Table 2.

Performance results for machine learning algorithms using leave-one-patient-out cross validation method.

Classifier	Accuracy (%)	Sensitivity (%)	Specificity (%)	Kappa	MCC	ROC_AUC
Decision tree	68.72	69.16	68.30	0.3744	0.3745	0.75
Discriminant analysis	72.15	84.40	66.33	0.4429	0.476	0.78
Gentle boost	73.52	72.10	75.12	0.4703	0.4713	0.82
k-nearest-neighbors	67.58	69.74	65.84	0.3516	0.3537	0.76
Logistic regression	74.79	79.19	71.55	0.4959	0.5016	0.85
Naive Bayes	63.70	68.29	60.95	0.274	0.283	0.65
Artificial neural network	78.08	85.55	73.21	0.5616	0.5745	0.89
Support vector machine	77.81	85.37	72.91	0.5562	0.5693	0.89