Arrhythmias & electrophysiology
PVCuRe: A machine-learning–based tool to predict left ventricular systolic function recovery in patients undergoing ablation for premature ventricular contraction
Authors' conclusion
This ML-based tool, built on widely available variables, accurately estimates the probability of LVEF recovery after PVC ablation and may support clinical decision making and patient counseling.
Study context
- Population
- Patients with left ventricular ejection fraction (LVEF) <50% who underwent successful PVC ablation at international referral centerspatients with LVEF <50% who underwent successful PVC ablation at 3 international referral centers, external validation in an independent cohort from 3 additional centers
- Participants
- 256 patients
- Predictors
- Routinely available clinical, echocardiographic, and electrocardiographic variables; predictors selected using the Boruta algorithm; 5 machine-learning (ML) models trained (best: Random Forest)
- Predicted outcome
- LVEF recovery after PVC ablation
Methodological detailsdevelopment and external validation of a machine-learning model, 10-fold cross-validation
- Development and external validation of a machine-learning model
- 10-fold cross-validation
Model performance
As reported in the abstract
Area under the curve
Random Forest model, internal test set (95% confidence interval = 0.79-0.98)
0.88
95% confidence interval = 0.72-0.95External validation
0.83
- Calibration, Hosmer-Lemeshow
- P = .562
Further findings
Key predictors
- Baseline PVC burden, QRS duration in sinus rhythm, and preprocedural LVEF
Original publication
PVCuRe: A machine-learning–based tool to predict left ventricular systolic function recovery in patients undergoing ablation for premature ventricular contraction. Heart Rhythm. Published 4 September 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →