Table 2.

Comparison of evaluation and calibration among the different models.

Model	Error Rate of Test Data Set	Accuracy	Precision	MCC	F1 Score	AUROC in the Test Set	Brier Score
Random forest model	21.4%	0.79	0.72	0.45	0.56	0.83 (0.79–0.87)	0.15
Decision tree	26.7%	0.73	0.59	0.30	0.44	0.71 (0.66–0.77)	0.19
XGBoost	25.0%	0.75	0.60	0.36	0.52	0.81 (0.76–0.85)	0.18
ANN	25.0%	0.75	0.67	0.33	0.42	0.79 (0.74–0.84)	0.19
Multivariable logistic regression	22.9%	0.77	0.67	0.41	0.54	0.81 (0.79–0.83)	0.16
SOFA score	25.5%	0.74	0.67	0.30	0.39	0.74 (0.68–0.80)	0.17
SAPS II score	23.2%	0.77	0.71	0.39	0.49	0.77 (0.71–0.82)	0.17
Charlson score	28.4%	0.72	0.73	0.16	0.13	0.69 (0.63–0.74)	0.19

MCC: worst value −1 and best value +1. F1 score, accuracy, and precision: worst value 0 and best value 1. The Brier score is a combined measure of discrimination and calibration that ranges between 0 and 1, where the best score is 0 and the worst is 1. ANN, artificial neural network; MCC, Matthews correlation coefficient; AUROC, area under the receiver operating characteristic curve; SOFA, Sequential Organ Failure Assessment; SAPS II, Simplified Acute Physiology Score.