Table 3.

Performance of the machine learning in the training and testing cohorts.

Model name	Accuracy	AUC	95% CI	Sensitivity	Specificity	PPV	NPV	Precision	Recall	F1 Score	Threshold	Data-cohort
SVM	0.992248	0.999356	0.9978 - 1.0000	1.000000	0.989691	0.969697	1.000000	0.969697	1.000000	0.984615	0.228051	train
SVM	0.727273	0.754821	0.5900 - 0.9196	0.727273	0.727273	0.470588	0.888889	0.470588	0.727273	0.571429	0.302342	test
KNN	0.775194	0.853254	0.7894 - 0.9171	0.812500	0.762887	0.530612	0.925000	0.530612	0.812500	0.641975	0.400000	train
KNN	0.772727	0.825069	0.7002 - 0.9500	0.636364	0.818182	0.538462	0.870968	0.538462	0.636364	0.583333	0.400000	test
RandomForest	0.976744	0.999517	0.9984 - 1.0000	1.000000	0.969072	0.914286	1.000000	0.914286	1.000000	0.955224	0.300000	train
RandomForest	0.613636	0.641873	0.4490 - 0.8348	0.636364	0.606061	0.350000	0.833333	0.350000	0.636364	0.451613	0.300000	test
ExtraTrees	1.000000	1.000000	nan - nan	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	train
ExtraTrees	0.704545	0.782369	0.6301 - 0.9347	0.727273	0.696970	0.444444	0.884615	0.444444	0.727273	0.551724	0.300000	test
XGBoost	0.945736	0.984214	0.9639 - 1.0000	0.937500	0.948454	0.857143	0.978723	0.857143	0.937500	0.895522	0.380025	train
XGBoost	0.704545	0.673554	0.4862 - 0.8609	0.636364	0.727273	0.437500	0.857143	0.437500	0.636364	0.518519	0.293565	test
LightGBM	0.968992	0.975515	0.9502 - 1.0000	0.937500	0.979381	0.937500	0.979381	0.937500	0.937500	0.937500	0.333712	train
LightGBM	0.772727	0.716253	0.5387 - 0.8938	0.545455	0.848485	0.545455	0.848485	0.545455	0.545455	0.545455	0.279661	test
MLP	0.930233	0.973260	0.9488 - 0.9978	0.968750	0.917526	0.794872	0.988889	0.794872	0.968750	0.873239	0.307806	train
MLP	0.818182	0.884298	0.7831 - 0.9855	0.909091	0.787879	0.588235	0.962963	0.588235	0.909091	0.714286	0.214518	test
LR	0.914729	0.975515	0.9521 - 0.9989	0.968750	0.896907	0.756098	0.988636	0.756098	0.968750	0.849315	0.276470	train
LR	0.727273	0.768595	0.6065 - 0.9306	0.727273	0.727273	0.470588	0.888889	0.470588	0.727273	0.571429	0.116137	test

AUC, area under the curve; CI, confidence interval; NPV, negative predictive value; PPV, positive predictive value.