Table 2.

Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and F-measure for the machine learning model and the CURB-65 (confusion, urea, respiratory rate, blood pressure, and 65 years of age or older) score, with different cut-offs.

Cut-off	Sensitivity	Specificity	PPV	NPV	F-measure
CURB-65 score >0.5	0.89	0.66	0.05	1.00	0.10
XGBoost score >0.06	0.89	0.75	0.36	0.99	0.43
CURB-65 score >1.5	0.53	0.93	0.14	0.99	0.22
XGBoost score >0.34	0.53	0.97	0.63	0.95	0.58
CURB-65 score >2.5	0.06	1.00	0.40	0.98	0.11
XGBoost score >0.89	0.06	1.00	0.95	0.90	0.12