Comparison of the AUCs among machine learning models for WCH/WUCH prediction by definition A. RF yielded the greatest AUCs. The grey dashed line represents baseline AP, which is defined by the total number of events over the total number of data in the data set. ‘Times’ refers to the AP value of each individual model divided by the baseline AP. (A) ROC curves and AUROCs for internal validation. (B) ROC curves and AUROCs for external validation. (C) PR curves and APs for internal validation. (D) PR curves and APs for external validation. ANN, artificial neural network; AP, average precision; AUC, area under the curve; AUROC, area under the receiver operating characteristic curve; CI, confidence interval; LR, logistic regression; PR, precision-recall; RF, random forest; WCH/WUCH, white-coat hypertension/white-coat uncontrolled hypertension; XGboost, eXtreme Gradient Boosting.