贝叶斯网络在中医证素辨证体系中的应用

中医学所说的“证”是一个非线性的开放复杂系统。中医的“辨证”属于认知科学、思维科学。在揭示辨证原理与规律的基础上,构建以证素为核心的辨证新体系,是统一辨证方法,提高辨证水平的需要。采用经验建模与计算建模相结合,将贝叶斯网络用于中医辨证诊断数据中症状与证素间隶属关系、证素之间组合关系的研究,运算结果表明其与中医专家经验有很高的吻合性。贝叶斯网络是对辨证进行信息挖掘处理的一种较好方法,但仍未能全面反映中医辨证的思维能力。

Application of Bayesian network in syndrome differentiation system of traditional Chinese medicine

The concept of syndrome in traditional Chinese medicine (TCM) is a nonlinear, open and complicated huge system. Syndrome differentiation in TCM belongs to cognitive and noetic science. To establish a new syndrome differentiation system based on the key elements of the syndrome is necessary for TCM practitioners to promote differentiation ability and reach consensus on differentiation method. With combination of experience and computation models, the Bayesian network was used in the study of the relationship between the key elements of syndrome and the symptoms, and the relationship among different key elements, in which the computing diagnosis result was identical to the result from an experienced TCM doctor. The study showed that Bayesian network is a good method to deal with the information of symptoms and signs for syndrome differentiation, but it is also not to reflect comprehensively the thinking ability of TCM doctors in doing syndrome differentiation.