Fuzzy cluster analysis of trace element content and therapeutic effect in traditional Chinese medicine
Dong SF,Han LQ,Zhao WX,Dong HB,Liu JH,Lei JT
DOI: https://doi.org/10.3321/j.issn:1673-8225.2007.40.031
2007-01-01
Abstract:AIM: It is difficult to distinguish drug similarity among different traditional Chinese medicines by the discrete contents of trace elements in them. This study was designed to investigate the interrelation of the trace element contents in these medicines and their therapeutic effect by means of blurred classification, verify the similarity of the medicine effect and their roles in the quantitative evaluation of these medicines. METHODS: The experiment was finished in the Laboratory of Traditional Chinese Medicine Trace Elements Analysis, Administration of Traditional Chinese Medicine of Jilin Province from October 2005 to November 2006. The contents of potassium, sodium, calcium, magnesium, iron, chromium, nickel, manganese, copper and zinc in 11 kinds of traditional Chinese medicines, such as Albizia flower, milkvetch root, grassleaf sweetflag rhizome, thinleaf milkwort root, hawthorn fruit, ginkgo leaf, ginseng, longan pulp, cattail pollen, spine date seed and sappan wood were determined by atomic absorption spectrophotometry. Combining with the Zn/Cu radio and Mg/Ca radio of the range analysis, the standard processing of statistical target was carried out through the data extreme difference method. The relative coefficients of the subjects investigated were calculated by the dot product method and the relative fuzzy matrix was obtained. RESULTS: The results showed that the cluster analysis in 7 kinds of medicines (Albizia flower, grassleaf sweetflag rhizome, thinleaf milkwort root, ginseng, longan pulp, cattail pollen and spine date seed) was similar when the relative coefficient was 0.50; the cluster analysis in 5 kinds of medicines (Albizia flower, grassleaf sweetflag rhizome, longan pulp, cattail pollen and spine date seed) was similar when the relative coefficient was 0.65; the cluster analysis in 3 kinds of medicines (Albizia flower, grassleaf sweetflag rhizome and cattail pollen) was similar when the relative coefficient was 0.75. The contents of metal elements were rich in these medicines. And the greater the relative coefficient was, the more similar the therapeutic effect was. CONCLUSION: The fuzzy cluster analysis model, established by mathematics and chemometrics method, can normalize the discrete data and define the fuzzy matrix of the therapeutic effects. It has provided a new scientific method for exploring the drug similarity and evaluating the quality of traditional Chinese medicines.