Table 5.

The performances of the models in the first scenario.

Model	AUROC	Brier score	Precision	Recall	F1 score
LoR	0.832 (0.797–0.857)	0.146	0.545	0.742	0.628
SVM	0.831 (0.797–0.856)	0.147	0.544	0.742	0.628
RF	0.854 (0.825–0.881)	0.137	0.572	0.763	0.653
NN	0.832 (0.791–0.853)	0.147	0.550	0.746	0.633
Ho-Pham et al.	0.823 (0.781–0.851)	0.149	0.489	0.813	0.610
OSTA	0.654 (0.606–0.702)	0.205	0.722	0.353	0.473