Research Of Modeling Method Based On Multi-Agent Technology In Medical Assistant Diagnosis System

With the development of society, people require better medical and health undertakings. In order to improve the medical services and avoid misdiagnosis farthest in the process of diagnosis, it needs to establish a computer-aided diagnosis system. The main purpose of this paper is to find a suitable method and correct tools for designing a Medical Assistant Diagnosis System based on MAS.By analyzing the main Multi-agent technology nowadays, this paper sums up a suit of modeling method oriented multi-agent system with the specific characteristics of the medical field. This method is mainly based on MaSE methods, adding the social organization ideas, covering aspects of the modeling process, and presenting a specific modeling tools and modeling language against all links.In addition, In the course of the study, a medical assist diagnose system is established using this method. This system is based on "request / services" cooperation model and interacts with message through the establishment of acquaintances. With pulp disease as an example, this system carries out the diagnose agent and knowledge base agent. During them, diagnostic reasoning is based on weighted fuzzy logic and gets the answer using relevancy searching- strategy; at the same time, knowledge base is expressed by production rule.Finally, through a number of specific cases, the system has basically achieved the goal of avoiding misdiagnosis. In addition, through the realization of the system, it also proves that this method can help software developers create a rational and efficient system of medical assist diagnosis.