基于隐结构法的慢性阻塞性肺疾病稳定期常见证候要素的研究

目的 运用隐结构法提取慢性阻塞性肺疾病稳定期常见证候要素。方法 用统一制定的临床信息采集表收集慢性阻塞性肺疾病稳定期病人的症状信息, 建立数据库并进行数据的录入与预处理, 用隐结构法分析数据, 并结合中医专业知识对获得的隐结构模型进行诠释。结果 慢性阻塞性肺疾病稳定期隐结构模型中共计有30个隐变量和70个隐类, 经结合专业知识诠释, 初步获取有意义的隐类23个。慢性阻塞性肺疾病稳定期涉及的病位类证候要素为:肺、脾、肾、胃、心, 病性类证候要素为:阴虚、阳虚、气虚、痰 (湿) 、血瘀、气滞、热。结论 通过隐结构法分析慢性阻塞性肺疾病稳定期的前瞻性无监督数据发现, 隐结构法的模型构建与中医理论有相似之处, 且比有监督的数据分析有更好的客观性。本研究为进一步建立慢性阻塞性肺疾病稳定期的证候诊断标准提供了依据。Objective To find out the common syndrome factors of chronic obstructive pulmonary disease (COPD) at stable stage by applying latent structure method.Methods The symptom information were collected from the patients with COPD at stable stage by using a unified collection table of clinical information, and then a database was established, in which the data were imputed and retreated.All data were analysed with latent structure method, and the constructed latent structure model of COPD was interpreted based on professional knowledge of Chinese Medicine.Results There were totally 30 latent variables and 70 latent categories in the latent structure model of COPD at stable stage.After the interpretation with professional knowledge, there were 23 significant latent categories obtained.The syndrome factors related to diseases locations of COPD at stable stage included lung, spleen, kidney, stomach and heart.The syndrome factors related to disease nature included yin deficiency, yang deficiency, qi deficiency, phlegm (dampness) , blood stasis, qi stagnation and heat.Conclusion The results of analysis on the prospective non-supervising data of COPD at stable stage indicate that the modeling by latent structure method is similar to TCM theory, and have a higher objective compared with the analysis on supervising data.The research has given some supports for working out the diagnostic standard of COPD at stable stage.