The Area under the ROC Curve and Its Competitors

The area under the receiver operating characteristic (ROC) curve is a popular measure of the power of a (two-disease) diagnostic test, but it is shown here to be an inconsistent criterion: tests of indistinguishable clinical impacts may have different areas. A class of diagnosticity measures (DMs) of proven optimality is proposed instead. Once a regret(-like) measure of diagnostic uncertainty is agreed upon, the associated DM is uniquely defined and, indeed, calculable from the ROC curve configuration. Two scaled variants of the ROC are introduced and used to advantage in the analysis. They may also be helpful to students of medical decision making.

[1]  Comments on the logistic function in ROC analysis: applications to breast cancer detection. , 1982, Methods of information in medicine.

[2]  J. Hanley,et al.  The meaning and use of the area under a receiver operating characteristic (ROC) curve. , 1982, Radiology.

[3]  J. Hanley,et al.  A method of comparing the areas under receiver operating characteristic curves derived from the same cases. , 1983, Radiology.

[4]  B. J. McNeil,et al.  Critical Questions Regarding a New Diagnostic Technology: A Case Study Using Computerized Tomography of the Head , 1983 .

[5]  R M Centor,et al.  A Visicalc Program for Estimating the Area Under a Receiver Operating Characteristic (ROC) Curve , 1985, Medical decision making : an international journal of the Society for Medical Decision Making.

[6]  R M Centor,et al.  An Evaluation of Methods for Estimating the Area Under the Receiver Operating Characteristic (ROC) Curve , 1985, Medical decision making : an international journal of the Society for Medical Decision Making.

[7]  C Brownie,et al.  Comparing indicators of health or nutritional status. , 1986, American journal of epidemiology.

[8]  J. Hilden,et al.  Decision Tables and logic in Decision Analysis , 1986, Medical decision making : an international journal of the Society for Medical Decision Making.

[9]  J Hilden,et al.  Optimistic bias in the assessment of sensitivity and specificity. , 1986, Journal of chronic diseases.

[10]  J. Habbema,et al.  Prognosis in medicine: An analysis of its meaning and rôles , 1987, Theoretical medicine.

[11]  George A. Diamond,et al.  ROC Steady , 1987, Medical decision making : an international journal of the Society for Medical Decision Making.

[12]  L. Irwig,et al.  Evaluating obstetric risk scores by receiver operating characteristic curves. , 1988, American journal of epidemiology.

[13]  W. England,et al.  An Exponential Model Used for optimal Threshold selection on ROC Curues , 1988, Medical decision making : an international journal of the Society for Medical Decision Making.

[14]  E. DeLong,et al.  Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. , 1988, Biometrics.

[15]  P Glasziou,et al.  Test Selection Measures , 1989, Medical decision making : an international journal of the Society for Medical Decision Making.

[16]  W Zucchini,et al.  On the statistical analysis of ROC curves. , 1989, Statistics in medicine.

[17]  J. Kardaun,et al.  Comparative Diagnostic Performance of Three Radiological Procedures for the Detection of Lumbar Disk Herniation , 1990, Methods of Information in Medicine.

[18]  P S Heckerling.  Information content of diagnostic tests in the medical literature. , 1990, Methods of information in medicine.

[19]  Contributions to a rational diagnosis and treatment of lumbar disk , 1990 .