| Rhubarb has a long history and high medicinal value as a well-known traditional medicinal material in China. The dried rhizome and roots of Rheum palmatum L, Rheum tanguticum Maxim. ex Balf, and Rheum officinale Baill are three kinds of official rhubarb in the Chinese Pharmacopoeia. Due to increasing dosages of rhubarb and the decrease of wild rhubarb, the cultivated official rhubarb can no longer meet the needs of the market in recent years. Some species of rhubarbs such as Rheum franzenbachii Munt, Rheum emodii Wall, and Rheum hotaoense C. Y. Cheng et C. T. Kao (a species of rhubarb named after persons’ names) are used as medicines instead of official rhubarb, which has a large impact on the clinical utility of rhubarb as unofficial rhubarb. The unofficial rhubarb has weaker effects than official rhubarb as a purgative and only has styptic and anti-inflammatory effects in practice. Therefore, to control the quality of the rhubarb medicinal material and ensure the safety and effectiveness of the rhubarb medication, it is necessary to identify the rhubarb accurately and rapidly. Identification of rhubarb with traditional Chinese medicine fingerprint technology has been reported widely, but these methods need chemical pretreatment of the samples before the measurement, and the analysis time is too long. Moreover, the operation is so complicated that the pretreatment technology depends on the test results to a certain extent.Therefore, the development of a method without chemical pretreatment that will be able to quickly and accurately identify and characterize rhubarb has an important significance.Terahertz radiation, or T-rays, usually refers to the frequency of the electromagnetic wave in0.1-10THz (1THz=1012Hz), which corresponds0.03-3mm in wavelength. Terahertz radiation is located between the microwave and infrared regions. Most samples contain abundant physical and chemical information in the terahertz band. Therefore, we can effectively identify the structure and configuration of the sample using terahertz time-domain spectroscopy (THz-TDS) technology. However, when some samples have no obvious characteristic absorption peaks in the terahertz band, we can extract the useful information from the terahertz spectrum of the sample with the aid of chemometrics.Chemometrics methods such as least squares support vector machine (LS-SVM), support vector machine (SVM), principal component analysis-linear discriminant analysis (PCA-LDA), fuzzy rule-building expert systems (FuRES), and partial least squares (PLS) were used to establish qualitative analysis model for the identification of41official and unofficial rhubarb in this paper. The aim of the present work is to assess the feasibility of THz-TDS combined with chemometrics methods to identify the genuine and counterfeit rhubarbs, which provides research ideas for the quality control of rhubarb in the process of production. The details are as follows:1. Terahertz time-domain spectroscopy (THz-TDS) combined with emphatic orthogonal signal correction (EOSC) and least squares support vector machine (LS-SVM) has been applied to establishing qualitative analysis model for identifying41official and unofficial rhubarb samples. The correction factors of model have been determined by bootstrapped Latin-partitions. The spectra were pretreated by autoscaling and Savitzky-Golay smoothing, a comparison with two pretreatment methods to the influences of the classified accuracy was implemented. The results showed that the identification accuracy of97.8±1.6%could be accomplished by using the pretreatment methods of autoscaling and EOSC, which is higher than the classified accuracy of87.5±3.0%by using Savitzky-Golay method. The proposed method was proved to be a convenient, non-polluting, accurate, and nondestructive approach for identifying rhubarb samples. The develop procedure can be easily implemented for controlling quality in rhubarb production.2. Terahertz time-domain spectroscopy (THz-TDS) and chemometrics methods have been applied to identifying41official and unofficial rhubarb samples in this paper. The qualitative analysis model of principal component analysis-linear discriminant analysis (PCA-LDA) and support vector machine (SVM) were established to predict unknown rhubarb samples. The predictive ability and stability of the model were evaluated using bootstrapped Latin-partitions method with50bootstraps and4Latin-partitions. Four standard kernels were used to evaluate the SVM for the effectiveness of the classification results, simple linear kernel, the polynomial kernel, radial basis function (RBF) kernel, and sigmoid kernel. The results showed that the identification accuracy of100%could be accomplished by the models of PCA-LDA and SVM using simple linear kernel, which is higher than the classified accuracy of57.5±1.0%by the using sigmoid kernel for the SVM. Moreover, the satisfactory results showed that the classification accuracy of99.7±0.2%and99.9±0.1%were achieved by using the polynomial kernel and RBF kernel for the SVM, respectively. The proposed method provided a simple, rapid, non-chemical pretreatment and environment friendly pathway for the rhubarb discrimination, which can also used for the recognition and sorting of the samples with the similar chemical composition.3. Terahertz time-domain spectroscopy (THz-TDS) technology as a new non-destructive testing method has been applied to identify41official and unofficial rhubarb samples in the present work. The THz time domain spectra of rhubarb samples were preprocessed and then used to establish an identification model by using fuzzy rule-building expert systems (FuRES). The model was validated using bootstrapped Latin-partitions (BLPs) method with10bootstraps and4Latin-partitions. The obtained results showed that the10model has good predictive ability with respect to the classification accuracy of94.8±0.5%and95.2±0.1%by using the preprocessing methods of Savitzky-Golay (S-G) first derivative combined with either one of two orthogonal signal correction (OSC) methods, respectively. The proposed method showed that the THz-TDS combined with chemometrics can be used to identify genuine and counterfeit Chinese herbal medicines, as well as official and unofficial rhubarbs.4. Terahertz time-domain spectroscopy (THz-TDS) combined with partial least squares (PLS) has been used for the identification of the41official and unofficial rhubarb samples in this paper. The component number of the PLS model was evaluated by leave-one-out cross-validation (LOOCV). The results showed that the identification accuracy of90%could be accomplished by using the pretreatment methods of Savitzky-Golay (S-G) first derivative, detrending, standard normal transformation (SNV), autoscaling, and mean centering, which was higher than the classified accuracy of80%obtained without any preprocessing for the time domain spectra. The minimum values of the root-mean squared error of cross-validation (RMSECV) and root-mean squared error of prediction (RMSEP) were achieved by using mean centering method, which were0.0766and0.1690, respectively. The obtained results showed that the combination of PLS algorithm and THz-TDS technology was applied for the recognition of genuine and counterfeit Chinese herbal medicines, as well as official and unofficial rhubarbs. The proposed method proved to be rapid, simple, non-pollution and non-invasive, which is suitable to be developed as a promising tool for quality control of Chinese herbal medicines.The application effects of different spectral preprocessing and modeling methods were further studied in the paper. The feasibility of the methods was proved that terahertz time-domain spectra and absorption spectra combined with chemometrics were applied for the identification of rhubarbs. At the same time, satisfactory results were obtained. In practical application, it is needed to select the corresponding modeling and spectral preprocessing methods according to the samples. The experiment provided a certain reference for the application of THz-TDS technology in noninvasive identification of Chinese herbal medicines. |
PDF Full Text Request