page 2  (14 pages)
to previous section1
3to next section

Expanding the text here will generate a large amount of data for your browser to display

Knowledge Organisation in Medical KBS

Construction?

A. Th. Schreiber z G. van Heijst z G. Lanzola x M. Stefanelli x

z University of Amsterdam, Social Science Informatics
Roetersstraat 15, NL-1018 WB Amsterdam, The Netherlands
E-mail: fschreiber,gertjang@swi.psy.uva.nl

x University of Pavia, Medical Informatics
Via Abbiategrasso 209, I-27100, Pavia, Italy
E-mail: giordano@ipaimed1.unipv.it, stefa@ipvccn.bitnet

Abstract

In this paper we present an analysis of the structure of medical knowledge for the purpose of the construction of knowledge based systems. We discern three different views on knowledge that each have a different but complementary function with regard to reusability. We show that each of these views suggests distinct activities in the knowledge engineering process. The reported knowledge organisation is a key element of the the games methodology for the construction of medical knowledge based systems.

1 Introduction

In this paper, we discuss structuring principles of medical domain knowledge in knowledgebased applications. We take a knowledge-level [Newell, 1982] viewpoint, abstracting from potential symbolic representations of the knowledge. We distinguish three views on medical knowledge, namely (i) the vocabulary of basic medical terms, (ii) the knowledge types distinguished in an application, comprising the ontology of the application, and (iii) the knowledge roles, pointing to the way in which knowledge is used during problem solving. Multiple views on medical knowledge are useful for a number of purposes. In this paper we focus on two aspects:

ffl The identification of generic, reusable knowledge elements.

?The research reported here was carried out in the course of the GAMES-II project. This project is partially funded by the AIM Programme of the Commission of the European Communities as project number A2034. The partners in this project are SAGO (Florence, Italy), Foundation of Research and Technology (Crete, Greece), Geneva University Hospital (Switzerland), the University of Amsterdam (The Netherlands), University College of London (UK), the University of Pavia (Italy) and the University of Ulm (Germany).
This paper reflects the opinions of the authors and not necessarily those of the consortium.