The application of unified and combined recognition decision rules to the multistage diagnosis problem

This paper deals with the decision rules of multistage diagnosis in the case when two kinds of information are available: empirical data and expert rules. For this problem combined and unified diagnostic algorithms are derived. In the first approach the algorithm uses both sets of data simultaneously. In the second however, one set of data is transformed into the second. The proposed algorithms were applied practically to the computer-aided diagnosis of acute abdominal pain and results of their classification accuracy were compared with existing methods.