampered by modest test-retest reproducibility. A novel artificial intelligence (AI) method based on deep learning provides fully automated measurements of LV global longitudinal strain (GLS) and may improve the clinical utility of echocardiography by reducing user-related variability. The aim of this study was to assess within-patient test-retest reproducibility of LV GLS measured by the novel AI method in repeated echocardiograms recorded by different echocardiographers and to compare the results to manual measurements.

Methods: Two test-retest data sets (n = 40 and n = 32) were obtained at separate centers. Repeated recordings were acquired in immediate succession by 2 different echocardiographers at each center. For each data set, 4 readers measured GLS in both recordings using a semiautomatic method to construct test-retest interreader and intrareader scenarios. Agreement, mean absolute difference, and minimal detectable change (MDC) were compared to analyses by AI. In a subset of 10 patients, beat-to-beat variability in 3 cardiac cycles was assessed by 2 readers and AI.

Results: Test-retest variability was lower with AI compared with interreader scenarios (data set I: MDC = 3.7 vs 5.5, mean absolute difference = 1.4 vs 2.1, respectively; data set II: MDC = 3.9 vs 5.2, mean absolute difference = 1.6 vs 1.9, respectively; all P < .05). There was bias in GLS measurements in 13 of 24 test-retest interreader scenarios (largest bias, 3.2 strain units). In contrast, there was no bias in measurements by AI. Beat-to-beat MDCs were 1,5, 2.1, and 2.3 for AI and the 2 readers, respectively. Processing time for analyses of GLS by the AI method was 7.9 ± 2.8 seconds.

Conclusion: A fast AI method for automated measurements of LV GLS reduced test-retest variability and removed bias between readers in both test-retest data sets. By improving the precision and reproducibility, AI may increase the clinical utility of echocardiography.

Keywords: Artificial intelligence; Echocardiography; Left ventricular function; Reproducibility; Strain.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Artificial Intelligence
Deep Learning*
Echocardiography / methods
Humans
Reproducibility of Results
Stroke Volume
Ventricular Dysfunction, Left* / diagnostic imaging
Ventricular Function, Left