Table 2.

Results from test data for top 3 models predicting short- and long-term mortality.

	AUC	Sensitivity	Specificity	PPV	NPV	Number of variables
3-day Mortality
Naive bayes	0.91 [0.91—0.91]	0.92 [0.91—0.92]	0.78 [0.78—0.79]	0.03 [0.3–0.3]	0.99 [0.99—0.99]	15
Linear discriminant analysis	0.93 [0.93—0.93]	0.89 [0.86—0.89]	0.83 [0.82—0.83]	0.04 [0.4–0.5]	0.99 [0.99—0.99]	15
Logistic regression	0.93 [0.93—0.93]	0.85 [0.83—0.86]	0.85 [0.84—0.89]	0.04 [0.4–0.4]	0.99 [0.99—0.99]	15
10-day mortality
Linear discriminant analysis	0.91 [0.90—0.91]	0.90 [0.87—0.93]	0.78 [0.78—0.79]	0.1 [0.09—0.11]	0.99 [0.99—0.99]	10
Logistic regression	0.91 [0.89—0.93]	0.90 [0.87—0.93]	0.79 [0.79—0.79]	0.1 [0.09—0.11]	0.99 [0.99—0.99]	10
Quadratic discriminant analysis	0.90 [0.90 –0.90]	0.91 [0.87—0.93]	0.77 [0.76—0.77]	0.1 [0.08—0.10]	0.99 [0.99—0.99]	10
30-day Mortality
Linear discriminant analysis	0.90 [090—0.90]	0.90 [0.87—0.92]	0.78 [0.77- 0.79]	0.19 [0.18–0.21]	0.99 [0.99—0.99]	10
Quadratic discriminant analysis	0.91 [0.89—0.91]	0.89 [0.86—0.91]	0.76 [0.75—077]	0.18 [0.17—0.19]	0.99 [0.99—0.99]	10
Gradient boosting classifier	0.92 [0.92—0.92]	0.86 [0.84—0.89]	0.82 [0.82—0.83]	0.22 [0.21—0.24]	0.99 [0.99—0.99]	10
365-day mortality
Gradient boosting classifier	0.88 [0.88—0.89]	0.82 [0.81—0.83]	0.77 [0.76—0.77]	0.44 [0.43—0.46]	0.96 [0.95—0.99]	10
Light gradient boosting machine	0.89 [0.89—0.89]	0.80 [0.80—0.81]	0.81 [0.80—0.82]	0.46 [0.44—0.49]	0.95 [0.95—0.98]	15
Quadratic discriminant analysis	0.87 [0.87—0.89]	0.85 [0.84—0.89]	0.74 [0.73—0.75]	0.40 [0.40—0.41]	0.96 [0.95—0.99]	15

AUC, mean area under receiver operating curve. The numbers are presented as mean with 95%-confidence; PPV, positive predictive value; NPV, negative predictive value.