Enter the sample sizes N(D+) with disease and N(D-) without disease, and the measured AUROC. This calculator returns the lower bound LB and upper bound UB of the confidence interval.
Instructions: Enter parameters in the green cells. Answers will appear in the blue box below.
AUROC
%
Expected area under the ROC curve
N(D+)
Sample size with disease
N(D-)
Sample size without disease
CL
%
Confidence level
The standard normal deviate for α = Zα =
Q1 = AUROC / (2-AUROC) =
Q2 = 2(AUROC)2 / (1+AUROC) =
Standard Error = SE = √( (AUROC * (1-AUROC)) + ((N(D+)-1) * (Q1-AUROC2) ) + ((N(D-)-1) * (Q2-AUROC2)) / (N(D+) * N(D-)) ) =
Lower bound = AUROC - (Zα * SE) =
Upper bound = AUROC + (Zα * SE) =
N =
Lower bound =
Upper bound =
Reference:
Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology. 1982;143(1):29-36.