Hypothesis testing of diagnostic accuracy for multiple readers and multiple tests an anova approach with dependent observations
- 1 January 1995
- journal article
- research article
- Published by Taylor & Francis in Communications in Statistics - Simulation and Computation
- Vol. 24 (2) , 285-308
- https://doi.org/10.1080/03610919508813243
Abstract
A method is proposed for comparing the accuracy of diagnostic tests that rely on a reader's subjective interpretation of the results. An ANOVA approach is applied where the dependencies between the readers' estimates of diagnostic accuracy are handled by adjusting the usual F statistic for theestimated correlation. The distribution of the resulting test statistic is evaluated. The problem is particularly relevant to diagnostic radiology where multi-reader studies of the indices associated with the ROC curve serve important roles in evaluating the efficacy of diagnostic tests.Keywords
This publication has 26 references indexed in Scilit:
- Signal DetectabilityMedical Decision Making, 1991
- The Efficacy of Diagnostic ImagingMedical Decision Making, 1991
- Nonparametric Estimation of Degenerate ROC Data Sets Used for Comparison of Imaging SystemsInvestigative Radiology, 1990
- On the statistical analysis of ROC curvesStatistics in Medicine, 1989
- Analyzing a Portion of the ROC CurveMedical Decision Making, 1989
- A family of nonparametric statistics for comparing diagnostic markers with paired or unpaired dataBiometrika, 1989
- Estimating Pr(X < Y) in Categorized Data using "ROC" AnalysisBiometrics, 1988
- ROC Methodology in Radiologic ImagingInvestigative Radiology, 1986
- Basic principles of ROC analysisSeminars in Nuclear Medicine, 1978
- Concerning the Effect of Intraclass Correlation on Certain Significance TestsThe Annals of Mathematical Statistics, 1947