Abstract
Variable selection methods can simplify the modelling process and improve the accuracy of models constructed for rapid monitoring using near-infrared (NIR) spectroscopy. In this study, NIR spectroscopy was applied in combination with chemometrics to determine the icariin, salvianolic acid B, and puerarin content, as well as the relative density through the concentration process of a Yangxinshi tablet (YXST) water extraction solution. Partial least squares with several variable selection algorithms were used for the modelling. Variable selection methods, including synergy interval partial least squares (Si-PLS), a genetic algorithm, competitive adaptive reweighted sampling (CARS), and their combination, were comparatively applied to calibrate the regression model. The performances of the models obtained were systematically evaluated according to the correlation coefficients of prediction (Rp), the relative standard error of prediction (RSEP), and the residual predictive deviation (RPD). For icariin, salvianolic acid B, and puerarin, the Si-PLS model showed optimal results. The Rp, RSEP, and RPD of Si-PLS were 0.9905, 7.84%, and 7.38 for icariin; 0.9781, 9.48%, and 4.80 for salvianolic acid B; and 0.9668, 12.81%, and 3.92 for puerarin, respectively. For the relative density, the CARS-Si-PLS acquired the best prediction results. The Rp, RSEP, and RPD of CARS-Si-PLS were 0.9725, 0.47%, and 4.31 for the relative density, respectively. This study demonstrated that variable selection methods can simplify the modelling process and improve the accuracy of the models. In addition, the methods developed in this study are suitable for a quality monitoring of the concentration process of YXST water extraction solution, and they provide reference for real-time monitoring of the concentration processes used in the preparation of other types of traditional Chinese medicine.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
