Construction of Receiver Operating Characteristic Curves when Disease Verification Is Subject to Selection Bias
Robert J. Gray, Colin B. Begg, Robert A. Greenes · Medical Decision Making · 1984
The estimates of ROC curves, which are frequently used in the assessment of diagnostic tests, may be biased if the sample is restricted to subjects whose disease status has been definitively verified. A method to provide an unbiased estimate of the ROC curve under these sample conditions is proposed. The new method requires information on the probability distribution of test results in the population from which the verified sample is drawn. It is illustrated using data from a study of computed tomography for fever of uncertain origin.
Cited by 104 works. doi:10.1177/0272989x8400400204
Authors
References
- The meaning and use of the area under a receiver operating characteristic (ROC) curve.
- Basic principles of ROC analysis
- Maximum-likelihood estimation of parameters of signal-detection theory and determination of confidence intervals—Rating-method data
- ROC Analysis Applied to the Evaluation of Medical Imaging Techniques
- Problems of Spectrum and Bias in Evaluating the Efficacy of Diagnostic Tests
- Evaluation of diagnostic systems : methods from signal detection theory
- Assessment of Diagnostic Tests When Disease Verification is Subject to Selection Bias
- A prospective study of computed tomography, ultrasound, and gallium imaging in patients with fever.
- A prospective evaluation of computed tomography and ultrasound of the pancreas.