Focus Topic: Artificial Intelligence and Echocardiography
Clinical Investigations
Deep Learning for Improved Precision and Reproducibility of Left Ventricular Strain in Echocardiography: A Test-Retest Study

https://doi.org/10.1016/j.echo.2023.02.017Get rights and content
Under a Creative Commons license
open access

Highlights

  • Deep-learning AI provides efficient automated GLS measurements in echocardiograms.

  • Deep-learning AI produces consistent GLS measurements in repeated echocardiograms.

  • Automated GLS measurements using deep learning improve test-retest reproducibility.

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.

Graphical abstract

Graphical abstract: Study design and main results: Two test-retest data sets were collected at 2 different centers (data set I and data set II). For both data sets, 2 different echocardiographers acquired separate recordings in immediate succession in each patient to create a test-retest pair of images (upper panel). For each patient, these test-retest recordings were analyzed by 4 readers who measured global longitudinal strain (GLS) using a commercially available semiautomatic method. This resulted in 12 interreader and 4 intrareader scenarios. In addition, GLS was measured using an artificial intelligence (AI) method based on deep learning. The test-retest measurement bias and minimal detectable change for all inter- and intrareader scenarios and the AI scenario are presented for both data sets (lower panels). Artificial intelligence–based measurements eliminated test-retest bias between readers and resulted in reduced test-retest variability.

  1. Download : Download high-res image (176KB)
  2. Download : Download full-size image

Keywords

Left ventricular function
Echocardiography
Strain
Reproducibility
Artificial intelligence

Abbreviations

AI
Artificial intelligence
ASE
American Society of Echocardiography
GLS
Global longitudinal strain
ICC
Intraclass correlation coefficient
LV
Left ventricle, ventricular
LVEF
Left ventricular ejection fraction
MDC
Minimal detectable change
ROI
Region of interest

Cited by (0)

Drs. Salte and Østvik contributed equally to this work.

This work was supported by the Centre for Innovative Ultrasound Solutions (a Norwegian Research Council center for research-based innovation, project no. 237887), by the Norwegian Health Association, the Central Norway regional health authority, South-Eastern Norway regional health authority, the national program for clinical therapy research (project no. 2017207), and ProCardio Center for Innovation (no. 309762) from the Norwegian Research council. The HUNT Echocardiography study was funded by the Liaison Committee for Education, Research and Innovation in Central Norway and grants from the Simon Fougner Hartmann Family Fund, Denmark.

Conflicts of Interest: None.