Cardiovascular magnetic resonance (CMR) is the gold standard to measure left ventricular ejection fraction (LVEF), and novel artificial intelligence-based automatic analyses have been proposed for less user interaction and time saving. However, whether automatic LVEF delivers similar information for risk stratification remains unknown.

To investigate the prognostic value for all-cause mortality of LVEF measured by stress CMR using a fully automatic machine learning algorithm without human correction.

Between 2016 and 2018, all consecutive patients referred for vasodilator stress CMR were included and followed for the occurrence of all-cause death. A fully automatic machine learning algorithm was trained on 3700 scans and validated on 1719 unseen CMR studies to identify end-diastolic and end-systolic phases and segment LV volumes from short-axis cine images. The algorithm combines multiple deep learning networks for detection and segmentation with active contours approach. Manual and automatic LVEF were compared with Pearson correlation and Bland-Altman analysis. Cox regressions were performed to determine the prognostic value of automatic LVEF.

Among 9883 patients included to this study, automatic LVEF was successfully computed in 9848 (99.6%) patients (66.7% male, mean age 66.4±12.2 years). The agreement between manual and automatic LVEF was good (bias=−0.01%, 95% limits of agreement, −6.7% to 6.7%; Pearson's correlation r=0.94). A total of 512 (5.2%) deaths were observed during a median (IQR) follow-up period of 4.5 (3.7–5.2) years. Both manual and automatic volumetric assessments showed similar impact on outcome in univariate analyses (manual LVEF: hazard ratio [HR], 0.96 [99.9%CI 0.95–0.97]; P<0.001; manual LVEF: HR, 0.95 [99.9%CI, 0.94–0.96]; P<0.001) and multivariable analyses (manual LVEF: HR, 0.96 [99.9%CI, 0.95–0.97]; P<0.001; automatic LVEF: HR, 0.96 [99.9% CI, 0.95–0.97]; P<0.001). Automatic LVEF showed an incremental prognostic value to predict all-cause mortality compared to a multivariable model including traditional risk factors, the presence of inducible ischemia and LGE (C-statistic improvement: 0.02; NRI=0.223; IDI=0.211; all P<0.001) (Fig. 1).

Automatic LVEF is equally predictive of all-cause mortality compared to manual LVEF and has an incremental prognostic value compared to traditional risk factors, and other stress CMR parameters.