Modeling Of Inquiry Diagnosis For Chronic Gastritis In Traditonal Chinese Medicine By Complex System Methods

Traditional Chinese Medicine (TCM) inquiry refers to differentiation syndromes based on the symptoms informations which mainly by asking questions of the patient or the patient’s companion. While the diagnostic process lead to a lot of subjectivity which is hard to be convinced by people. It is urgent to make TCM diagnosis more objective. Fortunately, the characteristics of TCM include integrality, nonlinearity and dynamics is conforming to the characteristics of a complex system. This research set up complex system modeling of chronic gastritis TCM physician and the main work and contributions are show as follows:1. In order to eliminate irrelevance and redundancy within symptoms, Feature selection combine with rout set and mutual information is applied to get syndrome subset according to each symdrom. The results shows it is fulfill with the of TCM theory about chronic gastritis syndromes.2. Community detecting methord is used to partition the selected symptoms into modules. The result has got the reasonable explanation of disease location and property with symdroms by the theory of traditional Chinese medicine.3. The probability graph model is applied to complex system modeling analysis of TCM. Probability graph model is excellent discribe lack-regularity causality of complex system. In this paper, Probability graph model is used to construct multi-label classification model of chronic gastritis of TCM syndrome differentiation. Finally, it performed outstanding to the most multi-label algorithms and get the average prediction of82.5%.4. An information collection and auxiliary syndrome differentiation system of traditional Chinese medicine is designed based on the VS2005platform and access database. The design fully considers the man-machine engineering to get the customer satisfaction system, and the vast TCM information is organized into blocks reasonable. The system implementation for several functions including information collection, data processing, and assist syndrome differentiation.