Abstract:
|
The objectives of many clinical studies are to evaluate the diagnostic performance of medical devices or procedures. For some of these studies, multiple cases and multiple readers may be involved and the primary statistical analysis will focus on sensitivity and specificity and whether a reader can correctly identify the case as positive or negative compared to clinical truth. In this study, we evaluate some commonly used study designs and statistical methods for evaluating Multiple-Reader Multiple-Case (MRMC) studies with binary study endpoints.
|