Please wait a minute...

Frontiers of Optoelectronics

Front. Optoelectron.    2018, Vol. 11 Issue (3) : 296-305     https://doi.org/10.1007/s12200-018-0804-2
RESEARCH ARTICLE |
Discrete combination method based on equidistant wavelength screening and its application to near-infrared analysis of hemoglobin
Tao Pan1(), Bingren Yan1, Jiemei Chen2, Lijun Yao1()
1. Department of Optoelectronic Engineering, Jinan University, Guangzhou 510632, China
2. Department of Biological Engineering, Jinan University, Guangzhou 510632, China
Download: PDF(688 KB)   HTML
Export: BibTeX | EndNote | Reference Manager | ProCite | RefWorks
Abstract

A wavelength selection method for discrete wavelength combinations was developed based on equidistant combination-partial least squares (EC-PLS) and applied to a near-infrared (NIR) spectroscopic analysis of hemoglobin (Hb) in human peripheral blood samples. An allowable model set was established through EC-PLS on the basis of the sequence of the predicted error values. Then, the wavelengths that appeared in the allowable models were sorted, combined, and utilized for modeling, and the optimal number of wavelengths in the combinations was determined. The ideal discrete combination models were obtained by traversing the number of allowable models. The obtained optimal EC-PLS and discrete wavelength models contained 71 and 42 wavelengths, respectively. A simple and high-performance discrete model with 35 wavelengths was also established. The validation samples excluded from modeling were used to validate the three models. The root-mean-square errors for the NIR-predicted and clinically measured Hb values were 3.29, 2.86, and 2.90 g·L−1, respectively; the correlation coefficients, relative RMSEP, and ratios of performance to deviation were 0.980, 0.983, and 0.981; 2.7%, 2.3%, and 2.4%; and 4.6, 5.3, and 5.2, respectively. The three models achieved high prediction accuracy. Among them, the optimal discrete combination model performed the best and was the most effective in enhancing prediction performance and removing redundant wavelengths. The proposed optimization method for discrete wavelength combinations is applicable to NIR spectroscopic analyses of complex samples and can improve prediction performance. The proposed wavelength models can be utilized to design dedicated spectrometers for Hb and can provide a valuable reference for non-invasive Hb detection.

Keywords near-infrared (NIR) spectroscopy      equidistant combination-partial least squares (EC-PLS)      allowable model set discrete combination models      hemoglobin     
Corresponding Authors: Tao Pan,Lijun Yao   
Just Accepted Date: 10 April 2018   Online First Date: 08 May 2018    Issue Date: 31 August 2018
 Cite this article:   
Tao Pan,Bingren Yan,Jiemei Chen, et al. Discrete combination method based on equidistant wavelength screening and its application to near-infrared analysis of hemoglobin[J]. Front. Optoelectron., 2018, 11(3): 296-305.
 URL:  
http://journal.hep.com.cn/foe/EN/10.1007/s12200-018-0804-2
http://journal.hep.com.cn/foe/EN/Y2018/V11/I3/296
Service
E-mail this article
E-mail Alert
RSS
Articles by authors
Tao Pan
Bingren Yan
Jiemei Chen
Lijun Yao
Fig.1  NIR spectra of 300 human peripheral blood samples
wavelength/nm N F RMSEPM/(g·L−1) RP,M
400–2498 1050 6 4.39 0.952
400–1800 and 2100–2300 842 8 3.80 0.965
Tab.1  Parameters and prediction effects of PLS models with the entire scanning region and the unsaturated region
Fig.2  SG derivative spectra (d = 2, p = 2, 3, m = 31) of all samples at 400–1880 and 2100–2300 nm
Fig.3  RMSEPM of the local optimal models for each m distinguished by different orders of derivative
d p m F RMSEPM/(g·L−1) RP,M
0 6 41 10 3.24 0.973
1 2 17 11 3.18 0.976
2 2 31 11 3.14 0.977
3 3 49 12 3.26 0.971
Tab.2  Effects of the local optimal SG-PLS model in 400–1880 and 2100–2300 nm corresponding to each order of derivative
Fig.4  RMSEPM of the local optimal models with EC-PLS for (a) initial wavelength, (b) number of wavelength, and (c) number of wavelength gaps
method wavelengths models F RMSEPM/(g·L−1) RP,M
SG-PLS 400–1880 and 2100–2300 nm 11 3.14 0.977
EC-PLS I=1230 nm, N=71, G=6 7 2.67 0.983
DC-PLS N=42 8 2.55 0.985
N=35 8 2.62 0.984
Tab.3  Parameters and prediction effects of the SG-PLS model, optimal EC-PLS model, and two discrete combination models
Fig.5  First 200 values of RMSEPM with EC-PLS
Fig.6  RMSEPM and number of adopted wavelengths of the optimal discrete combination model corresponding to each allowable model set
Fig.7  Wavelength combinations of the optimal EC-PLS model and optimal discrete combination model labeled in the average spectrum of the samples
Fig.8  RMSEPM of each discrete combination model in the case of S=40
Fig.9  Relationship between the predicted and measured values of the validation samples for the optimal EC-PLS model
Fig.10  Relationship between the predicted and measured values for the selected discrete combination models with (a) N=42 and (b) N=35
method N RMSEP/(g·L−1) RP RRMSEP RPD
EC-PLS 71 3.29 0.980 2.7% 4.6
DC-PLS 42 2.86 0.983 2.3% 5.3
35 2.90 0.981 2.4% 5.2
Tab.4  Validation effects of the optimal EC-PLS model and two discrete combination models
1 Cantor C R, Schimmel P R. Biophysical Chemistry.New York: W. H. Freeman and Company, 1980
2 Anand I, McMurray J J V, Whitmore J, Warren M, Pham A, McCamish M A, Burton P B. Anemia and its relationship to clinical outcome in heart failure. Circulation, 2004, 110(2): 149–154
https://doi.org/10.1161/01.CIR.0000134279.79571.73 pmid: 15210591
3 Reeves J T, Leon-Velarde F. Chronic mountain sickness: recent studies of the relationship between hemoglobin concentration and oxygen transport. High Altitude Medicine & Biology, 2004, 5(2): 147–155
https://doi.org/10.1089/1527029041352090 pmid: 15265336
4 Weatherall D J, Edwards J A, Donohoe W T. Haemoglobin and red cell enzyme changes in juvenile myeloid leukaemia. British Medical Journal, 1968, 1(5593): 679–681
https://doi.org/10.1136/bmj.1.5593.679 pmid: 4966603
5 Machovec K A, Jaquiss R D B, Kaemmer D D, Ames W A, Homi H M, Walczak R J Jr, Lodge A J, Jooste E H. Cardiopulmonary bypass strategy for a cyanotic child with hemoglobin SC disease. The Annals of thoracic surgery, 2016, 101(6): 2373–2375
https://doi.org/10.1016/j.athoracsur.2015.09.046 pmid: 27211949
6 Messina A, Fogliani A M. Alexithymia in oncological patients: the role of hemoglobin, malignancy type and tumor staging. European Neuropsychopharmacology, 2011, 21(8): S174–S175
https://doi.org/10.1016/S0924-977X(11)70227-9
7 Phrommintikul A, Haas S J, Elsik M, Krum H. Mortality and target haemoglobin concentrations in anaemic patients with chronic kidney disease treated with erythropoietin: a meta-analysis. Lancet, 2007, 369(9559): 381–388
https://doi.org/10.1016/S0140-6736(07)60194-9 pmid: 17276778
8 Vályi-Nagy I, Kaffka K J, Jákó J M, Gönczöl E, Domján G. Application of near infrared spectroscopy to the determination of haemoglobin. Clinica Chimica Acta, 1997, 264(1): 117–125
https://doi.org/10.1016/S0009-8981(97)00085-5 pmid: 9267708
9 Lee Y, Lee S, In J, Chung S H, Yon J H. Prediction of plasma hemoglobin concentration by near-infrared spectroscopy. Journal of Korean Medical Science, 2008, 23(4): 674–677
https://doi.org/10.3346/jkms.2008.23.4.674 pmid: 18756056
10 Shan X, Chen L, Yuan Y, Liu C, Zhang X, Sheng Y, Xu F. Quantitative analysis of hemoglobin content in polymeric nanoparticles as blood substitutes using Fourier transform infrared spectroscopy. Journal of Materials Science: Materials in Medicine, 2010, 21(1): 241–249
https://doi.org/10.1007/s10856-009-3864-4 pmid: 19730991
11 Macknet M R, Allard M, Applegate R L 2nd, Rook J. The accuracy of noninvasive and continuous total hemoglobin measurement by pulse CO-Oximetry in human subjects undergoing hemodilution. Anesthesia and Analgesia, 2010, 111(6): 1424–1426
https://doi.org/10.1213/ANE.0b013e3181fc74b9 pmid: 21048100
12 Butwick A, Hilton G, Carvalho B. Non-invasive haemoglobin measurement in patients undergoing elective Caesarean section. British Journal of Anaesthesia, 2012, 108(2): 271–277
https://doi.org/10.1093/bja/aer373 pmid: 22116296
13 Jiang J H, Berry R J, Siesler H W, Ozaki Y. Wavelength interval selection in multicomponent spectral analysis by moving window partial least-squares regression with applications to mid-infrared and near-infrared spectroscopic data. Analytical Chemistry, 2002, 74(14): 3555–3565
https://doi.org/10.1021/ac011177u pmid: 12139068
14 Du Y P, Liang Y Z, Jiang J H, Berry R J, Ozaki Y. Spectral regions selection to improve prediction ability of PLS models by changeable size moving window partial least squares and searching combination moving window partial least squares. Analytica Chimica Acta, 2004, 501(2): 183–191
https://doi.org/10.1016/j.aca.2003.09.041
15 Chen H Z, Pan T, Chen J M, Lu Q P. Waveband selection for NIR spectroscopy analysis of soil organic matter based on SG smoothing and MWPLS methods. Chemometrics and Intelligent Laboratory Systems, 2011, 107(1): 139–146
https://doi.org/10.1016/j.chemolab.2011.02.008
16 Pan T, Chen Z H, Chen J M, Liu Z Y. Near-infrared spectroscopy with waveband selection stability for the determination of COD in sugar refinery wastewater. Analytical Methods, 2012, 4(4): 1046–1052
https://doi.org/10.1039/c2ay05856a
17 Pan T, Liu J M, Chen J M, Zhang G P, Zhao Y. Rapid determination of preliminary thalassaemia screening indicators based on near-infrared spectroscopy with wavelength selection stability. Analytical Methods, 2013, 5(17): 4355–4362
https://doi.org/10.1039/c3ay40732b
18 Pan T, Li M M, Chen J M, Xue H Y. Quantification of glycated hemoglobin indicator HbA1c through near-infrared spectroscopy. Journal of Innovative Optical Health Sciences, 2014, 7(4): 1350060
19 Chen J M, Ai T, Pan T, Yao L J, Xia F G. AO–MW–PLS method applied to rapid quantification of teicoplanin with near-infrared spectroscopy. Journal of Innovative Optical Health Sciences, 2017, 10(1): 1650029
20 Yao L J, Xu W Q, Pan T, Chen J M. Moving-window bis-correlation coefficients method for visible and near-infrared spectral discriminant analysis with applications. Journal of Innovative Optical Health Sciences, 2018, 11(2): 1850005
21 Pan T, Li M, Chen J. Selection method of quasi-continuous wavelength combination with applications to the near-infrared spectroscopic analysis of soil organic matter. Applied Spectroscopy, 2014, 68(3): 263–271
https://doi.org/10.1366/13-07088 pmid: 24666942
22 Yao L, Lyu N, Chen J, Pan T, Yu J. Joint analyses model for total cholesterol and triglyceride in human serum with near-infrared spectroscopy. Spectrochimica Acta, Part A: Molecular and Biomolecular Spectroscopy, 2016, 159: 53–59
https://doi.org/10.1016/j.saa.2016.01.022 pmid: 26827178
23 Xie J, Pan T, Chen J M, Chen H Z, Ren X H. Joint optimization of Savitzky-Golay smoothing models and partial least squares factors for near-infrared spectroscopic analysis of serum glucose. Chinese Journal of Analytical Chemistry, 2010, 38(3): 342–346
24 Savitzky A, Golay M J E. Smoothing and differentiation of data by simplified least squares procedures. Analytical Chemistry, 1964, 36(8): 1627–1639
https://doi.org/10.1021/ac60214a047
25 Guo H S, Chen J M, Pan T, Wang J H, Cao G. Vis-NIR wavelength selection for non-destructive discriminant analysis of breed screening of transgenic sugarcane. Analytical Methods, 2014, 6(21): 8810–8816
https://doi.org/10.1039/C4AY01833H
Related articles from Frontiers Journals
[1] Yimei HUANG, Hongqin YANG, Yuhua WANG, Shusen XIE, Zhouyi GUO, Songhao LIU. In vivo experimental study of optical characteristics of human acupuncture points[J]. Front Optoelec Chin, 2011, 4(2): 223-227.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed