Rapid Discrimination of Dendrobium officinale from Different Geographical Origins Using Chemometric Methods

With the growing market demand for Dendrobium officinale, consumers are increasingly concerned about its quality. Distinct geographical environments lead to significant variations in the content of bioactive compounds such as polysaccharides and alkaloids, which in turn result in considerable differences in market value. However, the abundance of brands and uneven product quality, coupled with the prevalence of counterfeits, pose challenges to quality assurance. Chemometrics, with its strength in analyzing large datasets, has been widely applied in the geographical traceability of medicinal materials. In this study, near-infrared (NIR) spectroscopy combined with chemometric approaches including SVM, PLS-DA, and KNN was employed to develop origin discrimination strategies for D. officinale. By optimizing the training set proportion, the results revealed that the traceability model constructed using the DT preprocessing method in combination with KNN exhibited the best performance, achieving 100% classification accuracy. These findings provide a highly accurate and efficient technique for the geographical authentication of D. officinale.