可见-近红外光谱测定血红蛋白的等效波段选择

刘振尧; 潘涛

doi:10.3788/OPE.20122010.2170

您当前的位置：

首页 >

文章列表页 >

可见-近红外光谱测定血红蛋白的等效波段选择

现代应用光学 | 更新时间：2020-08-12

- 可见-近红外光谱测定血红蛋白的等效波段选择
- Equivalent waveband selection of VIS-NIR spectroscopic measurement for hemoglobin
- 光学精密工程 2012年20卷第10期页码：2170-2175
- 作者机构：
  
  暨南大学光电信息与传感技术广东普通高校重点实验室,广东广州,510632
- 作者简介：
- 基金信息：
  
  国家自然科学基金资助项目(No.61078040)();广东省科技计划资助项目(No.2009B030801239,No.2009A030301002)()
- DOI：10.3788/OPE.20122010.2170
  中图分类号： O657.33
- 收稿日期：2012-05-29，
  
  修回日期：2012-07-05，
  
  纸质出版日期：2012-10-10
- 稿件说明：
移动端阅览
刘振尧, 潘涛. 可见-近红外光谱测定血红蛋白的等效波段选择[J]. 光学精密工程, 2012,20(10): 2170-2175

LIU Zhen-yao, PAN Tao. Equivalent waveband selection of VIS-NIR spectroscopic measurement for hemoglobin[J]. Editorial Office of Optics and Precision Engineering, 2012,20(10): 2170-2175
刘振尧, 潘涛. 可见-近红外光谱测定血红蛋白的等效波段选择[J]. 光学精密工程, 2012,20(10): 2170-2175 DOI： 10.3788/OPE.20122010.2170.

LIU Zhen-yao, PAN Tao. Equivalent waveband selection of VIS-NIR spectroscopic measurement for hemoglobin[J]. Editorial Office of Optics and Precision Engineering, 2012,20(10): 2170-2175 DOI： 10.3788/OPE.20122010.2170.

摘要

将可见-近红外光谱和改进的移动窗口偏最小二乘(MWPLS)方法应用于人类血红蛋白(HGB)无试剂快速检测的高精度波段优选。为了避免模型评价失真

提出了一种新的模型评价体系。首先

从全体205个样品中随机抽取70个作为检验集

余下的135个作为建模集

并划分为具有相似性的定标集(80个样品)和预测集(55个样品)共50次;其次

对每一次划分都分别建模和优化

使得模型具有稳定性;最后

利用检验集对优选出的模型进行再次检验。实验结果表明:可见-短波近红外波段400~1100 nm可以作为人体全血HGB的信息波段;进一步采用MWPLS方法从400~1100 nm中选出全局最优波段为492~890 nm

并得到包含77个等效波段的模型空间。以492~890 nm为例

检验效果预测均方根偏差(V-SEP)、预测相关系数(V-RP)和相对预测均方根偏差(V-RSEP)分别为2.58 g L

-1

、0.988和1.97%

得到的样品的HGB预测值与临床实测值吻合精度很高

可望应用于临床。

Abstract

The VIS-NIR spectroscopy combined with the improved Moving Window Partial Least-square (MWPLS) method was applied to a high accurate waveband selection for the rapid no-reagent determination of Hemoglobin (HGB) in human whole blood. A new modeling evaluation system was proposed to avoid the evaluation distortion. First

seventy samples were randomly selected from a total of 205 samples as the validation set

the remaining 135 samples were used as the modeling set

and the modeling set was divided into similar calibration (80 samples) and prediction (55 samples) sets for a total of 50 times.Then

modeling and optimization were performed in each division to get stable model. Finally

the optimized model was validated again using the validation set. Experimental results indicate that the VIS-short NIR region 400-1 100 nm can be used as the information waveband of HGB in human whole blood

the global optimal waveband 492-890 nm is further selected from 400-1100 nm with MWPLS method

and a model space including 77 equivalent wavebands is obtained. By taking the 492-890 nm for an example

validation effects V-SEP

V-RP

and V-RSEP are 2.58 g L

-1

0.988

and 1.97%

respectively. It concludes that HGB prediction values of the samples are highly close to the clinic measured values

which may be used in clinical diagnosis.

关键词

Keywords

references

陈星旦. 近红外光谱无创生化检验的可能性[J]. 光学精密工程,2008,16(5):759-763. CHEN X D. Possibility of noninvasive clinical biochemical examination by near infrared spectroscopy [J]. Opt. Precision Eng., 2008, 16(5):759-763. (in Chinese)[2] 谢军,潘涛,陈华舟,等. 血糖近红外光谱分析的Savitzky-Golay平滑模式与PLS因子数的联合优选[J]. 分析化学,2010, 38(3):342-346. XIE J, PAN T, CHEN H ZH, et al.. Joint optimization of Savitzky-golay smoothing models and partial least squares factors for near-infrared spectroscopic analysis of serum glucose [J]. Chinese Joumal of Analytical Chemistry, 2010, 38(3):342-346. (in Chinese)[3] 陈洁梅, 潘涛, 陈星旦. 二阶导数光谱预处理在用FTIR/ATR方法定量测定葡萄糖-6-磷酸和果糖-6-磷酸中的应用[J]. 光学精密工程, 2006, 14(1):127. CHEN J M, PAN T, CHEN X D. Application of second derivative spectrum prepares in quantification measuring glucose-6-phosphate and fructose-6-phosphate using a FTIR/ATR method [J]. Opt. Precision. Eng., 2006,14 (1):127. (in Chinese)[4] ISTVAN V N, KAROLY J K, JANOS M J, et al.. Application of near infrared spectroscopy to the determination of haemoglobin [J]. Clinica Chimica Acta, 1997, 264(1):117-125.[5] LEE Y, LEE S, IN J Y, et al.. Prediction of plasma hemoglobin concentration by near infrared spectroscopy [J]. J Korean Med Sci, 2008, 23(4): 674-677.[6] SHAN X Q, CHEN L G, YUAN Y, et al.. Quantitative analysis of hemoglobin content in polymeric nanoparticles as blood substitutes using Fourier transform infrared spectroscopy [J]. J Mater Sci, 2010, 21(1): 241-249.[7] 曹璞, 潘涛, 陈星旦. 小型近红外玉米蛋白质成分分析仪器设计的波段选择[J]. 光学精密工程, 2007, 15(12):1952-1958. CAO P, PAN T, CHEN X D. Choice of wave band in design of minitype near-infrared corn protein content analyzer [J]. Opt. Precision Eng.,2007, 15(12):1952-1958. (in Chinese)[8] 黄富荣, 潘涛, 张甘霖,等. 应用近红外漫反射光谱快速测定土壤锌含量[J]. 光学精密工程, 2010, 18(3): 586-592. HUANG F R, PAN T, ZHANG G L, et al.. Rapid measurement of zinc contents in soils by near-infrared diffuse reflectance spectroscopy [J]. Opt. Precision Eng., 2010, 18(3): 586-592. (in Chinese)[9] PAN T, HASHIMOTO A, KANOU M. Development of a quantification system of ionic dissociative metabolites using an FT-IR/ATR method [J]. Bioprocess and Biosystems Engineering, 2003, 26(2):133-139.[10] JIANG J H, BERRY R J, SIESLER H W,et al.. Wavelength interval selection in multicomponent spectral analysis by moving window partial least-squares regression with applications to mid-infrared and near-infrared spectroscopic data [J]. Analytical Chemistry, 2002, 74(14):3555-3565.[11] CHEN H Z, PAN T, CHEN J M, et al.. Waveband selection for NIR spectroscopy analysis of soil organic matter based on SG smoothing and MWPLS methods [J]. Chemometrics and Intelligent Laboratory Systems, 2011, 107(1):139-146.[12] PAN T, CHEN Z H, CHEN J M, et al.. Near-infrared spectroscopy with waveband selection stability for the determination of COD in sugar refinery wastewater [J]. Analytical Methods, 2012, 4(4):1046-1052.

浏览量

312

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

光电容积脉搏波无创测量人体血液成分的评估

结合近红外光谱法的肝血流参数测量

血红蛋白浓度的光学测量方法与装置研究

探测海洋目标的光学遥感器工作波段选择