Adaptation and abstraction as steps towards case-based reasoning in the real medical world: case-based selection...

In this paper we describe an approach to making suitable case-based reasoning methods for real medical world problems. As an example: for the class of therapeutic problems, we choose therapy advice as antibiotics for patients in an intensive care unit, who have an infectious disease as an additional complication. As rapid advice is needed and the agent is unknown, we use an expected agent spectrum based on medical background knowledge and known resistances, which will both be adapted with the results of the laboratory. Case-based reasoning retrieval methods provide the advice for similar former patients. The old solutions are adapted to be applicable to the new medical situation of the current patient. Because of the large and incrementally increasing number of cases, we use prototypes as a structural aid. We present some experimental results of studies about the performance of our prototype design.