Case-based reasoning for antibiotics therapy advice: an investigation of retrieval algorithms and prototypes

We have developed an antibiotics therapy advice system called ICONS for pat ients in an intensive care unit (ICU) who have caught an infection as addit ional complication. Since advice for such critically ill patients is needed very quickly and as the actual pathogen still has to be identified by the laboratory, we use an expected pathogen spectrum based on medical backgroun d knowledge and known resistances. The expected pathogen spectra and the re sistance information are periodically updated from laboratory results. To s peed up the process of finding suitable therapy recommendations, we have ap plied case-based reasoning (CBR) techniques. As all required information sh ould always be up to date in medical expert systems, new cases should be in crementally incorporated into the case base and outdated ones should be upd ated or erased. For reasons of space limitations and of retrieval time an i ndefinite growth of the case base should be avoided. To fulfill these requi rements we propose that specific single cases should be generalised to more general prototypical ones and that subsequent redundant cases should be er ased. In this paper, we present evaluation results of different generation strategies for generalised cases (prototypes). Additionally, we compare mea sured retrieval times for two indexing retrieval algorithms: simple indexin g, which is appropriate for small and medium case bases, and tree-hash retr ieval, which is advantageous for large case bases. (C) 2001 Elsevier Scienc e B.V All rights reserved.