METHOD OF COMPONENT CONCENTRATION DETECTION BASED ON REFERENCE WAVELENGTH

• US 20110131021A1
• Filed: 06/03/2009
• Published: 06/02/2011
• Est. Priority Date: 06/06/2008
• Status: Active Grant

• Current Assignee: Tianjin Sunrise Technologies Development Co., Ltd.
• Original Assignee: Tanjin Sunshine Qptics Technologies Company Limited
• Inventors: Chen, Wenliang, Xu, Kexin
• Granted Patent: US 8,996,338 B2
• Time in Patent Office: Days
• US Class Current: 703/2
• CPC Class Codes: A61B 5/14532 for measuring glucose, e.g. by tissue impedance measurement
• CPC Class Codes: G01N 21/359 using near infrared light

Abstract

A method of detecting a concentration of a target component by using a reference wavelength comprises: defining a wavelength at which a light intensity is insensitive to the variation of the target component concentration as a reference wavelength for the target component; detecting spectra at both the reference wavelength and a further measuring wavelength; processing the spectrum detected at the further measuring wavelength, with the spectrum detected at the reference wavelength as an inner reference, to obtain a characteristic spectrum including specific information of the target component; building a calibration model between the characteristic spectrum and the concentration of the target component; and determining the concentration of the target component based on the calibration model.

Claims

1. A method of detecting a concentration of a target component in a sample, including steps of:

1) defining a wavelength at which a light intensity is insensitive to the variation of the target component concentration as a reference wavelength for the target component;

2) detecting spectra at the reference wavelength and a further measuring wavelength;

3) processing the spectrum detected at the further measuring wavelength, with the spectrum detected at the reference wavelength as an inner reference, to obtain a characteristic spectrum including specific information of the target component;

4) building a calibration model between the characteristic spectrum and the concentration of the target component; and

5) determining the concentration of the target component based on the calibration model.

2. The method according to claim 1, wherein step 1) includes:

analyzing the relation between the transmitted light intensity and the concentration variation of the target component according to relations between absorption and scattering coefficients and the concentration of the target component, wherein a wavelength at which the rate of light intensity variation is zero is defined as the reference wavelength.

3. The method according to claim 2, wherein the reference wavelength λ

_r is defined as follows;

for a pure absorption sample, α

_g(λ

_r)+α

_w(λ

_r)·

k=0; and

for a scattering sample,

4. The method according to claim 1, wherein step 1) includes:

changing the concentration of the target component in the sample and detecting respective spectra to calculating differential spectra among the samples with different concentrations; and

defining a wavelength at which the differential spectra intersect the zero value as the reference wavelength for the target component.

5. The method according to claim 1, wherein the reference wavelength for the glucose in the aqueous glucose solution is 1525 nm.

6. The method according to claim 1, wherein a wavelength at which the light intensity is most sensitive to the variation of the target component concentration is selected as the best measuring wavelength.

7. The method according to claim 1, wherein in step 3), the specific information of the target component is obtained as follows:

for a pure absorption sample, the specific information including the variation of the target component concentration is;

8. The method according to claim 1, wherein to perform temperature correction on the sample spectra, after step 3) the method further includes steps of:

expressing a difference spectral signal between two samples as a sum signal of a difference signal induced by the temperature variation and a difference signal induced by the concentration variation of the target component, wherein the difference signal induced by the concentration variation of the target component is zero at the reference wavelength for the target component;

calculating the temperature variation when measuring the samples according to the difference signal at the reference wavelength, to get the actual temperature when measuring the samples; and

correcting the light intensity at the further measuring wavelength, to get the measured spectrum corresponding to the temperature condition of the calibration model.

9. The method according to claim 1, wherein the calibration model is built by the Partial Least Squares (PLS) method or the net analyte signal method.