A regression modelling framework for receiver operating characteristic curves in medical diagnostic testing

摘要: SUMMARY Receiver operating characteristic curves (ROC's) are used to evaluate diagnostic tests when test results not binary. They describe the inherent capacity of for distinguishing between truly diseased and nondiseased subjects. Although methodology estimating comparing ROC'S is well developed, date no general framework exists evaluating covariate effects on ROC'S. We formulate a regression model which allows covariates accuracy be succinctly summarised. Such might include, example, characteristics patient or environment, type severity disease. The models shown arise naturally from some classic continuous ordinal data. Regression parameters fitted using an equation approach. method illustrated data study multiformat photographic images scintigraphy.