双光路测量结合净信号处理方法在葡萄糖浓度定量分析中的应用

发布时间：2018-01-04 17:34

本文关键词：双光路测量结合净信号处理方法在葡萄糖浓度定量分析中的应用　出处：《光谱学与光谱分析》2017年09期 　论文类型：期刊论文

【摘要】：在近红外无创血糖测量中,由于人体血糖浓度变化引起的光强变化非常小,光谱易受到测量仪器以及被测对象自身生理变化的影响。背景变化的影响在测量过程中是无法消除的,只能通过有效的方法来降低这种变化带来的影响,其中选择光学性质与待测物相似,通过扣除相似参考物的光谱来修正该影响是目前常用的手段之一。为减小背景变化对光谱信号的干扰,准确地提取葡萄糖的特异性信号,提出了一种双光路测量结合净信号(NAS)处理的参考测量方法,采用双光路双检测器系统同步测量参考物和被测样品的光谱,然后基于参考物的光谱构建噪声空间计算样品光谱中的葡萄糖的净信号,并在纯吸收和强散射介质中开展了葡萄糖的建模实验,最后,结合二维相关光谱(2DCOS)分析技术和偏最小二乘回归(PLSR)模型对该方法的有效性进行了评价。二维相关分析的结果表明,与直接扣除参考背景的方式相比,净信号处理能突出葡萄糖的特异信号;而PLSR模型的预测结果表明,双光路的效果要优于单光路扣除背景光谱,糖在水溶液和20%-Intralipid溶液中的预测均方根误差(RMSEP)分别降低了35.25%和37.95%;而双光路结合净信号处理后,两种溶液的预测精度又分别提高了26.11%和14.84%。这表明,双光路测量和净信号分析相结合的方法能更有效地提取葡萄糖的特征信息,提高模型的预测精度,从而为无创血糖检测的实现提供更多可能。
[Abstract]:In the near infrared noninvasive blood glucose measurement, the light intensity change caused by the change of blood glucose concentration is very small. Spectrum is easy to be affected by the measurement instrument and the object itself physiological changes. Background changes in the measurement process can not be eliminated, only through effective methods to reduce the impact of this change. The optical property is similar to that of the object to be tested, and it is one of the commonly used methods to correct the influence by deducting the spectrum of the similar reference, in order to reduce the interference of the background change to the spectral signal. In this paper, the specific signal of glucose is extracted accurately, and a method of reference measurement is proposed, which combines double optical path measurement with net signal NASN processing. The spectra of the reference and the sample are measured synchronously by using the dual optical path and detector system, and then the noise space is constructed based on the spectrum of the reference to calculate the net signal of glucose in the spectrum of the sample. The modeling experiments of glucose were carried out in pure absorption and strong scattering media. The effectiveness of the method is evaluated by using 2D correlation spectroscopy and partial least squares regression model. The results of two-dimensional correlation analysis show that the proposed method is effective. Compared with the direct subtraction of reference background, the net signal processing can highlight the specific signal of glucose. The prediction results of PLSR model show that the effect of double optical path is better than that of single optical path subtraction of background spectrum. The root-mean-square error (RMS) of sugar in aqueous solution and 20% -Intralipid solution decreased by 35.25% and 37.95, respectively. However, the prediction accuracy of the two solutions was improved by 26.11% and 14.84, respectively, after the combination of two optical paths with net signal processing. The combination of dual optical path measurement and net signal analysis can effectively extract the characteristic information of glucose and improve the prediction accuracy of the model, thus providing more possibilities for the realization of non-invasive blood glucose detection.
【作者单位】：天津大学精密测试技术及仪器国家重点实验室;
【基金】：国家自然科学基金项目(81471698,81401454) 国家高技术研究发展计划(2012AA022602) “111”计划项目(B07014) 天津市重大科技专项(13ZCDGX01161)资助
【分类号】：O657.33;R446.11
【正文快照】： 引言现代光谱技术在人体成分定性定量分析领域得到广泛应用,光学方法被认为是最有前景的无创血糖测量方法[1-2]。近红外光由于具有较强的组织穿透性、相对高的能量强度以及高灵敏度的光电检测器,成为了比较理想的生物检测波段[3]。但葡萄糖在近红外区域的吸收较弱,且与水、脂

【相似文献】