A Hierarchical Diagnosis Model for Syndrome Prediction in TCM of Post-Hepatitic Cirrhosis

Abstract

Owing to the typical vagueness of treatment processes in traditional Chinese medicine (TCM), determining the final diagnosis result is very difficult. In this paper, a novel hierarchical diagnosis model (HDM) for syndrome prediction problems in TCM is constructed, incorporating rough set theory, expert judgement, and an attribute hierarchical model (AHM). Firstly, using the concept of the attribute hierarchical model, a hierarchical model with three levels - patient level, symptom level and diagnosis level - is built to predict the syndrome of a new case. Secondly, by integrating the positive region, support degree in rough set theory and expert experience, the subjective and objective parts of each attribute on the three levels are acquired. Thirdly, the synthetic attribute-weights at diagnosis level are obtained to predict the syndrome. Finally, an example is studied to illustrate the detailed implementation process of the HDM approach, and a real post-hepatitic cirrhosis dataset is also diagnosed to evaluate the performance of the model. The encouraging results show the effectiveness of the presented method. 1