Case-based Adaptations in Medicine Focusing on Hypothyroidism.pdf

Case-based Adaptations in Medicine Focusing on Hypothyroidism Rainer Schmidt, Olga Vorobieva, Lothar Gierl Institut für Medizinische Informatik und Biometrie University of Rostock, Germany [rainer.schmidt, olga.vorobieva, lothar.gierl] @medizin.uni-rostock.de Abstract. So far, Case-based Reasoning has not become as successful in medicine as in some other application domains. One, probably the main reason is that on the one side in Case-based Reasoning the adaptation problem could not be solved in a general way. On the other side in medicine adaptation is often more difficult than in other domains, because usually more and complex features have to be considered. Recently we have developed we have developed some programs for endocrinology therapy support, especially for hypothyroidism. In this paper, we do not present them in detail, but focus on adaptation. We do not only summarise experiences with adaptation in medicine, but we want to elaborate typical medical adaptation problems and hope to indicate possibilities how to solve them. Keywords: Case-Based Reasoning, Adaptation, Thyroid, Hypothyroidism 1 Introduction Case-based Reasoning (CBR) has become a successful technique for knowledge-based systems in many domains, while in medicine some more problems arise to use this method. A CBR system has to solve two main tasks [1]: The first one is the retrieval, which is the search for or the calculation of most similar cases. For this task much research has been undertaken. The basic retrieval algorithms for indexing [2], Nearest Neighbor match [3], pre-classification [4] etc. have already been developed some years ago and have been improved in the recent years. So, actually it has become correspondingly easy to find sophisticated CBR retrieval algorithms adequate for nearly every sort of application problem. The second task, the adaptation means modifying a solution of a former similar case to fit for a current problem. If there are no important differences between a