Left Ventricular Ejection Fraction Reporting Variability and AI-Assisted Reproducibility: A Multicentre Analysis

Artificial intelligence (AI)-assisted assessment of left ventricular ejection fraction (LVEF) has been increasingly adopted in clinical practice. In this study we aimed to assess the reliability and reproducibility of AI-assisted LVEF assessment in a diverse, real-world, multicentre setting.

We conducted a retrospective multicentre study involving 354 cardiac magnetic resonance examinations. A standardized LVEF reassessment (M-LVEF) was performed using QMass 8.1 (Medis Medical Imaging, Leiden, Netherlands) and systematically compared with original report-derived LVEF values (R-LVEF) generated using vendor-specific AI tools. For interobserver reproducibility assessment, 3 operators independently analyzed 30 randomly selected cases using fully manual and AI-assisted modes, the latter also performed with QMass 8.1 software. Operator A performed 2 measurements in both modes for intraobserver analysis.

In overall and single-centre analyses, the consistency between R-LVEF and M-LVEF was good or excellent (intraclass correlation coefficient ≥ 0.86), but the 95% limits of agreement all exceeded ± 5%. In 44.3% of cases > 5% differences between R-LVEF and M-LVEF were observed, and 17.5% showed > 10% differences. The AI-assisted method not only significantly reduced LVEF assessment time compared with manual analysis, but also showed superior reproducibility. This was evidenced by greater interobserver agreement among 3 operators (intraclass correlation coefficients, 0.968, 0.974, 0.984 for AI vs 0.913, 0.924, 0.971 for manual, respectively) and greater intraobserver reproducibility, all with significantly lower coefficients of variation (5.6%, 8.8%, 9.3% vs 2.5%, 3.8%, 4.4%, respectively; 3.5% vs 1.0% [P < 0.05).

There are individual differences in AI-assisted LVEF assessment, but it has significant advantages in efficiency and reproducibility, making AI a worthwhile tool to facilitate cardiac magnetic resonance quantification.