【摘要】 利用傅里叶变换红外光谱(FTIR)和衰减全反射(ATR)技术,建立了人体血液血红蛋白(HGB)的快速定量分析方法。采集人体全血样品8个,用常规化学方法测定血红蛋白浓度作为光谱校正模型的参考化学值。每个样品用蒸馏水溶血,分别配制成2倍、3倍、4倍、5倍、6倍稀释的溶血液样品,和全血样品一起共得到6组48个样品用于光谱测定。基于11点Savitzky-Golay平滑的二阶导数光谱,采用MLR分析和交叉检验对每组样品分别建立定量模型。每组分别采用全谱(4000—600cm-1)、指纹领域(1800—800cm-1)建立MLR模型,并建立每个波数的一元线性回归模型,从中遴选效果最好的单点模型。结果表明,每组样品的最优单点模型都有良好的预测效果。直接测定的全血样品组的最优单点模型的采用波数、交叉检验均方误差(RMSECV)、相对交叉检验均方误差(RRMSECV)、预测相关系数(Rp)分别为1759cm-1、4.9g/L、3.6%、0.825。更多还原

【Abstract】 The rapid quantitative analysis method of the human blood hemoglobin (HGB) was developed by a Fourier transform infrared (FTIR) spectrometer and attenuated total reflection (ATR) techniques. Eight samples of the human blood were collected,hemoglobin concentration was measured by the conventional chemical method,and it was as the reference chemical value of the calibration model for the spectrum. Each sample with distilled water hemolysis,were configured to 2 times,3 times,4 times,5 times,6 times dilute hemolytic solution sample respectively,and the whole blood samples had been together a total of 6 groups of 48 samples for spectrometry. Based on the second derivatives of the spectra were calculated by using 11 points Savitzky-Goray smoothing,the MLR analysis and the cross-validation,the quantitative models of each sample group were established respectively. To each group the MLR model was established by using the whole region (4000—600cm-1) and the fingerprint region (1800—800cm-1) respectively.The linear regression model of the each wavenumber was established,and the best single-point model was selected from above models. The results showed that the best single-point models of each sample group had a good prediction effect. To the best single-point model for the whole blood sample group which by direct determination,the adopting wavenumber,the root mean square error cross validation (RMSECV),the relative root mean square error cross validation (RRMSECV),the prediction correlation coefficient (Rp) were 1759cm-1,4.9g/L,3.6%,0.825,respectively.更多还原