近红外光谱法快速分析利巴韦林片剂的含量

Development of a near IR method for rapid determination of ribavirin tabletes

目的:建立测定利巴韦林片剂含量的近红外光谱(NIR)快速分析方法。方法:以全国不同企业生产的43批利巴韦林片建立校正集,采集其近红外漫反射光谱;通过偏最小二乘法进行回归,经内部交叉验证,建立校正模型,进而对预测集的17批利巴韦林片样品进行分析。定量模型的预处理方法为一阶导数和矢量归一化法,波长范围为11914.3～10371.5,9796.8～9091,6877.1～5565.7,5334.3～4022.9 cm^(-1)。结果:定量模型的浓度范围为0.37～0.75 mg·mg^(-1),RSD小于10%,内部交叉验证决定系数r=99.61,内部交叉验证均方差(RMSECV)为0.0119,外部验证预测均方差(RMSEP)为0.0198,预测值与真值的相关系数为0.9948。结论:该法具有快速、简便、结果准确的特点,可用于药品的快速检验。

Objective ：To set up an NIR method for rapid determination of Ribavirin tablets. Method：Collect 43 ribavirin tablet samples from different pharmaceutical factories in various locations in this country,and scan their NIR spectra. Then use Partial Least Square （PLS） to correlate these data and get the best - fitted model. The validation set was set up by 17 samples. Applying this model to predict the validation set and detect the samples. For quantitative analysis, the first derivative combining vector normalization method was used for pre - treatment of the spectra,the wavelength range is 11914.3 - 10371.5,9796. 8 -9091,6877.1 -5565.7,5334.3 -4022.9 cm^-1. Results ： For the quantitative determination, the concentration range was 0 - 0. 75 mg·mg ^- 1, the RSD was less than 10%. The determination coefficients r is 99.61 as indicated from the cross -validation, the true mean predictive error RMSECV （ root mean square error of cross validation） is 0.0119. The mean prediction error RESEP （ root mean square error of prediction） obtained from the validation is 0. 0198, the correlation coefficient of the true value and prediction value is 0. 9948. Conclusions ： The method is simple, rapid and accurate, and can be applied to drug rapid determination.