Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/85917
Title: Application of Near-infrared Spectroscopy in the quality control and chemical analysis of Chinese herbal medicines
Authors: Lau, Ching Ching
Degree: Ph.D.
Issue Date: 2013
Abstract: The use of herbal medicine in Asian countries has a long history, but the concern for quality and efficacy of herbal medicine in the light of modern science has just been started in the last few decades. Quality control of Chinese Herbal Medicine (CHM) aims to ensure their consistency, safety and efficacy. As the active ingredient(s) of the herbs are usually not known, current quality control practices focus on consistency of chemical compositions and identification of the herbs. Microscopic identification and chromatography analysis are the conventional methods used, but these methods are time demanding, and sometimes, the information obtained is not sufficient for identification beyond doubt. Near-infrared Spectroscopy (NIRS) may be an advantage as it is quick, simple and non-destructive. Validity of the NIR based method is established from statistical correlation extracted from the results of many samples. Moreover, NIR can be used for online monitoring during the manufacturing of herbal products to improve the quality control strategy. The aim of this work is to examine the feasibility of using NIRS in analysing chemical composition of CHM and develop appropriate procedures and data analysis algorithms for quality control of CHM. In this work, NIRS was used to evaluate the quality of three CHM, Purariae Radix, Coptidis Rhizoma and Ganoderma, in terms of differentiation of the species, prediction of markers contents and biological effect. A systematic procedure for building up quantitative and classification models is proposed in the study of establishing NIR quantitation models for Puerariae Radix using conventional PLS. This improves the validity of the quantitation model being built.
A major drawback for the application of NIRS in the analysis of CHM is that usually only those components of content higher than 1% can be detected. However, in many herbs, the content of the marker could be lower than this. In order to extend the applicability of the NIR to different CHM, a better algorithm, Genetic Algorithm Partial Least Square regression (GA-PLS), is developed by selecting appropriate wavelength regions which correlate more with the parameters to be predicted. The case of Ganoderma shows the algorithm is capable to develop model for prediction of markers with low content (<0.1%). The models developed for Rhizoma Coptidis are more robust when the sample size is small. Generally speaking, people concerns more on the efficacy of CHM, while the current quality control practices focuses on consistency. To examine the possibility of predicting biological activity from the NIR spectra (which is, of course affected by components inside the herbs or herbal products), experiments were conducted to exam the possibility of predicting the anti-oxidant effect of Ganoderma methanolic extracts from the NIR spectra of the raw herbs. A good correlation model was established using GA-PLS. This study clearly demonstrated that NIR spectroscopy with suitable chemometric techniques could be used for identification of the CHM, quantitative analysis of the selected chemical compounds in the CHM as well as evaluation of related biological effects. The GA-PLS modeling algorithm developed widens the scope of the application as more subtle correlations in the data can be extracted. As a conclusion, NIR spectroscopy could be an effective tool for better quality control of CHM.
Subjects: Herbs -- Therapeutic use -- Testing.
Medicine, Chinese.
Hong Kong Polytechnic University -- Dissertations
Pages: xxiv, 251 leaves : ill. ; 30 cm.
Appears in Collections:Thesis

Show full item record

Page views

39
Last Week
0
Last month
Citations as of Mar 24, 2024

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.