Visual Data Analysis with Support Vector Machine for the Analysis of Three Processed Aconitum Drugs

Herb medicines (HMs) are becoming increasingly popular all over the world. However, due to misuse or confusion of HMs, some toxic effects and even poisonings have been reported, raising questions about their safety and efficacy. Processed Chuanwu (the mother root of Aconitum carmichaelii), Fuzi (the lateral root of A. carmichaelii) and Caowu (the root of A. kusnezoffii) have been used in traditional Chinese medicine for several thousand years. In this study, eight batches of Heishunpian (processed Fuzi), ten bathes of Zhichuanwu (processed Chuanwu) and nine bathes of Zhicaowu (processed Caowu) were collected. By optimizing the extraction and analytical conditions, a reliable and precise HPLC method coupled with mass spectrometry has been proposed for the quantification of eleven major alkaloids for these drugs. Difference of alkaloids contents among these Aconitum drugs was not directly observed, but the underlying information was revealed by support vector machine analysis then the discrimination became feasible. This method and quantization results could provide an objective and technical platform for the differentiation and quality control of three processed Aconitum drugs.