School of Basic Medical Science, Guangzhou University of Chinese Medicine, Guangzhou, China.
J Altern Complement Med. 2020 Aug;26(8):729-737. doi: 10.1089/acm.2020.0008. Epub 2020 Jul 2.
The aim of this study was to establish a quantitative syndrome differentiation model with logistic regression analysis for phlegm and blood stasis syndrome (PBSS) in coronary heart disease (CHD) to offer methodology guidance for the quantitative syndrome differentiation of Traditional Chinese Medicine (TCM). Tongue, face, and pulse information of each subject was obtained using the TCM-intelligent diagnosis instruments. Logistic regression model was used to construct the syndrome diagnosis model. The area under receiver operating characteristic curve (ROC-AUC) was used to evaluate the diagnostic value of the model. Among the 141 subjects, 83 belonged to the PBSS group, and 58 belonged to the non-PBSS group. The independent indexes used to predict PBSS in patients with CHD were length of the crack (LC) ( = 0.002), number of ecchymosis (NE) ( < 0.001), length of philtrum (LEP) ( = 0.022), and right hand pulse h1 (Rh) ( = 0.021). The expression of combining predictor L in this study was L = LC +57.58 NE +4.53 LEP +2.68 Rh. The ROC curve analysis indicated that the AUC values of LC, NE, LEP, and Rh were 0.646, 0.710, 0.619, and 0.613, respectively. The AUC = 0.825 of the syndrome diagnosis model was the largest. The quantitative study of TCM syndrome based on logistic regression analysis provides a good method for the objective analysis and application of TCM syndrome.
本研究旨在建立一个基于逻辑回归分析的冠心病痰瘀证定量辨证模型，为中医定量辨证提供方法学指导。通过中医智能诊断仪器获取每个受试者的舌、面、脉信息。使用逻辑回归模型构建证候诊断模型。采用受试者工作特征曲线下面积（ROC-AUC）评估模型的诊断价值。在 141 例患者中，83 例属于痰瘀证组，58 例属于非痰瘀证组。预测冠心病患者痰瘀证的独立指标为裂纹长度（LC）（=0.002）、瘀斑数量（NE）（<0.001）、人中长度（LEP）（=0.022）和右手脉 h1（Rh）（=0.021）。本研究中联合预测因子 L 的表达为 L=LC+57.58 NE+4.53 LEP+2.68 Rh。ROC 曲线分析表明 LC、NE、LEP 和 Rh 的 AUC 值分别为 0.646、0.710、0.619 和 0.613，而证候诊断模型的 AUC 值为 0.825。基于逻辑回归分析的中医证候定量研究为中医证候的客观分析和应用提供了一种较好的方法。
