Summary of early-onset PE prediction previous studies with methods they used.

MImp.: Missing imputation technique, R: Removal technique, EM: Expected maximization, M: Mean Imputation, FFS: Forward Feature Selection, BFS: Backward Feature Selection, CBF: Correlation Based Feature Selection, PCA: Principal Componant Analysis, Imb.: class imbalance method, OS: oversampling, WT: class-weight adjustment, ST: SMOTE, LR: Logistic regression, EN: Elastic Net, DT: Decision Tree, RF: Random Forest, SVM: Support vector machine, ANN: Artificial Neural Network, GB: Gradient Boosting, EL: Ensemble Learning, NB: Naïve Bayes, KNN: K Nearest Neighbor, KS: Kappa statistics, SN: Sensitivity, SP: Specificity, ACC: Accuracy, PR: Precision, GM: G-mean, FM: F-measure, BR: Brier Score.