| Summary: | Thesis (M.Sc.)--Memorial University of Newfoundland, 2009. Computer Science Includes bibliographical references (leaves 114-117) Coordination and negotiation among agents are often necessary for medical diagnoses when a community of experts is called to be involved in a joint diagnosis and treatment for a patient. However, most of existing diagnostic systems are single-agent and rule-based, using probability theory or Bayesian networks. Fuzziness and uncertainty of concepts, facts and rules should also be considered to meet the needs of practical medical diagnoses, especially for Traditional Chinese Medicine. -- In this thesis, a novel model of a multi-agent diagnosis helping system (MADHS) is given, where distributed knowledge-based systems are considered as cooperative agents in medical diagnoses. A final diagnosis compatible with both patient's anamnesis and existing medical principles can be reached through a joint decision-making procedure in this model. Fuzziness and uncertainty are incorporated into inference techniques to form the reasoning mechanism of the agents. The model is implemented to create a prototype system using Java, Java Agent Development Framework (JADE), Java Expert System Shell (JESS) and NRC FuzzyJ Toolkit. The prototype system has been tested by medical cases, especially those in Traditional Chinese Medicine (TCM). -- It is anticipated that the proposed model and methodologies will be widely used in many applicative areas, such as multi-agent medical diagnosis, medical helping, and other automatic diagnosis and decision-making systems.
|
|---|