Prediction of response after cardiac resynchronization therapy with machine learning

•We developed and compared nine predictive models to predict CRT response using ML approaches.•All ML models outperformed the latest guidelines, and the ridge regression model showed the highest predictive power.•A website tool was established to facilitate clinicians make predictions of CRT response around implantation.•LBBB, LVESD, and history of PCI were identified as the most crucial predictors of CRT response in the ML models.