(in Englisch)
Methodological research is carried out in the following areas:
Knowledge representation deals with the formal modelling of expert knowledge in a computer program. Important questions in this respect concern the given degree of structuralization of the medical domain under consideration, the necessity to include vagueness of medical terms and uncertainty of medical conclusions into the chosen formal representation, as well as the extent and completion of the respective knowledge domain.
Knowledge acquisition and machine learning concerns the problem of how to acquire domain-specific knowledge according to the chosen representation. Usually, human experts are asked to provide this information but other methods slowly evolve. Automatic "learning from examples" attempts to analyze verified patient records and to establish associations between medical entities, to express these associations in numerical form, and to extract new, clinically useable, rules from given patient data.
Knowledge base consistency checking becomes necessary if the knowledge base grows to such an extent that the semantic consistency of the stored knowledge cannot be guaranteed and logical contradictions in the stored knowledge may cause wrong results.
Reasoning mechanisms are inference methods which draw medical conclusions from given patient data by means of the stored medical knowledge. Most important is the selection of the appropriate formal approach with respect to the given medical domain. One differentiates methods to infer logical conclusions (e.g., propositional and predicate logic, three-valued logic, fuzzy logic, non-monotonic logic) and to combine medical evidence (e.g., Bayes theorem, certainty factors, Dempster-Shafer theory of evidence).
Human-computer interaction forms the basis for practical usage of medical expert systems and strongly influences the acceptance of these systems in practical settings. An important component of essentially every medical expert system is its ability to explain its reasoning.
Evaluating the accuracy and acceptance of expert systems is essential before a routine application of the developed system can be started. Evaluation methods comprise accuracy and utility evaluations, evaluations of the transferability and portability of the system, sensitivity analyses, and analyses of the cost effectiveness.