Method and device for quantitatively analyzing optical characteristics, volume and spatial distribution of fingerprint sweat gland tube based on OCT (Optical Coherence Tomography)

Abstract

The invention provides an OCT (Optical Coherence Tomography)-based quantitative analysis method and device for optical characteristics, volume and spatial distribution of a fingerprint sweat gland duct, and the method comprises the steps: extracting a light attenuation coefficient of a sweat gland duct region from a fingertip OCT signal by using a depth resolution light attenuation method on the basis of collecting an OCT image of a fingertip; extracting a sweat gland duct area by using a deep learning model, performing three-dimensional reconstruction, and calculating the volume of a sweat gland duct; the spatial distribution uniformity of sweat gland tubes is analyzed by adopting a Voronoi diagram, and the spatial arrangement characteristics and uniformity of the sweat gland tubes are quantified by drawing a regional area distribution histogram.

Description

Technical Field

[0001] The present invention belongs to the technical fields of medical imaging and quantitative analysis of biological tissues, and specifically relates to a method and device for quantitatively analyzing the optical properties, volume and spatial distribution of fingerprint sweat gland ducts based on OCT. Background Art

[0002] Current status of research on sweat gland structure 1. Sweat gland function and research challenges: Sweat glands are important appendages of the skin and are mainly responsible for regulating body temperature and excretion. Traditional sweat gland research methods, such as histopathological sections and functional tests, can provide certain structural and functional information. However, these methods usually have the disadvantages of high invasiveness and low resolution, and cannot perform non-destructive and comprehensive sweat gland structural analysis, especially in the study of fine structures at the microscopic level. There are significant limitations.

[0003] Advantages and challenges of OCT technology 2. Optical coherence tomography (OCT) technology has the characteristics of high resolution, non-destructive and real-time imaging, and its tissue penetration depth reaches 1-2 mm. It is an effective tool for analyzing the skin and its accessory structures. However, intuitive observation of OCT images is difficult to fully reveal the optical and spatial distribution characteristics of sweat gland ducts. Therefore, quantitative analysis methods combining deep learning and mathematical modeling have become an important means to solve this problem. Summary of the invention

[0004] Therefore, in view of the defects and shortcomings of the prior art, as well as actual needs, the present invention proposes a quantitative analysis method for the optical properties, volume and spatial distribution of fingerprint sweat gland ducts based on OCT, aiming to systematically study the optical properties, volume characteristics and spatial distribution laws of sweat gland ducts. The optical attenuation coefficient of the sweat gland duct area is extracted from the OCT signal by the depth-resolved light attenuation method, and then its optical properties are quantitatively analyzed. The U-Net segmentation technology based on deep learning is used to accurately extract the sweat gland duct area and perform three-dimensional reconstruction to calculate its volume characteristics. Furthermore, the Voronoi diagram is used to analyze the uniformity of the spatial distribution of the sweat gland duct, and the spatial arrangement characteristics are revealed by drawing the area distribution histogram of each region. This method provides a non-destructive, accurate and efficient quantitative analysis tool for the study of the structure and function of fingerprint sweat gland ducts.

[0005] The technical solution specifically adopted by the present invention to solve the technical problem is: A quantitative analysis method for the optical characteristics, volume and spatial distribution of fingerprint sweat gland ducts based on OCT: Based on the OCT images collected from the fingertips, the optical attenuation coefficient of the sweat gland duct area is extracted from the fingertip OCT signal using the depth-resolved optical attenuation method; the sweat gland duct area is extracted and three-dimensionally reconstructed using a deep learning model, and the volume of the sweat gland duct is calculated; the Voronoi diagram is used to analyze the spatial distribution uniformity of the sweat gland duct, and the spatial arrangement characteristics and uniformity of the sweat gland duct are quantified by drawing the regional area distribution histogram.

[0006] Furthermore, the light attenuation coefficient is calculated using a single scattering model.

[0007] Furthermore, the method of extracting the sweat gland duct area using a deep learning model and performing three-dimensional reconstruction is specifically as follows: performing image segmentation on the original OCT cross-sectional image of each fingertip skin using a U-Net deep learning model obtained by training with a labeled data set to extract the sweat gland duct area; and performing three-dimensional reconstruction on the two-dimensional slices after U-Net segmentation to construct a three-dimensional structure of the sweat gland duct.

[0008] Furthermore, the volume of the sweat gland duct is calculated by using a three-dimensional voxel calculation method, and the total volume pixel value is obtained by calculating the pixel value and physical size of each voxel, thereby obtaining the volume of the sweat gland duct.

[0009] Furthermore, the use of Voronoi diagram to analyze the uniformity of spatial distribution of sweat gland ducts is specifically as follows: in the distribution of sweat gland duct areas, a Voronoi diagram is generated through a center point, the area of ​​each sub-area is calculated, and the spatial distribution uniformity of sweat gland ducts of different groups is analyzed through the relationship between maximum intensity projection and fingerprint texture.

[0010] Furthermore, the method of quantifying the spatial arrangement characteristics and uniformity of the sweat gland ducts by drawing a regional area distribution histogram is specifically as follows: the histogram is obtained by drawing on the basis of calculating the area of ​​each sub-region.

[0011] Furthermore, the method of drawing a regional area distribution histogram to quantify the spatial arrangement characteristics and uniformity of the sweat gland ducts is specifically as follows: Gaussian fitting is performed on the envelope of the histogram to obtain Gaussian fitting parameters, including: the amplitude, mean and standard deviation of the Gaussian function, and the arrangement characteristics and uniformity of the spatial distribution of the sweat gland ducts are quantitatively evaluated through the shape and parameters of the fitting curve.

[0012] And, a device for quantitatively analyzing the optical characteristics, volume and spatial distribution of fingerprint sweat gland ducts based on OCT, comprising: An OCT system for acquiring OCT images of fingertips; A light attenuation analysis module, for extracting the light attenuation coefficient of the sweat gland duct area from the fingertip OCT signal acquired by the OCT system using a depth-resolved light attenuation method; The volume analysis module is used to extract the sweat gland duct area based on the fingertip OCT image using a deep learning model and perform three-dimensional reconstruction, and calculate the volume of the sweat gland duct; The spatial distribution analysis module is used to analyze the spatial distribution uniformity of sweat gland ducts based on fingertip OCT images using Voronoi diagrams, and to quantify the spatial arrangement characteristics and uniformity of sweat gland ducts by drawing regional area distribution histograms.

[0013] Furthermore, the light attenuation analysis module, the volume analysis module and the space distribution analysis module are arranged in a computer system including a memory and a processor.

[0014] Furthermore, the volume analysis module uses the U-Net deep learning model obtained by training the labeled data set to perform image segmentation on each original OCT cross-sectional image of the fingertip skin to extract the sweat gland duct area; the two-dimensional slices after U-Net segmentation are three-dimensionally reconstructed to construct the three-dimensional structure of the sweat gland duct; the volume adopts the three-dimensional voxel calculation method, and the total volume pixel value is obtained by calculating the pixel value and physical size of each voxel, thereby obtaining the volume of the sweat gland duct; The spatial distribution analysis module generates a Voronoi diagram through the center point in the distribution of the sweat gland duct area, calculates the area of ​​each sub-area, and analyzes the spatial distribution uniformity of different groups of sweat gland ducts through the relationship between the maximum intensity projection and the fingerprint texture; a histogram is obtained by drawing on the basis of calculating the area of ​​each sub-area; and Gaussian fitting is performed on the envelope of the histogram to obtain Gaussian fitting parameters, including: the amplitude, mean and standard deviation of the Gaussian function, and the arrangement characteristics and uniformity of the spatial distribution of the sweat gland ducts are quantitatively evaluated through the shape and parameters of the fitting curve.

[0015] Compared with the prior art, the present invention and its preferred solution realize quantitative analysis of the light attenuation characteristics, volume and spatial distribution of fingerprint sweat gland ducts with high accuracy and reliability. By combining deep learning with Voronoi diagram technology, non-destructive and real-time high-resolution analysis of sweat gland ducts can be performed, providing a new technical means for the study of the structure and function of sweat gland ducts, and has broad application prospects.

[0016] For example, in further research, it was found that the light attenuation characteristics of fingerprints of normal people and diabetic patients were quantitatively evaluated using OCT signals based on the depth-resolved light attenuation method, and it was found that there were obvious differences in the sweat gland ducts of the fingerprint epidermis. U-net segmentation based on deep learning was used to extract the sweat gland ducts of the internal fingerprints of patients at different stages of diabetes, and the volume of the sweat gland ducts of different diabetes courses was quantitatively analyzed and compared. It was found that with the progression of diabetes, the volume of the sweat gland ducts also decreased. In addition, the absence of sweat gland ducts leads to a decrease in the uniformity of their spatial distribution, which can be reflected in the changes in the size and shape of the Voronoi diagram area.

[0017] Therefore, the solution provided by the present invention is expected to be used to guide the assessment of the progress of diabetes. Since this solution is a non-destructive detection solution, it helps to reduce costs and improve experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments: Figure 1 It is a block diagram of the OCT system; Figure 2 This is a flow chart of obtaining a three-dimensional image of a sweat gland duct according to an embodiment of the present invention, wherein (a) is a step and (b) is an effect diagram; Figure 3 A diagram showing the obvious differences in the volumes of the sweat gland ducts corresponding to different groups of samples obtained through the embodiment of the present invention; Figure 4 A diagram showing the obvious difference in light attenuation of fingertip slices corresponding to different groups of samples obtained through the embodiment of the present invention; Figure 5 The following are the evaluation results of the uniformity of the distribution of three groups of sweat gland ducts based on the Voronoi diagram obtained by the embodiment of the present invention. Column (a) is a top view of the distribution of sweat gland ducts; column (b) is a Voronoi diagram of the distribution of sweat gland ducts; column (c) is a distribution diagram of the area size of each region in the Voronoi diagram. DETAILED DESCRIPTION

[0019] In order to make the features and advantages of this patent more obvious and easy to understand, the following embodiments are specifically described in detail as follows: It should be noted that the following detailed descriptions are illustrative and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used in this specification have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0020] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0021] The embodiment of the present invention provides a sweat gland duct assessment solution based on optical coherence tomography (OCT), which mainly includes the following designs: a. The optical attenuation coefficient of the sweat gland duct area is extracted from the OCT signal using the depth-resolved optical attenuation method; b. Based on U-Net deep learning segmentation technology, the sweat gland duct area is extracted and 3D reconstructed to calculate the volume of the sweat gland duct; c. The Voronoi diagram was used to analyze the spatial distribution uniformity of the sweat gland ducts, and the regional area distribution histogram was drawn to quantify the spatial arrangement characteristics and uniformity of the sweat gland ducts.

[0022] As a preferred solution of this embodiment, the light attenuation coefficient calculation in step a adopts a single scattering model, and the result is visualized using pseudo color mapping to facilitate further analysis and evaluation.

[0023] The U-Net segmentation technology in step b includes data labeling, model training and three-dimensional reconstruction of segmentation results. The volume calculation method is based on the three-dimensional voxel calculation method, and the volume of the sweat gland duct is obtained by calculating the pixel value and physical size of each voxel.

[0024] The volume of the sweat gland duct after three-dimensional reconstruction was calculated using the voxel accumulation method to calculate the pixel value and physical size of each voxel and obtain the total volume pixel value.

[0025] The Voronoi diagram analysis in step c quantifies the spatial arrangement characteristics and uniformity of the sweat gland ducts by calculating the area of ​​each sub-region and drawing a distribution histogram.

[0026] In the Voronoi diagram calculation, the distribution of sweat gland duct areas is used to generate a Voronoi diagram through the center point, the area of ​​each sub-area is calculated, and the uniformity of its spatial distribution is evaluated.

[0027] Based on the purpose of the present invention, in this embodiment, the light attenuation coefficient of the sweat gland duct area is extracted from the OCT image signal using the depth-resolved light attenuation method, which can quantitatively analyze the changes in its optical properties. By analyzing the light attenuation coefficients at different depths, the optical property differences of the sweat gland duct in the tissue can be effectively revealed, providing an in-depth understanding of the sweat gland duct structure.

[0028] The U-Net segmentation technology based on deep learning can accurately extract the sweat gland duct area from the OCT image and perform three-dimensional reconstruction. By calculating the volume characteristics of the reconstructed sweat gland duct, its changes under different conditions can be quantitatively evaluated. This method is efficient and accurate, and can provide strong support for the morphological analysis of the sweat gland duct and provide a reference for related clinical research. It should be noted here that U-Net is not the only deep model that can achieve the purpose of the present invention. Other existing models widely used in medical image processing can generally achieve the purpose of image segmentation required by the present invention. The use of U-Net is only a preferred implementation scheme.

[0029] The Voronoi Diagram is a mathematical model based on spatial distribution. The present invention introduces the Voronoi Diagram technology into OCT image analysis, which can accurately quantify the spatial distribution characteristics of sweat gland ducts and improve the understanding of their spatial structure. By analyzing the distribution and shape changes of the regions in the Voronoi Diagram, the arrangement of sweat gland ducts in space can be effectively revealed, thereby providing an in-depth analysis tool for the function of sweat gland ducts.

[0030] By calculating the area distribution of the sweat gland ducts and drawing the corresponding distribution histogram, the arrangement characteristics and uniformity of the sweat gland ducts in space are revealed, providing a quantitative basis for studying the functions of the sweat gland ducts and their changes under different conditions.

[0031] A specific implementation method and technical details of the embodiment of the present invention are further introduced below in conjunction with the accompanying drawings: 1. OCT system composition and working principle 1. As Figure 1 As shown in the figure, the OCT system includes modules such as SLD light source, reflector, focusing lens, NI-DAQ acquisition card, etc. The goal of the system design is to achieve high-resolution three-dimensional imaging and obtain detailed information of the sweat gland duct area. The system provides high-precision, non-destructive imaging of the sweat gland duct structure through optical coherence tomography technology.

[0032] Sweat gland duct region extraction and 3D reconstruction like Figure 2 As shown, the embodiment of the present invention uses a U-Net segmentation method based on deep learning to extract the sweat gland duct area in the OCT image and perform three-dimensional reconstruction. The specific steps include: (I) Use the LabelMe tool to manually annotate the sweat gland ducts and epidermis locations in fingerprints and construct an annotated dataset; (II) Use the U-Net deep learning model to train the labeled dataset, learn image features and optimize model hyperparameters; (III) Segment the input original image based on the learned features to obtain the segmentation map of the sweat gland duct and fingerprint; (IV) extracting the sweat gland duct area from the original OCT cross-sectional image based on each sweat gland duct segmentation map; (V) The segmented two-dimensional slices are reconstructed three-dimensionally to construct the three-dimensional structure of the sweat gland duct.

[0033] Through the above steps, three-dimensional visualization of the sweat gland duct area can be achieved, providing basic data for subsequent volume analysis and optical property analysis.

[0034] Quantitative analysis of sweat gland duct volume Based on the U-Net segmentation technology, this paper extracts the sweat gland duct area and performs three-dimensional reconstruction. The specific steps are as follows: Data labeling: Use the LabelMe tool to label the sweat gland duct area in the fingerprint and construct a training data set; U-Net model training: train the U-Net model and optimize segmentation parameters; Three-dimensional reconstruction and volume calculation: The segmentation results are reconstructed in three dimensions, the volume of the sweat gland ducts is calculated based on the three-dimensional voxel calculation method, and statistical analysis is performed.

[0035] This method can efficiently and accurately calculate the volume characteristics of sweat gland ducts and provide quantitative data for subsequent functional analysis.

[0036] The obvious differences in the volumes of the sweat gland ducts corresponding to the different groups of samples obtained by the embodiment of the present invention are shown as follows: Figure 3 As shown in the validation test of the association between sweat duct volume and diabetes, we found significant differences in sweat duct volume among the three groups of patients with diabetes progression. In diabetic patients with diabetes and without neuropathy, the number of larger sweat ducts was a major component and became less prominent as the severity of diabetes increased. At the same time, Figure 3 As shown in the box plot in , the volume of the sweat gland ducts decreases as the severity of diabetic neuropathy increases. This indicates that the reduction in the volume of the sweat gland ducts may lead to serious clinical consequences, including no sweat response or reduced sweat production. Therefore, the number and volume of the sweat gland ducts may be key parameters for the progression of diabetic neuropathy. Similarly, the number and volume of the sweat gland ducts also have the potential to be important indicators for evaluating other diseases or abnormal conditions related to sweat gland function. The solution provided by the present invention provides an effective way to accurately and quantitatively extract such indicators.

[0037] 4. Light attenuation characteristics analysis The present invention uses the depth-resolved light attenuation method to extract the light attenuation coefficient of the sweat gland duct area from the OCT signal. This method is based on the single scattering model, determines the response of the light attenuation coefficient to different depths through numerical simulation, and conducts experimental verification on uniform and stratified samples to achieve accurate estimation of the attenuation coefficient in biological tissues. The calculation formula is as follows: in, Representative The light attenuation coefficient of each pixel is 2, where the factor 2 is the round trip path of the light in the tissue twice. Represents the pixel value size, The light intensity of the current pixel, Represents the sum of the light intensity from this point to the bottom of the image.

[0038] The algorithm steps include: Read image and preprocess: read OCT image, apply Gaussian blur to reduce noise, convert to grayscale to extract light intensity information, and perform threshold processing to separate the area of ​​interest. Read segmentation mark: read segmentation mark from the file to determine the area of ​​interest in the OCT image. Calculate light attenuation coefficient: use the above formula to calculate the light attenuation coefficient of each pixel, which reflects the degree of light attenuation in the tissue. Normalization and pseudo-color mapping: normalize the light attenuation coefficient matrix so that its value is within a certain range (such as 0-255), and apply pseudo-color mapping for easy visualization and analysis. Analyze light attenuation coefficient: analyze the overall change of light attenuation coefficient, including calculating the mean, standard deviation and vertical profile to evaluate tissue characteristics. This step can quantitatively analyze the light attenuation characteristics of the sweat gland duct area and reveal its optical characteristic changes.

[0039] The obvious difference in light attenuation of fingertip slices corresponding to different groups of samples obtained by the embodiment of the present invention is shown as follows: Figure 4 As shown in the validation test of the association between sweat gland duct volume and diabetes, we found that Figure 4 In the figure, the light attenuation diagrams of healthy people (a) and (c) and diabetic patients (b) and (d) show significant differences. Therefore, analyzing the overall changes in the light attenuation coefficient is expected to be used to evaluate tissue characteristics and possible lesions.

[0040] Voronoi diagram to characterize the uniformity of sweat gland duct area The present invention uses Voronoi Diagram to quantitatively analyze the spatial distribution of sweat gland ducts. The spatial distribution uniformity of sweat gland ducts in different groups is analyzed through the relationship between maximum intensity projection (MIP) and fingerprint texture. Figure 5 Shown are the evaluation results of the uniformity of the distribution of sweat gland ducts of the three groups based on the Voronoi diagram. Figure 5 (a) is a top view of the distribution of sweat gland ducts, where the diameter of the point represents the volume of the sweat gland duct. Figure 5 (b) is the Voronoi diagram of the sweat gland duct distribution. In order to accurately evaluate the spatial uniformity of the sweat gland duct, the spatial distribution of the sweat gland duct obtained based on the Voronoi diagram is plotted into a histogram. Figure 5 (c) is the corresponding distribution diagram of the area size of each region in the Voronoi diagram. By performing Gaussian fitting on the envelope of the histogram, the Gaussian fitting parameters are obtained, where: a is the amplitude (peak value) of the Gaussian function; b is the mean (center position) of the Gaussian function; σ is the standard deviation (bandwidth) of the Gaussian function.

[0041] Figure 5(c) shows the differences in regional characteristics of the Voronoi diagrams of different groups of sweat gland ducts. In this figure, the Voronoi diagram Gaussian fitting curve of the group of sweat gland ducts in row (I) shows a high and relatively concentrated peak value near 30, indicating that the distribution of the sweat gland duct area is relatively uniform. In contrast, the Gaussian fitting curve of the group of sweat gland ducts in row (II) shows a lower peak and right-skewed distribution, and the mean moves to the right, showing an increase in the spatial area of ​​some sweat gland ducts, which is due to the reduction of some sweat gland ducts. It is particularly noteworthy that the Gaussian fitting curve of the group of sweat gland ducts in row (III) shows a wide and flat shape, reflecting the loss of a large number of sweat gland ducts.

[0042] In general, the absence of sweat ducts leads to a decrease in the uniformity of their spatial distribution, which is reflected in the changes in the size and shape of the Voronoi diagram area. Through the regional area distribution of Gaussian fitting, it can be observed that the uniformity of the spatial distribution of sweat ducts is directly related to the size and shape of the regional area. The arrangement characteristics and uniformity of the spatial distribution of sweat ducts can be quantitatively evaluated by the shape and parameter changes of the fitting curve. The volatility of the regional area is usually reflected by the size of the standard deviation (σ), and a more uniform distribution is usually manifested as a higher and more concentrated peak in the histogram. This quantitative characterization and feedback of differences are expected to be applied in the judgment of different diabetes processes or at least have important reference value.

[0043] In addition, based on the method scheme introduced above in this embodiment, the present invention also provides a design of a corresponding device scheme, wherein, as a necessary component, the OCT system is an existing device, and at least one computer system that interacts with the OCT system is also required to be provided, on which the calculation and display of the above light attenuation analysis, volume analysis and spatial distribution analysis are implemented in the form of a computer program, including but not limited to the process of building and training a deep learning model.

[0044] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0045] The above shows and describes the basic principles, main features and advantages of the present disclosure. Those skilled in the art should understand that the present disclosure is not limited by the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present disclosure. Without departing from the spirit and scope of the present disclosure, the present disclosure may have various changes and improvements, and these changes and improvements fall within the scope of the present disclosure to be protected.

[0046] This patent is not limited to the above-mentioned optimal implementation mode. Anyone can derive other various forms of quantitative analysis methods and devices for optical properties, volume and spatial distribution of fingerprint sweat gland ducts based on OCT under the inspiration of this patent. All equal changes and modifications made according to the scope of the patent application of the present invention should be covered by this patent.

Claims

1. A quantitative analysis method for optical properties, volume and spatial distribution of fingerprint sweat gland ducts based on OCT, characterized by: Based on the OCT images collected from the fingertips, the light attenuation coefficient of the sweat gland duct area was extracted from the fingertip OCT signal using the depth-resolved light attenuation method. The sweat gland duct area was extracted and reconstructed in three dimensions using a deep learning model, and the volume of the sweat gland duct was calculated. The Voronoi diagram was used to analyze the spatial distribution uniformity of the sweat gland duct, and the spatial arrangement characteristics and uniformity of the sweat gland duct were quantified by drawing the regional area distribution histogram.

2. The method for quantitatively analyzing the optical properties, volume and spatial distribution of fingerprint sweat gland ducts based on OCT according to claim 1, characterized in that: The light attenuation coefficient is calculated using a single scattering model.

3. The method for quantitatively analyzing the optical properties, volume and spatial distribution of fingerprint sweat gland ducts based on OCT according to claim 1, characterized in that: The method of extracting the sweat gland duct area using a deep learning model and performing three-dimensional reconstruction is specifically as follows: using the U-Net deep learning model obtained by training with a labeled data set to perform image segmentation on the original OCT cross-sectional image of each fingertip skin to extract the sweat gland duct area; and performing three-dimensional reconstruction on the two-dimensional slices after U-Net segmentation to construct a three-dimensional structure of the sweat gland duct.

4. The method for quantitatively analyzing optical properties, volume and spatial distribution of fingerprint sweat gland ducts based on OCT according to claim 3, characterized in that: The volume of the sweat gland duct is calculated by using a three-dimensional voxel calculation method, and the total volume pixel value is obtained by calculating the pixel value and physical size of each voxel, thereby obtaining the volume of the sweat gland duct.

5. The method for quantitatively analyzing the optical properties, volume and spatial distribution of fingerprint sweat gland ducts based on OCT according to claim 1, characterized in that: The method of using Voronoi diagram to analyze the uniformity of spatial distribution of sweat gland ducts is specifically as follows: in the distribution of sweat gland duct areas, a Voronoi diagram is generated through a central point, the area of ​​each sub-area is calculated, and the uniformity of spatial distribution of sweat gland ducts of different groups is analyzed through the relationship between maximum intensity projection and fingerprint texture.

6. The method for quantitatively analyzing the optical properties, volume and spatial distribution of fingerprint sweat gland ducts based on OCT according to claim 5, characterized in that: The method of quantifying the spatial arrangement characteristics and uniformity of the sweat gland ducts by drawing a regional area distribution histogram is specifically as follows: the histogram is obtained by drawing on the basis of calculating the area of ​​each sub-region.

7. The method for quantitatively analyzing the optical properties, volume and spatial distribution of fingerprint sweat gland ducts based on OCT according to claim 1, characterized in that: The method of drawing a regional area distribution histogram to quantify the spatial arrangement characteristics and uniformity of the sweat gland ducts is specifically as follows: Gaussian fitting is performed on the envelope of the histogram to obtain Gaussian fitting parameters, including: the amplitude, mean and standard deviation of the Gaussian function, and the arrangement characteristics and uniformity of the spatial distribution of the sweat gland ducts are quantitatively evaluated through the shape and parameters of the fitting curve.

8. An OCT-based quantitative analysis device for optical properties, volume and spatial distribution of fingerprint sweat gland ducts, characterized in that: include: An OCT system for acquiring OCT images of fingertips; A light attenuation analysis module, for extracting the light attenuation coefficient of the sweat gland duct area from the fingertip OCT signal acquired by the OCT system using a depth-resolved light attenuation method; The volume analysis module is used to extract the sweat gland duct area based on the fingertip OCT image using a deep learning model and perform three-dimensional reconstruction, and calculate the volume of the sweat gland duct; The spatial distribution analysis module is used to analyze the spatial distribution uniformity of sweat gland ducts based on fingertip OCT images using Voronoi diagrams, and to quantify the spatial arrangement characteristics and uniformity of sweat gland ducts by drawing regional area distribution histograms.

9. The device for quantitatively analyzing optical properties, volume and spatial distribution of fingerprint sweat gland ducts based on OCT according to claim 8, characterized in that: The light attenuation analysis module, the volume analysis module and the space distribution analysis module are arranged in a computer system including a memory and a processor.

10. The device for quantitatively analyzing optical properties, volume and spatial distribution of fingerprint sweat gland ducts based on OCT according to claim 8, characterized in that: The volume analysis module uses the U-Net deep learning model obtained by training the annotated data set to perform image segmentation on the original OCT cross-sectional image of each fingertip skin to extract the sweat gland duct area; the two-dimensional slices segmented by U-Net are three-dimensionally reconstructed to construct the three-dimensional structure of the sweat gland duct; the volume adopts the three-dimensional voxel calculation method, and the total volume pixel value is obtained by calculating the pixel value and physical size of each voxel, thereby obtaining the volume of the sweat gland duct; The spatial distribution analysis module generates a Voronoi diagram through the center point in the distribution of the sweat gland duct area, calculates the area of ​​each sub-area, and analyzes the spatial distribution uniformity of different groups of sweat gland ducts through the relationship between the maximum intensity projection and the fingerprint texture; a histogram is obtained by drawing on the basis of calculating the area of ​​each sub-area; and Gaussian fitting is performed on the envelope of the histogram to obtain Gaussian fitting parameters, including: the amplitude, mean and standard deviation of the Gaussian function, and the arrangement characteristics and uniformity of the spatial distribution of the sweat gland ducts are quantitatively evaluated through the shape and parameters of the fitting curve.

Images