Table II.

Evaluation indicators for the Support Vector Machine (SVM) classifier in the training set and the validation sets.

				ROC
Datasets	C-index	Brier score	Log rank P-value	AUC	Sensitivity	Specificity	PPV	NPV
Training set (TCGA, n=390)	0.942	0.021	4.72×10−12a	0.905	0.847	0.946	0.914	0.829
Validation set 1 (GSE25204, n=85)	0.899	0.066	7.98×10−7a	0.941	0.783	0.903	0.75	0.918
Validation set 2 (GSE27290, n=58)	0.842	0.063	6.39×10−6a	0.902	0.875	0.905	0.778	0.95

ROC, receiver operating characteristic; AUC, area under the receiver operating characteristic curve; PPV, positive prediction value; NPV, negative prediction value.

^aSignificant P-value.