RÉSUMÉ
Objective: To establish a near-infrared diffuse reflectance spectroscopy(NIRS) identification model for crude products,counterfeit products and processed products of Calamina by principal component analysis(PCA) and support vector machine(SVM) algorithm. Method: NIRS of crude products,counterfeit products and processed products of Calamina were collected,the characteristic spectrum segments were selected,the preprocessing method and the optimum principal component number were optimized,and the PCA-SVM qualitative model was established. Result: The characteristic spectrum segment of analysis model was 7 500-4 000 cm-1.Spectra were preprocessed by the first-order derivative method(FD).The optimum principal component number was 5. And the optimum internal parameters of SVM[penalty factor(c)=0.25 and kernel function parameter(g)=8] were screened by applying the grid search algorithm.In the PCA-SVM qualitative model,the prediction accuracy rate was 100%for the 5-fold cross validation,and the prediction accuracy rates also were 100%both for training set and test set. Conclusion: PCA-SVM analysis model of NIRS for Calamina samples has a high prediction accuracy rate,and it can be used for the rapid and nondestructive identification of crude products,counterfeit products and processed products of Calamina by combining the diffuse reflection technique on solid powder.
RÉSUMÉ
OBJECTIVE:To establish the method for rapid determination of nuezhenoside in Salvia miltiorrhiza.METHODS:The contents of salvianolic acid B,rosmarinic acid,lithospermic acid and tanshinone (tanshinone Ⅱ A+tanshinone Ⅰ +cryptotanshinone) were determined by HPLC (as reference value).Quantitative model for the contents of above components was established by partial least square (PLS)-NIR spectra.According to the reference value,143 samples were collected and the spectrum was pretreated by first-order derivative.The optimal range of wave band for salvianolic acid B,rosmarinic acid,oxalic acid and tanshinone were 6 773.98-3 981.12,6 670.85-3 996.54,8 544.66-3 936.28,8 188.06-3 875.31 cm-1.RESULTS:The methodology for the content determination of salvianolic acid B,rosmarinic acid,oxalic acid and tanshinone were in line with the requirement.The coefficient (R2) of internal cross validation for quantitative correction model of salvianolic acid B,rosmarinic acid,oxalic acid and tanshinone were 0.919 0,0.832 2,0.821 5,0.925 6.The deviation of corrected mean square roots were 4.46,0.48,1.34,0.71,respectively.The R2 values of external validation were 0.852 6,0.957 3,0.819 3,0.953 1,respectively;and mean square root of prediction error (RMSEP)were 9.77,0.28,0.94,0.63,respectively.CONCLUSIONS:The method is rapid,accurate,simple and pollution-free,and can be used for rapid determination of multi-maker ingredients in S.miltiorrhiza.
RÉSUMÉ
Objective: To develop a model for rapid and non-destructive determination of amoxilcillin sodium and sulbactam sodium for injection using the analysis of near infrared diffuse reflectance spectroscopy (NIR) and chemometrics.Methods: Totally 41 batches of commercial samples and 20 batches of laboratory samples were analyzed by NIR and the legal methods.The first derivative and vector normalization were selected as the preprocessing methods and 8 720-5 446 cm-1 was selected as the frequency range.Results: The quantitative model was constructed based on 16 batches of commercial samples and 15 batches of laboratory samples (0.75 g) and the content ranged from 4.45% to 61.82% for amoxilcillin and 15.75% to 30.25% for sulbactam.The root mean square errors of cross validation (RMSECV), determination coefficients (R 2) and root mean square errors of prediction (RMSEP) respectively was 0.858 , 0.998 1 and 0.936 for amoxilcillin, and respectively was 0.541 , 0.988 1 and 0.423 for sulbactam.The model was tested based on 5 batches of commercial samples and 5 batches of laboratory samples (0.75 g) and the results well met with those of the legal methods with difference ≤ 1.5 %.The model also applied in 10 batches of commercial samples (1.5 g) and 2 batches of samples from the other manufacturers.Conclusion: The non-destructive quantitative NIR methods are accurate with good reproducibility, and applicable for the rapid analysis of amoxilcillin sodium and sulbactam sodium for injection.
RÉSUMÉ
OBJECTIVE: To establish a nondestructive and rapid analysis method to measure the critical parameter, mixing uniformity, in the production of Ginkgo leaf dispersible tablets in order to control the quality of products. METHODS: The near-infrared diffuse reflectance spectroscopy (NIRDRS) spectra of the pre-mix and the total mixed Ginkgo leaf dispersible tablets (Wuhan Jianmin Suizhou Pharmaceutical Co., Ltd.) were obtained. The mixing uniformity of samples was monitored by quantitative analysis model and standard deviation method combined with the Mahalanobis distance method, respectively. Meanwhile, lab-made mixing samples and dynamic simulation mixing process were used to validate the feasibility of Mahalanobis distance method. RESULTS: The results obtained from quantitative analysis model and standard deviation method were consistent with each other, and the Mahalanobis distance method was feasible for monitoring the mixing process. CONCLUSION: The method is rapid, simple, and can be used for online and sideline analysis in drug production process.
RÉSUMÉ
This study was aimed to establish a qualitative model of near-infrared spectroscopy in order to accurately and rapidly identify several mineral Chinese medicinals from fossil including Os Draconis, Dens Draconis, Fossil Shell of Spirifer, and Fossil Crabs. The near-infrared diffuse reflectance spectroscopy combined with OPUS software was used to analyze the spectral characteristics of these samples. The pattern recognition method was explored through cluster analysis. And the accuracy of the model was verified. The results showed that these mineral Chinese medicinals from fossil had their characteristics absorption so that they can be quickly and accurately differentiated from each other through pattern recognition method. It was concluded that based on near-infrared spectroscopic mod-eling, these mineral Chinese medicinals from fossil including Os Draconis, Dens Draconis, Fossil Shell of Spirifer, and Fossil Crabs can be quickly and accurately identified.
RÉSUMÉ
OBJECTIVE: To establish a rapid method based on near-infrared diffuse reflectance spectroscopy (NIR) to monitor the blending process of traditional Chinese medicinal powder on-line. METHODS: NIR spectra were collected during the blending processes of mixtures consisting of Red Peony Root powder and defatted Spine Date Seed powder. Multivariate calibration models based on PLS regression were developed to correlate the collected spectra with the content of Red Peony Root powder. The optimal calibration model was applied to monitor the blending process in real time with noninterference. RESULTS: After spectral pretreatment using multiplicative signal correction (MSC), first derivative and Savitzky-Golay filter, Red Peony Root powder contents were quantified using PLS algorithm in the spectral ranges of 5900-5950 cm-1, 6700-6750 cm-1 and 6950-7100 cm-1. The correlation coefficient of the calibration model was 0.9990. The RMSEC, RMSEP and RMSECV were 1.14, 1.16 and 1.26, respectively. CONCLUSION: It was demonstrated that a miniature near-infrared analyzer could be used to monitor the contents of the constituents and the blending homogeneity of traditional Chinese medicinal powder in real time, and to rapidly determine the end-point of blending process. In addition, it can be used to judge whether the physical state of powder mixture conforms to the standard. Therefore, NIR spectroscopy is considered to be an appropriate PAT tool for end-point control of blending process of traditional Chinese medicinal powder.
RÉSUMÉ
OBJECTIVE: To develop a method for rapid determination of proanthocyanidins and total phenolics in the seedpod of Nelumbo nucifera Gaertn by near infrared diffuse reflectance spectroscopy. METHODS: Calibration models were established using PLS (partial least squares) method, while the contents of proanthocyanidins and total phenolics were measured by UV. RESULTS: The range of 7500-6900 cm-1 of near infrared spectra was selected. Different spectra pretreatment methods were compared, and it was shown that second derivative pretreatment and Savitzky-Golay smoothing filter methods could extract spectra information thoroughly. The correlation coefficient (r) of proanthocyanidins was 0.9621, SEC (st andard deviation of the calibration sets) was 0.64, and SEP (st andarddeviation of the prediction sets) was 0.94. The correlation coefficient (r) of total phenolics was 0.9254, SEC was 0.97, and SEP was 1.26. CONCLUSION: The near infrared diffuse reflectance spectroscopy method can be used to analyze the main active components in the seedpod of Nelumbo nucifera Gaertn rapidly, so it can be used routinely by the manufacturers for the quality control of the raw materials.
RÉSUMÉ
Objective To determine ecdysterone content in Radix Achyranthis Bidentalae by near-infrared diffuse reflectance spectroscopy (NIRDRS) combined with partial least squares (PLS). Methods NIRDRS and PLS. Results The correlation coefficient of the quantitative mathematics model between the prediction and the true values was 0.948 9. It is feasible to appling the above established method to determination of chemical constituents in Chinese materia medica. Conclusion Compared with other methods, the method is simple, rapid, high efficient, low expenditure, no pollution and with preparation undone and is suit for large quantity of unknown samples.
RÉSUMÉ
Object To establish a new method for the identification of Yangti (Chinese herb from Rumex L., Polygonaceae) by near infrared diffuse reflectance spectrometry. Methods Cluster analysis and discriminative analysis were adopted for their identification. Results The method can identify crude Yangti to a certain degree with results coincident with that of the traditional phytotoxnomy. Conclusion This method can be used for the rapid and accurate differentiation of crude drug of Rumex L..
RÉSUMÉ
AIM: To establish an identification method of Fritillaria based on near-infrared diffuse reflectance spectroscopy(NIRDRS). METHODS: Seven certified Fritillaria,i.e.,F.przewalskii Matim,F.cirrhosa D.Don,F.unibraacteate Hsiaaoet.K.C.Hsia,F.thunbergii Mig,F.pallidiflora,F.ussuriensis Matin and F.kupehensis Hsiaet K.C.Hsia,and three fake "Fritillaria",i.e.,F.thunbergii Varchekiugensis Hsiaet K.C.Hsia,Tulipa edulis and Iphigeniaindica were dried,grinded and sieved before their NIRDRS spectra were detected.Cluster analysis,convolution transform-visualization-similarity analysis were further employed to(identify) the NIRDRS of the ten herbal medicines. RESULTS: The zero-order NIRDRS spectra of the ten herbal medicines were too similar to be identified respectively.The cluster analysis helped differentiate the three fake "Fritillaria" from seven certified Fritillaria,but the identification between the certified Fritillaria remained dissatisfactory.The convolution transform-visualization-similarity analysis magnified and then quantified the minute differences between the certified Fritillaria,which significantly improved their identification. CONCLUSION: NIRDRS is a direct,non-destructive analytical method of solid samples.Coupled with information treatment techniques,DIRDRS provides a new identification method of crude traditional Chinese medicines.