摘要
目的探讨应用近红外光谱技术结合化学计量学算法对不同产地丹参及其伪品的鉴别效果。方法收集陕西、山东、四川、河北等9个产地的265批丹参样品,以及市场常见的续断、牛蒡根和甘西鼠尾3种丹参伪品61批,利用SabIR漫反射光纤探头采集样品粉末的近红外漫反射光谱,并采用多元散射校正、标准正则变换、S-G平滑、Norris平滑、一阶微分及二阶微分等对光谱进行预处理。应用TQ Analyst分析系统,采用主成分分析结合马氏距离的判别分析方法,确定主成分数,建立丹参产地鉴别和真伪鉴别模型,并采用四重交叉验证测试模型预测性能。结果采用MSC+一阶+Norris的光谱预处理方式,确定最佳主成分数为10,建立模型并经过验证,产地鉴别准确率达到98%以上,真伪鉴别准确率达到100%。结论该方法可快速、准确鉴别丹参产地及其伪品,是对现有丹参鉴别方法的科学补充。
Objective To explore the effect of near infrared reflectance spectroscopy combined with chemometrics algorithm on the identification of Salvia miltiorrhiza and its counterfeits from different places.Methods A total of 265 batches of Salvia miltiorrhiza samples were collected from 9 producing areas in Shaanxi,Shandong,Sichuan and Hebei,and 61 batches of three common counterfeit Salvia miltiorrhiza products were collected.The near infrared diffuse reflection spectrum of the sample powder was collected by SabIR diffuse reflection fiber probe.Multiple scattering correction,standard canonical transformation,S-G smoothing,Norris smoothing,first-order differentiation and second-order differentiation were used to preprocess the spectra.TQ Analyst analysis system was applied,principal component analysis combined with Mahalanobis distance discriminant analysis method was used to determine the number of principal components,establish the origin identification and authenticity identification models of Salvia miltiorrhiza,and four-fold cross-validation was used to test the prediction performance of the model.Results Using MSC+first-order+Norris spectral preprocessing method,the optimum principal component number was determined to be 10,the model was established and verified,and the accuracy of origin identification was more than 98%,and the accuracy of authenticity identification was 100%.Conclusion This method can quickly and accurately identify the origin of Salvia miltiorrhiza and its counterfeit products,which is a scientific supplement to the existing methods.
作者
张延莹
程子航
ZHANG Yanying;CHENG Zihang(Tianjin Tasly Modern Chinese Medicine Resources Co.,LTD,Tianjin300410,China;School of Mathematics and Statistics,Nanjing University of Science and Technology,Jiangsu Province,Nanjing210094,China)
出处
《中国当代医药》
CAS
2024年第27期4-8,共5页
China Modern Medicine
基金
天津市智能制造专项资金项目(20192083)。
关键词
近红外光谱
丹参
主成分分析
判别分析
Near infrared reflectance spectroscopy
Salvia miltiorrhiza
Principle component analysis
Discriminative analysis
