基于浮动基准法的无创血糖检测实验应用研究

发布时间：2018-05-28 16:27

  本文选题：浮动基准法 + 葡萄糖浓度检测　； 参考：《光谱学与光谱分析》2012年03期

【摘要】：近红外无创血糖浓度测量面临着信号微弱、仪器信噪比低、血液中其他成分干扰以及人体生理环境不断变化等因素影响,致使血糖信号难以辨识这一难题。浮动基准法针对葡萄糖浓度变化对吸收和散射效应的影响,选择吸收效应和散射效应相互抵消的位置获得基准光谱,在吸收和散射效应引起最大综合影响的位置获得测量光谱,并采用基准光谱对测量光谱进行修正,从而消减样品背景状态变化以及仪器噪声、漂移等干扰。通过应用实验评估浮动基准法对提高模型预测精度和稳定性的有效性。通过对处理前后的预测结果进行比较分析得出:应用浮动基准法数据处理之后,交互验证均方根误差(RMSEP)最大改进比率34.7%,实验结果表明浮动基准法能够有效消减样品自身状态变化以及仪器噪声、漂移等干扰因素的影响,较大幅度地提高了模型的预测性能和稳定性,以期突破无创血糖浓度测量的障碍。
[Abstract]:Near-infrared noninvasive blood glucose concentration measurement is faced with weak signal, low signal-to-noise ratio of instrument, interference of other components in blood and changing physiological environment of human body, which makes it difficult to identify the blood glucose signal. In view of the effect of glucose concentration variation on absorption and scattering effect, the floating reference method selects the position where the absorption effect and scattering effect cancel each other to obtain the reference spectrum, and obtains the measurement spectrum at the position where the absorption and scattering effects cause the most comprehensive effect. The reference spectrum is used to modify the measurement spectrum to reduce the background state change of the sample and the interference of instrument noise and drift. The effectiveness of the floating benchmark method in improving the prediction accuracy and stability of the model is evaluated by experiments. By comparing and analyzing the prediction results before and after processing, it is concluded that the floating datum method is used to process the data. The experimental results show that the floating reference method can effectively reduce the influence of disturbance factors such as instrument noise and drift. The predictive performance and stability of the model are greatly improved in order to break the barrier of noninvasive blood glucose concentration measurement.
【作者单位】： 天津市生物医学检测技术及仪器重点实验室;天津市先石光学技术有限公司博士后创新实践基地;天津大学精密测试技术及仪器国家重点实验室;
【基金】：国家自然科学基金重点项目(60938002) 天津市生物医学检测技术及仪器重点实验室开放基金项目资助
【分类号】：G649.1
，

本文编号：1947405