Fig. 1From: Comparison of discrimination and calibration performance of ECG-based machine learning models for prediction of new-onset atrial fibrillationAÂ AUC values for the varying sample sizes (original event fraction in the training set). Error bars represent the 95% CI around the mean. BÂ AUC values for the varying sample sizes (perfectly balanced training set). CÂ ICI values for the varying sample sizes (original event fraction in the training set). DÂ ICI values for the varying sample sizes (perfectly balanced training set)Back to article page