Share Email Print

Proceedings Paper

Data correction method based on the floating reference point for the noninvasive glucose sensing
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

As an effective noninvasive method for glucose doesn't come into clinical realization due to the weakness of glucose unique signal and complexity of background noise, a method based on a floating reference point and a measuring point, where the diffuse reflectance intensity is insensitive and most sensitive to the variation of glucose concentration, respectively, is applied. In this paper, the data processing method based on the information of reference point was investigated to improve the precision of glucose sensing. The diffuse reflectance of intralipid solution with different glucose concentration in different source-detector distances was obtained by Monte-Carlo simulation. And the radial region selection of reference position and measuring position were discussed. Then in order to simulate the actual measurement condition, the random noise and linear drift were added on the simulated spectra. And the spectra in the proper measuring region corrected by that in the reference point were used to build the multivariate model. Further more, the corresponding optical probe was designed according to the distribution of light intensity in the radial distance and an in vitro experiment about intralipid solution with different glucose concentration was conducted to verify the effect of the data correction based on the information from the reference point. Results showed that, three different measuring regions should be determined in the wavelength of 1100nm-1700nm according to the wavelength characteristic of reference point. And the measuring region should be about 0.2-0.3mm far away from the reference region. For the simulation and in vitro experiment, after the correction by the information from the reference point, the prediction error for glucose was reduced by 46.2% and 23.2%, respectively.

Paper Details

Date Published: 20 February 2009
PDF: 9 pages
Proc. SPIE 7186, Optical Diagnostics and Sensing IX, 71860O (20 February 2009); doi: 10.1117/12.808738
Show Author Affiliations
Rong Liu, Tianjin Univ. (China)
Bin He, Tianjin Univ. (China)
Kexin Xu, Tianjin Univ. (China)
Yue Yang, Tianjin Univ. (China)

Published in SPIE Proceedings Vol. 7186:
Optical Diagnostics and Sensing IX
Gerard L. Coté, Editor(s)

© SPIE. Terms of Use
Back to Top