基于贝叶斯网络的肺癌证候研究

Research on Lung Cancer Syndromes Based on Bayesian Networks

目的:利用贝叶斯网络方法分析肺癌的中医证候。方法:制定《肺癌证候观察表》,观察225例患者的横断面证候,运用贝叶斯网络模型进行数据研究。结果:以症状之间的关联性及关联强度为基础,概括出证候要素,包括病机要素(9个)、病位要素(5个)及病机要素之主要症状与次要症状。结论:基于贝叶斯网络研究的中医证候符合中医基本理论,具有一定的科学性、客观性。

Objective： To analysis traditional Chinese medicine syndromes of lung cancer by using Bayesian networks method. Methods： To observe the cross-sectional syndrome of 225 cases of patients with lung cancer by formulating＂lung cancer syndrome observation table＂,to study the data by means of Bayesian network model.Results： The syndrome elements,including pathogenesis factors（ 9）,disease location elements（ 5） and main symptoms and minor symptoms of the pathogenesis factor were summarized on the basis of the correlation between symptoms and associated strength. Conclusion： The research about Chinese syndromes based on Bayesian network was coincidence with the basic theory of TCM,that shows certain scientific and objective.