In Journal of the American Society of Echocardiography : official publication of the American Society of Echocardiography
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 further, to compare the results to manual measurements.
METHODS AND RESULTS : Two test-retest datasets (n=40 and n=32) were obtained at separate centers. Repeated recordings were acquired in immediate succession by two different echocardiographers at each center. For each dataset, four readers measured GLS in both recordings using a semi-automatic method to construct test-retest inter-reader and intra-reader 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 three cardiac cycles was assessed by two readers and AI. Test-retest variability was lower with AI compared to inter-reader scenarios (dataset I: MDC 3.7 vs 5.5, mean absolute difference 1.4 vs 2.1; dataset II: MDC 3.9 vs 5.2, mean absolute difference 1.6 vs 1.9, all p<0.05). There was bias in GLS measurements in 13 of 24 test-retest inter-reader 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 two 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 datasets. By improving the precision and reproducibility, AI may increase the clinical utility of echocardiography.
Salte Ivar M, Østvik Andreas, Olaisen Sindre H, Karlsen Sigve, Dahlslett Thomas, Smistad Erik, Eriksen-Volnes Torfinn Kirknes, Brunvand Harald, Haugaa Kristina H, Edvardsen Thor, Dalen Håvard, Lovstakken Lasse, Grenne Bjørnar
2023-Mar-16
Artificial intelligence, Echocardiography, Left ventricular function, Machine learning, Repeatability, Reproducibility, Strain
