Table 4.

Prediction performances for logistic regression (GLM) and stochastic gradient boosting (GBM). In parentheses 95 percent confidence intervals. Positive class “malignant”.

	AUC (inner) 1	AUC 2	PPV 3	NPV 4	Accuracy 5
GLM	0.66	0.66	0.65	0.61	0.63
	(0.65–0.67)	(0.63–0.70)	(0.63–0.67)	(0.57–0.64)	(0.60–0.65)
GBM	0.67	0.67	0.65	0.59	0.62
	(0.66–0.67)	(0.64–0.69)	(0.63–0.67)	(0.57–0.62)	(0.61–0.64)

¹ Average Area Under the receiver operating Characteristic from the inner cross-validation; ² Average Area Under the receiver operating Characteristic from the outer cross-validation; ³ Average Positive Predictive Value; ⁴ Average Negative Predictive Value; ⁵ Average Accuracy.