Skip to main content
Fig. 2 | BMC Medical Informatics and Decision Making

Fig. 2

From: A novel estimator for the two-way partial AUC

Fig. 2

Graphical representation of the two-way partial AUC. In both panels, Area A represents the tpAUC. Panel a shows the ROC curve defined in terms of FPR and TPR (sensitivity) where the sum of Areas A and B represents the partial AUC focusing on the FPR, and \(u_0 = c_{fpr} = S_G\left(S_F^{-1}(b_{se})\right)\) corresponds to the FPR value at the sensitivity threshold of \(b_{se}\) and \(u_1 = b_{fpr} = 1 - b_{sp}\). Panel b shows the ROC curve defined in terms of specificity and sensitivity where the sum of Areas A and B represents the partial AUC focusing on specificity, and \(c_{sp} = 1 - S_G\left(S_F^{-1}(b_{se})\right)\) corresponds to the specificity value at the sensitivity threshold of \(b_{se}\)

Back to article page