JPH10325794A - Method and device for determining glucose concentration - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately determine glucose concentration, by determining the amount of glucose based on a measurement result in three wavelength regions for measuring the absorption originating from the OH group of glucose molecule and the absorption originating from CH group. SOLUTION: For example, a light-emitting diode 1 emits near infrared rays that allows the first harmonic of a molecule to be observed, has a small influence of water absorption, and has wavelength characteristics for covering a wavelength region of 1,480-1,880 nm. An interference filter 2 disperses the near infrared rays into a wavelength region of 1,550-1,650 nm for measuring absorption originating from the OH group of glucose molecule, a wavelength region of 1,480-1,550 nm for measuring absorption originating from the NH group of an organism constituent, and a wavelength region of 1,650-1,880 nm for measuring absorption originating from the CH group. Then, light is condensed onto the end face of an optical fiber 4 by a lens 3 and is guided to an object to be measured, transmission light or diffusion reflection light is detected by a photodiode 5, an absorption signal is measured, and glucose concentration is measured based on the light with three central wavelengths.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] The present invention relates to a glucose concentration in a living tissue or a glucose concentration in a body fluid such as blood, serum, plasma, cell fluid, saliva, tears, sweat, urine, etc. for health care and treatment of diseases. The present invention relates to a quantification method for measuring a concentration, and particularly to a quantification method for a glucose concentration using a spectroscopic analysis technique in a near infrared region.

[0002]

2. Description of the Related Art Near-infrared spectroscopy is used in various fields including agriculture, food and petrochemicals in recent years, since it does not require any special operation on a sample and can perform nondestructive and quick measurement. It has become. Compared to the spectroscopic analysis in the mid-infrared region, the spectroscopic analysis in the near-infrared region generally has a small absorption spectrum of water in the near-infrared region, which makes it possible to analyze aqueous solutions that are difficult in the mid-infrared region. High ability to penetrate living organisms ・ It is often not necessary to prepare special samples for measurement, but on the other hand, analysis in the mid-infrared region captures absorption due to normal vibration of molecules. In the near-infrared region, it is possible to capture harmonics or combined tones of the reference vibration observed due to the inharmonicity of molecular vibrations,
The absorption in the near infrared region is caused by the participation of a hydrogen atom in a CH group, OH
Signal level is as low as about 1/100 compared to the mid-infrared region. Is a universal entity in living organisms, and it has the disadvantage that the assignment of the absorption peak is often unclear because the signal of this molecular bond is captured. Due to this disadvantage, when performing quantitative or qualitative analysis in the near-infrared region, it is difficult to perform accurate analysis using an analysis method based on peak positions and peak heights as in the case of mid-infrared analysis.

In order to solve this problem, recently, a method of analyzing a measured spectrum using a statistical analysis technique, for example, a multivariate analysis technique such as linear multiple regression analysis (MLR), principal component regression analysis, or PLS regression analysis, A technique called so-called chemometrics is used. This method is an analysis method that has spread rapidly with the development of personal computers. By using a statistical method based on numerical analysis, it is possible to carry out quantitative and qualitative analysis for practical use even with an absorption signal with a small SN ratio in the near infrared region. Becomes

[0004] Against this background, the quantification of glucose concentration in living organisms using near-infrared light has attracted much attention in recent years. This is because non-invasive quantification that does not require blood collection is possible. Among them, the following are typical examples. U.S. Pat. No. 4,655,225: Non-invasive spectroscopic method and apparatus This patent discloses at least one wavelength selected from 1575, 1765, 2100, 2270 ± 15 nm and no absorption of glucose. Some or insignificant from 1000 nm to 27
A technique for quantifying glucose from a reference wavelength selected from 00 nm is disclosed, and 1100 nm is selected as the reference wavelength in the examples. This reference wavelength 1
100 nm is a wavelength belonging to an absorption region of the second harmonic,
Compared with the first overtone or reference vibration absorption region to which 1575, 1765, 2100, and 2270 nm belong, the signal magnitude is about 1/100 or less, and therefore, there is no glucose absorption or an insignificant reference. It can be called wavelength.

[0005] Conversely, this is the second or third harmonic.
This indicates that it is difficult to select a reference wavelength from a region other than a harmonic region such as an overtone. 1 and 2 show the absorption spectra of distilled water and solid (powder) glucose, albumin, and cholesterol measured in the first overtone region and the second overtone region, respectively.
In the region where the overtones and their combined sounds exist, there is no wavelength where there is no absorption by glucose, and there is only a place where there is relatively little absorption. Given that the absorption of other biological components is superimposed, it is practically impossible to determine the glucose concentration without considering the contribution of other biological components. Think there is.
Further, 1575, 1765, 2100, 2270 nm
As a reason for the selection of the wavelength, it is shown that the peak of the spectrum obtained by measuring solid-state glucose exists at the above four wavelengths, but the absorption peak of glucose in an aqueous solution dissolved in a body fluid remains in the solid state. We should consider it impossible to exist. It is considered that a molecule containing a large number of OH groups, such as glucose, is affected by hydrogen bonds due to water molecules in water, as described later, so that the peak of the absorption spectrum is shifted. 157 related to OH group
It is natural to consider that absorption peaks at two wavelengths of 5,2100 nm make accurate measurement difficult in an aqueous solution due to wavelength shift.

[0006] US Pat. No. 5,028,787:
Non-Invasive Measurement of Blood Sugar This U.S. Patent shows a technique for quantifying glucose in the range of 600 nm to 1100 nm;
An example in which glucose is quantified using two wavelengths of ± X nm is shown. Specifically, 945 nm and 1015 nm
(That is, X = 35) is shown. Since 980 nm, which has an important meaning here, is a wavelength indicating the absorption peak of the second harmonic of water, in this embodiment, glucose is quantified using two wavelengths selected with the absorption peak of water interposed. Will be. 945 nm and 101
It is shown that a good correlation with the glucose concentration is obtained as the reason for selecting 5 nm.

[0007] US Pat. No. 5,070,874:
Non-invasive determination of glucose concentration in a patient's body This U.S. patent discloses a method for non-invasive measurement of blood glucose by quantifying glucose from the nth derivative of a spectrum obtained by changing the wavelength in a narrow range around 1660 nm. Have been. Specifically, the absorbance lo measured at around 1660 nm
The second derivative of g1 / T is derived, and glucose is quantified using multivariate analysis. As will be described in detail later, in the experiment conducted by us, the regression coefficient obtained by PLS analysis of the second derivative spectrum in the first harmonic region has a peak near 1660 nm. However, the characteristic absorption around 1660 nm is derived from the CH group of glucose, and the quantitative analysis focusing on the absorption at 1660 nm derived from the CH group of glucose is a quantitative analysis focusing on the absorption derived from the OH group of glucose. The result is that the measurement accuracy is much worse than the analysis. The CH group is a molecular bond ubiquitously present not only in glucose but also in other biological components, and its absorption is observed with a peak concentrated in a region around 1700 nm as is apparent from FIG. C
This is probably because it is difficult to separate the absorption derived from the H group from the absorption derived from the CH group of glucose, and the absorption derived from the CH group of glucose is smaller than other biological components in the first place. However, it does not disclose the contribution of other biological components, and it is difficult to quantify the glucose concentration in this respect as well.

US Pat. No. 5,222,496: Near-Infrared Glucose Sensor This US Pat. No. 5,222,496 quantifies glucose by using a reference light within 1600 nm having an absorption peak derived from the OH group of glucose and within 60 nm from the wavelength. It is disclosed. Specifically, an example is shown in which glucose is quantified from a reference wavelength selected in the range of 1630 nm to 1660 nm and the absorption wavelength of glucose at 1600 nm. The reason for using a reference wavelength within the range of 1600 nm to 60 nm is that light having similar wavelengths has almost the same optical properties such as a refractive index, so that a signal with little influence of disturbance can be obtained. In addition, US Pat.
As in the specification of US Pat. No. 5,225, it shows that a peak of a spectrum obtained by measuring solid-state glucose exists at 1600 nm. However, as in the conventional example, 160
In the vicinity of 0 nm, the absorption of the glucose component forms a broad absorption band rather than an absorption peak, and in some cases, it is considered that the glucose signal is hard to be clearly separated by reference light selected from within 30 nm. In addition, this prior art does not disclose the contribution of other biological components, and in this respect, it is difficult to quantify the glucose concentration.

[0009]

As described above, the prior art generally determines the concentration of glucose in a living body by absorbing glucose with respect to reference light.
However, as described above, the absorption spectrum in the near-infrared region is generated by molecules that biological components basically have, such as OH group, NH group, and CH group. Because absorption is observed,
The absorption signal at a certain wavelength should be considered to be more or less superimposed with the absorption of various biological components.
For example, the absorption region derived from the NH group exists in the broad absorption region derived from the OH group of glucose, and the absorption derived from the OH group of glucose overlaps with the absorption of other biological components. Exists. Therefore, it can be said that quantification of glucose cannot be established without considering the effects of biological components other than glucose and the effects of disturbance components such as temperature, scattering, and optical path length. Understanding the phenomenon by relating the state of dissolved glucose in the living body to the near infrared spectrum,
It is necessary to understand the influence of protein components and fat components in living tissues or body fluids that cause a large disturbance in connection with the near-infrared spectrum.

According to the present invention, based on such knowledge, when quantifying the concentration of glucose in a biological component such as a biological tissue or a body fluid which changes every moment, the above-described disturbance factors are taken into account to accurately determine the glucose concentration. It is an object of the present invention to provide a quantification method capable of performing quantitative analysis of a concentration and an apparatus therefor.

[0011]

The method of quantifying glucose concentration according to the present invention is a method of quantifying glucose concentration in living tissue or body fluid using light absorption in the near-infrared region. A first wavelength range for measuring the absorption of the biological component derived from the OH group, a second wavelength range for measuring the absorption of the biological component derived from the NH group, and measuring the absorption of the biological component derived from the CH group. It is characterized in that glucose is quantified based on the measurement results in at least three wavelength ranges of the third wavelength range.

The above three wavelength ranges are preferably set to a range where the influence of absorption of water molecules derived from living body water is relatively small in order to eliminate the influence of living body water. It is preferable that the wavelength ranges be adjacent to each other in terms of the configuration of the light source and the detecting means. The three wavelength ranges are wavelength ranges in which the first overtone of the molecule can be observed, and the influence of water absorption is relatively small from 1480 nm to 188 nm.
When selecting within the 0 nm wavelength range, the first wavelength range for measuring the absorption derived from the OH group of the glucose molecule is 15 nm.
The wavelength range for measuring the absorption derived from the NH group of the biological component is from 1480 nm to 1550 nm.
nm, the wavelength range for measuring the absorption derived from the CH group is 16
It is good to be 50 nm to 1880 nm, and when it is selected in the wavelength range of 1000 nm to 1300 nm where the second harmonic of the molecule can be observed and the influence of water absorption is relatively small, the absorption derived from the OH group of the glucose molecule is selected. The wavelength range for measurement is from 1050 nm to 1130 nm,
The wavelength range for measuring the absorption derived from the NH group of the biological component may be from 1000 nm to 1050 nm, and the wavelength range for measuring the absorption derived from the CH group may be from 1130 nm to 1300 nm.

In the quantification, glucose is quantified by multivariate analysis of a spectrum signal obtained by continuously measuring each wavelength in the selected three adjacent bands. In addition to selecting at least one wavelength from each of the three wavelength ranges, a predetermined range of wavelength bands may be selected to determine the glucose concentration. At this time, at least one of the above wavelengths or wavelength bands is obtained by converting a continuous spectrum measured by quantitative analysis of a plurality of samples into a principal component regression analysis or a PL spectrum.
If the wavelength is around the positive or negative peak of the regression coefficient obtained by the S regression analysis, a favorable result can be obtained.

A wavelength band of 1600 ± 40 nm from the first wavelength band or at least one wavelength within the wavelength band is used,
A wavelength band of 1530 ± 20 nm or at least one wavelength within the wavelength band from the third wavelength range.
85 ± 20 nm or 1715 ± 20 nm or 174
Use of any one of the wavelength bands of 0 ± 20 nm or at least one wavelength within the wavelength band can also provide suitable results.

When measuring the glucose concentration in a living tissue or a body fluid, it is preferable to determine the glucose concentration based on light of at least three central wavelengths having a predetermined half-width selected from the above three wavelength ranges. In this case, the half-value width is about 70% or less of the peak value obtained by performing a principal component regression analysis or a PLS regression analysis on the continuous spectrum measured in the analysis of a plurality of samples and obtaining a regression coefficient for each wavelength calculated by the above analysis method. It is preferable to use a wavelength width indicating a value. In addition, the center wavelength of at least one wavelength is set to 1560.
It is also preferable that the width is set within a range from nm to 1640 nm and the half width is set to 60 nm or less.

[0016] Based on data obtained by performing pre-processing of obtaining a difference or a quotient at a wavelength in the near infrared region from an absorption signal obtained as a result of measurement in three wavelength ranges or a value obtained by converting the absorption signal into an absorbance or a transmittance. The pre-processing in this case may include a regression coefficient for a plurality of principal component factors calculated by performing a principal component regression analysis or a PLS regression analysis on a continuous spectrum measured by analyzing a plurality of samples. A favorable result can be obtained by taking a difference or a quotient at a wavelength near the intersection of the two, and further as a wavelength for pretreatment,
1550 ± 10 nm or 1650 in the first overtone region
Wavelength selected from ± 10 nm, 106 in second harmonic region
It is preferable to use a wavelength selected from 0 ± 10 nm or 1130 ± 10 nm.

The apparatus for quantifying glucose concentration according to the present invention comprises a near-infrared light source, a detecting means for detecting near-infrared light, and a near-infrared light emitted from the near-infrared light source introduced into a living tissue or body fluid. Then, a guiding hand for guiding the near-infrared light transmitted or diffusely reflected by the living tissue or body fluid to the detecting means, and a calculating hand for calculating a signal obtained by the detecting means and converting the signal into a glucose concentration. The detection means measures the wavelength range for measuring the absorption derived from the OH group of the glucose molecule, the wavelength range for measuring the absorption derived from the NH group of the biological component, and the absorption derived from the CH group of the biological component. It is characterized in that it detects near-infrared light in at least three wavelength ranges of the wavelength range.

The near-infrared light source has a first wavelength range for measuring absorption derived from OH groups of glucose molecules, a second wavelength range for measuring absorption derived from NH groups of biological components,
It has a light-emitting characteristic covering at least three wavelength ranges of the third wavelength range for measuring absorption derived from the CH group of the biological component, and even if a device equipped with spectral means for spectrally separating each wavelength range is used, Alternatively, a first wavelength range for measuring the absorption derived from the OH group of the glucose molecule,
A plurality of light sources having emission characteristics in at least three wavelength ranges of a second wavelength range for measuring absorption derived from H groups and a third wavelength range for measuring absorption derived from CH groups of biological components. May be used.

[0019]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS As described above, when quantifying a glucose component in a living tissue or a body fluid, the state of the biological component in the living tissue or the body fluid and a change in absorption of near-infrared light caused by the state are determined. Need to figure out. The phenomenon recognition in performing glucose quantification in a living body will be described below. The absorption in the near infrared region is mainly due to CH
Group, OH group, and NH group.
-O, C = O group or SH, PH group is observed,
Even in a living body, the absorption of OH, CH, and NH groups may be considered. However, for hair, the SH group also has an important meaning.

The wavelength to be used for the qualitative or quantitative determination of a substance by near-infrared spectroscopy is determined by the characteristics of the measured object, the magnitude of the signal of the target quantitative substance, the light intensity of the light source, Judgment will be made comprehensively taking into account the characteristics of the device, but when quantifying glucose, the absorption signal at a wavelength that has specific absorption compared to substances such as proteins, lipids, and water present in the living body It is self-evident that the above-described prior art is also based on the idea. However, considering the molecular structure of glucose (C ₆ H ₁₂ O ₆ ) and the glucose absorption spectrum measured in the solid state, It is considered best to capture the absorption originating from the OH group of glucose as the specific absorption signal.
The absorption peak derived from the OH group of glucose in the solid state is
As shown in FIG. 1, the first overtone region (1350 to 1880n)
m), it exists as a broad peak from 1525n to around 1640 mm, and the second overtone region (900 to 1350
nm) from 1000 nm to 1140 as shown in FIG.
It exists as a broad peak around nm. The absorption of glucose OH groups that can be observed in this region
Compared to the peak of the group (1450 nm for the first harmonic and 980 nm for the second harmonic), the molecular vibration state of the OH group in the solid state and the molecular vibration state of the OH group in the aqueous solution are present on the longer wavelength side. Is affected by the very large hydrogen bonds of water molecules, especially in aqueous solutions. It should be considered that the absorption derived from the OH group of glucose in an aqueous solution is shifted due to the change in interaction between molecules and the effect of hydrogen bonding by water molecules.

In an aqueous solution, the absorption spectrum derived from the OH group of glucose is 153 as in the solid state.
The fact that there is no peak around 0 nm to 1600 nm and the absorption becomes flat due to the shift means that glucose
This means that the data pre-processing by the differential operation (first-order differentiation, second-order differentiation) usually used for the quantitative analysis in the near-infrared region in the quantification method focusing on the H group may not be suitable. In other words, although absorption originating from the OH group of glucose exists around 1530 nm to 1600 mm, the change with respect to wavelength is small, so that it is difficult to capture the absorption signal by the differential operation, that is, the operation of obtaining the amount of change with respect to wavelength. is there.

The characteristics of a body fluid or a biological spectrum are as follows: The absorption spectrum of a trace substance in a living body in the near-infrared region is hidden by a large absorption of water, but is surely present by being superimposed on the absorption. This is characteristically shown in the regression coefficient or principal component obtained by PLS regression analysis. The absorption observed in the near infrared region is mainly CH, OH, and NH.
Since it is a group, it is affected by hydrogen bonding in aqueous solution.
The magnitude of the effect differs depending on the characteristics of the molecule.Those with a low molecular weight and many OH groups, such as glucose, are greatly affected, but the effects on high-molecular-weight proteins such as albumin are relatively small. .

Therefore, in order to determine glucose in the living body, the spectrum information (measured normal biological component spectrum) on which the spectrum of the other component in the living body is superimposed is obtained.
It turns out that it is important to find a specific absorption band in which glucose absorption behaves differently from other biological components.
In addition, since signals of other biological components (hereinafter referred to as “disturbance components”) are also superimposed on the specific absorption band, a reliable measurement method has been established to obtain information for removing the contribution of the disturbance components. Think it is important to.

In order to consider the influence of disturbance components, an experiment was conducted on a system in which the three components of glucose, albumin, and water in bovine serum were artificially fluctuated at the first and second harmonics.
That is, a quantitative experiment of glucose concentration was performed using albumin as a disturbance component. -Experiment A- (Glucose-albumin-water fluctuation system No. 1
5 ml of an aqueous solution in which glucose was dissolved in distilled water and 15 ml of an aqueous solution in which albumin was dissolved in distilled water were added to 80 ml of bovine serum to prepare a sample, and the glucose concentration in the sample was 30 mg / dl, 93 mg. / Dl, 1
55mg / dl, 280mg / dl, 530mg / d
1, 1030 mg / dl, albumin concentration 2.24 g / dl, 2.84 g / dl, 3.44 g / d
Samples were prepared by picking 15 combinations out of a total of 6 × 5 = 30 combinations of 5 levels of 1, 4.64 g / dl and 5.84 g / dl. As can be seen from the preparation method, the sample is adjusted so that only three components, glucose, albumin, and water, fluctuate.

In this experiment, the concentration of both glucose and albumin fluctuates, so that the quantification of glucose cannot be determined simply by the relationship between glucose and water or albumin and water. For spectrum measurement and regression analysis, a sample was placed in a glass cell having a thickness of 1 mm.
Using 0, the averaging was performed 128 times, under the conditions of a resolution of 16, a detector DTGS KBr, and a white light source. Prior to the measurement of the spectrum of the sample, the measurement of the reference signal is performed by converting the absorption signal measured by the reference signal stored in the memory of the spectrometer (FT-IR) into the absorbance and obtaining the spectrum data obtained by multivariate analysis on the market. PLS regression analysis was performed using software. In the wavelength range used for the PLS regression analysis, the harmonic of the first harmonic is observed.
It is 50 nm to 1800 nm.

The reason that albumin was selected as a disturbance component is that it is thought that in addition to being the most common protein component present in blood, information for estimating the characteristics of the protein component constituting the living body can be obtained. is there. The result of the glucose quantification in Experiment A is an estimation based on the number of principal components 7,
Correlation coefficient 0.996 for standard curve creation, standard error SEP2
8.1 mg / dl, calibration curve test correlation coefficient 0.992,
Standard error was SEP 38.1 mg / dl.

FIG. 3 shows a regression coefficient using the seventh principal component. Compared to the spectrum of each single component shown in FIG. 1, a negative peak around 1510 nm derived from the NH group of the albumin molecule, 1 peak derived from the OH group of the glucose molecule,
From the positive peak around 580 nm and the negative peak around 1700 nm derived from the CH group of the albumin molecule, it can be seen that the glucose concentration is being quantified.

The existence of a positive peak near 1580 nm derived from the OH group of the glucose molecule in the experimental results seemingly contradicts that the OH group spectrum of glucose in an aqueous solution has a broad absorption spectrum shifted from that of a solid. Seems like a conclusion. However, looking at the results of FIG. 4 in which the same spectrum data was subjected to second-order differentiation processing and analysis, it can be seen that from about 1580 nm to 157 nm to 157 nm.
A peak with a large regression coefficient is observed at 0 mm, and if the absorption is at 1580 nm, the absorption peak to be observed there is naturally shifted. From this, it is considered that the absorption around 1580 nm of glucose shows a relatively flat absorption characteristic with a shifted peak, and an interpretation consistent with the above results is established.

The reason why the regression coefficient having a peak at 1580 nm is calculated in the analysis of the absorption spectrum without performing the differentiation process is that the absorption from the OH group of glucose in this vicinity is broad and the absorption from the adjacent NH group and the CH group are close to each other. This is because it is affected by absorption from the origin. In other words, it can be interpreted that the wavelength that is least affected by the two exists near 1580 nm.

-Experiment B- (Glucose-albumin-water fluctuation system 2nd harmonic) 5 ml of an aqueous solution in which glucose was dissolved in distilled water and 15 ml of an aqueous solution in which albumin was dissolved in distilled water were added to 80 ml of bovine serum. And the glucose concentration in the sample was 35 mg / dl, 136 mg / dl, 2
20mg / dl, 412mg / dl, 750mg / dl
5 levels, albumin concentration of 2.6 g / dl, 3.0 g
/ Dl, 3.3 g / dl, 4.0 g / dl, 5.4 g /
dl = 5 levels = 5 × 5 = 25 combinations, 13
Samples were prepared by picking up combinations of the types. In addition,
The detector was cooled with liquid nitrogen and measured. PLS regression analysis was performed on the absorption spectrum smoothed by a moving average of 17 points. The wavelength range used for the PLS regression analysis was from 900 nm where the second harmonic was observed to 13 nm.
50 nm.

The results of the glucose quantification in Experiment B are estimated based on the number of principal components 7, and the correlation coefficient is 0.981 for the preparation of a calibration curve, the standard error is 53.1 mg / dl for SEP, and the correlation coefficient is 0.959 for the calibration curve test. And the standard error was SEP 77.2 mg / dl. FIG. 5 shows a regression coefficient using up to the seventh principal component. As compared with the spectrum of each of the three components shown in FIG. 2, a negative peak around 1020 nm derived from the NH group of the albumin molecule, and 1 representing the OH group derived from the glucose molecule.
From the positive peak around 070 nm to 1120 nm and the negative peak around 1150 nm derived from the CH group of the albumin molecule, it can be seen that the glucose concentration is being quantified. That is, the result in the second overtone region has a similarity to the result obtained in the first overtone region.

From the results of the above-described glucose quantification experiment with disturbance fluctuations in the first and second overtone regions, the glucose concentration in the living body was found to be due to the disturbance component and the glucose absorption spectrum, especially the absorption spectrum derived from the OH group of glucose. It can be seen from the relationship that the measurement is highly reliable and a highly accurate measurement method can be established. As the specific absorption band of glucose used for quantitative analysis, absorption of glucose OH groups appearing at a position different from that of water OH groups is effective as described above. It is only necessary to grasp the absorption of the NH group and CH group.

In setting a wavelength range including absorption of NH group, OH group (of glucose), and CH group in a region excluding a portion of large absorption caused by OH group of water from each harmonic region. , 1480 nm to 1 in the first overtone region
In the range of 880 nm, the absorption range derived from the OH group is 1
The absorption range from 550 nm to 1650 nm, from the NH group is preferably from 1480 nm to 1550 nm, and the absorption range from the CH group is from 1650 nm to 1880 nm.
In the overtone region, in the range of 1000 nm to 1300 nm, the absorption range derived from the OH group is 1050 nm to 113 nm.
0 nm, absorption range derived from NH group is 1000 nm to 10 nm
50 nm, absorption range derived from CH group is 1130 nm to 1
The thickness may be 300 nm.

One of the disturbance factors other than the biological component is a temperature change of the living body or the measurement sample. Temperature change does not pose a major problem if measurement is performed in an appropriate environment because humans are constant temperature animals, but temperature can also be used as an explanatory variable for quantifying glucose concentration. Further, since water molecules are most affected by temperature changes in living tissues or body fluids, the effects of temperature changes can be reduced by quantification in a region where the influence of water molecule absorption is small.

For the quantitative analysis, it is better to perform the analysis for each harmonic region. However, if the fluctuation using the maximum and minimum values at the wavelength to be used is normalized, the analysis can be performed for a plurality of harmonic regions. A straddling absorption signal can also be used in the same row. The glucose concentration can be quantified by irradiating the measurement object with light containing a multi-wavelength component or monochromatic light, and detecting the reflected light or transmitted light.In this case, the glucose concentration can be measured before reaching the measurement object. It is necessary to perform a spectral operation after the light is reflected or transmitted, and for this spectral operation, an interference filter, a diffraction grating, a Fourier transform method, or the like can be used.

The glucose concentration quantifying device of the present invention will be described below. As described above, the glucose concentration quantifying device of the present invention comprises a near-infrared light source, a detecting means for detecting near-infrared light, Introducing near-infrared light emitted from an infrared light source into a living tissue or body fluid, guiding the near-infrared light transmitted or diffusely reflected through the living tissue or body fluid to the detecting means, And a calculating means for converting a signal to be converted into a glucose concentration, wherein the detecting means measures a wavelength range for measuring absorption derived from OH groups of glucose molecules and an absorption derived from NH groups of biological components. It is characterized by detecting near-infrared light in at least three wavelength ranges of a wavelength range for measurement and a wavelength range for measuring absorption derived from a CH component of a biological component. Performing determination of glucose using the absorption signal of the near-infrared light by using a calibration formula which are input in advance.

As is apparent from the above description, the near-infrared light source is capable of quantifying glucose in a relatively narrow wavelength range, and in some cases, is capable of quantifying glucose at several wavelengths. For this reason, light emitting diodes (LEDs) are suitable. The light emitting diodes that can be used in the first and second overtone regions include InP-based light emitting diodes in the current technology, and GaAs or GaAl in the third overtone region.
There are As-based light emitting diodes. Light-receiving elements that can be used in the first and second overtone regions include InGaAsP-based light-receiving elements, and silicon light-receiving elements in the third overtone area.

Further, depending on the characteristics of the light emitting diode, there is a case where it is not necessary to use the spectral means. In particular, since the half width can be set relatively large in a range where the absorption derived from the OH group of the glucose molecule is measured, a spectral means such as an interference filter can be omitted. When spectroscopic means is required, various spectroscopic means can be used as described above, but generally, an interference filter is preferably used.

As the inducing means, it is suitable to use an optical fiber when non-invasively measuring a living tissue to determine the glucose concentration. In particular, when quantification is performed in the first harmonic region, the degree of arrival of light in the search direction is controlled by the optical fiber interval, and the measurement accuracy can be improved by a method of quantifying the glucose concentration in the subcutaneous tissue at an arbitrary search. .

As a simple method of controlling the optical fiber interval, there is a method of setting the distance by the center-to-center distance obtained by bringing an optical fiber of an arbitrary diameter into contact, that is, by changing the diameter of the optical fiber used for receiving and emitting light. The analysis using such an optical fiber is performed by the light receiving and emitting fibers 8 and 9.
15 is applied to the surface of the object to be measured at a right angle as shown in FIG. Note that the angle is not limited to a right angle as long as the transmitted light path can be specified.

The light emitted from the light emitting fiber 8 to the object to be measured is transmitted to the optical path A to the adjacent light receiving fiber 9, the optical path B to the light receiving fiber 9 farther away, and the optical path C to the light receiving fiber 9 farther away. The light that has passed through the optical path enters the light receiving fiber 9. Now, comparing the optical path lengths of the optical paths A, B, and C, if the optical fibers having the same diameter are used for light receiving and emitting, the optical paths B, C can be considered to be almost twice or three times the optical path A. Although the transmission of light in a substance such as a living body that is optically opaque and scattered does not strictly follow Lambert-Beer's law, the amount of transmitted light rapidly decreases according to the optical path length, and the optical path length is doubled. Becomes 1/10 depending on the scattering condition
In the following, if it is tripled, it will be about 1/10 or less of that. Therefore, the transmitted light focused on the light receiving fiber 9 can be basically considered to be the sum of the transmitted light from the adjacent light emitting fibers 8.

It is well known that the operation of receiving and emitting light is facilitated by bundling not only one optical fiber but also a plurality of optical fibers, and the above-mentioned light emitting fiber 8 and light receiving fiber 9 are bundled on the measurement side. At the same time, by bundling and branching the ends opposite to the measurement side between the light emitting fibers 8 and the light receiving fibers 9, a fiber bundle convenient for measurement can be manufactured. The optical fiber used has a diameter of several μm to several thousand μm.
(Several mm) can be used.
The use of optical fibers of m to 1000 μm is suitable. The optical fiber to be used may be a thin fiber coated with an arbitrary wire diameter.

As a specific living tissue, there is a human dermis as shown in FIG. Human dermis is 100
The tissue has a thickness of about 1 mm sandwiched between the epidermis b having a thickness of about μm and a subcutaneous tissue C composed of a large amount of fat cells.
For the analysis of the chemical components in the dermis, it is suitable to use a bundle of optical fibers in which the distance between the centers of the light receiving and emitting fibers is adjusted to 250 to 1000 μm. For example, by performing spectroscopic analysis using an optical fiber bundle having a light receiving / emitting distance (center-to-center distance) of 500 μm, it is possible to analyze the glucose component in the dermis. Since the path of the light transmitted through the skin tissue varies depending on the physical properties of the tissue and the wavelength of the light used for analysis, it is preferable to perform preliminary experiments and numerical simulations in advance to determine the light receiving / emitting distance used for measurement. The glucose concentration of the obtained light absorption signal is calculated by numerical calculation in the calculation unit. Measurement of glucose concentration in the living body, in particular, accurate determination of glucose concentration in the living tissue, is a software that grasps the state of glucose in the living body and the influence of disturbance components as an absorption signal in the near infrared region, It becomes possible only by combining many optimal factors, such as hardware that enables measurement of an accurate absorption spectrum with high reproducibility and selection of a stable measurement site from a medical point of view.

[0044]

【Example】

-Example 1-In this example, glucose in bovine serum was quantitatively analyzed as a biological component, and the absorption region derived from NH, OH, and CH groups from 1510 to 1880 nm was measured and analyzed as a continuous spectrum. Was done. The bovine serum sample was the same as the above-described experiment of the three-component variation system of glucose, albumin, and water. For 80 ml of bovine serum, 5 ml of an aqueous solution in which glucose was dissolved in distilled water and an aqueous solution in which albumin was dissolved in distilled water were used. A sample was prepared by adding 15 ml. The sample has a glucose concentration of 39 mg / dl and 93 mg / d in the sample.
1,155mg / dl, 280mg / dl, 530mg
/ Dl, 6 levels of 1030 mg / dl and albumin concentrations of 2.24 g / dl, 2.84 g / dl, 3.44 g /
From the total of 6 × 5 = 30 combinations of 5 levels of dl, 4.64 g / dl and 5.84 g / dl, 15 combinations were picked up to prepare samples.

For the spectrum measurement, a sample was placed in a glass cell having a thickness of 1 mm, and the averaging was performed 128 times using a Magna 850 manufactured by Nicolet, resolution 16, detector DTGS.
The spectrum data measured under the conditions of KBr and a white light source and converted into absorbance was subjected to PLS regression analysis using a commercially available multivariate analysis software with a glucose concentration as a target variable and an absorption spectrum as an explanatory variable. PLS as described above
The wavelength range used for regression analysis was 1510 nm to 188
0 nm.

The results of the glucose quantification in the experiment were estimated based on the number of principal components of 6, and a correlation coefficient of 0.996 for preparing a calibration curve, a standard error of 27.8 mg / dl for SEP, a correlation coefficient of 0.996 for a calibration curve test, Standard error SEP was 27.8 mg / dl. -Example 2-The experimental system in this example is the same as that in Example 1, except that glucose, albumin, and water in the bovine serum sample were varied. The software used for the multivariate analysis is the same as in the first embodiment.

In this embodiment, 1510, 1540, 159
Linear multiple regression analysis (MLR) was performed using the absorbance at 5 wavelengths of 0, 1655 and 1700 nm as explanatory variables and glucose concentration as the target variable. The results of the glucose quantification in the experiment were as follows: the correlation coefficient of the calibration curve was 0.980, and the standard error was SEP.
62.5 mg / dl, calibration coefficient 0.97
5. The standard error was SEP 68.0 mg / dl.

The wavelengths of 150, 1590, and 1700 nm were selected as the wavelengths showing the peak values of the regression coefficients shown in FIG. 3, and the wavelengths of 1540 and 1655 nm were selected as the intersections of the regression coefficients of a plurality of principal component factors. -Example 3-The experimental system in this example is the same as that in Example 1, except that glucose, albumin, and water in the bovine serum sample were varied. The software used for the multivariate analysis is the same as in the first embodiment.

In this embodiment, 150, 1590, 170
Linear multiple regression analysis (MLR) was performed using the absorbance at three wavelengths of 0 nm as an explanatory variable and glucose concentration as a target variable. The result of the glucose quantification was a correlation coefficient of 0.97 for preparing a calibration curve.
8, standard error SEP 65.8 mg / dl, correlation coefficient of calibration curve test 0.975, standard error SEP 69.0 mg / d
l.

The wavelengths of 150, 1590 and 1700 nm were selected as the wavelengths showing the peak values of the regression coefficients shown in FIG. Example 4 The experimental system in this example is the same as that in Example 1, except that three components of glucose, albumin, and water in the bovine serum sample were varied. The software used for the multivariate analysis is the same as in the first embodiment.

In this embodiment, 1525, 1590, 169
Linear multiple regression analysis (MLR) was performed using the absorbance at three wavelengths of 0 nm as an explanatory variable and glucose concentration as a target variable. The results of the glucose quantification in the experiment were as follows: a correlation coefficient of 0.983 for the preparation of a calibration curve, a standard error of 57.0 mg / dl, a correlation coefficient of the calibration curve test of 0.981, and a standard error of SEP60.1.
mg / dl.

The wavelength of 1590 nm is selected by selecting the wavelengths showing the peak values of the regression coefficients shown in FIG.
Was selected as the wavelength on the absorption side derived from the OH group from each peak. As described above, in the case of an absorption peak derived from an NH group or a CH group, a quantitative property may be improved by selecting a wavelength slightly shifted from the absorption derived from an OH group. -Example 5-In this example, glucose in bovine serum was quantitatively analyzed as a biological component, and the absorption region derived from NH, OH, and CH groups from 1480 to 1850 mm was measured and analyzed as a continuous spectrum. Was done. Glucose was quantified by changing five components of glucose, albumin, cholesterol, neutral fat and water in the bovine serum sample. The sample had a glucose concentration of 35 mg / dl, 85 mg / dl,
140mg / dl, 220mg / dl, 270mg / d
1,415mg / dl, 5l0mg / dl, 800mg
/ Dl, 985mg / dl, 1500mg / dl of 10
Level, albumin concentration 2.2g / dl, 2.3g / d
1, 2.4 g / dl, 2.5 g / dl, 2.8 g / d
1, 3.4 g / dl, 4.5 g / dl, 5.4 g / dl
8 levels, cholesterol concentration is 55mg / dl, 63
mg / dl, 70mg / dl, 75mg / dl, 83m
g / dl, 100 mg / dl, 135 mg / dl, 20
9 levels of 5 mg / dl and 350 mg / dl, neutral fat concentration of 10 mg / dl, 15 mg / dl, 20 mg / d
1, 70 mg / dl, 133 mg / dl, 250 mg /
Samples were prepared by picking up 45 kinds of combinations so as to obtain seven levels of dl and 480 mg / dl.

For the spectrum measurement, a sample was placed in a glass cell having a thickness of 1 mm, and the average was averaged 128 times using a Magna 850 manufactured by Nicolet, resolution 16, detector DTGS.
The spectrum data measured under conditions of KBr and a white light source and converted into absorbance was subjected to PLS regression analysis using glucose concentration as a target variable and absorption spectrum as an explanatory variable using commercially available multivariate analysis software. As described above, the wavelength range used for the PLS regression analysis was 1480 nm to 1850 nm.
nm.

The results of the glucose quantification in the experiment were estimated based on the number of principal components of 7, and the correlation coefficient of the calibration curve was 0.992, the standard error SEP was 48.7 mg / dl, the correlation coefficient of the calibration curve test was 0.991, Standard error SEP51. It was 1 mg / dl. -Example 6-The experimental system in this example is the same as that in Example 5, except that five components of glucose, albumin, cholesterol, neutral fat and water in the bovine serum sample were varied. The software used for the multivariate analysis is the same as in the fifth embodiment.

In this embodiment, 1520, 1540, 158
0, 1645, 1685, 1715, 1740 nm
Linear multiple regression analysis (MLR) was performed using the absorbance of wavelength as an explanatory variable and glucose concentration as an objective variable. The result of the glucose quantification in the experiment was the correlation coefficient 0.98
9, standard error SEP 55.6 mg / dl, correlation coefficient 0.988 for calibration curve test, standard error SEP 57.8 mg / d
l.

1520, 1580, 1685, 171
The selection of the wavelength of 5,1740 nm is from 1250 nm to 18
FIG. 6 obtained by performing PLS regression analysis on a range of 50 nm.
154 as the wavelength indicating the peak value of the regression coefficient
The wavelength of 0,1645 nm was selected as the point of intersection of the regression coefficients due to a plurality of principal component factors. -Example 7-The experimental system in this example is the same as that in Example 5, except that the five components of glucose, albumin, cholesterol, neutral fat and water in the bovine serum sample were varied. The software used for the multivariate analysis is the same as in the fifth embodiment.

In this embodiment, 1520, 1580, 168
Linear multiple regression analysis (MLR) was performed using the absorbance at three wavelengths of 5 nm as an explanatory variable and the glucose concentration as a target variable. The results of the glucose quantification in the experiment were as follows: a correlation coefficient of 0.983, a standard error of SEP 68.0 mg / dl, a correlation coefficient of a calibration curve test of 0.983, and a standard error of SEP 69.2.
mg / dl.

The wavelengths of 1520, 1580 and 1685 nm were selected as the wavelengths showing the peak values of the regression coefficients shown in FIG. -Example 8-The experimental system in this example is the same as that in Example 5, except that the five components of glucose, albumin, cholesterol, neutral fat, and water in the bovine serum sample were varied. The software used for the multivariate analysis is the same as in the fifth embodiment.

In this embodiment, 1525, 1580, 168
Linear multiple regression analysis (MLR) was performed using the absorbance at three wavelengths of 0 nm as an explanatory variable and glucose concentration as a target variable. The results of the glucose determination in the experiment were as follows: a correlation coefficient of 0.987 for preparing a calibration curve, a standard error SEP of 60.0 mg / dl, a correlation coefficient of a calibration curve test of 0.987, and a standard error of SEP 61.0.
mg / dl.

The wavelength of 1590 nm is selected by selecting the wavelength indicating the peak value of the regression coefficient shown in FIG.
The wavelengths of 1525 nm and 1690 mm were selected as the wavelengths on the absorption side derived from OH groups from the respective peaks. -Example 9-The experimental system in this example is the same as that in Example 5, except that the five components of glucose, albumin, cholesterol, neutral fat, and water in the bovine serum sample were varied. The software used for the multivariate analysis is the same as in the fifth embodiment.

In this embodiment, 1510 nm to 1820 n
The data obtained by measuring the absorption region derived from the NH group, OH group, and CH group up to m as a continuous spectrum is savitzky-
The analysis was performed by performing a second-order differentiation process using Golay's differentiation method. For analysis, PLS regression analysis was performed using the second derivative as an explanatory variable and glucose concentration as a target variable. The result of the glucose quantification in the experiment is the estimation based on the number of principal components 6, and the correlation coefficient of the calibration curve is 0.986, and the standard error is SEP63.5 m.
g / dl, the correlation coefficient of the calibration curve test was 0.983, and the standard error was SEP 69.0 mg / dl.

Example 10 In this example, glucose in bovine serum was quantitatively analyzed as a biological component, and the absorption region derived from NH, OH, and CH groups from 1000 to 1300 nm was defined as a continuous spectrum. Measured and analyzed. The bovine serum sample is the same as the above-described experiment of the three-component variation system of glucose, albumin, and water. For 80 ml of bovine serum, 5 ml of an aqueous solution obtained by dissolving glucose in distilled water and an aqueous solution obtained by dissolving albumin in distilled water are used. A sample was prepared by adding 15 ml. The sample had a glucose concentration of 35 mg / dl, 136 mg /
dl, 220mg / dl, 412mg / dl, 750m
g / dl, albumin concentration of 2.6 g / dl,
3.0 g / dl, 3.3 g / dl, 4.0 g / dl,
Thirteen combinations were picked up from five combinations of 5 levels of 5.4 g / dl and 5 × 5 = 25 combinations to prepare samples.

For spectrum measurement, a sample was placed in a glass cell having a thickness of 1 mm, and the average was measured 128 times using a Magna 850 manufactured by Nicolet, resolution 16, detector MCT /
A. The spectrum data measured under the condition of a white light source and converted into the absorbance was subjected to PLS regression analysis using a commercially available multivariate analysis software with the glucose concentration as the target variable and the absorption spectrum as the explanatory variable.

As described above, the wavelength range used for the PLS regression analysis is from 1000 nm to 1300 nm. The result of the glucose quantification in the experiment is an estimation based on the number of principal components 6, and the correlation coefficient for preparing a calibration curve is 0.980, and the standard error is SEP54.2.
mg / dl, the correlation coefficient of the calibration curve test was 0.962, and the standard error was SEP 74.8 mg / dl.

Example 11 The experimental system in this example is the same as that in Example 5, except that the five components of glucose, albumin, cholesterol, neutral fat and water in the bovine serum sample were varied. The software used for the multivariate analysis is the same as in the fifth embodiment. In this example, preprocessing of spectral data was performed by taking the difference between the measured absorbance at each wavelength and the absorbance at 1540 nm. The wavelength range used for the analysis was 1480 nm to 1
880 nm.

Each wavelength data subjected to the preprocessing is described as an explanatory variable,
PLS regression analysis was performed using glucose concentration as a target variable. The result of the glucose quantification in the experiment is the estimation based on the number of principal components of 8, and the correlation coefficient of the calibration curve is 0.997, and the standard error is SE.
P28.9 mg / dl, correlation coefficient 0.99 for calibration curve test
6, standard error SEP was 31.9 mg / dl.

The reason why the wavelength of 1540 nm was selected was that it was selected as the intersection of the regression coefficients of a plurality of principal component factors as shown in FIG. -Example 12-The experimental system in this example is the same as that in Example 10, except that three components of glucose, albumin, and water in the bovine serum sample were varied. The software used for the multivariate analysis is the same as in the fifth embodiment.

In this embodiment, the measured absorbance at each wavelength is 1
Preprocessing of the spectral data was performed by taking the difference in absorbance at 060 nm. The wavelength range used for the analysis is
It is from 1000 nm to 1300 nm. PLS regression analysis was performed using the wavelength data subjected to the preprocessing as an explanatory variable and glucose concentration as a target variable.

The results of the glucose quantification in the experiment were estimated based on the number of principal components of 6, and the correlation coefficient of the calibration curve was 0.978, the standard error SEP was 56.8 mg / dl, the correlation coefficient of the calibration curve test was 0.961, The standard error was 76.1 mg / dl. The reason why the wavelength of 1540 nm was selected was that it was selected as an intersection where regression coefficients of a plurality of principal component factors intersect as shown in FIG.

Embodiment 13 As shown in FIG. 7, the apparatus for quantifying glucose concentration in a living body in this embodiment interferes with near-infrared light emitted from a light-emitting means (light-emitting diode 1) as a near-infrared light source. Filter 2
, And the light condensed on the end face of the optical fiber 4 by the lens 3 as a light condensing means is guided to an object to be measured, and the transmitted light of the object to be measured is a photodiode as a detecting means. 5, the absorption signal was measured, and the calculating means calculated the glucose concentration.

The three interference filters are arranged on concentric circles of the disk, and a center wavelength and a half width corresponding to the characteristics of the interference filter can be obtained by rotating the interference filters by the motor 6. The detected absorption signal is converted into an absorbance, and the glucose concentration is calculated based on a preset calibration curve. The light emitting diode has a center wavelength of 1600 nm and a half width of 160 nm.
Was used. The three types of interference filters have a center wavelength of 1530 nm and a half width of 10 n.
m, a half-width of 60 nm at a center wavelength of 1585 nm, and a half-width of 10 nm at a center wavelength of 1680 nm were used. The reason for setting the center wavelength and the half width of the interference filter having the center wavelength of 1585 nm and the half width of 60 nm is OH of glucose.
In order to capture a signal derived from the group, the peak derived from the OH group of the regression coefficient obtained in Example 5 was referred to. Example 5
FIG. 8 shows a partial enlargement of the regression coefficient. The regression coefficient of the principal component 7 has a maximum value at about 1580 nm, and 1555 nm and 1615 nm take 70% of the maximum value. Therefore, the center wavelength is 1555n
The half width was set to 60 nm at 1585 mm, which is the midpoint between m and 1615 nm. By irradiating near-infrared light with a center wavelength and a half-value width set by this method, it is possible to perform measurement equivalent to measuring a large number of points by measuring only one point.

The irradiating end 7 of the optical fiber is used perpendicularly to the object to be measured. At the irradiation end 7, the light emitting side 8 and the light receiving side 9 of the optical fiber are alternately arranged.
Near-infrared light that has been irradiated and transmitted through the living tissue is received by the light-receiving fiber 9 and guided to the detector 5. The diameter of the optical fiber used for light emission and light reception is 500 μm, and the distance between the centers of the light reception and light emission fibers is also 500 μm.

The calibration curve for calculating the glucose concentration is obtained by statistically analyzing the relationship between the true value of the glucose concentration obtained by another standard method and the absorbance according to the present invention using the user of the present invention or a plurality of persons as subjects. It is obtained by analyzing by When preparing a calibration curve from the user of the present invention, using a glucose tolerance test, after eating and drinking a glucose-containing substance, blood sampling and signal measurement according to the present invention may be performed at regular time intervals, or during life. The signal measurement may be performed by selecting a time when the blood glucose level is stabilized.

The glucose concentration obtained by the standard method for analyzing the collected blood and the signal stored at the time of measurement are compared to form data, and a calibration curve is created by the analyzing means built in the calculating means. A multiple regression analysis method is suitable for creating a calibration curve for individuals from absorption signals at several points as in this embodiment.

Embodiment 14 The apparatus for quantifying the concentration of glucose in a living body in this embodiment has the same configuration as that of Embodiment 13 shown in FIG. The difference is 3
The center wavelength and the half-value width of the interference filters are 1520 nm, the half-value width is 10 nm, and the center wavelength is 1580.
A half-width of 40 nm, a half-width of 1085 nm and a center wavelength of 1,865 nm was used. The reason for setting the center wavelength and the half width of the interference filter having the center wavelength of 1580 nm and the half width of 40 nm was based on the peak wavelength of the regression coefficient obtained in Example 5. The regression coefficient of the main component 7 in Example 5 had a maximum value at about 1520, 1580, and 1685 nm, and the half width of the interference filter was set based on the width of each peak. By selecting a wavelength at which the regression coefficient has a peak, a stable value can be measured even for a relatively steep peak such as 1520 mm or 1685 nm by using a half width of about 10 nm. Since the peak is broad at 1680 nm, the half width was set to be as wide as 40 nm.

By irradiating near-infrared light having a center wavelength and a half-value width set by this method, it is possible to perform measurement equivalent to measuring a large number of points by measuring only one point. -Example 15-As shown in Fig. 9, the apparatus for quantifying glucose concentration in a living body in the present example is configured such that three infrared light sources (semiconductor lasers) 1 serving as near-infrared light sources emit near-infrared light into a concave mirror 1.
0, the light condensed on the end face of the optical fiber 4 is guided to the object to be measured, and the transmitted signal of the object to be measured is detected by the photodiode 5 which is a detecting means to measure an absorption signal.
The computing means calculates the glucose concentration. The detected absorption signal is converted into an absorbance, and the glucose concentration is calculated based on a preset calibration curve.

The semiconductor laser has a center wavelength of 1530 n.
m, an InGaAsP-based device having a half-value width of 10 nm, a device having a center wavelength of 1,585 nm, a device having a half-value width of 30 nm, and a center wavelength of 16
A device having a half width at half maximum of 10 nm was used. As the half-width of the semiconductor laser, the half-width may be used as a characteristic of the element itself, or an interference filter may be provided on an irradiation surface of the light-emitting diode to adjust the half-width to an arbitrary value. The optical fiber is the same as in the thirteenth embodiment.

Embodiment 16 As shown in FIG. 10, the apparatus for quantifying the concentration of glucose in a living body according to the present embodiment emits light from a light emitting means 1 which is a halogen lamp and a part of the body of a subject ( For example, the optical fiber 4 for guiding the diffused transmitted light to the flat field type diffraction grating 20 while the light is guided to the forearm skin tissue B and the light subjected to the spectral operation by the diffraction grating 20 is received by the array type photodiode 5 to measure the absorption signal. The blood glucose level is calculated by the calculating means CPU. Arithmetic unit CPU
Converts the absorption signal into an absorbance and calculates the glucose concentration of the subject using a preset calibration curve. FIG.
Reference numeral 21 denotes a reflection mirror, 22 denotes a slit disposed between the diffraction grating 20 and the optical fiber 4, and 23 denotes an A / D converter.

As shown in FIG. 11, the optical fibers 4 used here are such that each light emitting side fiber 8 is surrounded by light receiving side fibers 9 located at each corner of a hexagonal pattern. Receiving fiber 9 and 1
In the set composed of the two light-emitting fibers 8, one light-receiving fiber 9 is shared by a pair adjacent in the X-axis direction, and two light-receiving fibers 9 are shared by a pair adjacent in the Y-axis direction.
Is shared. In each set, the center distance L between the light emitting fiber 8 and the light receiving fiber 9 is set to 0.5 mm (each diameter is 200 microns). When selectively extracting spectral information from the dermis of the subject's cortex, the center-to-center distance L is 0.1 to 2 mm,
In particular, it is preferable to set the thickness in the range of 0.2 to 1 mm.
The irradiation end 7 of the optical fiber 4 is pressed perpendicularly to the skin surface of the forearm of the subject. Optical fiber bundle 4
It is preferable to use a pressure gauge or a fixture to press the skin against the skin surface at a predetermined pressure. 1 from 1350nm
FIG. 12 shows an example of an absorption spectrum of forearm skin tissue in a wavelength range of 850 nm.

The glucose concentration in the subject's blood was determined using the above apparatus. A 30-year-old healthy male subject was rested for 30 minutes, then taken with a partially hydrolyzed starch preparation, Torayan G (manufactured by Shimizu Pharmaceutical Co., Ltd.), and ingested glucose equivalent to 75 g. Then, the blood glucose level was measured every ten minutes from the start of rest using fingertip blood using a simple blood glucose meter Novo Assist (manufactured by Novo Nordisk Pharma). The forearm spectrum was measured four times every 5 minutes after blood collection. (The glucose concentration in the forearm skin tissue has a time delay having individual differences with respect to the blood glucose concentration, but a time delay of 5 minutes was appropriate in this subject.) Blood glucose measurement by blood sampling started resting Until 90 minutes later, the spectral measurement of the forearm was 5
Minutes. The calibration curve was prepared using PLS regression analysis and multiple regression analysis with the measured blood glucose value as the target variable and the spectrum measured 5 minutes later as the explanatory variable. The blood sugar level of the subject in this actual measurement was 89 to 134 mg /
dl.

FIG. 1 shows the regression coefficients obtained by the PLS regression analysis.
3 is shown. In the PLS regression analysis, the first harmonic is observed using the cross-validation method.
In the wavelength range of 350 nm to 1850 nm, glucose concentration was used as a target variable, and absorbance was used as an explanatory variable. The shape of this regression coefficient is similar to the regression coefficient (FIGS. 3 and 6) obtained in the experiment using the serum system shown in the previous example and having a peak near 1600 nm. It can be seen that is determined. The result of this PLS regression analysis is an estimation of the number of principal components 7, and the correlation coefficient for creating a calibration curve is 0.993, the standard error SEP is 1.9 mg / dl,
The correlation coefficient of the calibration curve test was 0.988, and the standard error SEP was 2.6 mg / dl.

1550 nm, 164 in the same data
FIG. 14 shows a result obtained by applying measured data to a calibration curve obtained by performing multiple regression analysis by a cross-validation method using three wavelengths of 0 nm and 1690 nm. The results obtained by the three-wavelength multiple regression analysis were as follows: a correlation coefficient of 0.957 for preparing a calibration curve, a standard error of SEP 4.8 mg / dl, a correlation coefficient of a calibration curve test of 0.949, and a standard error of SEP 53 mg.
/ Dl.

[0083]

As described above, in the present invention, the concentration of glucose in a living body is measured in a wavelength range for measuring the absorption of glucose molecules derived from OH groups in the near-infrared region, and the absorption of biological components derived from NH groups is determined. And a quantitative range of glucose from the measurement results in at least three wavelength ranges of a wavelength range for measuring the absorption derived from the CH group of the biological component, and a specific absorption by glucose. In addition to using the most appropriate absorption from the OH group,
Since the biological components as disturbances are taken into account, the accuracy of the quantification of the glucose concentration can be greatly improved.

By setting the above three wavelength ranges to a range where the influence of water molecules derived from living body water is relatively small, it is easy to eliminate the influence of living body water.
The two wavelength ranges are mutually adjacent wavelength ranges, eliminating the need for a light source that emits light in a wide wavelength range.
Since the detection sensitivities of the detection means can be made uniform, quantification becomes easy.

Here, the above three wavelength ranges correspond to the first wavelength of the molecule.
When selecting in the wavelength range where overtones can be observed and the influence of water absorption is relatively small from 1480 nm to 1880 nm, the first wavelength range for measuring absorption derived from the OH group of glucose molecule is 1550 nm. 1650
nm, the wavelength range for measuring the absorption derived from the NH group of the biological component is from 1480 nm to 1550 nm, and the wavelength range for measuring the absorption derived from the CH group is from 1640 nm to 1880.
If the wavelength is selected from the range of 1000 nm to 1300 nm in which the second harmonic of the molecule can be observed and the influence of water absorption is relatively small, a glucose molecule can be obtained. The wavelength range for measuring the absorption derived from the OH group of
30 nm, the wavelength range for measuring the absorption derived from the NH group of the biological component is from 1000 nm to 1050 nm, and the wavelength range for measuring the absorption derived from the CH group is from 1130 nm to 13 nm.
Good results can be obtained when the thickness is set to 00 nm.

In quantification, glucose may be quantified by multivariate analysis of spectral signals obtained by continuously measuring each wavelength in the three selected adjacent regions. Further, in addition to selecting at least one wavelength from each of the three wavelength ranges, a predetermined range of wavelength bands may be selected to determine the glucose concentration. At this time, at least one of the above-mentioned wavelengths or wavelength bands has a positive or negative peak around a regression coefficient obtained by principal component regression analysis or PLS regression analysis of a continuous spectrum measured by quantitative analysis of a plurality of samples. As for the wavelength, a favorable result can be obtained.

Using a wavelength band of 1600 ± 40 nm from the first wavelength band or at least one wavelength within the wavelength band, the second
A wavelength band of 1530 ± 20 nm or at least one wavelength within the wavelength band from the third wavelength range.
85 ± 20 nm or 1715 ± 20 nm or 174
A good quantitative result can be obtained by using any one of the wavelength bands of 0 ± 20 nm or at least one wavelength within the wavelength band. In the measurement of glucose concentration in a biological tissue or a body fluid, the above three It is preferable to determine the glucose concentration based on light of at least three central wavelengths having a predetermined half width selected from the wavelength range. An easily available light source can be used.

In this case, the half value width is obtained by analyzing a continuous spectrum measured in the analysis of a plurality of samples by principal component regression analysis or PLS.
If a regression analysis is performed and a wavelength width showing a value of about 70% or less of a peak value taken by a regression coefficient for each wavelength calculated by the analysis method is used, a better result can be obtained, and at least one wavelength can be obtained. Center wavelength 1560nm
And the half width is set to 60 nm or less.

Based on data obtained by performing a pre-process of obtaining a difference or a quotient at a wavelength in the near-infrared region from an absorption signal obtained as a result of measurement in three wavelength ranges or a value obtained by converting the absorption signal into an absorbance or a transmittance. The preprocessing in this case may include a regression coefficient for a plurality of principal component factors calculated by performing a principal component regression analysis or a PLS regression analysis on a continuous spectrum measured by analyzing a plurality of samples. The difference or quotient at the wavelength near the intersection of the two may be obtained. Further, as a wavelength for preprocessing, 1550 is used in the first harmonic region.
Wavelength selected from ± 10 nm or 1650 ± 10 nm, and 1060 ± 10 nm or 11 in the second harmonic region
It is preferable to use a wavelength selected from 30 ± 10 nm.

The apparatus for quantitatively determining glucose concentration according to the present invention comprises a near-infrared light source, a detecting means for detecting near-infrared light, and a device for introducing near-infrared light emitted from the near-infrared light source into living tissue or body fluid. Then, a guiding hand for guiding the near-infrared light transmitted or diffusely reflected by the living tissue or body fluid to the detecting means, and a calculating hand for calculating a signal obtained by the detecting means and converting the signal into a glucose concentration. The detection means measures the wavelength range for measuring the absorption derived from the OH group of the glucose molecule, the wavelength range for measuring the absorption derived from the NH group of the biological component, and the absorption derived from the CH group of the biological component. Since it detects near-infrared light in at least three wavelength ranges, the quantification of glucose concentration with high accuracy can be easily performed.

The near-infrared light source includes glucose molecules O
A first wavelength range for measuring the absorption derived from the H group, a second wavelength range for measuring the absorption derived from the NH group of the biological component, and a second wavelength range for measuring the absorption derived from the CH group of the biological component. Third
It has a light emission characteristic covering at least three wavelength ranges of the above wavelength range, and may also be provided with a spectroscopic means for splitting light into each wavelength range, or for measuring absorption derived from OH group of glucose molecule. A first wavelength range, and a second wavelength range for measuring absorption derived from the NH group of the biological component,
A light source having a plurality of light sources having emission characteristics in at least three wavelength ranges of a third wavelength range for measuring absorption derived from a CH group of a biological component may be used. Without using a light source having a light emission characteristic at a wide wavelength such as a halogen lamp, it is possible to determine glucose using an element having a light emission characteristic at a relatively narrow wavelength such as a light emitting diode.

[Brief description of the drawings]

FIG. 1 is an absorption spectrum diagram of a biological material in a first overtone region.

FIG. 2 is an absorption spectrum diagram of a biological substance in a second overtone region.

FIG. 3 is an explanatory diagram of a regression coefficient in an experimental system in which three components of glucose, albumin, and water are varied in a first harmonic region.

FIG. 4 is an explanatory diagram of a regression coefficient of a second derivative in an experimental system in which three components of glucose, albumin, and water are changed in a first harmonic region.

FIG. 5 is an explanatory diagram of regression coefficients in an experimental system in which three components of glucose, albumin, and water are varied in a second overtone region.

FIG. 6 is an explanatory diagram of regression coefficients in an experimental system in which five components of glucose, albumin, cholesterol, neutral fat, and water are varied in the first overtone region.

FIG. 7 is a schematic view of an example of the device according to the present invention.

8 is a partially enlarged view of the regression coefficient of FIG.

FIG. 9 is a schematic view of another example of the device according to the present invention.

FIG. 10 is a schematic view of still another example of the device according to the present invention.

FIG. 11 is a cross-sectional view of the above optical fiber.

FIG. 12 is an explanatory diagram of an example of an absorption spectrum of skin tissue.

FIG. 13 is an explanatory diagram of regression coefficients obtained by PLS regression analysis.

FIG. 14 is an explanatory diagram of a calibration curve of multiple regression analysis and actually measured data.

FIG. 15 is a diagram illustrating the optical path of near-infrared light emitted from a fiber to living tissue.

FIG. 16 is a conceptual diagram of human skin tissue.

Claims (18)

[Claims] 1. A method for quantifying glucose concentration in living tissue or body fluid using light absorption in the near-infrared region, comprising a first wavelength range for measuring absorption of glucose molecules derived from OH groups. Based on measurement results in at least three wavelength ranges of a second wavelength range for measuring the absorption derived from the NH group of the biological component and a third wavelength range for measuring the absorption derived from the CH group of the biological component. A method for quantifying glucose concentration, characterized in that glucose is quantified by performing the method. 2. The method for quantifying glucose concentration according to claim 1, wherein the three wavelength ranges are set within a range in which the influence of absorption of water molecules derived from living body water is relatively small. 3. The method according to claim 1, wherein the three wavelength ranges are adjacent to each other. 4. The three wavelength ranges are selected within a wavelength range in which the first overtone of the molecule can be observed and within a wavelength range from 1480 nm to 1880 nm where the influence of water absorption is relatively small, and the OH group of the glucose molecule is selected. The first wavelength range for measuring the absorption originating from 1550 nm to 1650 n
m, the wavelength range for measuring the absorption derived from the NH group of the biological component is from 1480 nm to 1550 nm, and the wavelength range for measuring the absorption derived from the CH group is from 1650 nm to 1880 n.
The method for quantifying glucose concentration according to any one of claims 1 to 3, wherein m is defined as m. 5. The three wavelength ranges are selected within a wavelength range where the second overtone of the molecule can be observed and within a wavelength range of 1000 nm to 1300 nm where the influence of water absorption is relatively small, and the OH group of the glucose molecule is selected. The wavelength range for measuring the absorption originating from 1050 nm to 1130 nm, and the wavelength range for measuring the absorption originating from the NH group of the biological component is 1
The method for quantifying glucose concentration according to any one of claims 1 to 3, wherein the wavelength range for measuring absorption derived from a CH group is from 000 nm to 1050 nm and from 1130 nm to 1300 nm. 6. The quantitative determination of glucose by multivariate analysis of a spectrum signal obtained by continuously measuring each wavelength in the selected three adjacent bands. 4. The method for quantifying a glucose concentration according to 1. 7. The method for quantifying glucose concentration according to claim 1, wherein the quantification of glucose concentration is performed by selecting at least one wavelength or one wavelength band from each of the three wavelength regions. . 8. A positive or negative peak of a regression coefficient obtained by principal component regression analysis or PLS regression analysis of a continuous spectrum measured by quantitative analysis of a plurality of samples, at least one of the wavelengths or wavelength bands. The method for quantifying glucose concentration according to claim 7, wherein the wavelength is a peripheral wavelength. 9. A wavelength band of 1600 ± 40 nm from the first wavelength band or at least one wavelength within the wavelength band, and a wavelength band of 1530 ± 20 nm or at least one wavelength within the wavelength band from the second wavelength band. From the third wavelength range.
685 ± 20 nm or 1715 ± 20 nm or 17
5. The method for quantifying glucose concentration according to claim 4, wherein any one wavelength band of 40 ± 20 nm or at least one wavelength within the wavelength band is used. 10. A method for measuring glucose concentration in a living tissue or a body fluid, wherein the glucose concentration is determined based on light of at least three central wavelengths having a predetermined half-width selected from the three wavelength ranges. The method for quantifying a glucose concentration according to any one of claims 1 to 5, which is characterized in that: 11. A continuous spectrum measured by analyzing a plurality of samples is subjected to principal component regression analysis or PLS regression analysis, and shows a value of about 70% or less of a peak value obtained by a regression coefficient for each wavelength calculated by the analysis method. The method for quantifying glucose concentration according to claim 10, wherein the wavelength width is a half width. 12. The central wavelength of at least one wavelength is 156.
11. The semiconductor device according to claim 10, wherein the half-value width is set within a range of 0 nm to 1640 nm and the half width is set to 60 nm or less.
A method for quantifying glucose concentration according to the above. 13. A pre-processed data which takes a difference or a quotient at a wavelength in a near-infrared region from an absorption signal obtained as a measurement result in three wavelength ranges or a value obtained by converting the absorption signal into an absorbance or a transmittance. The method for quantifying glucose concentration according to any one of claims 1 to 12, wherein the quantification is performed based on: 14. As preprocessing, a continuous spectrum measured by analyzing a plurality of samples is subjected to principal component regression analysis or PLS.
14. The method according to claim 13, wherein a difference or a quotient at a wavelength near an intersection between regression coefficients for a plurality of principal component factors calculated by regression analysis is calculated. 15. The wavelength for preprocessing is 1550 ± 10 nm or 1650 ± 10n in the first harmonic region.
m, 1060 ± 10 in the second overtone region
14. The method for determining a glucose concentration according to claim 13, wherein a wavelength selected from nm or 1130 ± 10 nm is used. 16. A near-infrared light source, detecting means for detecting near-infrared light, and introducing near-infrared light emitted from the near-infrared light source into living tissue or body fluid to transmit or pass through the living tissue or body fluid. A guiding means for guiding the diffused and reflected near-infrared light to the detecting means, and a calculating means for calculating a signal obtained by the detecting means and converting the signal into a glucose concentration, wherein the detecting means comprises OH of glucose molecules. At least three wavelength ranges of a wavelength range for measuring the absorption derived from the group, a wavelength range for measuring the absorption derived from the NH group of the biological component, and a wavelength range for measuring the absorption derived from the CH group of the biological component. An apparatus for detecting glucose concentration, which detects near-infrared light in a region. 17. The near-infrared light source is OH of glucose molecule.
A first wavelength range for measuring the absorption derived from the group, a second wavelength range for measuring the absorption derived from the NH group of the biological component, and a second wavelength range for measuring the absorption derived from the CH group of the biological component. 3
17. The apparatus for quantifying glucose concentration according to claim 16, wherein said apparatus has a light emission characteristic covering at least three wavelength ranges of said wavelength range, and further comprises spectral means for spectrally separating each wavelength range. 18. The near-infrared light source is OH of glucose molecule.
A first wavelength range for measuring the absorption derived from the group, a second wavelength range for measuring the absorption derived from the NH group of the biological component, and a second wavelength range for measuring the absorption derived from the CH group of the biological component. 3
The glucose concentration quantifying device according to claim 16, comprising a plurality of light sources having emission characteristics in at least three wavelength ranges of the wavelength range.