Key result
AI assessment of LVEF via left ventricular opacification shows >0.95 correlation with manual measurement.
Why the study?
Manual delineation of the endocardium on LVO has observer variability, and AI has the potential to improve reproducibility when assessing LVEF.
Does an AI model improve the reproducibility and accuracy of LVEF assessment based on left ventricular opacification compared to manual measurement?
Population
1305 echocardiograms of 797 patients plus 50 prospective internal and 42 retrospective external validation patients
Comparison
AI-based assessment of LVEF vs manual sonographer measurements
Design
Retrospective model development with prospective internal and retrospective external validation cohorts
Authors
Loading...
May reduce observer variability in LVEF assessment; leaves open prospective validation before clinical adoption.
Observational (n=889)
Yes
Does an AI model improve the reproducibility and accuracy of LVEF assessment based on left ventricular opacification compared to manual measurement?
Effect estimate: ICC > 0.95
An AI model for assessing LVEF using left ventricular opacification demonstrated high reliability and low error compared to manual measurements, potentially reducing observer variability.
Zhu et al. (2024) conducted an observational in Patients undergoing echocardiography with left ventricular opacification (n=889). Artificial intelligence (AI) model vs. Manual measurements by sonographers was evaluated on Differences between LV function determined by AI and sonographers (median absolute error, spearman correlation, and intraclass correlation coefficient) (ICC > 0.95). An artificial intelligence model for assessing left ventricular ejection fraction via left ventricular opacification showed excellent reliability versus manual measurement (MAE 2.5-2.7%, ICC > 0.95).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: