COVID-19 ICU mortality prediction: a machine learning approach using SuperLearner algorithm

Table 2 Training and test validation performances

		Sens	Spec	Balanced accuracy	F	ROC
Model 1	Training performance	0.74	0.74	0.72	0.77	0.73
	External validation performance (Veneto cohort)	0.69	0.68	0.68	0.76	0.65
	External validation performance (Lombardia cohort)	0.70	0.69	0.70	0.78	0.70
Model 2	Training performance	0.78	0.77	0.78	0.79	0.85
Model 2	External validation performance	0.67	0.65	0.66	0.76	0.65
Model 3	Training performance	0.90	0.89	0.90	0.91	0.93
Model 3	External validation performance	0.84	0.83	0.83	0.88	0.81

Sens Sensitivity, Spec Specificity, Balanced Accuracy, F1 statistics, and ROC are reported for each model

ISSN: 2731-3786