摘要
目的建立枇杷叶药材中浸出物含量的近红外定量模型,快速测定枇杷叶中浸出物含量。方法用热浸法测定146批枇杷叶中浸出物含量,同时采集其近红外光谱数据,采用多元校正散射法加二阶导数法预处理,结合偏最小二乘法建立枇杷叶中浸出物含量的定量模型,采用28批验证集样品对模型进行验证。结果所建立的校正模型,内部交叉验证决定系数达到0.984 28,校正均方差为0.387,预测均方差为0.659,内部交叉验证均方差为1.836 52。结论该近红外定量模型稳定,准确可靠,可快速对枇杷叶中浸出物含量进行测定。
Objective To develop a method for the determination of ethanol extracts content in Eriobotrya japonica bynear-infrared (NIR) spectroscopy. Methods The ethanol extract contents of 146 samples were determined by hot waterextraction and ethanol precipitation method. NIR spectrums were measured. The multivariate calibration method and Secondderivative test were used for spectral pre-processing. Calibration model of the ethanol extract content was established by the partialleast squares regression analysis. Twenty-eight batches of samples were tested to validate the model. Results The correlationcoefficients, the Root-mean-square error of calibration, the Root-mean-square error of prediction and the Root-mean-square errorof cross-validation of the calibration model for ethanol extract content were 0. 984 80, 0. 381, 0. 661 and 1. 836 52, respectively. Conclusion The method is steady, accurate, and can be used to predict ethanol extract contents of Eriobotrya japonicarapidly.
出处
《医药导报》
CAS
北大核心
2014年第2期232-236,共5页
Herald of Medicine
基金
广东省科技计划项目(2009B030801044)
关键词
枇杷叶
浸出物含量
近红外光谱
Eriobotrya japonica
Ethanol extracts content
Near-infrared spectrum