基于隱結構模型分析的心血管疾病中醫問診證候分類研究

目的探討心血管疾病的中醫問診證候分類特征,為中醫證候診斷標準的建立提供客觀依據。方法利用中醫心系問診采集量表,采集大樣本心血管疾病臨床病例,根據問診信息的"有、無"分別賦值"1、0",建立問診數據庫;基于隱結構分析,找出規律,建立隱結構模型;人機結合,對部分隱變量進行綜合聚類分析和類的細分;基于模型結果,分析心血管疾病的臨床中醫問診證候分類特征。結果心血管疾病中醫問診證候以心氣虛、心陽虛、氣陰兩虛、痰濕、血瘀、氣滯、心火亢盛、津液虧虛為主,并兼見胃氣上逆、腎氣不固、脾胃虛寒等證,其出現率依次為46%、23%、34%、18%、19%、39%、14%、1.7%、19%、27%、25%,并提示了這些證候與關系密切的各問診癥狀之間的相關性(包括出現的概率和互信息)。結論隱結構分析方法能為中醫證候的分類提供定性定量依據,并提示綜合聚類分析和類的細分方法的應用能進一步明確隱變量與變量之間的定量關系,從而為臨床中醫證候標準的建立提供依據。Objective To provide some objective basis for establishment of TCM diagnostic cri researching the syndromes classification of TCM inquiry about cardiovascular disease.Methods A samples clinical cases were collected with TCM heart-system inquiry information scales, "Yes, N assigned "1, 0", then the interrogation database was established.We analyzed the data based o structure model, then find the regular pattern to establish the model.With the combination of m machine, some latent variables were treated with comprehensive cluster analysis analysis an subdivision.We analyzed the clinical characteristics of TCM inquiry syndromes.Results The ma syndromes on cardiovascular disease were as follows:heart-qi deficiency, heart-yang deficiency, defic heart yin and qi, phlegm, blood stasis, qi stagnation, hyperactivity of heart-fire, body fluid de stomach-qi go against, kidney-qi deficiency, cold and deficiency of spleen and stomach, and the occ rate was 46%, 23%, 34%, 18%, 19%, 39%, 14%, 1.7%, 19%, 27%, 25% respectively.It showed the co between these syndromes and the various symptoms of inquiry (including the probability and information) .Conclusion The latent structure analysis method can provide a quantitative basis and pr the classification of TCM, and the study demonstrated comprehensive cluster analysis and class sub application can further clarify the relationship between the hidden variables and the variables, thus basis for establishing TCM syndrome criteria.