Skip to main content
. 2014 Mar 1;34(3):204–214. doi: 10.1089/jir.2013.0075

FIG. 5.

FIG. 5.

The use of ROC curves and AUC values in evaluation of the sensitivity and specificity of the diagnostic biomarkers. (A) E3. (B) E1/E3 ratio. (C) CCL22. (D) IL-15. Numbers within the tables/curves are AUC values for any 2 correspondent clinical conditions. The plots in the middle and right columns show representative curves for the lowest and highest AUC in the correspondent group, respectively. An AUC value of 1 is indicative of 100% sensitivity and specificity (a perfect test), and a value of 0.5 is indicative of 50% sensitivity and specificity (an entirely worthless test). The horizontal axes on the curves are presenting 100% minus specificity (the highest specificity of 100% is located at zero). ROC, receiver-operating characteristic curves; AUC, area under the curve.