WO2017167030A1 - 一种人皮肤光谱的建模方法以及高拟合度的多个皮肤参数的数学建模方法 - Google Patents
一种人皮肤光谱的建模方法以及高拟合度的多个皮肤参数的数学建模方法 Download PDFInfo
- Publication number
- WO2017167030A1 WO2017167030A1 PCT/CN2017/077001 CN2017077001W WO2017167030A1 WO 2017167030 A1 WO2017167030 A1 WO 2017167030A1 CN 2017077001 W CN2017077001 W CN 2017077001W WO 2017167030 A1 WO2017167030 A1 WO 2017167030A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- skin
- layer
- light
- dermis
- volume fraction
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/11—Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
Definitions
- the present invention relates to computational biology, to a method for establishing an analytical model of skin spectrum, and to a modeling method for constructing skin biological parameters using the model, in particular to a method for modeling human skin spectrum and a plurality of skin parameters with high fitness Mathematical modeling approach.
- the interaction between light and matter causes the electronic transition of the atomic and molecular energy levels inside the material, which causes the absorption, reflection, and scattering of light to change in wavelength and intensity information.
- the spectrometer can be used to detect and process such changes. Compared with other analytical methods, spectral detection is non-destructive, highly sensitive, and highly accurate, and is widely used in the detection and identification of various materials.
- the skin rendering model is built to visually reflect skin structure and composition.
- This type of model is mainly used for skin rendering, emphasizing visual effects. Accuracy is not a key indicator. Therefore, there are problems in using it for quantitative analysis of skin parameters: 1 There are few skin parameters, only 4 component parameter variables, which cannot fully reflect the actual skin. The results used for quantitative analysis are not accurate enough. 2 Only the wavelength range of 400-600 nm of visible light is reflected, and the band of 600 nm or more cannot be fitted. (The Kubelka-Munk model does not consider horizontal scattering, and the result is higher than the real result; the Multipole model cannot be simulated in some cases of too thin skin layers due to its assumed discrete point source position problem, which often occurs at 590. -700nm wavelength range; Monte-Carlo model results are uncertain, difficult to optimize, and the required calculation time is very long) 3 only involves skin absorption parameters related to skin absorption, scattering related skin composition parameters. 4 does not contain skin structure parameters.
- a method of modeling a human skin spectrum and a mathematical modeling method for a plurality of skin parameters of high fitness filling in the blank of quantitative analysis of skin biology using spectral model, a set of skin parameters that can describe the spectral characteristics of the skin is analyzed.
- the virtual skin spectrum has a very good fit to the actual skin spectrum, which improves the accuracy of the analysis.
- a method of modeling a human skin spectrum comprising the steps of:
- Step 1 According to the characteristics of absorption, reflection, scattering and transmission when the visible light is irradiated to the skin, the skin is abstracted from top to bottom into a skin surface layer, a skin epidermis layer, a skin dermis layer and a subcutaneous tissue layer;
- Step 2 According to the light absorption, scattering coefficient and skin skin thickness variable parameters of the skin epidermis, the light reflection and light transmission equations in the skin epidermis are established, and the reflection in the skin epidermis is calculated. Rate and transmittance;
- Step 3 According to the light absorption and scattering coefficient of the dermis layer of the skin and the thickness parameter of the skin dermis layer, the light reflection equation in the dermis layer of the skin is established, and the reflectance in the dermis layer of the skin is calculated;
- Step 4 Calculate the reflectance of light in the two-layer structure of the skin epidermis and the dermis layer
- Step 5 Calculate the reflectance of light through the overall structure of the four layers of skin.
- step 1 an equation of light reflection and light transmission in the rough surface layer of the skin is established according to the radiation conduction equation and the reflection coefficient on the rough surface layer of the skin, and the reflectance at the surface layer of the skin is calculated. step.
- step two light reflection and light transmission in the skin epidermis are determined by two paths of light entering from above and light entering from the lower dermis, said light reflection and light transmission in the skin epidermis The equation is
- d epi is the thickness of the epidermal layer
- L air ⁇ L epi represents light entering the epidermal layer by air
- L derm ⁇ L epi represents light entering the epidermal layer from the dermis layer.
- step 3 only the way in which the light reflected light enters from the upper skin layer by light is considered, and the light reflection equation in the skin dermis layer is
- d derm is the thickness of the dermis layer
- L epi ⁇ L derm represents the light entering the dermis layer from the upper epidermal layer.
- step four the two layers of the light in the skin epidermis and the dermis layer
- the formula for calculating the reflectance in the structure is:
- R inter is the reflectivity of light in the two-layer structure of the skin epidermis and the dermis layer.
- a mathematical modeling method for a plurality of high-fit skin parameters based on a modeling method of human skin spectrum comprising the following steps:
- Step 1 Establish an equation representing the relationship between the reflectance and skin roughness and skin lightness in the skin surface layer
- Step 2 Establishing the absorption coefficient and epidermal melanin volume fraction, epidermal melanin concentration, epidermal layer brown melanin concentration, epidermal water volume fraction, epidermal lipid volume fraction, and epidermal layer in the epidermal layer of the skin An equation for the relationship between carotene concentration and epidermal cortical thickness;
- Step 3 Establishing the absorption rate and the dermal layer water volume fraction, the blood volume fraction, the hemoglobin concentration, the oxidized hemoglobin volume fraction in the blood, the deoxygenated hemoglobin volume fraction in the blood, the carbon monoxide hemoglobin volume fraction in the blood, and the blood. Equations for the relationship between hemoglobin volume fraction, hemoglobin volume fraction in blood, carotene concentration in dermis, bilirubin concentration in dermis, platelet volume fraction in blood, hemoglobin volume fraction in blood, and elastin volume fraction in dermis;
- Step 4 establishing an equation representing the relationship between the scattering coefficient and the skin collagen volume fraction, the skin collagen fiber radius or diameter, the skin collagen fiber bundle radius or diameter in the dermal layer of the skin;
- Step 5 Establish an equation representing the relationship between light reflectance, light transmittance, and skin thickness in the skin epidermis
- Step 6 Establish an equation representing the relationship between light reflectance and skin thickness in the dermis layer of the skin.
- the method further includes the step 7: comparing the skin parameters calculated in steps 1 to 4 with the actual skin parameters of the sample, obtaining a fitness degree, and determining whether the mathematical modeling methods of the plurality of skin parameters are accurate.
- step one the equation is
- f surf represents the reflectance of the skin surface in a single direction
- d ⁇ 0 represents the micro-element of the direction of reflection
- step two the equation is
- f me represents the melanin volume fraction of the epidermis layer
- c eu represents the melanin concentration of the epidermis layer.
- c ph represents the brown melanin concentration
- f lipid skin layer represents the volume fraction of oil
- Indicates the carotene absorption coefficient Indicates the baseline absorption coefficient of the skin.
- step three the equation is
- sulf represents the volume fraction of sulphur hemoglobin in the blood.
- S sulf represents the volume fraction of sulphur hemoglobin in the blood.
- Indicates the hemoglobin absorption coefficient Indicates the concentration of carotene in the dermis, Indicates the carotene absorption coefficient, c br represents the bilirubin concentration, Indicates the bilirubin absorption coefficient, and f plt represents the platelet volume fraction in the blood.
- H represents the hemoglobin volume fraction in the blood
- f ela represents the elastin volume fraction.
- Expresses the elastin absorption coefficient Indicates the baseline absorption coefficient of the skin.
- step four the equation is
- ⁇ s represents the dermal layer scattering coefficient
- Rayleigh scattering coefficient for the dermis It is the Mie scattering coefficient of the dermis layer.
- step five the equation is
- d epi is the thickness of the epidermal layer
- L air ⁇ L epi represents light entering the epidermal layer by air
- L derm ⁇ L epi represents light entering the epidermal layer from the dermis layer.
- step 6 the equation is
- d derm is the thickness of the dermis layer
- L epi ⁇ L derm represents the light entering the dermis layer from the upper epidermal layer.
- the present invention has the following beneficial effects:
- the present invention constructs a skin spectral model and a mathematical model of skin parameters through conduction analysis of light in the skin, establishing a link between skin spectrum and skin biological parameters.
- the virtual spectrum of a set of skin biological parameter simulations calculated by the mathematical model of the present invention is highly fitted to the actual spectral height, and the model is accurate and reliable.
- parameters related to spectral absorption of the skin related to the present invention there are 19 parameters related to spectral absorption of the skin related to the present invention, 3 parameters related to spectral scattering, 2 parameters related to diffuse reflection on the skin surface, 2 parameters related to skin thickness, and a total of 26-dimensional feature vectors, which can be simulated.
- the method has three functions: 1. Establishing a mathematical relationship between the volume fraction of a certain absorbed component of the skin on other volume fractions, ensuring the physical constraint of the volume fraction of each absorbed component of the skin; 2. Excluding the correction of the search direction The impact of the quantitative analysis process is not subject to external interference; 3, to ensure the correctness of the results of quantitative analysis of biological parameters related to skin light absorption.
- the method of the present invention accurately fits in the full range of visible light 400-700 nm, which can achieve the precision required for skin quantitative analysis.
- the skin spectral information on which the present invention is based can be obtained by a non-invasive acquisition method.
- Figure 1 is a schematic view showing the structure of the fitting degree of the present invention.
- the present invention constructs a four-layer structure of skin under the action of visible light (400-700 nm, which acts on the skin of humans up to 4 mm) (skin table)
- the model of the face, epidermis, dermis, and subcutaneous tissue abstracts the radiation conduction path analysis of the four forms of light (reflection, transmission, absorption, and scattering) on the four-layer structure model of the skin.
- An optical analysis model between the skin spectrum and reflection, absorption, scattering, and transmission is formed.
- the model has two characteristics: 1. In the structural model of the skin, the thickness of the epidermis and dermis of the skin is used as a structural variable, highlighting the influence of these two variables on the spectrum; 2.
- the model covers light and skin.
- the four modes of action are complete optical models.
- RTE Radiation Transfer Equation
- Electromagnetic waves propagate in the medium and are in the direction x Emitted energy Said. a bunch of along The amount of change in emissivity after passing through a small length of medium (dS(x)) when the energy propagated in the direction passes through the position at x. Can be written as:
- the radiation conduction equation is:
- the skin is abstracted into a multi-layer structure, each layer having independent, multiple absorption and scattering media;
- the skin is infinite in the direction perpendicular to the thickness extension, and may be infinite or infinite in a direction parallel to the thickness extension;
- the skin is a plane-parallel structure, that is, the optical properties of the skin change only in parallel with the direction in which the thickness extends, and the optical properties of the skin at the same depth are completely identical;
- the scattering medium in the skin is small-scale collagen fiber and large-scale collagen fiber bundle.
- the scattering of small-scale collagen fiber can be approximated by spherical Rayleigh scattering, which causes the dermis layer to have both spherical Rayleigh and cylindrical Mie. scattering.
- the purpose of establishing a light propagation model in the skin structure is to be able to calculate the overall skin reflectance.
- the calculation of the overall skin reflectance is composed of three parts: 1. skin surface reflection and transmission; 2. overall reflection of the multilayer structure inside the skin; Reflection of the overall structure of the skin.
- the modeling method of the skin spectrum of the present invention includes the following steps,
- Step 1 According to the characteristics of absorption, reflection, scattering and transmission when the visible light is irradiated to the skin, the skin is abstracted from the top to the bottom of the rough layer, the skin epidermis, the skin dermis layer and the subcutaneous tissue layer.
- Surface layer no actual thickness, infinitely thin, located at the outermost layer, connected to the external environment, lower to the skin layer;
- Epidermis the actual first layer of skin, with a limited thickness, connected to the rough surface layer, and connected to the dermis layer;
- Dermis layer the actual second layer of the skin, with a limited thickness, connected to the epidermis layer, and connected to the subcutaneous tissue;
- Subcutaneous tissue layer The subcutaneous tissue has an infinite thickness and does not describe any composition, its function is to absorb all light that enters the subcutaneous tissue from the dermal layer.
- Step 2 According to the radiation conduction equation and the light absorption and scattering coefficients in the skin surface layer, an equation of light reflection and light transmission in the skin surface layer is established, and the reflectance and absorption rate in the skin surface layer are calculated.
- the reflection and transmission of the skin surface is calculated by the light reflection model of the surface as:
- r surf is a reflection coefficient of the skin surface, and is composed of surface roughness ⁇ surf and surface luminance ⁇ surf .
- the layer does not absorb the energy of any light, its reflectance and transmittance have the following relationship:
- Step 3 According to the light absorption and scattering coefficient of the skin epidermis, the light reflection and light transmission equations in the skin epidermis are established, and the reflectance and absorption rate in the skin epidermis are calculated.
- the skin layer is capable of absorbing and scattering light entering the layer, the thickness of the layer affecting the total amount of light absorbed and scattered during propagation, and the refractive index of the upper layer (air) and the lower layer of the epidermis (dermis), as well as light.
- the layer from which the epidermis enters will also affect the reflection and transmission of the layer. Therefore, the calculation of the reflection and transmittance of the layer requires simultaneous calculation of the upper illumination (light entering from the air, indicated by the + sign) and lower illumination (the light from the dermis layer). Enter, - sign indicates) two cases:
- Step 4 According to the light absorption and scattering coefficient of the dermis layer of the skin, the light reflection equation in the dermis layer of the skin is established, and the reflectance in the dermis layer of the skin is calculated and calculated.
- the dermis layer also absorbs and scatters the incoming light.
- the thickness of the dermis affects the total amount of light absorbed and scattered. However, since it is assumed that the light entering the subcutaneous tissue is completely absorbed, it will not return to the dermis layer, so it is not necessary to consider Illumination (light entering from the subcutaneous tissue) and dermal transmission, the formula is as follows:
- Step 5 Calculate the reflectance of light in the two-layer structure of the skin epidermis and the dermis layer. Since the inside of the skin is a multi-layered structure, light can be reflected between the epidermis layer and the dermis layer. The light may leave the skin after being internally reflected multiple times.
- the reflectivity of the multilayer structure is calculated as follows:
- R inter is the reflectance of the two-layer structure of the epidermal layer and the dermis layer of the skin, For the reflectivity of light from the air into the skin layer, For the transmittance of light from the air into the skin layer, For the reflectivity of light from the dermis into the epidermal layer, For the transmission of light from the dermis into the epidermal layer, The reflectivity of light entering the dermis from the epidermis.
- Step 6 Calculate the reflectance of light through the overall structure of the four layers of skin. After calculating the skin surface reflection, transmission, and overall internal reflection of the skin, the overall skin reflection can be calculated. After the light is irradiated to the skin, first part of the light is reflected by the surface of the skin, and all the unreflected light enters the inside of the skin and interacts with the inside of the skin (absorption, scattering). Finally, the light that is not absorbed by the inside of the skin escapes from the inside of the skin through reflection. :
- R skin R surf +T surf R inter (7)
- R skin is the total reflectance of the four-layer structure of the skin
- R surf is the reflectance of the skin surface
- T surf is the transmittance of the skin surface
- R inter is the reflection of the two layers of the skin epidermis and dermis rate.
- the skin biological parameters related to light sensitivity were found, and the skin parameter set corresponding to the skin structure model was formed, and the correlation of skin parameters, skin structure and light action form was established. Furthermore, by modeling the three light action forms and skin parameters, a set of mathematical expressions is established to establish the relationship between skin parameters and optical properties. In the present invention, as much as possible related skin biological parameters are involved.
- a mathematical modeling method for a plurality of skin parameters that makes the model data sufficient and improves the accuracy of the algorithm, that is, a high degree of fit, the steps are as follows:
- Step 1 An equation representing the relationship between the reflectance and skin roughness and skin lightness in the rough surface layer of the skin is established.
- the skin surface is abstracted into an infinite number of symmetrical microscopic V-shaped grooves, each of which is specularly reflected by light and can be calculated by geometric optics.
- the reflection of the final skin surface is statistically obtained from the reflection of all microscopic V-shaped grooves.
- the distribution of the V-shaped grooves and the reflection intensity of each surface affect the total surface reflection, and the V-shaped grooves also block the incident light and the reflected light, so the surface is reflected in a single direction f surf
- Equation (8) where ⁇ i , ⁇ o are the incident light, the opposite direction of the observation direction, n is the macro surface normal vector, and h is the half vector of ⁇ i , ⁇ o .
- V-shaped grooves are randomly modeled by parametric, controlled by the average depth-to-space ratio of all grooves, usually using a Beckmann distribution, and the total reflection of the final skin surface in all directions is reflected in a single direction at all viewing angles. Points:
- R surf is the skin surface reflectance
- f surf is the reflectance at a single viewing angle
- d ⁇ o is the micro-element of the viewing angle
- Step 2 Establishing the absorption coefficient and epidermal melanin volume fraction, epidermal melanin concentration, epidermal layer brown melanin concentration, epidermal water volume fraction, epidermal lipid volume fraction, and epidermal layer in the epidermal layer of the skin The equation for the relationship between carotene concentrations.
- Formula (10) where Indicates the absorption coefficient of the epidermal layer, f me represents the melanin volume fraction of the epidermis layer, and c eu represents the melanin concentration of the epidermis layer.
- Step 3 Establishing the absorption coefficient and the dermal layer water volume fraction, the blood volume fraction, the hemoglobin concentration, the oxidized hemoglobin volume fraction in the blood, the deoxygenated hemoglobin volume fraction in the blood, the carbon monoxide hemoglobin volume fraction in the blood, and the blood.
- S co represents the volume fraction of carbon monoxide hemoglobin in the blood. Represents the carbon monoxide hemoglobin absorption coefficient, and S met represents the methemoglobin volume fraction in the blood.
- sulf represents the volume fraction of sulphur hemoglobin in the blood.
- S sulf represents the volume fraction of sulphur hemoglobin in the blood.
- Indicates the hemoglobin absorption coefficient Indicates the concentration of carotene in the dermis, Indicates the carotene absorption coefficient, c br represents the bilirubin concentration, Indicates the bilirubin absorption coefficient, and f plt represents the platelet volume fraction in the blood.
- H represents the hemoglobin volume fraction in the blood
- f ela represents the elastin volume fraction.
- Expresses the elastin absorption coefficient Indicates the baseline absorption coefficient of the skin.
- Step 4 establishing an equation representing the relationship between the scattering coefficient and the skin collagen volume fraction, the skin collagen fiber radius, and the skin collagen fiber bundle diameter in the dermal layer of the skin;
- Step 5 Establish an equation representing the relationship between light reflectance, light transmittance, and skin thickness in the skin epidermis.
- Scattering is a phenomenon in which light travels in the interaction with the skin, and the direction of light propagation is changed. The change in direction does not follow Snell's law (refraction law).
- the main scattering component in the skin is collagen, and the skin has two different scattering: Rayleigh scattering and Mie scattering.
- Rayleigh scattering is mainly produced by small-scale collagen fiber structure, while Mie scattering mainly produces large-scale collagen fiber bundles (bounded by multiple bundles of fibers with the same orientation).
- the scattering coefficient of dermis is composed of 3 Parameters indicate: collagen volume fraction, collagen fiber radius, collagen fiber bundle radius. According to biological studies, the scattering coefficient of the epidermis in this skin model is approximated by the dermal scattering coefficient, ie the two layers have the same scattering coefficient.
- ⁇ s is the dermal layer scattering coefficient, Rayleigh scattering coefficient for the dermis, It is the Mie scattering coefficient of the dermis layer.
- the scattering theory mainly describes that the scattering coefficient ⁇ s of the medium can be obtained by multiplying the single collagen cross-section scattering coefficient ⁇ ss and its number density N, while the collagen cross-section scattering coefficient ⁇ ss can be multiplied by the cross-sectional area by the scattering efficiency ⁇ sca A is obtained.
- the Mie scattering equation calculate the Mie scattering efficiency, calculate the anisotropy index, and obtain the Mie scattering coefficient, and finally obtain the formula (12).
- Step 5 Establish an equation representing the relationship between light reflectance, light transmittance, and skin thickness in the skin epidermis.
- Step 6 Establish an equation representing the relationship between light reflectance and skin thickness in the dermis layer of the skin.
- the method further comprises the following steps: comparing the skin parameters calculated in steps 1 to 4 with the actual skin parameters of the sample, and calculating the fitness degree, It is judged whether the mathematical modeling method of the above plurality of skin parameters is simulated accurately, and can be used for quantitative analysis of skin parameters.
- Specific steps include: 1. Measuring the various wavelengths of the target skin Reflectance samples, and pre-processed to increase smoothness and reduce measurement noise; 2. Guess a set of initial skin biological parameters; 3. Calculate reflectance and transmission of skin surface, epidermis, and dermis according to the steps Rate; 4.
- the steps calculate the reflectance of the three layers of the skin epidermis, dermis and subcutaneous tissue; 5. Calculate the reflectivity of the four-layer structure of the skin; 6. Compare the calculated skin reflectance and the measured The distance between the skin reflectances; 7. Calculate the change in the distance between the skin reflectance and the measured skin reflectance calculated for each parameter change in the selected parameter set; 8. Select the next set according to the change in distance Skin biological parameters, and repeat steps 3 through 8, until a set of biological parameters with the smallest distance is obtained.
- the fitting curve is shown in Fig. 1. It can be seen from the fitting curve that the fitting degree of the method is very high.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Optimization (AREA)
- Computational Mathematics (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Databases & Information Systems (AREA)
- Algebra (AREA)
- Operations Research (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
一种人皮肤光谱的建模方法以及高拟合度的多个皮肤参数的数学建模方法，通过建立两个层级的分析模型，即皮肤光谱分析模型和皮肤参数数学模型，从而在皮肤光谱与光敏感的皮肤生物学参数之间建立起关联，实现由一组皮肤参数模拟出虚拟光谱，然后通过虚拟光谱与实际光谱利用信息处理技术进行优化迭代，能够找到一组最优解，达到指定的拟合度标准，实现皮肤生物学量化分析的目的。
Description
本发明涉及计算生物学，涉及建立皮肤光谱的分析模型的方法以及利用该模型构建皮肤生物学参数的建模方法，特别是涉及人皮肤光谱的建模方法以及高拟合度的多个皮肤参数的数学建模方法。
光与物质相互作用会引起物质内部原子及分子能级的电子跃迁，使物质对光的吸收、反射、散射等在波长及强度信息上发生变化，光谱仪可用于检测并处理这类变化。与其他分析方法相比，光谱检测具有非破坏、高灵敏、高精准的特点，因而在各类材料的检测和鉴定方面得到广泛的应用。
生物学与光学的研究表明，人类皮肤是由多种生物学成分组成，其中的一些成分对光谱作用敏感，各成分具有特定的可见光光学性质，并且，这些皮肤成分含量已由生物学方法进行了测定。因此，形成一种籍于皮肤的可见光谱的数据处理方法实现皮肤的生物学成分量化分析是可行的。但在建模方面，要求可见光光学作用分析完整、皮肤参数尽量齐全，才能体现皮肤参数的光学性质细节，达到可见光全波段所需的量化分析精度要求。
目前，在皮肤参数量化分析领域主要存在以下手段，1)基于图像的皮肤分析。由于是平面的图像分析，其所能测得参数的与光谱的立体数据相比准确性有较大差距。所期望的其他的皮肤成分参数和皮肤结构参数无法获得。2)基于生物阻抗的皮肤水份检测。此种检测成分参数单一且精度不够。3)基于超声波的皮肤超声影像诊断系统。此种方式主要是对皮肤的结构组织进行定性观察，无法进行皮肤参数量化分析。4)基于X射线的三维断层成像的皮肤CT影像分析系统。此种方式主要是对皮肤的结构组织
进行定性观察，无法进行皮肤参数量化分析。5)皮肤光谱检测系统。属于研究项目。此种方式是利用光谱进行某波段的直观比较观察，没有形成信息处理模型。
同时，建立皮肤渲染模型能够直观反映皮肤结构和组分。该类模型主要用于皮肤渲染，强调视觉效果，精度不是关键指标，因而将其用于皮肤参数量化分析存在问题有：①皮肤参数少，仅有4个成分参数变量，不能完整反映皮肤实际，用于量化分析的结果精度不够。②仅反映可见光的400-600nm波长区间，600nm以上波段无法拟合。(Kubelka-Munk模型由于未考虑水平方向散射，其结果会高于真实结果；Multipole模型由于其假设的离散点光源位置问题，在某些过薄皮肤层情况下无法模拟，该情况经常发生于590-700nm波长区间；Monte-Carlo模型结果具有不确定性，难以优化，且所需计算时间很长)③仅涉及皮肤的吸收相关的皮肤成分参数，散射相关的皮肤成分参数。④不包含皮肤结构参数。
因此，为了实现精确的皮肤生物学参数的量化目的，急需设计一种基于皮肤光谱进行皮肤参数量化分析的方案，可见光波段分析完整、皮肤生物学特征参数尽可能充足地被涉及，从而达到对不同人、不同部位差异尽可能多的分析维度和精度，并形成大数据处理的基础。
发明内容
本发明的目的是提供一种人皮肤光谱的建模方法以及高拟合度的多个皮肤参数的数学建模方法。填补了利用光谱模型进行皮肤生物学量化分析的空白，分析出可以描述皮肤光谱特征的一组皮肤参数，其虚拟出的皮肤光谱与实际皮肤光谱的拟合度非常高，提高分析的精确性。
为实现上述发明目的，本发明提供的技术方案是：
一种人皮肤光谱的建模方法，所述方法包括以下步骤：
步骤一 根据可见光照射皮肤时吸收、反射、散射、透射的特征，将皮肤从上至下抽象为皮肤表面层、皮肤表皮层、皮肤真皮层、皮下组织层四层；
步骤二 根据皮肤表皮层的光吸收、散射系数以及皮肤表皮层厚度变量参数，建立在皮肤表皮层中光反射和光透射方程，计算在皮肤表皮层的反射
率和透射率；
步骤三 根据皮肤真皮层的光吸收、散射系数以及皮肤真皮层厚度变量参数，建立在皮肤真皮层中光反射方程，计算在皮肤真皮层的反射率；
步骤四 计算光在皮肤表皮层和皮肤真皮层的两层结构中的反射率；
步骤五 计算光通过四层皮肤的总体结构中的反射率。
进一步地，在步骤一和步骤二之间还包括根据辐射传导方程和在皮肤粗糙表面层的反射系数，建立在皮肤粗糙表面层中光反射和光透射的方程，计算在皮肤表面层的反射率的步骤。
进一步地，所述在皮肤表面层中光反射和光透射的方程为[Rsurf,Tsurf]＝LSI(rsurf)，其中，Rsurf为表面的光反射率，Tsurf为表面的光透射率，LSI代表了用于计算皮肤反射率以及透射率的一系列计算公式，rsurf为皮肤表面的反射系数。
进一步地，在步骤二中，在所述皮肤表皮层中光反射和光透射由光从上方空气中进入和光从下方真皮层进入两种途径所决定，所述在皮肤表皮层中光反射和光透射的方程为
进一步地，在步骤三中，只考虑光反射光由光从上方表皮层中进入的途径，所述在皮肤真皮层中光反射方程为
进一步地，在步骤四中，所述光在皮肤表皮层和皮肤真皮层的两层结
构中的反射率计算公式为：
其中，Rinter为光在皮肤表皮层和皮肤真皮层的两层结构中的反射率。
进一步地，在步骤五中，所述总体结构中的反射率公式为Rskin＝Rsurf+TsurfRinter。
一种根据人皮肤光谱的建模方法建立的高拟合度的多个皮肤参数的数学建模方法，，包括以下步骤：
步骤一 建立表示在皮肤表面层中的所述反射率和皮肤粗糙度以及皮肤光亮度之间关系的方程；
步骤二 建立表示在皮肤表皮层中的所述吸收系数与表皮层黑色素体积分数、表皮层优黑素浓度、表皮层褐黑素浓度、表皮层水分体积分数、表皮层油脂体积分数、表皮层中胡萝卜素浓度、表皮层皮层厚度之间关系的方程；
步骤三 建立表示在皮肤真皮层中的所述吸收率与真皮层水分体积分数、血液体积分数、血红蛋白浓度、血液中氧化血红蛋白体积分数、血液中脱氧血红蛋白体积分数、血液中一氧化碳血红蛋白体积分数、血液中高铁血红蛋白体积分数、血液中硫化血红蛋白体积分数、真皮层胡萝卜素浓度、真皮层胆红素浓度、血液中血小板体积分数、血液中血红蛋白体积分数、真皮层弹性蛋白体积分数之间关系的方程；
步骤四 建立表示在皮肤真皮层中所述散射系数与皮肤胶原蛋白体积分数、皮肤胶原蛋白纤维半径或直径、皮肤胶原蛋白纤维束半径或直径之间关系的方程；
步骤五 建立表示皮肤表皮层中光反射率、光透射率与表皮厚度之间关系的方程；
步骤六 建立表示皮肤真皮层中光反射率与表皮厚度之间关系的方程。
进一步地，还包括步骤七：将步骤一至步骤四中计算的皮肤参数与样本实际的皮肤参数比较，求拟合度，判断所述多个皮肤参数数学建模方法是否精确。
进一步地，在步骤一中，所述方程为
进一步地，在步骤二中，所述方程为
其中，
表示表皮层吸收系数，fme表示表皮层黑色素体积分数，ceu表示表皮层优黑素浓度，
表示优黑素吸收系数，cph表示褐黑素浓度，
表示褐黑素吸收系数，
表示表皮水分体积分数，
表示水分吸收系数，flipid表示表皮层油脂体积分数，
表示油脂吸收系数，
表示表皮层胡萝卜素浓度，
表示胡萝卜素吸收系数，
表示皮肤基线吸收系数。
进一步地，在步骤三中，所述方程为
其中，
表示真皮层吸收系数，
表示真皮层水分体积分数，
表示水分吸收系数，fblood表示血液体积分数，Soxy表示血液中含氧血红蛋白体积分数即血氧浓度，cblood表示血红蛋白浓度，
表示含氧血红蛋白吸收系数，Sdeoxy表示血液中脱氧血红蛋白体积分数，
表示脱氧血红蛋白吸收系数，Sco表示血液中一氧化碳血红蛋白体积分数，
表示一
氧化碳血红蛋白吸收系数，Smet表示血液中高铁血红蛋白体积分数，
表示高铁血红蛋白吸收系数，Ssulf表示血液中硫化血红蛋白体积分数，
表示硫化血红蛋白吸收系数，
表示真皮层胡萝卜素浓度，
表示胡萝卜素吸收系数，cbr表示胆红素浓度，
表示胆红素吸收系数，fplt表示血液中血小板体积分数，
表示血小板吸收系数，H表示血液中血红蛋白体积分数，fela表示弹性蛋白体积分数，
表示弹性蛋白吸收系数，
表示皮肤基线吸收系数。
进一步地，在步骤四中，所述方程为
进一步地，在步骤五中，所述方程为
其中，
为从空气进入表皮层的光反射率，
为从空气进入表皮层的光透射率，
为从真皮层进入表皮层的光反射率，
为光从真皮层进入表皮层的光透射率，
为表皮层吸收系数，
为表皮层散射系数，depi为表皮层厚度，Lair→Lepi代表光由空气进入表皮层，Lderm→Lepi代表光由真皮层进入表皮层。
进一步地，在步骤六中，所述方程为
其中，
为从表皮层进入真皮层的光反射率，
为真皮层吸收系数，
为真皮层散射系数，dderm为真皮层厚度，Lepi→Lderm代表光由上方表皮层进入真皮层。
采用上述技术方案，本发明具有如下有益效果：
第一，本发明通过光在皮肤中的传导分析，构建了皮肤光谱模型和皮肤参数数学模型，建立了皮肤光谱与皮肤生物学参数之间的联系。利用本发明的数学模型计算出的一组皮肤生物学参数仿真的虚拟光谱与实际光谱高度拟合，模型准确可靠。
第二，本发明相关的皮肤的光谱吸收相关的参数19个，光谱散射相关的参数3个，皮肤表面漫反射相关的参数2个，皮肤厚度相关参数2个，共计26维特征向量，能够仿真皮肤光谱的细节，模拟真实的皮肤情况，
精确度高，从而实现了皮肤生物学参数量化分析计算的目的，并可作为皮肤大数据的处理的基础。
第三，本发明所建立的皮肤表皮层、真皮层吸收系数与皮肤光吸收相关的生物学参数之间的数学模型中，引入了如下形式的“余量体积分数”进行归一化处理，使得在优化分析过程中能够得到合理的最优解。
该方法有3个作用：1、建立了皮肤某一吸收成分的体积分数对其他体积分数影响的数学关系，保证了皮肤各个吸收成分体积分数的物理约束；2、排除了搜索方向的矫正所造成的影响，保证了量化分析过程不会受到外部干扰；3、确保了皮肤光吸收相关的生物学参数量化分析结果的正确性。
第四，本发明的方法在可见光400-700nm的全波段内精确拟合，可以达到皮肤量化分析所需要的精度。
第五，本发明所基于的皮肤光谱信息可由无创的采集方式得到。
图1为本发明拟合度的结构示意图。
为了使本发明的目的、技术方案及优点更加清楚明白，下面结合附图及实施例，对本发明进行进一步详细说明。应当理解，此处所描述的结构图及具体实施例仅用以解释本发明，并不用于限定本发明。
实施例1
基于辐射传导理论方程应用的需要，本发明构建了在可见光(400-700nm，作用于人的皮肤深度最大4mm)作用下皮肤的四层结构(皮肤表
面、表皮层、真皮层、皮下组织)模型，抽象了光的四种作用形式(反射、透射、吸收、散射)在皮肤四层结构模型上的辐射传导路径分析。综合形成了皮肤光谱与反射、吸收、散射、透射之间的光学分析模型。该模型有两个特点：1、在皮肤的结构模型中，将皮肤的表皮层、真皮层厚度作为结构变量，突出了这两个变量对光谱的影响；2、该模型涵盖了光与皮肤的四种作用形式，是完整的光学模型。
在现有技术中，辐射传导方程(RTE，Radiative Transfer Equation)，该理论叙述了电磁波在介质中传播时，电磁波会：在“吸收”(Absorption)中损失能量，在“激发”(Emission)中获得能量，在“散射”(Scattering)中重新分配能量。
电磁波在介质传播中，其在位置x沿方向
传播的能量可由辐射率
表示。一束沿
方向传播的能量在经过位置在x时，在经过一小段介质(dS(x))后，辐射率的变化量
可写为：
辐射传导方程为：
根据理论方程，建立皮肤中的光辐射传导方程，皮肤模型假设：
1)皮肤被抽象为多层结构，每一层拥有独立、多种吸收、散射介质；
2)皮肤在垂直于厚度延伸方向是无穷的，在平行于厚度延伸的方向可以有穷或无穷；
3)皮肤为平面平行结构，即皮肤的光学性质只在平行于厚度延伸方向改变，位于同一深度的皮肤的光学特性完全一致；
4)每层的散射、吸收介质都是均匀分布的，即皮肤任意一层中，任何位置的光学特性都完全相同，拥有相同的含量和吸收、散射系数；
5)皮肤中各种介质的吸收、散射均相互独立；
6)皮肤中散射介质为小尺度的胶原蛋白纤维及大尺度的胶原蛋白纤维束，其中小尺度胶原蛋白纤维的散射可以近似为球状瑞利散射，造成真皮层同时拥有球形瑞利和圆柱形米氏散射。
建立皮肤结构中光传播模型的目的是为了能够计算皮肤总体反射率，皮肤总体反射率的计算由3部分组成：1、皮肤表面反射和透射；2、皮肤内部多层结构的总体反射；3、皮肤总体结构的反射。
本发明人皮肤光谱的建模方法包括以下步骤，
步骤一 根据可见光照射皮肤时吸收、反射、散射、透射的特征，将皮肤从上之下抽象为皮肤粗面层、皮肤表皮层、皮肤真皮层、皮下组织层四层。
表面层：无实际厚度，无限薄，位于最外层，上与外界环境、下与表皮层相连；
表皮层：为皮肤实际的第一层，拥有有限厚度，上连粗糙表面层、下与真皮层连接；
真皮层：为皮肤实际的第二层，拥有有限厚度，上连表皮层、下与皮下组织连接；
皮下组织层：皮下组织拥有无穷厚度，并不描述任何组成，其作用为吸收所有进入从真皮层透射进入皮下组织的光。
步骤二 根据辐射传导方程和在皮肤表面层的光吸收、散射系数，建立在皮肤表面层中光反射和光透射的方程，计算在皮肤表面层的反射率和吸收率。
皮肤表面的反射和透射由表面的光反射模型计算方程为：
[Rsurf,Tsurf]＝LSI(rsurf)， (1)
公式(1)其中，rsurf为皮肤表面的反射系数，由表面粗糙度σsurf和表面光亮度ρsurf组成。
由于该层并不吸收任何光的能量，故其反射率和透射率拥有如下关系：
Tsurf＝1-Rsurf。 (2)
步骤三 根据皮肤表皮层的光吸收、散射系数，建立在皮肤表皮层中光反射和光透射方程，计算在皮肤表皮层的反射率和吸收率。
表皮层能够对进入该层的光吸收和散射，该层厚度会对在传播中光被吸收、散射的总量造成影响，且表皮上层(空气)和表皮下层(真皮)的折射率，以及光从何层进入表皮层也会对该层反射、透射造成影响，故该层的反射、透射率的计算需要同时计算上照明(光从空气进入，+号表示)和下照明(光从真皮层进入，-号表示)两种情况：
公式(3)，其中
为表皮层吸收系数，
为表皮层散射系数，depi为表皮层厚度，Lair→Lepi代表光从空气中进入表皮层。
公式(4)，其中
为表皮层吸收系数，
为表皮层散射系数，depi为表皮层厚度，Lderm→Lepi代表光从真皮层中进入表皮层。
步骤四 根据皮肤真皮层的光吸收、散射系数，建立在皮肤真皮层中光反射方程，计算计算在皮肤真皮层的反射率。
真皮层也会对进入的光吸收和散射，真皮厚度对光被吸收、散射的总量造成影响，但由于假设了进入皮下组织的光完全被吸收，不会回到真皮层，故不用考虑下照明(光从皮下组织进入)以及真皮透射，公式如下：
公式(5)，其中
为真皮层吸收系数，
为真皮层散射系数，dderm为真皮层厚度，Lepi→Lderm代表光从表皮层中进入真皮层。
步骤五 计算光在皮肤表皮层和皮肤真皮层的两层结构中的反射率。由于皮肤内部是多层结构，故光能在表皮层和真皮层之间反射，光可能在被多次内部反射之后才离开皮肤，多层结构的反射率计算如下：
公式(6)，其中Rinter为皮肤表皮层和真皮层的两层结构的反射率，为光从空气进入表皮层的反射率，
为光从空气进入表皮层的透射率，
为光从真皮层进入表皮层的反射率，
为光从真皮层进入表皮层的透射率，
为光从表皮层进入真皮层的反射率。
步骤六 计算光通过四层皮肤的总体结构中的反射率。计算完皮肤表面反射、透射，以及皮肤内部总体反射后，可以计算皮肤总体反射。光在照射到皮肤后，首先一部分光被皮肤表面反射，未被反射的光全部进入皮肤内部，并与皮肤内部相互作用(吸收、散射)，最后未被皮肤内部吸收的光通过反射逃离皮肤内部：
Rskin＝Rsurf+TsurfRinter (7)
公式(7)，其中Rskin为皮肤四层结构的总反射率，Rsurf为皮肤表面的反射率，Tsurf为皮肤表面的透射率，Rinter为皮肤表皮层和真皮层两层结构的反射率。
实施例2
在上述皮肤光谱特征模型的基础上，找到与光敏感有关的皮肤生物学参数，形成了与皮肤结构模型相对应的皮肤参数集，建立了皮肤参数、皮肤结构、光作用形式的关联。进而通过对三种光作用形式与皮肤参数的建模，建立了一组数学表达式，搭建了皮肤参数与光学性质之间的关系，在本发明中尽可能多地涉及相关皮肤生物学参数，使模型数据量充足，提高算法精确度即高拟合度的多个皮肤参数的数学建模方法，其步骤如下：
步骤一 建立表示在皮肤粗糙表面层中的所述反射率和皮肤粗糙度以及皮肤光亮度之间关系的方程。
皮肤表面被抽象成无穷多的对称的微观V形沟槽，每个V形沟槽表面都对光造成镜面反射，并可以通过几何光学进行计算。最终皮肤表面的反射则由对所有的微观V形沟槽的反射进行统计得到。
V形沟槽的分布及每个表面反射强度对表面总反射均造成影响，而其V形沟槽相互之间也会对入射光、反射光造成遮挡，故将表面单一方向反射fsurf建模为表面反射强度ρsurf、V形沟槽分布D，几何衰减G以及菲涅尔表面漫反射F的乘积：
公式(8),其中ωi,ωo是入射光、观察方向的反方向，n是宏观表面法向量，h是ωi,ωo的半向量。
微观V形沟槽是通过参数化随机模拟的，由所有沟槽平均深度—间隔比来控制，通常使用Beckmann分布，最终的皮肤表面向各个方向的总反射则是对单一方向反射在所有观察角度上积分：
Rsurf＝∫2πfsurfdωo (9)
公式(9)，其中Rsurf为皮肤表面反射率，fsurf为单一观察角度上的反射率，dωo为观察角度的微元。
步骤二 建立表示在皮肤表皮层中的所述吸收系数与表皮层黑色素体积分数、表皮层优黑素浓度、表皮层褐黑素浓度、表皮层水分体积分数、表皮层油脂体积分数、表皮层中胡萝卜素浓度之间关系的方程。
由皮肤的假设可知，皮肤中各个吸收成分相互独立，因此吸收系数可以表示为每个成分吸收的线性组合，表皮层吸收系数与对应成分参数的数学关系可以如下：
公式(10)，其中
表示表皮层吸收系数，fme表示表皮层黑色素体积分数，ceu表示表皮层优黑素浓度，
表示优黑素吸收系数，cph表示褐黑素浓度，
表示褐黑素吸收系数，
表示表皮水分体积分数，
表示水分吸收系数，flipid表示表皮层油脂体积分数，
表示油脂吸收系数，
表示表皮层胡萝卜素浓度，
表示胡萝卜素吸收系数，
表示皮肤基线吸收系数。
步骤三 建立表示在皮肤真皮层中的所述吸收系数与真皮层水分体积分数、血液体积分数、血红蛋白浓度、血液中氧化血红蛋白体积分数、血液中脱氧血红蛋白体积分数、血液中一氧化碳血红蛋白体积分数、血液中高铁血红蛋白体积分数、血液中硫化血红蛋白体积分数、真皮层胡萝卜素浓度、真皮层胆红素浓度、血液中血小板体积分数、血液中血红蛋白体积分数、真皮层弹性蛋白体积分数之间关系的方程。
公式(11)，其中
表示真皮层吸收系数，
表示真皮层水分体积分数，
表示水分吸收系数，fblood表示血液体积分数，Soxy表示血液中含氧血红蛋白体积分数(血氧浓度)，cblood表示血红蛋白浓度，
表示含氧血红蛋白吸收系数，Sdeoxy表示血液中脱氧血红蛋白体积分数，
表示脱氧血红蛋白吸收系数，Sco表示血液中一氧化碳血红蛋白体积分数，
表示一氧化碳血红蛋白吸收系数，Smet表示血液中高铁血红蛋白体积分数，
表示高铁血红蛋白吸收系数，Ssulf表示血液中硫化血红蛋白体积分数，
表示硫化血红蛋白吸收系数，
表示真皮层胡萝卜素浓度，
表示胡萝卜素吸收系数，cbr表示胆红素浓度，
表示胆红素吸收系数，fplt表示血液中血小板体积分数，
表示血小板吸收系数，H表示血液中血红蛋白体积分数，fela表示弹性蛋白体积分数，
表示弹性蛋白吸收系数，
表示皮肤基线吸收系数。
步骤四 建立表示在皮肤真皮层中所述散射系数与皮肤胶原蛋白体积分数、皮肤胶原蛋白纤维半径、皮肤胶原蛋白纤维束直径之间关系的方程；
步骤五 建立表示皮肤表皮层中光反射率、光透射率与表皮厚度之间关系的方程。
散射是光在与皮肤相互作用中，光的传播方向被改变而产生的现象，
方向的改变并不遵循Snell定律(折射定律)。皮肤中的主要散射成分为胶原蛋白，并且皮肤具有两种不同的散射：瑞利散射以及米氏散射。瑞利散射主要由小尺度的胶原蛋白纤维结构产生，而米氏散射主要有大尺度的胶原蛋白纤维束(由多束具有相同走向的纤维绑定而成)产生，真皮的散射系数由3个参数表示：胶原蛋白体积分数，胶原蛋白纤维半径，胶原蛋白纤维束半径。根据生物学的研究，本皮肤模型中表皮的散射系数使用真皮散射系数近似，即两层拥有相同的散射系数。
公式(12)，其中σs为真皮层散射系数，
为真皮层瑞利散射系数，
为真皮层米氏散射系数。
散射理论主要描述了介质的散射系数σs可以通过单个胶原蛋白截面散射系数σss和其数量密度N的乘积获得，而胶原蛋白截面散射系数σss则可以通过散射效率σsca乘以其截面面积A获得。
根据物理瑞利散射理论,米氏散射方程，计算米氏散射效率，计算各向异性指数，得到米氏散射系数，最终得到公式(12)。
虽然表皮层中并不含有胶原蛋白，但根据生物—物理学家的研究结果，表皮层散射特性与真皮层散射特性无异，故表皮层散射也由真皮层散射近似。
步骤五 建立表示皮肤表皮层中光反射率、光透射率与表皮厚度之间关系的方程。
步骤六 建立表示皮肤真皮层中光反射率与表皮厚度之间关系的方程。
实施例3
优选地，在本发明高拟合度的多个皮肤参数的数学建模方法中，还包括步骤七：将步骤一至步骤四中计算的皮肤参数与样本实际的皮肤参数比较，计算拟合度，判断上述多个皮肤参数的数学建模方法是否模拟精确，可以用于皮肤参数的量化分析。具体步骤包括1.测量目标皮肤各个波段的
反射率样本，并进行预处理以增加平滑度和减少测量噪音；2.猜测一组初始皮肤生物学参数；3.根据所述步骤，分别计算皮肤表面、表皮层、真皮层的反射率及透射率；4.根据所述步骤，计算皮肤表皮层、真皮层以及皮下组织共三层结构的反射率；5.计算皮肤四层结构的反射率；6.比较计算得到的皮肤反射率和测量的皮肤反射率之间的距离；7.计算所选的参数组中，各个参数变化时计算得到的皮肤反射率和测量的皮肤反射率之间距离的变化；8.根据距离的变化选择下一组皮肤生物学参数，并重复步骤3至步骤8，直到得到拥有最小距离的一组生物学参数。其拟合曲线如图1所示，由拟合曲线可知，本方法拟合程度非常高。
以上所述实施例仅表达了本发明的实施方式，其描述较为具体和详细，但并不能因此而理解为对本发明专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本发明构思的前提下，还可以做出若干变形和改进，这些都属于本发明的保护范围。因此，本发明专利的保护范围应以所附权利要求为准。
Claims (15)
- 一种人皮肤光谱的建模方法，其特征在于：所述方法包括以下步骤：步骤一 根据可见光照射皮肤时吸收、反射、散射、透射的特征，将皮肤从上至下抽象为皮肤表面层、皮肤表皮层、皮肤真皮层、皮下组织层四层；步骤二 根据皮肤表皮层的光吸收、散射系数以及皮肤表皮层厚度变量参数，建立在皮肤表皮层中光反射和光透射方程，计算在皮肤表皮层的反射率和透射率；步骤三 根据皮肤真皮层的光吸收、散射系数以及皮肤真皮层厚度变量参数，建立在皮肤真皮层中光反射方程，计算在皮肤真皮层的反射率；步骤四 计算光在皮肤表皮层和皮肤真皮层的两层结构中的反射率；步骤五 计算光通过四层皮肤的总体结构中的反射率。
- 根据权利要求1所述的人皮肤光谱的建模方法，其特征在于，还包括根据辐射传导方程和在皮肤表面层的反射系数，建立在皮肤表面层中光反射和光透射的方程，计算在皮肤表面层的反射率的步骤。
- 根据权利要求2所述的人皮肤光谱的建模方法，其特征在于，所述在皮肤表面层中光反射和光透射的方程为[Rsurf,Tsurf]＝LSI(rsurf)，其中，Rsurf为表面的光反射率，Tsurf为表面的光透射率，LSI代表了用于计算皮肤反射率以及透射率的一系列计算公式，rsurf为皮肤表面的反射系数。
- 权利要求1所述的人皮肤光谱的建模方法，其特征在于，在步骤二中，在所述皮肤表皮层中光反射和光透射由光从上方空气中进入和光从下方真皮层进入两种途径所决定，所述在皮肤表皮层中光反射和光透射的方程为和
其中，
为从空气进入表皮层的光反射率，
为从空气进入表皮层的光透射率，
为从真皮层进入表皮层的光反射率，
为光从真皮层进入表皮层的光透射率，σα epi为表皮层 吸收系数，
为表皮层散射系数，depi为表皮层厚度，Lair→Lepi代表光由空气进入表皮层，Lderm→Lepi代表光由真皮层进入表皮层。
- 根据权利要求1所述的人皮肤光谱的建模方法，其特征在于，在步骤三中，只考虑光反射光由光从上方表皮层中进入的途径，所述在皮肤真皮层中光反射方程为其中，
为从表皮层进入真皮层的光反射率，
为真皮层吸收系数，
为真皮层散射系数，dderm为真皮层厚度，Lepi→Lderm代表光由上方表皮层进入真皮层。
- 根据权利要求1所述的人皮肤光谱的建模方法，其特征在于，在步骤四中，所述光在皮肤表皮层和皮肤真皮层的两层结构中的反射率计算公式为：
其中，Rinter为光在皮肤表皮层和皮肤真皮层的两层结构中的反射率。 - 根据权利要求1所述的人皮肤光谱的建模方法，其特征在于，在步骤五中，所述总体结构中的反射率公式为Rskin＝Rsurf+TsurfRinter。
- 一种根据权利要求1所述的人皮肤光谱的建模方法，建立的高拟合度的多个皮肤参数的数学建模方法，其特征在于，包括以下步骤：步骤一 建立表示在皮肤表面层中的所述反射率和皮肤粗糙度以及皮肤光亮度之间关系的方程；步骤二 建立表示在皮肤表皮层中的所述吸收系数与表皮层黑色素体积分数、表皮层优黑素浓度、表皮层褐黑素浓度、表皮层水分体积分数、表皮层油脂体积分数、表皮层中胡萝卜素浓度、表皮层皮层厚度之间关系的方程；步骤三 建立表示在皮肤真皮层中的所述吸收率与真皮层水分体积分数、血液体积分数、血红蛋白浓度、血液中氧化血红蛋白体积分数、血液中脱 氧血红蛋白体积分数、血液中一氧化碳血红蛋白体积分数、血液中高铁血红蛋白体积分数、血液中硫化血红蛋白体积分数、真皮层胡萝卜素浓度、真皮层胆红素浓度、血液中血小板体积分数、血液中血红蛋白体积分数、真皮层弹性蛋白体积分数之间关系的方程；步骤四 建立表示在皮肤真皮层中所述散射系数与皮肤胶原蛋白体积分数、皮肤胶原蛋白纤维半径或直径、皮肤胶原蛋白纤维束半径或直径之间关系的方程；步骤五 建立表示皮肤表皮层中光反射率、光透射率与表皮厚度之间关系的方程；步骤六 建立表示皮肤真皮层中光反射率与表皮厚度之间关系的方程。
- 根据权利要求8所述的高拟合度的多个皮肤参数的数学建模方法，其特征在于，还包括步骤七：将步骤一至步骤六中通过所述皮肤参数所虚拟的皮肤光谱与实际的皮肤光谱拟合。
- 根据权利要求8所述的高拟合度的多个皮肤参数的数学建模方法，其特征在于，在步骤一中，所述方程为其中fsurf表示皮肤表面对单一方向的反射率，dω0表示反射方向的微元。
- 根据权利要求8所述的高拟合度的多个皮肤参数的数学建模方法，其特征在于，在步骤二中，所述方程为
其中，表示表皮层吸收系数，fme表示表皮层黑色素体积分数，ceu表示表皮层优黑素浓度，
表示优黑素吸收系数，cph表示褐黑素浓度，
表示褐黑素吸收系数，
表示表皮水分体积分数，
表示水分吸收系数，flipid表示表皮层油脂体积分数，
表示油脂吸收系数，
表示表皮层胡萝卜素浓度，
表示胡萝卜素吸收系数，
表示皮肤基线 吸收系数。
- 根据权利要求8所述的高拟合度的多个皮肤参数的数学建模方法，其特征在于，在步骤三中，所述方程为
其中，
表示真皮层吸收系数，
表示真皮层水分体积分数，
表示水分吸收系数，fblood表示血液体积分数，Soxy表示血液中含氧血红蛋白体积分数即血氧浓度，cblood表示血红蛋白浓度，
表示含氧血红蛋白吸收系数，Sdeoxy表示血液中脱氧血红蛋白体积分数，
表示脱氧血红蛋白吸收系数，Sco表示血液中一氧化碳血红蛋白体积分数，
表示一氧化碳血红蛋白吸收系数，Smet表示血液中高铁血红蛋白体积分数，
表示高铁血红蛋白吸收系数，Ssulf表示血液中硫化血红蛋白体积分数，
表示硫化血红蛋白吸收系数，
表示真皮层胡萝卜素浓度，
表示胡萝卜素吸收系数，cbr表示胆红素浓度，
表示胆红素吸收系数，fplt表示血液中血小板体积分数，
表示血小板吸收系数，H表示血液中血红蛋白体积分数，fela表示弹性蛋白体积分数，
表示弹性蛋白吸收系数，
表示皮肤基线吸收系数。
- 根据权利要求8所述的高拟合度的多个皮肤参数的数学建模方法，其特征在于，在步骤四中，所述方程为其中，σs表示真皮层散射系数，
为真皮层瑞利散射系数，
为真皮层米氏散射系数。
- 根据权利要求8所述的高拟合度的多个皮肤参数的数学建模方法，其特征在于，在步骤五中，所述方程为和
其中，为从空气进入表皮层的光反射率，
为从空气进入表皮层的光透射率，
为从真皮层进入表皮层的光反射率，
为光从真皮层进入表皮层的光透射率，σα epi为表皮层吸收系数，
为表皮层散射系数， depi为表皮层厚度，Lair→Lepi代表光由空气进入表皮层，Lderm→Lepi代表光由真皮层进入表皮层。
- 根据权利要求8所述的高拟合度的多个皮肤参数的数学建模方法，其特征在于，在步骤六中，所述方程为
其中，为从表皮层进入真皮层的光反射率，
为真皮层吸收系数，
为真皮层散射系数，dderm为真皮层厚度，Lepi→Lderm代表光由上方表皮层进入真皮层。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610181986.6 | 2016-03-28 | ||
| CN201610181986.6A CN105786762B (zh) | 2016-03-28 | 2016-03-28 | 一种人皮肤光谱的建模方法以及高拟合度的多个皮肤参数的数学建模方法 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2017167030A1 true WO2017167030A1 (zh) | 2017-10-05 |
Family
ID=56392012
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2017/077001 WO2017167030A1 (zh) | 2016-03-28 | 2017-03-16 | 一种人皮肤光谱的建模方法以及高拟合度的多个皮肤参数的数学建模方法 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN105786762B (zh) |
| WO (1) | WO2017167030A1 (zh) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113723028A (zh) * | 2021-09-07 | 2021-11-30 | 北京理工大学 | 一种测定人体皮肤油脂中角鲨烯和臭氧反应速率常数的方法 |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106021178B (zh) * | 2016-03-28 | 2018-11-06 | 陈威 | 一种利用数学模型计算人皮肤胶原蛋白相关3个参数的方法 |
| CN105786762B (zh) * | 2016-03-28 | 2017-02-22 | 陈威 | 一种人皮肤光谱的建模方法以及高拟合度的多个皮肤参数的数学建模方法 |
| CN107411709B (zh) * | 2017-06-02 | 2020-08-25 | 皑高森德医疗技术（北京）有限责任公司 | 一种基于高光谱图像的皮肤生物学成分量化分析方法 |
| CN107752981B (zh) * | 2017-10-30 | 2019-08-06 | 温州医科大学 | 一种基于层状结构映射的光学成像方法 |
| CN108272533B (zh) * | 2017-12-26 | 2019-12-17 | 中国科学院苏州生物医学工程技术研究所 | 皮肤创面区域的皮肤建模方法 |
| CN108447085B (zh) * | 2018-02-11 | 2022-01-04 | 浙江大学 | 一种基于消费级rgb-d相机的人脸视觉外观恢复方法 |
| CN110826431B (zh) * | 2019-10-23 | 2023-04-25 | 南京航空航天大学 | 基于蒙特卡洛的可见光静脉显像方法 |
| US11982724B2 (en) * | 2020-10-06 | 2024-05-14 | Canon Medical Systems Corporation | MR signal processing apparatus |
| CN116012379B (zh) * | 2023-03-27 | 2023-09-08 | 皑高森德医疗器械(北京)有限责任公司 | 一种基于黑色素体积含量分布形态的白斑分级方法 |
| CN116019425B (zh) * | 2023-03-27 | 2023-05-23 | 皑高森德医疗器械(北京)有限责任公司 | 一种基于高光谱图像的数字皮肤模型及其应用 |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103970982A (zh) * | 2013-01-31 | 2014-08-06 | 施乐公司 | 用于从扩散反射数据估计生物物理学模型的生物学参数向量的基于对照的反演 |
| CN104395727A (zh) * | 2012-06-18 | 2015-03-04 | 富士胶片株式会社 | 皮肤的评价方法及皮肤评价装置 |
| CN105786762A (zh) * | 2016-03-28 | 2016-07-20 | 陈威 | 一种人皮肤光谱的建模方法以及高拟合度的多个皮肤参数的数学建模方法 |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8154612B2 (en) * | 2005-08-18 | 2012-04-10 | Qualcomm Incorporated | Systems, methods, and apparatus for image processing, for color classification, and for skin color detection |
| CN105380609A (zh) * | 2015-12-07 | 2016-03-09 | 江苏鼎云信息科技有限公司 | 基于多光谱的皮肤检测方法与系统 |
-
2016
- 2016-03-28 CN CN201610181986.6A patent/CN105786762B/zh active Active
-
2017
- 2017-03-16 WO PCT/CN2017/077001 patent/WO2017167030A1/zh active Application Filing
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104395727A (zh) * | 2012-06-18 | 2015-03-04 | 富士胶片株式会社 | 皮肤的评价方法及皮肤评价装置 |
| CN103970982A (zh) * | 2013-01-31 | 2014-08-06 | 施乐公司 | 用于从扩散反射数据估计生物物理学模型的生物学参数向量的基于对照的反演 |
| CN105786762A (zh) * | 2016-03-28 | 2016-07-20 | 陈威 | 一种人皮肤光谱的建模方法以及高拟合度的多个皮肤参数的数学建模方法 |
Non-Patent Citations (1)
| Title |
|---|
| FOTOUHI, M. ET AL.: "Skin Detection Using Contourlet Texture Analysis", COMPUTER CONFERENCE, 31 December 2009 (2009-12-31), XP031577897 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113723028A (zh) * | 2021-09-07 | 2021-11-30 | 北京理工大学 | 一种测定人体皮肤油脂中角鲨烯和臭氧反应速率常数的方法 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105786762B (zh) | 2017-02-22 |
| CN105786762A (zh) | 2016-07-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2017167030A1 (zh) | 一种人皮肤光谱的建模方法以及高拟合度的多个皮肤参数的数学建模方法 | |
| Krishnaswamy et al. | A biophysically‐based spectral model of light interaction with human skin | |
| Chen et al. | Hyperspectral modeling of skin appearance | |
| Fredriksson et al. | Inverse Monte Carlo method in a multilayered tissue model for diffuse reflectance spectroscopy | |
| US7440659B2 (en) | Depth-resolved reflectance instrument and method for its use | |
| Chatterjee et al. | Investigating the origin of photoplethysmography using a multiwavelength Monte Carlo model | |
| WO2017167031A1 (zh) | 一种利用数学模型计算人皮肤光吸收相关的19个生物学参数的方法 | |
| Yim et al. | A cell‐based light interaction model for human blood | |
| Duadi et al. | Dependence of light scattering profile in tissue on blood vessel diameter and distribution: a computer simulation study | |
| Vaudelle et al. | Light source distribution and scattering phase function influence light transport in diffuse multi-layered media | |
| Chatterjee et al. | Investigating optical path and differential pathlength factor in reflectance photoplethysmography for the assessment of perfusion | |
| WO2017193699A1 (zh) | 一种利用数学模型计算人皮肤胶原蛋白相关3个参数的方法 | |
| Nishidate et al. | Estimation of absorbing components in a local layer embedded in the turbid media on the basis of visible to near-infrared (VIS-NIR) reflectance spectra | |
| Ohtsuki et al. | Multiple-reflection model of human skin and estimation of pigment concentrations | |
| Yim et al. | On the modeling of light interactions with human blood | |
| Herrmann et al. | Monte-Carlo simulation of light tissue interaction in medical hyperspectral imaging applications | |
| Yamamoto et al. | Optical path-length matrix method for estimating skin spectrum | |
| Wang et al. | The optical properties of onion dry skin and flesh at the wavelength 632.8 nm | |
| CN100387930C (zh) | 皮下脂肪厚度测定方法及其测定装置 | |
| Chatterjee et al. | Investigating optical path in reflectance pulse oximetry using a multilayer Monte Carlo model | |
| Budidha et al. | Monte Carlo characterization of short-wave infrared optical wavelengths for biosensing applications | |
| Baranoski et al. | Multilayer modeling of skin color and translucency | |
| Lee et al. | Performance estimation of optical skin probe in short wavelength infrared spectroscopy based on Monte-Carlo simulation | |
| Meng et al. | Noninvasive optical monitoring of pulmonary embolism: a Monte Carlo study on visible Chinese human thoracic tissues | |
| Mroczka et al. | Modeling of light transmittance measurement in a finite layer of whole blood--a collimated transmittance problem in Monte Carlo simulation and diffusion model. |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 17773060 Country of ref document: EP Kind code of ref document: A1 |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 17773060 Country of ref document: EP Kind code of ref document: A1 |