Machine learning-based model for predicting major adverse cardiovascular and cerebrovascular events in patients aged 65 years and older undergoing noncardiac surgery

Table 4 Performance evaluation and comparison of six models

	Logistic regression	Decision Tree	XGBoost	Support Vector Machine	K-nearest Neighbor	Neural Network
AUC	0.891 [0.818–0.963]	0.824 [0.718–0.930]	0.903 [0.839–0.968]	0.901 [0.839–0.963]	0.803 [0.712–0.894]	0.789 [0.677–0.900]
Sensitivity	0.8261	0.7826	0.8261	0.9130	0.8261	0.6087
Specificity	0.8228	0.8608	0.9114	0.7342	0.6329	0.8987
Accuracy	0.824 [0.736–0.892]	0.843 [0.758–0.908]	0.892 [0.815–0.945]	0.775 [0.681–0.851]	0.677 [0.577–0.766]	0.833 [0.747–0.900]
Pos Pred Value	0.5758	0.6207	0.7308	0.5000	0.3958	0.6364
Neg Pred Value	0.9420	0.6207	0.9474	0.9667	0.9259	0.8875
F1 score	0.6786	0.6923	0.7755	0.6462	0.5352	0.622

ISSN: 1471-2318