Maximum Likelihood Estimation of the Accuracy Rates of Diagnostic Tests by Means of the EM Algorithm,
INTERFACE FOUNDATION OF NORTH AMERICA INC FAIRFAX STATION VA
Pagination or Media Count:
An EM algorithm has been developed for computing the maximum likelihood estimates and standard errors of the accuracy rates i.e., sensitivity and specificity of a new diagnostic test and an established reference test, based on the outcomes of both tests when applied to individuals sampled from an arbitrary number of populations with different prevalence rates of a given disease. This algorithm is heuristically appealing in that it also estimates the prevalence rate within each source population and aids the perception of the effects of numerical constraints imposed on some of the rate parameters. An example is given to illustrate the application of this algorithm to practical clinical situations.
- Statistics and Probability