论文部分内容阅读
建立了同时测定利福平片、异烟肼片、吡嗪酰胺片、异福片和异福酰胺片5种抗结核药物中的利福平(RMP)、异烟肼(INH)和吡嗪酰胺(PZA)含量的新方法,应用径向基神经网络(RBFNN)建立5种抗结核片剂药物样品的近红外光谱(NIRS)与其中RMP、INH和PZA含量间相关模型。模型以交互验证均方根误差(RMSECV)为评价标准,选择最有效光谱区域、对网络结构参数和扩展常数进行优化,得到最优定量分析模型。最优模型的RMSECV分别为0.0127、0.0104、0.0078,应用最优模型对预测集样品中RMP、INH和PZA含量进行预测,预测均方根误差(RMSEP)分别为0.0125、0.0109、0.0103。内部交互验证和外部验证均表明,该方法具有较高的准确度,能够满足5种抗结核药物生产中RMP、INH和PZA的同时检测精度要求。
To establish a method for the simultaneous determination of rifampicin (RMP), isoniazid (INH) and pyrazine in 5 kinds of antituberculosis drugs including rifampin, isoniazid, pyrazinamide, (NIRS) and the content of RMP, INH and PZA in drug samples of five anti-TB tablets using Radial Basis Function Neural Network (RBFNN). The RMSECV was used as the evaluation criterion to select the most efficient spectral region and optimize the network structure parameters and the extended constants to obtain the optimal quantitative analysis model. The RMSECV of the optimal model was 0.0127, 0.0104 and 0.0078, respectively. The optimal model was used to predict the contents of RMP, INH and PZA in the predictive set. The root mean square error of prediction (RMSEP) were 0.0125, 0.0109 and 0.0103 respectively. Both internal and external validation showed that this method has high accuracy and can meet the requirements of simultaneous detection of RMP, INH and PZA in five kinds of anti-TB drugs.