| The advance in study of traditional Chinese medicine (TCM) fingerprint with the application of chemometrics in recent years was reviewed in this article. Fingerprints of TCM from different botanical origin, location or processing methods were developed with the use of high performance liquid chromatography (HPLC), infrared (IR), near infrared (NIR) and excitation-emission matrix fluorescence (EEM) spectroscopy, respectively. Chemometrics were employed to analyze the chromatographic and spectral data, and develop the discriminating models, which could be used for the quality control of TCM.In the first chapter, the application and the significance of chemometrics in the study of TCM was reviewed. An introduction about the research foreground and advances were presented, where the definition, characteristic, necessity and significance, research contents and methods were summarized and briefly discussed. In addition, the application and prospects of chemometrics for fingerprint analysis of TCM were also presented and discussed.In chapter two, multi-wavelength fingerprints of Cassia seed, a TCM, were collected by HPLC at two wavelengths with the use of diode array detection. The two data sets of chromatograms were combined by the data fusion-based method. This data set of fingerprints was compared separately with the two data sets collected at each of the two wavelengths. It was demonstrated with the use of principal component analysis (PCA), that multi-wavelength fingerprints provided a much improved representation of the differences in the samples. Furthermore, partial least squares (PLS), back propagation artificial neural network (BP-NN), and radial basis function artificial neural network (RBF-NN) were effectively applied to predict the category of the four different samples in the test set.In the third chapter, raw and roasted Semen Cassiae seeds, were used as examples, to research and develop a method of classification analysis based on measurements of Fourier transform infrared (FTIR) spectral fingerprints. Eighty samples of the TCM were measured in the mid-infrared range,2000-400 cm-1 (KBr pellets), and the complex overlapping spectra were submitted for interpretation to:principal component analysis least squares support vector machine (PC-LS-SVM), kernel principal component analysis least squares support vector machine (KPC-LS-SVM), and (RBF-NN). The LS-SVM models were developed with an RBF kernel function and a grid search technique. Training models were constructed with the use of raw and first derivative spectra and these were then verified by another data set containing both raw and roasted spectral objects. It was demonstrated that the first derivative data set produced the best separation of the spectral objects. In general, satisfactory analytical performance was obtained with the PC-LS-SVM, and KPC-LS-SVM, and RBF-NN training models and with the classification of the verification spectral objects. With regard to chemometrics modelling, the performance of KPC-LS-SVM was somewhat more economical than that of the PC-LS-SVM model. It would appear that the latter relatively simple model would be sufficient to apply for most small to medium size FTIR fingerprint data sets but with larger matrices, the more complex models such as the RBF-ANN and KPC-LS-SVM may be more advantageous on computational basis.In the fourth chapter, Near-infrared spectroscopy (NIRS) was applied for direct and rapid collection of characteristic spectra from, Rhizoma Corydalis, a common traditional Chinese medicine (TCM) with the aim of developing a method for the classification of such substances according to their geographical origin. The powdered form of the TCM was collected from two such different sources, and their NIR spectra were pretreated by the wavelet transform (WT) method. A training set of such objects was modeled with the use of least-squares support vector machines (LS-SVM), radial basis function artificial neural networks (RBF-NN), discriminant partial least squares (DPLS) and K-nearest neighbors (KNN) methods. Although all four chemometrics models performed reasonably on the basis of spectral recognition and prediction criteria, the LS-SVM method performed best with over 95% success on both criteria. Thus, an NIR spectroscopic method supported by the WT-LS-SVM chemometrics modeling was recommended for application to classify TCM, Rhizoma Corydalis, samples according to their geographical origin.In the fifth chapter, the EEMs spectroscopy of Rhizoma Corydalis Decumbentis (RCD) and Rhizoma Corydalis (RC), two medicinal herbs of Papaveraceae family, were measured in the range ofλex= 215～395 nm,λem= 290～560 nm, which were regarded as three-way fingerprints of the herb medicine. The potential of EEM spectroscopy combined with two-way and three-way data processing methods to discriminate RCD, RC and their adulteration was studied. On the one hand, PCA was applied on EEMs to extract the two-way feature spectra, the first PCs of each sample, which were used to construct two-way fingerprints. PCA was again performed upon the data of the two-way fingerprints for exploratory analysis, which provided a clear distribution of the tested samples in a score projection plot. On the other hand, a PARAFAC analysis on the three-way fingerprints with F=4 also produced scores projection plots. It was demonstrated that the three-way fingerprints with PARAFAC produced the better separation of the RCD, RC and adulterations. In addition, the two-way fingerprints matrix based on PCA extraction and the PARAFAC score matrix was used as input for LDA, KNN, BPNN and RBFNN models. Finally, the EEM spectroscopy coupled with PARAFAC and BPNN/RBFNN was recommended for the application to discriminate RCD, RC and adulterations, and predict the category of the unknown samples.
|
PDF Full Text Request