论文部分内容阅读
利用傅立叶变换红外光谱(FTIR)和衰减全反射(ATR)技术,建立血清葡萄糖的快速定量分析方法。基于模型预测效果,提出了一种联合最佳单波段的补偿波段建模方法。在全谱(4000~600 cm-1)区间,根据移动窗口偏最小二乘法得到人体血清葡萄糖的最佳单个连续波段后,另外选择一个波段进行补偿建模,优选出最佳补偿波段。结果表明,这种补偿波段建模方法建模效果不仅仅远好于全谱,也好于最佳单个波段。用独立的实际样品对模型进行测试,补偿模型的预测均方根偏差(RMSEP)和预测相关系数(Rp)分别为0.782 mmol/L和0.944。
The rapid quantitative analysis method of serum glucose was established by Fourier Transform Infrared Spectroscopy (FTIR) and Attenuated Total Reflection (ATR). Based on the model prediction effect, a method of modeling the compensation band of the optimal single band is proposed. In the whole spectrum (4000 ~ 600 cm-1), the best single continuous wave band of human serum glucose is obtained by moving window partial least squares method, and another wave band is selected for compensation modeling. The results show that the modeling effect of this method is not only better than the full spectrum, but also better than the best single band. The independent real samples were used to test the model. The root mean square error of prediction (RMSEP) and predicted correlation coefficient (Rp) of the compensation model were 0.782 mmol / L and 0.944, respectively.