A method and system for quantitatively detecting carbon particles in macrophages, and a computer device

Abstract

The application belongs to the technical field of artificial intelligence, and discloses a macrophage intracellular carbon particle quantitative detection method, which comprises the following steps: recognizing a macrophage picture through a target detection network to obtain a macrophage; according to the obtained macrophage, quantitatively calculating the area proportion of carbon particles in the macrophage through a color recognition algorithm; and statistically analyzing the area proportion of the carbon particles in the macrophage to obtain a macrophage intracellular carbon particle quantitative result. The embodiment of the application realizes the functions from recognizing the macrophage to quantitatively calculating the proportion of the carbon particles in the macrophage based on a deep learning network algorithm and a binarization method. The embodiment of the application also discloses a macrophage intracellular carbon particle quantitative detection system and a computer device.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to a method for quantitative detection of carbon particles in macrophages, a system for quantitative detection of carbon particles in macrophages, and a computer device. Background Technology

[0002] Finding appropriate biomarkers for quantifying exposure to fine particulate matter (FPM) is a crucial step in assessing the hazards of FPM. Carbon particles, an important inorganic component of FPM, readily deposit in alveolar macrophages and are difficult to metabolize. Therefore, the amount of carbon particles deposited in macrophages is a potential biomarker for assessing FPM exposure. Currently, most methods rely on software such as ImageJ and Photoshop to manually count carbon particles within macrophages. Traditional quantitative methods are complex and cumbersome, wasting significant time and resources, and the results are highly subjective and unstable.

[0003] The role of machine learning in the field of biomedical image recognition is gradually becoming apparent. Based on this, how to use machine learning to propose a fast and efficient method for identifying carbon particles in macrophages is an urgent problem to be solved. Summary of the Invention

[0004] This invention provides a method and system for quantitative detection of carbon particles within macrophages, addressing the problem of quantifying the exposure of carbon particles within macrophages using manual counting methods in the prior art. To provide a basic understanding of some aspects of the disclosed embodiments, a brief summary is given below. This summary is not intended as a general commentary, nor is it intended to identify key / important components or to define the scope of protection of these embodiments. Its sole purpose is to present some concepts in a simple form as a prelude to the detailed description that follows.

[0005] According to a first aspect of the present invention, a method for quantitative detection of carbon particles in macrophages is provided.

[0006] In one embodiment, a method for quantitative detection of carbon particles in macrophages includes the following steps:

[0007] Macrophages are obtained by identifying images of macrophages using an object detection network;

[0008] Based on the obtained macrophages, the area ratio of carbon particles in the macrophages was quantitatively calculated using a color recognition algorithm.

[0009] Statistical analysis was performed on the area ratio of carbon particles within macrophages to obtain quantitative results of carbon particles within macrophages.

[0010] Optionally, the step of identifying macrophage images using a target detection network to obtain macrophages includes:

[0011] Import macrophage images;

[0012] Feature extraction networks were used to extract features from macrophage images.

[0013] Regions of interest are extracted from macrophage images using a region generation network to generate candidate bounding boxes and obtain macrophage mask images.

[0014] The macrophage mask image is input into the head network, which includes three branches: segmentation network, bounding box regression, and region classification. The head network classifies and regresses the bounding boxes and segments the mask results, outputting the feature data of the macrophage mask image.

[0015] Real-time analysis of macrophage mask image feature data is performed to conduct target detection and obtain macrophages.

[0016] Optionally, the step of extracting regions of interest from macrophage images and generating candidate boxes using a region generation network includes:

[0017] N candidate boxes are generated for each image, and the candidate boxes are mapped onto the feature map of the last convolutional layer of the object detection network.

[0018] Optionally, the step of extracting regions of interest from macrophage images and generating candidate boxes using a region generation network specifically includes:

[0019] The sliding window of the region generation network moves across each macrophage image, generating 15 anchors at the original image position corresponding to each feature pixel, with sizes of (8, 16, 32, 64, 132) and aspect ratios of (0.5, 1, 2).

[0020] Then the anchor is input into a regression class layer of size 1*1 to determine the category and position;

[0021] Finally, the anchor positions are adjusted using a non-maximum suppression algorithm, and the selected anchors are identified as candidate boxes for the region of interest.

[0022] Optionally, the step of quantitatively calculating the area ratio of carbon particles within macrophages using a color recognition algorithm based on the obtained macrophages includes:

[0023] The images of individual macrophages were sequentially segmented and binarized to distinguish carbon particles from macrophages;

[0024] The mask of a single macrophage is processed, and the pixel ratio of the macrophage is calculated by traversing the pixel one by one, and defined as the relative area of ​​the macrophage.

[0025] The pixel percentage of carbon particles is calculated using a pixel-by-pixel traversal method and defined as the relative area of ​​the carbon particles.

[0026] The area ratio of carbon particles within macrophages is calculated based on the relative areas of macrophages and carbon particles.

[0027] Optionally, the step of sequentially performing image segmentation and binarization processing on the single macrophage image specifically includes:

[0028] The image is divided into several small regions, and then pixel analysis is performed in the local regions to divide the image into high-contrast and low-contrast regions.

[0029] For high-contrast regions, the coordinates of each pixel in the image are recorded as (x, y). A w*w window is centered on this pixel, with the average grayscale value of the pixels within this window being m(x, y) and the standard deviation being s(x, y). The local threshold T(x, y) for this pixel is:

[0030] T(x,y)=m(x,y)[1+k(s(x,y) / R-1)]

[0031] Where R is a set value, k ranges from [0.2, 0.5], and T(x,y) is adjusted by m(x,y) and s(x,y);

[0032] For low-contrast, low-resolution regions, pixels are assigned a value of 0, and then the following formula is used for calculation:

[0033] T(x,y)=m(x,y)[1+p e-qm(x,y) +k(s(x,y) / R-1)]

[0034] p, q, k, and e are set values.

[0035] Optionally, the step of binarizing the image of a single macrophage further includes:

[0036] Thresholding segmentation is used to obtain a binary mask image, where pixels with a value of 1 are defined as carbon particle regions and pixels with a value of 0 are defined as macrophage regions.

[0037] Optionally, the step of statistically analyzing the area ratio of carbon particles within macrophages includes:

[0038] Statistical analysis yielded the mean, standard deviation, median, 25th percentile, and 75th percentile.

[0039] According to a second aspect of the present invention, a quantitative detection system for carbon particles in macrophages is provided.

[0040] In one embodiment, a system for quantitative detection of carbon particles within macrophages includes: a target detection network, a quantitative calculation module, and a statistical module; wherein,

[0041] The object detection network is configured to identify macrophage images and obtain macrophages.

[0042] The quantitative calculation module is configured to: quantitatively calculate the area ratio of carbon particles in macrophages based on the obtained macrophages using a color recognition algorithm;

[0043] The statistics module is configured to perform statistical analysis on the area ratio of carbon particles in macrophages to obtain quantitative results of carbon particles in macrophages.

[0044] According to a third aspect of the present invention, a computer device is provided.

[0045] In some embodiments, the computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the method described above.

[0046] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0047] Based on deep learning network algorithms and binarization methods, this system enables the identification of macrophages and the quantitative statistical analysis of the proportion of carbon particles within macrophages.

[0048] It reduces the complexity of relying on manual carbon particle counting, avoids subjective errors caused by humans, and is more in line with the rigor of scientific research.

[0049] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit the invention. Attached Figure Description

[0050] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0051] Figure 1 This is a flowchart illustrating a method for quantitative detection of carbon particles in macrophages according to an exemplary embodiment;

[0052] Figure 2 This is a schematic diagram illustrating the principle of a quantitative detection system for carbon particles in macrophages according to an exemplary embodiment;

[0053] Figure 3 This is a schematic diagram of the structure of a computer device according to an exemplary embodiment. Detailed Implementation

[0054] The following description and accompanying drawings fully illustrate specific embodiments described herein to enable those skilled in the art to practice them. Some embodiments may include or substitute parts and features of other embodiments. The scope of the embodiments herein encompasses the entire scope of the claims and all available equivalents thereof. Throughout this document, the terms “first,” “second,” etc., are used only to distinguish one element from another without requiring or implying any actual relationship or order between the elements. Indeed, a first element can also be referred to as a second element, and vice versa. Furthermore, the terms “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, such that a structure, apparatus, or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a structure, apparatus, or device. Without further limitation, an element defined by the phrase “comprising one…” does not exclude the presence of other identical elements in the structure, apparatus, or device that includes said element. The various embodiments described herein are presented in a progressive manner, with each embodiment focusing on its differences from other embodiments; similar or identical parts between embodiments can be referred to interchangeably.

[0055] The terms "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer" used in this document to indicate orientations or positional relationships are based on the orientations or positional relationships shown in the accompanying drawings. They are used solely for the convenience of describing the document and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. In the description herein, unless otherwise specified and limited, the terms "installed," "connected," and "linked" should be interpreted broadly. For example, they can refer to mechanical or electrical connections, or internal connections between two elements; they can be direct connections or indirect connections through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms according to the specific circumstances.

[0056] In this document, unless otherwise stated, the term "multiple" means two or more.

[0057] In this article, the character " / " indicates that the objects before and after it are in an "or" relationship. For example, A / B means: A or B.

[0058] In this article, the term "and / or" describes an association between objects, indicating that three relationships can exist. For example, A and / or B means: A or B, or A and B.

[0059] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0060] Figure 1 An embodiment of a method for quantitative detection of carbon particles in macrophages according to the present invention is shown.

[0061] In this embodiment, the method for quantitative detection of carbon particles in macrophages includes the following steps:

[0062] Step S1: Macrophage images are identified using a target detection network to obtain macrophages;

[0063] Step S2: Based on the obtained macrophages, the area ratio of carbon particles in the macrophages is quantitatively calculated using a color recognition algorithm.

[0064] Step S3: Statistical analysis is performed on the area ratio of carbon particles in macrophages to obtain quantitative results of carbon particles in macrophages.

[0065] In one embodiment, step S1, which involves identifying macrophage images using a target detection network to obtain macrophages, includes:

[0066] Step S11: Import macrophage images;

[0067] Step S12: Feature extraction network is used to extract features from macrophage images;

[0068] Step S13: Extract regions of interest (RoIs) from macrophage images using a region generation network (RPN), generate feature map candidate boxes, and obtain macrophage mask images;

[0069] Step S14: Input the macrophage mask image into the head network. The head network includes three branches: segmentation network, bounding box regression, and region classification. It completes the classification and regression of the bounding boxes and segments the mask results, and outputs feature data.

[0070] Step S15: Perform real-time analysis of the feature data of the macrophage mask image, perform target detection, and obtain macrophages.

[0071] Optionally, the object detection network is a Mask-RCNN network.

[0072] The test subjects in this application embodiment are images of well-stained, morphologically intact macrophages taken under an optical microscope, each image being unique. Optionally, the imported macrophage images are in .tif format.

[0073] Optionally, embodiments of this application further include a step of preprocessing the imported macrophage image, including: saving the original image in RGB format, and adjusting the image resolution to 1024*800 after processing; the image is saved and used as input for the next stage to generate candidate boxes.

[0074] Optionally, in step S12, this embodiment uses a ResNet50 backbone network for feature extraction from macrophage images, obtaining both low-level and high-level features, and performing better feature fusion. Using the ResNet50 structure as the backbone network can improve target detection accuracy by 1%-2% compared to the original ResNet, while maintaining the inference speed essentially unchanged. Optionally, the macrophage similarity threshold is adjustable. Optionally, before detection, a threshold is selected based on the user's requirements for macrophage morphology extraction. Depending on the selected threshold, a macrophage accuracy recognition range can be set; the lower the value, the higher the probability of being identified as a macrophage, and vice versa.

[0075] Optionally, step S13, which involves extracting regions of interest from macrophage images using a region generation network and generating candidate boxes, specifically includes generating N candidate boxes for each image and mapping the candidate boxes onto the feature map of the last convolutional layer of the object detection network.

[0076] Optionally, step S13, the step of extracting regions of interest from macrophage images and generating candidate boxes using a region generation network, specifically includes:

[0077] The sliding window of the region generation network moves across each macrophage image, generating 15 anchors at the original image position corresponding to each feature pixel, with sizes of (8, 16, 32, 64, 132) and aspect ratios of (0.5, 1, 2).

[0078] Then the anchor is input into a regression class layer of size 1*1 to determine the category and position;

[0079] Finally, the anchor position is adjusted using a nonmaximum suppression algorithm, and the final selected anchor is identified as the feature map of the region of interest.

[0080] Optionally, in step S15, the ROIAlign calculation method is used to perform real-time analysis of the macrophage mask image feature data, and target detection is performed to obtain macrophages.

[0081] In one embodiment, step S2, which involves quantitatively calculating the area ratio of carbon particles within macrophages using a color recognition algorithm based on the obtained macrophages, includes:

[0082] Step S21: Perform image segmentation and binarization on the single macrophage image sequentially to distinguish carbon particles and macrophages;

[0083] Step S22: Process the mask of a single macrophage, calculate the pixel ratio of the macrophage by traversing the pixel by pixel, and define it as the relative area of ​​the macrophage.

[0084] Step S23: Calculate the pixel ratio of carbon particles using a pixel-by-pixel traversal method and define it as the relative area of ​​carbon particles.

[0085] Step S24: Calculate the area ratio of carbon particles within macrophages based on the relative areas of macrophages and carbon particles.

[0086] Optionally, step S21, which involves sequentially performing image segmentation and binarization processing on the single macrophage image, specifically includes:

[0087] The image is divided into several small regions, and then pixel analysis is performed in the local regions to divide the image into high-contrast and low-contrast regions.

[0088] For high-contrast regions, the coordinates of each pixel in the image are recorded as (x, y). A w*w window is centered on this pixel, with the average grayscale value of the pixels within this window being m(x, y) and the standard deviation being s(x, y). The local threshold T(x, y) for this pixel is:

[0089] T(x,y)=m(x,y)[1+k(s(x,y) / R-1)]

[0090] Where R is a set value, k ranges from [0.2, 0.5], and T(x,y) is adjusted by m(x,y) and s(x,y);

[0091] For low-contrast, low-resolution regions, pixels are assigned a value of 0, and then the following formula is used for calculation:

[0092] T(x,y)=m(x,y)[1+p e-qm(x,y) +k(s(x,y) / R-1)],

[0093] p, q, k, and e are set values.

[0094] For example, R is set to 0.5, p to 2, q to 10, k to 0.25, and e to 2.72.

[0095] Optionally, the step of binarizing the image of a single macrophage further includes:

[0096] Thresholding segmentation is used to obtain a binary mask image, where pixels with a value of 1 are defined as carbon particle regions and pixels with a value of 0 are defined as macrophage regions.

[0097] In one embodiment, step S3, the step of statistically analyzing the area ratio of carbon particles in macrophages, includes: statistically analyzing to obtain the mean, standard deviation, median, 25% and 75% quartiles.

[0098] The average value is calculated as follows: X = (X1 + X2 + X3 + ... + Xn) / n, where n is the number of macrophages in the image and Xn is the area ratio of carbon particles in the nth macrophage.

[0099] The method for calculating variance is as follows:

[0100] Methods for calculating median and quartiles:

[0101] After sorting the values ​​in ascending order, the median divides the data into two 50% segments, with the median being the middle value. The quartiles divide the entire set of numbers into two 50% segments: the lower quartile (25%) divides the first 50% into two 25% segments, and the upper quartile (75%) divides the last 50% into two 25% segments. This application integrates multiple statistical methods, allowing users to choose the appropriate method based on their needs, increasing flexibility and openness, and enabling personalized data processing.

[0102] This application discloses a method for statistically analyzing the proportion of carbon particles within macrophages. Based on the Mask R-CNN deep learning algorithm and binarization methods, this invention achieves the function of identifying macrophages and quantitatively calculating the proportion of carbon particles within them. From the user's perspective, this invention combines multiple statistical methods to calculate various parameters such as the mean, standard deviation, median, and quartiles of the carbon particle proportion, allowing users to select the appropriate statistical method.

[0103] In one embodiment, such as Figure 2 As shown, a system for quantitative detection of carbon particles in macrophages is provided, including: a target detection network, a quantitative calculation module, and a statistical module; wherein, the target detection network is configured to: identify macrophage images to obtain macrophages; the quantitative calculation module is configured to: quantitatively calculate the area ratio of carbon particles in macrophages using a color recognition algorithm based on the obtained macrophages; the statistical module is configured to: statistically analyze the area ratio of carbon particles in macrophages to obtain the quantitative result of carbon particles in macrophages.

[0104] The working principle of the quantitative detection system for carbon particles in macrophages in this embodiment is the same as that of the quantitative detection methods for carbon particles in macrophages provided in the above embodiments, and will not be repeated here.

[0105] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 3 As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores static and dynamic information data. The network interface communicates with external terminals via a network connection. When the computer program is executed by the processor, it implements the steps in the above method embodiments.

[0106] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present invention and does not constitute a limitation on the computer device to which the present invention is applied. A specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0107] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0108] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the method embodiments described above.

[0109] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0110] This invention is not limited to the structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this invention is limited only by the appended claims.

Claims

1. A method for quantitatively detecting carbon particles in macrophages, characterized by, The method comprises the following steps: The macrophage picture is identified by the target detection network to obtain the macrophage, including: importing the macrophage picture; performing feature extraction on the macrophage picture by using a feature extraction network; a sliding window of an RPN network moves in each macrophage picture, and 15 anchors are generated for the original image position corresponding to each feature pixel, with sizes of (8, 16, 32, 64, 132) and aspect ratios of (0.5, 1, 2); then the anchors are input into a regression class layer with a size of 1*1 to determine the class and position; finally, the position of the anchor is adjusted by using a non-maximum suppression algorithm, the finally selected anchor is identified as a candidate frame of the region of interest, and a macrophage mask picture is obtained; the macrophage mask picture is input into a head network, the head network includes a segmentation network, a bounding box regression and a region classification, the classification and regression of the bounding box are completed, the mask result is segmented out, and macrophage mask picture feature data are output; the macrophage mask picture feature data are analyzed in real time, target detection is performed, and the macrophage is obtained; According to the obtained macrophage, the area proportion of the carbon particles in the macrophage is quantitatively calculated by a color recognition algorithm, including: sequentially performing image segmentation and binarization processing on a single macrophage image to distinguish the carbon particles and the macrophage; processing the mask of the single macrophage, and calculating the pixel proportion of the macrophage by using a pixel-by-pixel traversal method, and defining the pixel proportion as the relative area of the macrophage; calculating the pixel proportion of the carbon particles by using the pixel-by-pixel traversal method, and defining the pixel proportion as the relative area of the carbon particles; and calculating the area proportion of the carbon particles in the macrophage according to the relative areas of the macrophage and the carbon particles; the step of sequentially performing image segmentation and binarization processing on the single macrophage image specifically comprises: The picture is divided into a plurality of small block regions, and then pixel analysis is performed in the local regions to divide the picture into a high-contrast resolution region and a low-contrast resolution region; For the high-contrast resolution region, the coordinates of each pixel in the image are recorded as (x, y), a w*w size window is centered on the pixel, the average value of the pixel gray scale in the window is m(x, y), and the standard deviation is s(x, y); the local threshold T(x, y) of the pixel is: T(x, y) = m(x, y) [1 + k (s(x, y) / R - 1)] wherein R is a set value, the value range of k is [0.2, 0.5], and T(x, y) is adjusted by m(x, y) and s(x, y); For the low-contrast resolution region, the pixel is divided into 0, and the following formula is used for calculation: T(x,y) = m(x,y) [1 + p e-qm(x,y) + k(s(x,y) / R - 1)], p, q, k, and e are set values; The area proportion of the carbon particles in the macrophage is statistically analyzed to obtain the quantitative result of the carbon particles in the macrophage.

2. The method according to claim 1, wherein the step of binarizing the single macrophage image further comprises: threshold segmentation is used to obtain a binarized mask image, wherein the pixel with a value of 1 is defined as the carbon particle region, and the pixel region with a value of 0 is defined as the macrophage region. ​ 3. The method of claim 1, wherein the step of statistically analyzing the area proportion of the carbon particles in the macrophages comprises: statistical analysis to obtain the mean, standard deviation, median, 25% and 75% quartiles. comprises:

4. A system for quantitatively detecting carbon particles in macrophages, characterized by, a target detection network, a quantitative calculation module, and a statistical module, wherein the target detection network is configured to identify the macrophage images to obtain the macrophages, including: importing the macrophage images; performing feature extraction on the macrophage images using a feature extraction network; moving a sliding window of an RPN network in each macrophage image, and generating 15 anchors for each feature pixel corresponding to the original image position, with sizes of (8, 16, 32, 64, 132) and aspect ratios of (0.5, 1, 2); then inputting the anchors into a regression class layer with a size of 1*1 to determine the class and position; finally, adjusting the position of the anchors using a non-maximum suppression algorithm, and selecting the final anchors as the candidate boxes of the regions of interest to obtain the macrophage mask images; inputting the macrophage mask images into a head network including a segmentation network, a bounding box regression, and a region classification, to complete the classification and regression of the bounding boxes and segment the mask results, and output the macrophage mask image feature data; performing real-time analysis on the macrophage mask image feature data to execute target detection and obtain the macrophages; the quantitative calculation module is configured to quantitatively calculate the area proportion of the carbon particles in the macrophages according to the obtained macrophages by a color recognition algorithm, including: sequentially performing image segmentation and binarization processing on the single macrophage image to distinguish the carbon particles and the macrophages; processing the mask of the single macrophage, calculating the pixel proportion of the macrophage using a pixel-by-pixel traversal method, and defining the relative area of the macrophage; calculating the pixel proportion of the carbon particles using the pixel-by-pixel traversal method, and defining the relative area of the carbon particles; and calculating the area proportion of the carbon particles in the macrophages according to the relative areas of the macrophages and the carbon particles; the step of sequentially performing image segmentation and binarization processing on the single macrophage image specifically comprises: dividing the image into small blocks, and then performing pixel analysis in the local area to divide the image into high-contrast resolution areas and low-contrast resolution areas; for the high-contrast resolution areas, recording the coordinates of each pixel in the image as (x, y), taking the pixel as the center of a w*w window, the average value of the pixel gray scale in the window as m(x, y), and the standard deviation as s(x, y), and calculating the local threshold T(x, y) for the pixel: T(x, y) = m(x, y) [1 + k (s(x, y) / R - 1)] wherein R is a set value, the value of k is in the range of [0.2, 0.5], and T(x, y) is adjusted by m(x, y) and s(x, y); for the low-contrast resolution areas, dividing the pixels into 0, and then calculating using the following formula: p, q, k, and e are set values. T(x,y) = m(x,y) [1 + p e-qm(x,y) + k(s(x,y) / R - 1)], ​ The statistical module is configured to statistically analyze the area proportion of the carbon particles in the macrophages, and obtain a quantitative result of the carbon particles in the macrophages. 5.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-4 when the computer program is executed by the processor. The processor implements the steps of the method of any one of claims 1 to 3 when executing the computer program.

Images