近红外光谱结合偏最小二乘法测定福多斯坦成分含量

采用近红外漫反射光谱分析技术和偏最小二乘法对福多斯坦药物的有效成分进行定量分析测定,采集48个不同比例的福多斯坦样品近红外漫反射光谱,用一阶导数的光谱预处理方法,结合偏最小二乘法建立福多斯坦的定量分析模型.结果显示:交互验证均方根误差为0.003 57,相关系数R为0.994 77,预测均方根误差为0.003 89,平均回收率为99.63%(n=8),结果表明,用近红外光谱分析技术联合偏最小二乘法对福多斯坦进行定量分析结果准确可靠,方法简便快速.

Determination Active Ingredients of Fudosteine Using Near Infrared Diffuse Reflectance Spectroscopy Combined with Partial Least Squares

Near infrared reflectance spectroscopy （NIRS） and partial least squares （PLS） method were applied for the fast, non-destructive, quantitative analysis of fudosteine. The fudosteine quantitative analysis model was established by 48 different series near-infrared reflectance spectroscopy of fudosteine samples, combined with a derivative of the spectral pretreatment method and partial least squares. The cross-validation root mean square error was 0.003 57 and correlation coefficient R was 0.994 77; root mean square error of prediction was 0.003 89. The average recovery was 99.63% （n=8）. It is shown that near-infrared spectroscopy technology for fudosteine quantitative analysis is accurate and reliable. The method is simple and rapid, and it may be extended to industrial on-line analysis of such samples.