Classification and diagnosis of syndromes in Chinese medicine in the context of coronary heart disease model based on data mining methods

Abstract

Objective: To study on the classification and diagnostic of syndromes in Chinese medicine (TCM) based on the coronary heart disease model (CHD, myocardial ischemia) by application of clustering analysis in mathematical statistics methods. Methods: By application of combining disease with syndrome model, dynamically observed and recorded pathologic signs of animal models, a total of 172 frequencies of the signs were collected, and the variables indicators were analyzed by cluster analysis. Results: The results show that CHD model can be divided into four syndromes by cluster analysis. The four categories can cover the ratio of 71.05% models; it gets a diagnostic accuracy rate of 92.11%, which can be used as key points to diagnose various syndromes in CHD. Conclusion: Cluster analysis can help to classify the TCM syndromes reasonably and objectively. What more, it also can discover the pattern of the syndrome evolution, Thus to provide a theoretical basis for the standardization of TCM research. © 2010 Springer-Verlag.