Automated reasoning system in histopathologic diagnosis and prognosis of prostate cancer and its precursors

OBJECTIVE: This article presents the rationale and options offered to diagnostic and prognostic decision support systems for prostate pathology by automated reasoning capabilities. METHODS: The symbolic information used in diagnostic decision-making is systematically ordered, compared, numerically assessed in its probability, and combined such that a conclusion can be drawn. The framework for the processing of such symbolic information may be an expert system, an inference network or a case-based reasoning system. Automated reasoning is implemented by the use of a rule base and information flow control modules. RESULTS: Automated reasoning allows decision support systems to follow highly adaptive decision sequences, capable of handling contradictory evidence, exceptions in diagnostic clue expression, and nonmonotonic decision-making. CONCLUSIONS: Automated reasoning capability in diagnostic and prognostic decision support systems allows highly flexible decision development, very close to human decision procedures.