A Weak Signal Extraction Method for Human Blood Glucose Noninvasive Measurement using Near Infrared Spectroscopy

Background interference from optical absorption of matrix components, low spectral selectivity and low spectral sensitivity are the main interference factors for human blood glucose noninvasive measurement using near infrared (NIR) spectroscopy. In order to extract the weak glucose concentration information, a modified uninformative variable elimination (mUVE) method combined with successive projections algorithm (SPA) named as mUVE-SPA, is proposed. mUVE is used to eliminate matrix background and high-frequency noise by wavelet multi-resolution technology. SPA is followed to select variables with minimum colinearity by projection algorithm in a vector space. The proposed method was applied in two NIR spectra data sets (plasma samples experiment in vitro and human blood glucose noninvasive measurement experiment in vivo) respectively. The performance and adaptability of the proposed strategy were discussed. The results indicate that the proposed hybrid method can give an alternative path to extract weak glucose information and yield more parsimonious models with higher precision.

[1]  T. B. Blank,et al.  Noninvasive prediction of glucose by near-infrared diffuse reflectance spectroscopy. , 1999, Clinical chemistry.

[2]  Steven D. Brown,et al.  Wavelet analysis applied to removing non‐constant, varying spectroscopic background in multivariate calibration , 2002 .

[3]  Wang Yan,et al.  Application of Digital Fourier Filtering Pretreatment Method to Improving Robustness of Multivariate Calibration Model in Near Infrared Spectroscopy , 2007 .

[4]  Carlos Eduardo Ferrante do Amaral,et al.  Current development in non-invasive glucose monitoring. , 2008, Medical engineering & physics.

[5]  W. Cai,et al.  A variable selection method based on uninformative variable elimination for multivariate calibration of near-infrared spectra , 2008 .

[6]  G. W. Small,et al.  Determination of glucose in a biological matrix by multivariate analysis of multiple band-pass-filtered Fourier transform near-infrared interferograms. , 1997, Analytical chemistry.

[7]  Kexin Xu,et al.  Non-invasive glucose sensing with near-infrared spectroscopy enhanced by optical measurement conditions reproduction technique , 2005 .

[8]  R. Yu,et al.  An ensemble of Monte Carlo uninformative variable elimination for wavelength selection. , 2008, Analytica chimica acta.

[9]  D. Massart,et al.  Elimination of uninformative variables for multivariate calibration. , 1996, Analytical chemistry.

[10]  Riccardo Leardi,et al.  Application of genetic algorithm–PLS for feature selection in spectral data sets , 2000 .

[11]  Maria Fernanda Pimentel,et al.  Robust modeling for multivariate calibration transfer by the successive projections algorithm , 2005 .

[12]  M A Arnold,et al.  Phantom glucose calibration models from simulated noninvasive human near-infrared spectra. , 1998, Analytical chemistry.

[13]  Kirsten E. Kramer,et al.  Robust absorbance computations in the analysis of glucose by near-infrared spectroscopy , 2007 .

[14]  Li-na Li,et al.  APPLICATION OF DENOISING AND BACKGROUND ELIMINA TION BASED ON WAVELET TRANSFORM TO BLOOD GLUCOSE NONINVASIVE MEASUREMENT OF NEAR INFRARED SPECTROSCOPY: APPLICATION OF DENOISING AND BACKGROUND ELIMINA TION BASED ON WAVELET TRANSFORM TO BLOOD GLUCOSE NONINVASIVE MEASUREMENT OF NEAR INFRARED SPECTROSCO , 2009 .

[15]  J. I. Arana,et al.  Sugar Determination in Grapes Using NIR Technology , 2001 .

[16]  R. Leardi,et al.  Genetic algorithms applied to feature selection in PLS regression: how and when to use them , 1998 .

[17]  APPLICATION OF DENOISING AND BACKGROUND ELIMINATION BASED ON WAVELET TRANSFORM TO BLOOD GLUCOSE NONINVASIVE MEASUREMENT OF NEAR INFRARED SPECTROSCOPY , 2009 .

[18]  Dong Wang,et al.  Successive projections algorithm combined with uninformative variable elimination for spectral variable selection , 2008 .

[19]  Qing-bo Li,et al.  PRETREATMENT METHOD RESEARCH OF NEAR-INFRARED SPECTRA IN BLOOD COMPONENT NON-INVASIVE MEASUREMENT , 2009 .

[20]  C. Jarén,et al.  Selection of the Temperature in the Sugar Content Determination of Kiwi Fruit , 2005 .

[21]  X. Shao,et al.  A background and noise elimination method for quantitative calibration of near infrared spectra , 2004 .

[22]  M. C. U. Araújo,et al.  The successive projections algorithm for variable selection in spectroscopic multicomponent analysis , 2001 .

[23]  P. Arnal,et al.  WHITE ASPARAGUS HARVEST DATE DISCRIMINATION USING NIRS TECHNOLOGY , 2007 .

[24]  D. Massart,et al.  Application of wavelet transform to extract the relevant component from spectral data for multivariate calibration. , 1997, Analytical chemistry.

[25]  G. W. Small,et al.  Evaluation of nonlinear model building strategies for the determination of glucose in biological matrices by near-infrared spectroscopy , 1999 .