Agent-based reasoning in medical planning and diagnosis combining multiple strategies
2014 (English)In: International journal on artificial intelligence tools, ISSN 0218-2130, Vol. 23, no 01, Special issue, 1440004- p.Article in journal (Refereed) Published
Medical reasoning describes a form of qualitative inquiry that examines the cognitive (thought) processes involved in making medical decision. In this field the goal for diagnostic reasoning is assessing causes of observed conditions in order to make informed choices about treatment. In order to design a diagnostic reasoning method we merge ideas from a hypothetic-deductive method and the Domino model. In this setting, we introduce the so called Hypothetic-Deductive-Domino (HD-D) algorithm. In addition, a multi-agent approach is presented, which takes advantage of the HD-D algorithm for illuminating different standpoints in a diagnostic reasoning and assessment process, and for reaching a well-founded conclusion. This multi-agent approach is based on the so called Observer and Validating agents. The Observer agents are supported by a deductive inference process and the Validating agents are supported by an abductive inference process. The knowledge bases of these agents are captured by a class of possibilistic logic programs. Hence, these agents are able to deal with qualitative information. The approach is illustrated by a real scenario from diagnosing dementia diseases.
Place, publisher, year, edition, pages
World Scientific, 2014. Vol. 23, no 01, Special issue, 1440004- p.
Medical diagnosis, decision making, knowledge representation and reasoning
Research subject Computer Science
IdentifiersURN: urn:nbn:se:umu:diva-79513DOI: 10.1142/S0218213014400041ISI: 000332124600006OAI: oai:DiVA.org:umu-79513DiVA: diva2:642264