Method of determining a glucose concentration in a target by using near-infrared spectroscopy

• US 5,957,841 A
• Filed: 03/24/1998
• Issued: 09/28/1999
• Est. Priority Date: 03/25/1997
• Status: Expired due to Term

• Current Assignee: Panasonic Healthcare Holdings Co., Ltd.
• Original Assignee: Asahi Matsushita Electric Works Ltd. (Panasonic Holdings Corporation)
• Inventors: Maruo, Katsuhiko, Oka, Masami
• Primary Examiner(s): WINAKUR, ERIC FRANK
• Application Number:
  US09/046,580
• Time in Patent Office: 553 Days
• Field of Search: 600/316, 600/319, 600/322, 600/326, 600/473, 356/39, 439/63, 439/95
• US Class Current: 600/316
• CPC Class Codes: A61B 5/14532 for measuring glucose, e.g. by tissue impedance measurement
• CPC Class Codes: A61B 5/1455 using optical sensors, e.g. spectral photometrical oximeters
• CPC Class Codes: A61B 5/1495 Calibrating or testing of in-vivo probes
• CPC Class Codes: G01N 21/359 using near infrared light

Abstract

A glucose concentration in a living tissue as a target is determined by the following method. Near-infrared radiation is projected on the living tissue, and a resulting radiation emitted from the living tissue is received. A spectrum analysis of the resulting radiation is performed to detect a first absorption signal from a wavelength region, e.g., 1550 nm to 1650 nm, having an absorption peak of OH group derived from glucose molecule, a second absorption signal from a wavelength region, e.g., 1480 nm to 1550 nm, having an absorption peak of NH group in the living tissue, and a third absorption signal from a wavelength region, e.g., 1650 nm to 1880 nm, having an absorption peak of CH group in the living tissue. The glucose concentration is determined by a multivariate analysis of results of the spectrum analysis, in which the first, second and third absorption signals are used as explanatory variables, and the glucose concentration is a criterion variable. This method can predict the glucose concentration of the subject with an improved accuracy.

Claims

1. A method of determining a glucose concentration in a target by using near-infrared spectroscopy, said method comprising the steps of:

projecting near-infrared radiation on said target;

receiving resulting radiation emitted from said target;

performing spectrum analysis of the resulting radiation to detect at least one first absorption signal from a first wavelength region having an absorption peak of OH group derived from glucose molecule, at least one second absorption signal from a second wavelength region having an absorption peak of NH group in said target, and at least one third absorption signal from a third wavelength region having an absorption peak of CH group in said target; and

determining said glucose concentration by multivariate analysis of results of said spectrum analysis, in which said first, second and third absorption signals are used as explanatory variables, and said glucose concentration is a criterion variable,

wherein said first absorption signal is an absorbency at a first wavelength in said first wavelength region, said second absorption signal is an absorbency at a second wavelength in said second wavelength region, and said third absorption signal is an absorbency at a third wavelength in said third wavelength region, and wherein said first, second and third wavelengths are determined by a method comprising the steps of;

measuring a plurality of absorption spectrums in at least one sample;

performing multivariate analysis of said absorption spectrums to obtain a profile indicative of a relation between wavelength and regression coefficient; and

selecting as said first wavelength a wavelength substantially corresponding to a peak of said regression coefficient within said first wavelength region, selecting as said second wavelength a wavelength substantially corresponding to a peak of said regression coefficient within said second wavelength region, and selecting as said third wavelength a wavelength substantially corresponding to a peak of said regression coefficient within said third wavelength region.

2. The method as set forth in claim 1, wherein said first wavelength region is in a range of 1550 nm to 1650 nm, said second wavelength region is in a range of 1480 nm to 1550 nm, and said third wavelength region is in a range of 1650 nm to 1880 nm.

3. The method as set forth in claim 1, wherein said first wavelength region is in a range of 1050 nm to 1130 nm, said second wavelength region is in a range of 1000 nm to 1050 nm, and said third wavelength region is in a range of 1130 nm to 1300 nm.

4. The method as set forth in claim 1, wherein said first wavelength region is in a range of 1600±

40 nm, said second wavelength region is in a range of 1530±

20 nm, and said third wavelength region is in a range selected from the group consisting of 1685±

20 nm, 1715±

20 nm, and 1740±

20 nm.

5. The method as set forth in claim 1, wherein said plurality of absorption spectrums are measured by a method comprising the steps of:

preparing a plurality of test samples having different concentrations in a system including albumin, glucose, and water; and

measuring absorption spectrums of said test samples.

6. The method as set forth in claim 5, wherein fourth and fifth absorption signals are used as said explanatory variables in addition to said first, second and third absorption signals, said fourth and fifth absorption signals are absorbencies at fourth and fifth wavelengths, respectively, and wherein said fourth and fifth wavelengths are determined by the steps of:

performing said multivariate analysis of said absorption spectrums with respect to different principal components to obtain a plurality of profiles indicative of relations between wavelength and regression coefficient;

selecting as said fourth wavelength a wavelength substantially corresponding to an intersection of said profiles at the vicinity of a boundary between said first and second wavelength regions, and selecting as said fifth wavelength a wavelength substantially corresponding to an intersection of said profiles at the vicinity of a boundary between said second and third wavelength regions.

7. The method as set forth in claim 1, wherein said plurality of absorption spectrums are measured by a method comprising the steps of:

applying a glucose tolerance test to a subject; and

measuring absorption spectrums of said subject during said glucose tolerance test.

8. The method as set forth in claim 1, wherein said near-infrared radiation projected on said target essentially consists of a first near-infrared radiation having a center wavelength and a half-width within said first wavelength region, a second near-infrared radiation having a center wavelength and a half-width within said second wavelength region, and a third near-infrared radiation having a center wavelength and a half-width within said third wavelength region.

9. The method as set forth in claim 8, wherein said center wavelength and said half-width of said first near-infrared radiation are determined by the steps of:

preparing a plurality of test samples having different concentrations in a system including albumin, glucose, and water;

measuring absorption spectrums of said test samples;

performing a multivariate analysis of said absorption spectrums to obtain a profile indicative of a relation between wavelength and regression coefficient;

selecting as said center wavelength a wavelength substantially corresponding to a maximum value of said regression coefficient within said first wavelength region, and selecting as said half-width a wavelength region substantially corresponding to 70% or more of said maximum value within said first wavelength region.

10. The method as set forth in claim 9, wherein said first near-infrared radiation has said center wavelength within a range of 1560 nm to 1640 nm, and said half-width of 60 nm or less.

11. The method as set forth in claim 8, wherein said center wavelength and said half-width of said first near-infrared radiation are determined by the steps of:

applying a glucose tolerance test to a subject;

measuring absorption spectrums of said subject during said glucose tolerance test;

performing a multivariate analysis of said absorption spectrums to obtain a profile indicative of a relation between wavelength and regression coefficient;

selecting as said center wavelength a wavelength substantially corresponding to a maximum value of said regression coefficient within said first wavelength region, and selecting as said half-width a wavelength region substantially corresponding to 70% or more of said maximum value within said first wavelength region.

12. A method of determining a glucose concentration in a target by using near-infrared spectroscopy, said method comprising the steps of:

projecting near-infrared radiation on said target;

receiving resulting radiation emitted from said target;

performing spectrum analysis of the resulting radiation to detect at least one first absorption signal from a first wavelength region having an absorption peak of OH group derived from glucose molecule, at least one second absorption signal from a second wavelength region having an absorption peak of NH group in said target, and at least one third absorption signal from a third wavelength region having an absorption peak of CH group in said target; and

determining said glucose concentration by multivariate analysis of results of said spectrum analysis, in which said first, second and third absorption signals are used as explanatory variables, and said glucose concentration is a criterion variable,

wherein said first wavelength region is in a range of 1600±

40 nm, said second wavelength region is in a range of 1530±

20 nm, and said third wavelength region is in a range selected from the group consisting of 1685±

20 nm, 1715±

20 nm, and 1740±

20 nm.