Using credit scoring technique to evaluate the usefulness of an imagery examination

The development of exploration methods (imaging, assays, functional explorations) provides the physicians with many tests which can be used to settle a reliable diagnosis. To evaluate the individual diagnostic value of each test, a reference or gold standard certifying the true diagnosis must be used to compute the usual parameters, sensitivity and specificity. But every examination is part of the medical process which we must optimize in terms of health economy and a better minimum fare of the patient. It then seems more and more necessary to evaluate the utility usefulness of an examination in a medical decision process, and to analyze it as an information with questions such as : "Does this examination give any new information with regard to initial data ? Was this information predictible from these initial data ?". In this last case, the examination would be of no use. For this purpose, we propose a methodology using the credit scoring technique. This methodology has been used to study the benefit of a myocardial scintigraphy. This test was useful because It has given a new information (in 48% of the cases we have studied) which could'nt be predictable.