论文部分内容阅读
目的:采用近红外光谱技术定量分析葡萄糖注射液中葡萄糖的含量,用于含葡萄糖制剂的快速检测及相关产品中间体质量控制。方法:采用近红外光谱法进行建模,样品来源于企业生产的不同含量的葡萄糖注射液、实验室自制的不同浓度的葡萄糖溶液样品。选择液体测样附件,用偏最小二乘法建立定量模型,用各种含葡萄糖的溶液制剂进行验证,并考察模型对环境温度的耐受性。结果:建立的近红外偏最小二乘(PLS)定量模型,相关系数达到0.999 3,交叉验证均方根误差(RMSECV)为0.077 4,验证试验中,各类含葡萄糖的液体制剂(不含氯化钠)预测的结果和实验室测定的结果一致,在20~35℃温度范围内,预测偏差小于5%。结论:建立的模型可以准确地预测葡萄糖类产品中葡萄糖的含量,适用于大输液类生产厂家进行葡萄糖类产品中间体质量控制和成品的含量检测。
OBJECTIVE: To quantitatively analyze glucose in glucose injection by near-infrared spectroscopy for the rapid detection of glucose-containing preparations and the quality control of intermediates in related products. Methods: Near-infrared spectroscopy was used to model the samples from different concentrations of glucose injection and lab-made dextrose solutions of different concentrations. Select the liquid sample attachment, using the partial least squares method to establish a quantitative model, with a variety of glucose-containing solution preparations to verify and study the model of the ambient temperature tolerance. Results: Near infrared partial least squares (PLS) quantitative model was established, the correlation coefficient was 0.999 3, and the root mean square error of cross validation (RMSECV) was 0.077 4. In the validation test, various glucose-containing liquid formulations Sodium) predicted results consistent with the laboratory test results in the temperature range of 20 ~ 35 ℃, the predicted deviation of less than 5%. CONCLUSION: The established model can accurately predict the glucose content in glucose products and is suitable for the detection of intermediate quality of glucose products and finished products in large transfusion manufacturers.