Table 6.

Performance of the five algorithms on external testing dataset in predicting the occurrence of COVID-19 ARDS.

Models	AUC	Accuracy	Sensitivity	Specificity	F-measure (no-ARDS/ARDS)	Balance accuracy
Decision tree	0.99	0.97	1.00	0.96	0.98/0.93	0.98
Logistic regression	0.98	0.93	0.93	0.93	0.95/0.84	0.87
Random forest	0.97	0.92	0.79	0.95	0.95/0.79	0.83
Support vector machine	0.95	0.83	0.14	1.00	0.90/0.25	0.57
Deep neural networks	0.95	0.90	0.71	0.95	0.94/0.74	0.93