RU2545814C1 - Method of determining physical-biological parameters of skin and concentration of haemoglobin derivatives in blood - Google Patents

Abstract

FIELD: medicine.

SUBSTANCE: invention relates to the field of medicine and can be applied in medical diagnostics and therapy. A method of determining physical-biological parameters of skin and the concentration of haemoglobin derivatives in blood includes sending onto the skin polarised optical irradiation with a known spectrum, registration of signals of light diffusively-reflected by the skin with the polarisation, orthogonal to the polarisation of the sent onto skin radiation. Optical radiation is focused onto the skin surface into an irradiation spot, signals of light diffusively-reflected by the skin P(ρ, λ) in the spectral range λ=450-800 nm not less than from two ring areas on the skin surface, located at different distances ρ from the irradiation spot, are registered. Then the difference signals r(ρ, λ)=ln(P(ρ₀, λ)/P(ρ, λ)) are determined, where ρ₀ is a distance to the registration ring nearest to the irradiation spot, and the physical-biological parameters of the skin and the concentration of haemoglobin derivatives are determined by solution of an inverse problem with the application of analytical expressions, which approximate the dependence r(ρ, λ) on the determined parameters.

EFFECT: invention ensures accurate determination of parameters without the necessity to carry out calibrating changes and the increase of quantity of the determined parameters.

6 dwg,1 tbl

Description

The invention relates to the field of medical instrumentation.

A known method for determining the content of hemoglobin derivatives in the blood [1], based on measuring the optical density of a hemolyzed one percent blood solution at wavelengths in the range 450-650 nm and determining the concentrations of oxyhemoglobin, deoxyhemoglobin, carboxyhemoglobin and methemoglobin, from solving the system of equations

where D (λ _i ) is the optical density of the blood solution at the i-th wavelength; l is the thickness of the blood layer; ε _j (λ _i ) are the molar absorption coefficients of the j-th hemoglobin derivative at the i-th wavelength; C _j is the concentration of the jth derivative; n is the number of analyzed hemoglobin derivatives; m is the number of wavelengths used; i = 1, 2, ..., m. This method is not efficient and requires highly qualified medical staff, since it involves taking a blood sample and processing it in a transforming solution. In addition, taking a blood sample with a syringe is associated with injury to the patient and the risk of infection.

There are a number of methods for non-invasive analysis of hemoglobin levels in the blood, using the principles of LED pulse oximetry [2-4]. These methods include measuring the time dependence of the light flux passing through a layer of tissue with pulsating arterial blood; the exclusion of the constant component of the flow due to capillaries, veins and other non-pulsating tissue elements; determination of the optical density D (λ _i ) of arterial blood at characteristic wavelengths λ _i (i = 1, ..., m) of the analyzed hemoglobin derivatives and calculation of the concentrations C _j (j = 1, ..., n) of hemoglobin derivatives in arterial blood from the solution of the system equations of the form (1), where l is the path length of the photon in the tissue. The disadvantages of these methods include their binding to cardioscillations of blood flow in the tissue, which significantly limits their scope, as well as the effect on the measurement result of the width of the emission lines of LEDs and the scattering properties of the tissue, which determine the value of l and its spectral dependence. In addition, due to the substantial overlap of the absorption spectra of hemoglobin derivatives, a system of m equations (1) with n unknowns at m ~ n is poorly conditioned, and its solution is unstable to optical density measurement errors. As a result, these methods do not allow achieving the accuracy of the analysis of hemoglobin composition necessary for medical practice.

A known method and device for determining the physico-biological characteristics of the skin and mucous membranes in vivo [5], based on the measurement of signals (P) of diffuse reflection of tissue in the spectral regions of 460-490 nm, 520-590 nm, 650-690 nm and 900 950 nm; determination of optical density D = log (P _ref / P), where P _ref is the reference signal from the light-scattering standard; and calculating the concentrations of optically active tissue components (oxyhemoglobin, deoxyhemoglobin, melanin, collagen) from solving a system of linear equations (1) for the optical density of a multicomponent medium. In [6], a method essentially similar is described, characterized in that the optical density D (λ) is measured at wavelengths λ = 560, 578, 620 and 720 nm. These methods are based on the principle of additivity of optical densities, according to which the optical density of a multicomponent medium is equal to the sum of the optical densities of its individual components. Under the conditions of multiple scattering of light that occurs in biological tissues, this principle, strictly speaking, is not applicable, since the absorption of light by each component depends on the optical path traveled by light in the medium, taking into account its multiple scattering and absorption by all other components of the medium. In addition, these methods require calibration measurements and do not allow to determine the blood levels of dyshemoglobins (carboxy-, met- and sulfhemoglobin), which in some cases significantly reduces the accuracy of diagnosis and makes these methods not applicable for some pathological conditions of the human body.

There is also a method for determining the concentrations of optically active components of tissue [7], based on summing the probe radiation to the tissue and collecting the scattered light flux using optical fibers, measuring the reflectance of the tissue in the UV and visible regions of the spectrum, and modeling the theoretical reflectance spectrum of the tissue in relation to the experimental by selecting model parameters. This method involves the contact of the fiber-optic sensor with tissue, which leads to a change in its optical parameters (due to slight pressure) and distorts the measurement result. In addition, this method requires the use of a calibration diffuse reflector with known and non-changing optical parameters over time.

Closest to the claimed invention is a method for determining the concentration of total hemoglobin in the skin and the degree of oxygenation of the blood [8], including lighting the skin with polarized radiation with a known spectrum; registration of a multispectral image of the skin with polarization, orthogonal polarization of the illuminating radiation; converting the spectral layers of the skin image into the spectral values of its diffuse reflection coefficient by comparing the skin image with the image of a white diffuse reflector; and determining skin parameters from solving a system of equations for the spectral values of its diffuse reflection coefficient.

This method allows you to determine the parameters of the skin without contact with her measuring device. However, there remains a need for calibration measurements. In addition, in this method, light diffusely reflected from the skin is collected within the area of its illumination. The corresponding photons pass through the skin in a relatively small optical path, which is reflected in the low sensitivity of the recorded light fluxes to the concentrations of skin and blood components. In this regard, this method does not allow to determine pathological forms of hemoglobin, the absorption spectra of which substantially overlap with similar spectra of the main forms of hemoglobin.

The present invention is aimed at improving the accuracy of determining the physico-biological parameters of the skin and the concentration of hemoglobin derivatives in the blood, eliminating the need for calibration measurements and expanding the functionality of the method, by increasing the number of determined parameters.

To solve these problems in a method for determining the physico-biological parameters of the skin and the concentration of hemoglobin derivatives in the blood by sending polarized optical radiation with a known spectrum to the skin, recording signals of light diffusely reflected by the skin with polarization, orthogonal polarization of the radiation sent to the skin, the optical radiation is focused on the surface skin in the irradiation spot, the signals of diffusely reflected light of the skin P (ρ, λ) are recorded in the spectral range λ = 450-800 nm from at least two ring neoplasms of the skin surface located at different distances from the spot irradiation ρ; determining the difference signals r (ρ, λ) = ln (P (ρ ₀ , λ) / P (ρ, λ)), where ρ ₀ is the distance to the registration ring closest to the irradiation spot; and the physico-biological parameters of the skin and the concentration of hemoglobin derivatives in the blood are determined by solving the inverse problem using analytical expressions that approximate the dependence of r (ρ, λ) on the determined parameters.

Measurement of signals diffusely reflected by the skin of light at various distances from the skin irradiation area eliminates the need to use a reference reflector and increase the convenience of practical use of the method. The sensitivity of the method to the concentrations of skin and blood components is increased due to the larger optical path that light travels in the skin tissue. Analytical expressions linking the measured signals with the physico-biological parameters of the skin and the concentration of hemoglobin derivatives in the blood can improve the accuracy of diagnosis and increase the number of determined parameters.

The essence of this invention is illustrated using figures 1-6.

Figure 1 presents a possible view of a device for measuring the spectral and spatial characteristics of diffuse reflection of the skin.

Figure 2 shows the difference signals of light diffusely reflected by the skin at ρ ₀ = 0.4 mm and ρ = 0.8 (1), 1.2 (2), 1.6 (3) mm, modeled by the Monte Carlo method (points) and calculated analytically (curves) .

Figure 3 compares the known (abscissa axis) and reconstructed from signals r (ρ, λ) (ordinate axis) concentrations of oxyhemoglobin.

Figure 4 compares the known (abscissa axis) and reconstructed from signals r (ρ, λ) (ordinate axis) methemoglobin concentrations.

Figure 5 compares the known (abscissa axis) and reconstructed from signals r (ρ, λ) (ordinate axis) carboxyhemoglobin concentrations.

Figure 6 compares the known (abscissa axis) and reconstructed from signals r (ρ, λ) (ordinate axis) sulfhemoglobin concentrations.

One of the possible schemes of a non-contact optical meter of skin parameters is shown in figure 1. The meter contains a broadband radiation source 1, transmitting optical fibers 2 and 4, a monochromator 3, mutually orthogonal polarizers 5 and 8, a focusing lens device 6, a monochrome (black and white) CCD camera 9, a control unit for the radiation source, recording and processing signals 10 Using the lens device 6, you can provide any diameter of the irradiation spot, and using the CCD camera 9 to obtain a two-dimensional distribution of the flux of radiation scattered by the skin. Thus, measurements can be made remotely, i.e. without contact of the meter with the patient’s body. The angle of incidence of radiation on the skin should be selected so that the radiation reflected from the surface 7 does not fall into the camera lens, and only diffuse radiation reflected by the inner layers of the skin is recorded. The influence of radiation reflected from the surface on the recorded optical signals can also be eliminated by using mutually orthogonal polarizers 5 and 8 in the channels for sending and recording radiation. Since radiation reflected by the surface retains the original polarization, the use of crossed polarizers allows blocking this interference component of the signals.

Based on the considered scheme, it is possible to measure azimuthally averaged scattered radiation fluxes F (ρ, λ) at various distances ρ from the center of the irradiation spot. To eliminate the need to manufacture and use a calibration sample with optical parameters known in a wide spectral range and not changing with time, we will only operate with differential signals of diffuse reflection (DO)

- мощность излучения, попадающего в объектив камеры с участка кожи, ограниченной кольцом радиусом ρ и толщиной Δρ; ρ₀ - радиус ближайшего к области облучения приемного кольца.Where

- the power of the radiation entering the camera lens from a skin area bounded by a ring of radius ρ and thickness Δρ; ρ ₀ is the radius of the receiving ring closest to the irradiation region.

To date, the best method to calculate the light scattering characteristics of a multilayer medium without restrictions on its OD and experiment geometry is the Monte Carlo (MK) method. This method is based on repeated repetition of a numerical experiment to calculate a random photon trajectory in the medium under study, followed by a generalization of the results. Each photon is characterized by its own "weight", Cartesian coordinates, which determine its position in the medium, and directing cosines, which determine the direction of its motion. The initial "weight" of each photon is equal to one. When a photon “wanders” in a medium, its “weight” decreases due to Fresnel reflection from the surface of the medium, as well as absorption and scattering processes in the medium. The trajectory of the photon is traced until its “weight” becomes less than a predetermined value ( ^{10–4 was} used in our calculations), or the photon does not go beyond the boundaries under consideration. The determination of the initial conditions for introducing photons into the medium and the conditions for their detection for given geometric parameters of the experiment are based on the approach of beam optics.

To calculate the signals r (ρ, λ), the radial distribution of photons backscattered by the medium is important. The MK method allows one to obtain it in the form of an array H [i], each element of which is equal to the total "weight" of photons emitted from the medium at distances ρ _i from the center of the irradiation spot within the solid angle at which the camera lens is visible from the photon exit point. The wavelength of light in the MK method is implicitly specified through the optical parameters of the medium. For this, an optical model of the skin is used.

The skin is modeled as a medium consisting of two layers (epidermis and dermis) with the same light scattering parameters (β, g) and different absorption coefficients. The stratum corneum, due to its small optical thickness, plays an extremely insignificant role in the diffuse reflection of light; therefore, it is conditionally included in the epidermis. The anatomical areas of the dermis (nipple, reticular, superficial and deep plexus of the vessels) have no clear physical boundaries, no fundamental morphological differences, so they are all replaced by one homogeneous layer. The deeper layers of the skin (fatty layer and muscle tissue) practically do not participate in the process of light reflection with λ = 450-800 nm due to its strong attenuation by the overlying layers. Model parameters are: n - skin refractive index; β ′ (λ ₀ ) is the transport coefficient of scattering of connective tissue at λ ₀ = 600 nm; ρ _Mie is the fraction of Mie scattering in the total tissue scattering at λ = 400 nm; x is the spectral parameter of the transport scattering coefficient Mie; L _e is the thickness of the epidermis; f _m - volumetric concentration of melanin in the epidermis; C _bil is the concentration of bilirubin in the dermis; f _bl - volumetric concentration of capillaries in the dermis; d _v is the average diameter of the capillaries; C _tHb is the concentration of total hemoglobin in the blood; S is the degree of blood oxygenation. The optical parameters of the skin are calculated by the formulas

, ε_Hb и ε_bil - молярные коэффициенты поглощения оксигемоглобина, деоксигемоглобина и билирубина в мм^-1/(моль/л); µ_tHb=64500 г/моль - молярная масса гемоглобина; µ_bil=585 г/моль - молярная масса билирубина; t_dif=5 - отношение концентраций билирубина в крови и в окружающей ткани, в общем случае, зависящее от коэффициента диффузии билирубина через стенки кровеносных сосудов; α - поправочный коэффициент, учитывающий эффект локализованного поглощения света кровеносными сосудами [10]where β ′ and g are the transport scattering coefficient and the scattering anisotropy factor of the epidermis and dermis [9]; k _e and k _d are the absorption coefficients of the epidermis and dermis; k _t is the absorption coefficient of connective tissue (bloodless); k _bl is the absorption coefficient of blood; $${\epsilon }_{\mathrm{<figure-callout\; id="2"\; label="H\; b\; O"\; filenames="00000016.png,00000017.png"\; state="\{\{state\}\}">H\; </figure-callout>}b{O}_{}{}_2}$$

, ε _Hb and ε _bil are the molar absorption coefficients of oxyhemoglobin, deoxyhemoglobin and bilirubin in mm ^-1 / (mol / l); µ _tHb = 64500 g / mol - molar mass of hemoglobin; µ _bil = 585 g / mol - molar mass of bilirubin; t _dif = 5 - the ratio of the concentrations of bilirubin in the blood and in the surrounding tissue, in general, depending on the diffusion coefficient of bilirubin through the walls of blood vessels; α - correction factor that takes into account the effect of localized absorption of light by blood vessels [10]

The MK method, with all its advantages, is not convenient for practical use, since it does not allow the processing of experimental data in real time. In this regard, we have obtained expressions that allow us to calculate the optical signals r (ρ, λ) measured in the experiment in a simple analytical form, moreover, with the accuracy of the MK method. For this, random values of model parameters were generated from the ranges of their variations for moderately pigmented human skin: n = 1.4-1.5, β ′ (λ ₀ ) = 3-11 mm ^-1 , ρ _Mie = 0.1-0.6, x = 0.5-1.0 , L _e = 50-150 microns, f _m L _e = 0.5-10 microns, f _bl = 0.2-7%, C _tHb = 120-200 g / l, d _v = 5-60 microns, C _bil = 0.1- 50 mg / L, S = 40-98%. For each implementation of the model parameters, we calculated the absorption coefficients of the epidermis k _e (λ) and dermis k _d (λ), as well as the transport tissue scattering coefficient β ′ (λ) at 30 wavelengths from the spectral range λ = 450-800 nm. The Heni-Greenstein function with the empirical anisotropy parameter (2) was used as an indicator of tissue scattering.

In accordance with the generated values of n and _Le , as well as the calculated values of k _e (λ), k _d (λ), β ′ (λ) and g (λ), the signals r (ρ, λ) corresponding to a specific geometry were simulated by the MC method experiment. Thus, an ensemble of 10 ³ random realizations of r (ρ, λ) was formed. The spectral values of all of the optical parameters of the skin and their respective signals r (ρ, λ) are combined into a single data set covering the following _{ranges: n = 1.35-1.50, L e =} 50-150 m, k _e = 0.09-18 mm ^-1 , k _d = 0.02-3.0 mm ^-1 , β ′ = 0.5-8 mm ^-1 , β ′ / k _e = 0.15-35, β ′ / k _d = 0.7-150 (the range r (ρ) depends on the measurement geometry ) This approach to obtaining analytical expressions approximating the results of numerical calculations r (ρ) allows us to take into account the physical conditionality of the optical parameters of the tissue and the relationships between them characteristic of the spectral range under consideration, as well as to exclude combinations of optical parameters not encountered in reality.

Based on the simulated data, an analytical expression was searched that best approximates the dependence r (ρ, n, _Le , k _e , k _d , β, g). As a result, the following expression was obtained:

- глубина проникновения света в дерму (в диффузионном приближении). В качестве примера рассмотрим аппроксимацию формулой (8) сигналов r(ρ), рассчитанных методом МК для конкретной геометрии эксперимента. Будем полагать, что кожа освещается по нормали пучком света диаметром 0.2 мм, а диффузное излучение регистрируется с четырех концентрических колец толщиной 0.4 мм, расположенных на расстояниях ρ=0.4, 0.8, 1.2 и 1.6 мм от центра пятна облучения. Поскольку рассматриваемые размеры пятна облучения и области регистрации обратно-рассеянного кожей излучения во много раз меньше диаметров типичных объективов, то с каждого из колец камерой захватывается примерно одинаковая доля диффузного потока. В связи с этим разностные сигналы r(ρ)=ln(P(ρ₀=0.4 мм)/P(ρ)) практически не зависят от положения объектива относительно исследуемой поверхности кожи. Это обстоятельство позволяет при расчете сигналов r(ρ) учитывать все фотоны, вылетающие с приемных колец на поверхности кожи (во всех направлениях), и тем самым уменьшить число моделируемых траекторий фотонов в методе МК. Коэффициенты формулы (8), отвечающие рассматриваемой геометрии эксперимента, приведены в таблице.where a _{i, m} are approximation coefficients, s is a vector of geometric parameters of the experiment, $${\delta }_d=\mathrm{one}/\sqrt{3k_d\left(k_d+\beta \text{'}\text{'}\right)}$$

- the depth of penetration of light into the dermis (in the diffusion approximation). As an example, we consider the approximation by the formula (8) of the signals r (ρ) calculated by the MK method for a specific experiment geometry. We assume that the skin is normally illuminated by a 0.2 mm diameter light beam, and diffuse radiation is detected from four 0.4 mm thick concentric rings located at distances ρ = 0.4, 0.8, 1.2, and 1.6 mm from the center of the irradiation spot. Since the dimensions of the irradiation spot and the detection region of the back-scattered skin radiation are considered many times smaller than the diameters of typical lenses, approximately the same fraction of the diffuse flux is captured from each of the rings by the camera. In this regard, the difference signals r (ρ) = ln (P (ρ ₀ = 0.4 mm) / P (ρ)) are practically independent of the position of the lens relative to the studied surface of the skin. This circumstance allows us to take into account all the photons emitted from the receiving rings on the skin surface (in all directions) when calculating the r (ρ) signals, and thereby reduce the number of simulated photon trajectories in the MK method. The coefficients of formula (8) corresponding to the considered geometry of the experiment are given in the table.

Table The coefficients of the approximation formula (8) with the dimensions of the parameters of the medium [L _e ] = [mm], [β ′] = [mm ^-1 ], [k _e ] = [mm ^-1 ], [k _d ] = [mm ^-1 ] a _{i, m} ρ = 0.8 mm ρ = 1.2 mm ρ = 1.6 mm a _{i, m} ρ = 0.8 mm ρ = 1.2 mm ρ = 1.6 mm a _0,0 2,2908 4,2329 4.4594 a _7.1 -3.4910 -6.0053 -7.7473 a _1,1 -17,265 -30.815 -30,691 a _7.2 12,364 27,124 42,859 a _1,2 37,495 66,476 65,049 a _7.3 -48,980 -106,982 -169.41 a _1.3 -27,534 -48,489 -46.871 a _8.1 -1.9499 -1.7280 -0.9544 a _2.1 0.7718 3,3395 4.7531 a _8.2 0.8034 -0.2505 -1.7573 a _2.2 0.2150 -0,1003 -0.1574 a _8.3 -0.1845 0.1044 0.5347 a _2,3 -0.0223 -0,0008 0.0013 a _9.1 1,1811 0.7599 -0.7378 a _3,1 2,0686 4,2509 6,3013 a _9.2 6.3160 12,919 21,199 a _3.2 -0.2983 -0.5931 -0.7892 a _9.3 -2.6891 -5.4070 -9.0289 a _3.3 0,0313 0,0593 0,0683 a _10.1 0.8898 2,5010 4,1066 a _4.1 -0.0025 -0.0050 -0.0075 a _10,2 2,3165 2,4289 2,6298 a _4.2 0.0000 0.0000 0.0001 a _10.3 -0.7669 0.4727 1,5329 a _4.3 0.0000 0.0000 0.0000 a _11.1 -0.8712 -0.5328 0.3335 a _5.1 -0.4900 -4.0883 -6.0548 a _11,2 0.3440 0.3311 0,0583 a _5.2 -0.7413 0,0572 0.1783 a _11.3 -0.0448 -0.0661 -0.0414 a _5.3 0,1260 0,0297 0,0258 a _12.1 -0.6436 -1.3295 -1.1761 a _6.1 2,5517 3.7464 4,3218 a _12,2 -3,2607 -4,8008 -7.7719 a _6.2 -3.0923 -6,9977 -10.842 a _12.3 0.6348 -0.1790 0,0452 a _6.3 1.7461 3,8648 6.0903 - - - -

The results of calculating r (ρ) by the MK method and by formula (8) (with the coefficients found) for ρ = 0.8, 1.2 and 1.6 mm differ on average by 1.0, 1.6 and 2.5%, respectively. The increase in the approximation error r (ρ) with increasing ρ is associated with the statistical “noise” inherent in the MC method, which increases with increasing distance from the photon entry point into the medium.

Thus, expressions (2) - (8) make it possible to quickly and with high accuracy calculate the experimentally measured signals in the approximation of a two-layer medium simulating skin tissue. As an example of such calculations, Fig. 2 shows the dependences r (ρ, λ) calculated numerically and analytically of the following values of the model parameters: n = 1.45, β ′ (λ ₀ ) = 6.3 mm ^-1 , ρ _Mie = 0.4, x = 0.6, L _e = 100 μm, f _m = 3%, C _bil = 4 mg / L, f _bl = 1%, d _v = 15 μm, C _tHb = 150 g / L, S = 75%. It can be seen that the difference between them is within the error of the MK method.

The inverse problem is solved by minimizing the discrepancy between the experimental and the spectral-spatial profile r (ρ, λ) predicted using the theoretical model

- экспериментальные и расчетные сигналы ДО кожи, x=(x_m) - вектор подлежащих определению модельных параметров, N_λ - количество длин волн оптического зондирования, N_ρ - количество пространственных колец регистрации рассеянного излучения, ω_k и ν_ik - весовые коэффициенты, обратно пропорциональные среднеквадратичным погрешностям измерений.where r (ρ _i , λ _k ) and $$\tilde{r}\left(x,{\rho }_i,{\lambda }_k\right)$$

- experimental and calculated signals BEF the skin, x = (x _m ) is the vector of the model parameters to be determined, N _λ is the number of wavelengths of optical sounding, N _ρ is the number of spatial rings for recording scattered radiation, ω _k and ν _ik are weight coefficients, back proportional to rms measurement errors.

At the stage of obtaining the approximating formula (8), we modeled an ensemble of 10 ³ random realizations of r (ρ, λ). The calculation was performed for 30 values of λ from the range of 450-800 nm. The geometric parameters of the measurement scheme at which the signals r (ρ, λ) were calculated are indicated above. The ranges of variations of the model parameters corresponding to the simulated ensemble of realizations r (ρ, λ) completely cover the real values of the parameters of moderately pigmented skin. In this regard, this ensemble can be used to assess the accuracy of solving the inverse problem. For this, all implementations of r (ρ, λ) were sorted out and minimization of the residual (9) was performed for each. The coefficients ν _{ik were} taken equal to 0.5 for λ _k <600 nm and 1.0 for λ _k ≥600 nm. The most important parameters of the dermis were considered: the concentration of total hemoglobin in the tissues of the dermis F _tHb = f _bl C _tHb (g / l); hemoglobin saturation with oxygen S (%); the concentration of bilirubin C _bil (mg / l); the parameters C _s (mm ^-1 ) and ν of the spectral dependence of the transport coefficient of tissue scattering β ′ (λ) = C _s · (632 / λ) ^ν in the range of 600-700 nm. The average errors in the recovery of these parameters are δF _tHb = 6.2%, δS = 2.0, δC _bil = 6.4%, δC _s = 4.0% and δν = 4.3%.

Consider the possibility of an express analysis of hemoglobin concentrations in the blood of the proposed method. We will consider on the basis of the skin model described above, in which, in addition to hydroxy- (HbO ₂ ) and deoxyhemoglobin (Hb), we additionally take into account the presence of carboxy- (COHb), meth- (MetHb) and sulfhemoglobin (SHb) in the blood. Given these hemoglobin derivatives, the spectral dependence of the blood absorption coefficient is described by the following expression:

, ε_Hb, ε_MetHb, ε_COHb, ε_SHb - молярные коэффициенты поглощения производных гемоглобина в мм^-1 (моль/л). В остальном модель оставим без изменений. На основе данной модели методом МК рассчитаны 350 реализаций потоков диффузно отраженного кожей излучения в спектральном диапазоне 450-800 нм с шагом 5 нм. Расчет r(ρ, λ) выполнялся для геометрических параметров эксперимента, соответствующих ρ₀=0.4 мм и ρ=0.8, 1.2, 1.6 мм. Диапазоны вариаций модельных параметров, соответствующие полученному ансамблю реализаций r(ρ, λ) указаны выше, с тем отличием, что концентрация билирубина в тканях дермы C_bil варьируется в диапазоне 0-5 мг/л (нормальный уровень билирубина), а концентрация кровеносных сосудов f_bl в типичном для нормальной (не опухолевой) кожи в диапазоне 1-3%. Для концентраций MetHb, COHb и SHb выбраны следующие диапазоны: C_MetHb=1-20%, C_COHb=1-20%, C_SHb=0.2-10%.where C _tHb is the concentration of total hemoglobin in the blood (g / l); µ _tHb = 64500 g / mol - molar mass of hemoglobin; S, C _MetHb , C _COHb , C _SHb — relative concentrations of HbO ₂ , MetHb, COHb and SHb; $${\epsilon }_{\mathrm{<figure-callout\; id="2"\; label="H\; b\; O"\; filenames="00000016.png,00000017.png"\; state="\{\{state\}\}">H\; </figure-callout>}b{O}_{}{}_2}$$

, ε _Hb , ε _MetHb , ε _COHb , ε _SHb are the molar absorption coefficients of hemoglobin derivatives in mm ^-1 (mol / l). The rest of the model will remain unchanged. Based on this model, the MK method was used to calculate 350 realizations of the fluxes of diffusely reflected radiation by the skin in the spectral range 450-800 nm with a step of 5 nm. The calculation of r (ρ, λ) was performed for the geometric parameters of the experiment, corresponding to ρ ₀ = 0.4 mm and ρ = 0.8, 1.2, 1.6 mm. The ranges of variations of model parameters corresponding to the ensemble of realizations r (ρ, λ) are indicated above, with the difference that the concentration of bilirubin in the tissues of the dermis C _bil varies in the range 0-5 mg / l (normal bilirubin level), and the concentration of blood vessels f _bl in typical for normal (not tumor) skin in the range of 1-3%. The following ranges were selected for the concentrations of MetHb, COHb, and SHb: C _MetHb = 1-20%, C _COHb = 1-20%, C _SHb = 0.2-10%.

For each implementation r (ρ, λ), model parameters were restored by minimizing functional (9). Figure 3-6 compares the exact and reconstructed from the signals r (ρ, λ) values of the concentrations of HbO ₂ , MetHb, COHb and SHb for the considered variation in skin parameters, and also shows the correlation coefficients ρ between the exact and reconstructed concentrations. It can be seen that the measurements under consideration are highly sensitive to the presence in the blood of all hemoglobin derivatives. Moreover, the use of differential signals r (ρ, λ) to determine the hemoglobin composition eliminates the need for calibration of the measuring device, which increases the usability of the device and eliminates the additional cost of periodic calibration. The developed methods for calculating r (ρ, λ) allow the interpretation of experimental data in real time. In this case, the sensitivity of the signals r (ρ, λ,) to small variations in hemoglobin composition can be easily increased by increasing the measurement base, i.e. the distance between the tissue irradiation area and the scattered light registration area.

Thus, the proposed method for determining the parameters of the skin and blood is the basis of high-precision and easy-to-use optical meters that have such important advantages as non-invasiveness, non-contact, non-calibration and speed measurements. In this regard, the developed method can find wide application in medical diagnostics and therapy.

Literature

1. RF patent No. 2140083, G01N 33/52, G01N 33/72; 10/20/1999.

2. Pulse Oximetry / R. Ortega [et. al.] // New Engl. J. Med. 2011. V.364, No. 16: e33.

3. Buinevicius R.P., Webster J.G. A three-wavelength pulse oximeter for carboxyhemoglobin determination // J. Iranian Association of Electrical and Electronics Engineers. 2008. V.5, No. 2. P.1-7.

4. S.J. Barker, J.J. Badal The measurement of dyshemoglobins and total hemoglobin by pulse oximetry // Current Opinion in Anaesthesiology. 2008. V.21, No. 6. P.805-810.

5. Pat. RF No. 2234853, A61B 005/05; 12/26/2002.

6. Pat. No. US 2005/0049467 A1, 03.03.2005.

7. Patents No. US 7,570,988 B2, A61B 6/00; Aug. 4, 2009.

8. Non-invasive skin oxygenation imaging using a multi-spectral camera system: effectiveness of various concentration algorithms applied on human skin / J. H. G. M. Klaessens [et. al.] // IFMBE Proceedings. 2009. V.25, No. 2. P.725-728.

9. Jacques S.L. Origins of tissue optical properties in the UVA, visible, and NIR // Advances in Optical Imaging and Photon Migration. 1996. V.2. P.364-369.

10. Barun V.V., Ivanov A.P. Evaluation of the contribution of localized absorption of light by blood vessels to the optical properties of biological tissue // Optics and Spectrum. 2004. V. 96, No. 6. P.1019-1024.

Claims (1)

A method for determining the physico-biological parameters of the skin and the concentration of hemoglobin derivatives in the blood by sending polarized optical radiation with a known spectrum to the skin, recording signals of diffusely reflected light with polarization on the skin, orthogonal polarization of the radiation sent to the skin, characterized in that the optical radiation is focused on the skin surface into the irradiation spot, the signals of diffuse-reflected light of the skin P (ρ, λ) are recorded in the spectral range λ = 450-800 nm from at least two annular regions to it on the skin surface located at various distances ρ from the irradiation spot; determining the difference signals r (ρ, λ) = ln (P (ρ ₀ , λ) / P (ρ, λ)), where ρ ₀ is the distance to the registration ring closest to the irradiation spot; and the physico-biological parameters of the skin and the concentration of hemoglobin derivatives in the blood are determined by solving the inverse problem using analytical expressions that approximate the dependence of r (ρ, λ) on the determined parameters.

Images