Logic engineering and clinical dilemmas

Abstract

There are many challenges to designers wishing to build computer systems to help doctors in their everyday decision making. Systems must be flexible, easy to maintain, robust, sound and safe. It is not easy to meet all these requirements simultaneously. General medical practice exemplifies these problems in quite an extreme form; we discuss how the use of logic programming techniques has facilitated the design and implementation of decision support systems for this medical setting. Although classical logic provides a strong formal foundation for the work, practical decision making requires significant extensions. Principled extensions for constructing arguments under uncertainty, and for framing and taking decisions, are outlined.

This work has been supported by the European Commission under its Advanced Informatics in Medicine and Esprit Basic Research programmes, and by the UK Science and Engineering Research Council.