Table 5.

Performance summary of the supervised machine learning models with the highest Youden's index.

Models with the best Youden index	VRE (503/3860 = 13.03%)			Models with the best Youden index	MRSA (332/4446 = 7.47%)
	Log. Reg.	XGBoost	Rand. Forest		Log. Reg.	XGBoost	Rand. Forest
Training AUC	0.76	0.77	0.77	Training AUC	0.65	0.66	0.66
Testing AUC	0.80	0.77	0.77	Testing AUC	0.66	0.66	0.70
Testing sensitivity	0.80	0.73	0.75	Testing sensitivity	0.70	0.67	0.76
Testing specificity	0.66	0.65	0.66	Testing specificity	0.55	0.60	0.59
Youden index	0.46	0.39	0.41	Youden index	0.24	0.27	0.34
Threshold opt. bound	0.15	0.15	0.20	Threshold opt. bound	0.20	0.10	0.30
Models with the best Youden index	CRE (53/3661 = 1.45%)			Models with the best Youden index	MDRO (792/4503 = 17.59%)
	Log. Reg.	XGBoost	Rand. Forest		Log. Reg.	XGBoost	Rand. Forest	Dec. Tree
Training AUC	0.70	0.76	0.79	Training AUC	0.72	0.86	0.87	0.76
Testing AUC	0.78	0.72	0.72	Testing AUC	0.70	0.87	0.89	0.81
Testing sensitivity	0.73	0.73	0.64	Testing sensitivity	0.56	0.75	0.82	0.89
Testing specificity	0.73	0.77	0.79	Testing specificity	0.75	0.82	0.83	0.65
Youden index	0.45	0.50	0.42	Youden index	0.30	0.57	0.65	0.54
Threshold opt. bound	0.025	0.015	0.05	Threshold opt. bound	0.50	0.30	0.40	0.30