Anti-double-stranded DNA antibody indirect immunofluorescence method picture interpretation method, equipment and medium

Abstract

The invention provides an anti-double-stranded DNA antibody indirect immunofluorescence method picture interpretation method and device and a medium, and the method comprises the following steps: obtaining a to-be-collected slide which is obtained by treating a to-be-detected serum sample through an anti-double-stranded DNA indirect immunofluorescence method; performing image acquisition on the to-be-acquired slide to obtain an acquired image set corresponding to the to-be-acquired slide; performing image processing on the acquired image set to obtain a first target image set; performing image processing on the first target image set to obtain a cell region corresponding to each first target image in the first target image set; and performing identification processing on the cell region corresponding to each first target image to obtain a sample report of the to-be-collected slide. According to the method, a series of steps of image acquisition, processing, cell region identification and the like are adopted for processing, so that the traditional manual operation is replaced, the automation of the interpretation process is realized, and the interpretation accuracy and efficiency are improved.

Description

Technical Field

[0001] The present application relates to the field of cell classification technology, and in particular to a method, device and medium for interpreting images using an indirect immunofluorescence method using an anti-double-stranded DNA antibody. Background Art

[0002] A commonly used method for detecting anti-double-stranded DNA (anti-dsDNA) antibodies is indirect immunofluorescence (IIF). This method uses Crithidia luciliae as a substrate. Crithidia luciliae cells contain a structure called the kinetoplast, which exclusively contains double-stranded DNA. When the test serum contains anti-dsDNA antibodies, these antibodies bind to the double-stranded DNA on the kinetoplast. Detection with a fluorescently labeled secondary antibody then causes the kinetoplast region to fluoresce. The presence of anti-dsDNA antibodies in the serum sample is determined by the presence of kinetoplast fluorescence and the pattern of fluorescence (e.g., homogeneous).

[0003] In the existing technology, the judgment of kinetochores being positive or negative mainly relies on the naked eye observation and subjective experience of the inspectors. Since the interpretation process relies on subjective judgment, different inspectors may have different understandings of the intensity, morphology, and whether the fluorescence is positive, resulting in inconsistent interpretation results. Manual observation and interpretation of cells on the slide one by one takes a lot of time, especially when processing large batches of samples, which is very inefficient. Manual interpretation is easily affected by factors such as fatigue, emotions, and experience level, resulting in low accuracy and repeatability of the interpretation and unstable results. For laboratories that need to test a large number of samples, manual interpretation is a very heavy task.

[0004] Therefore, although indirect immunofluorescence combined with Crithidia luciliae substrate is a classic method for detecting anti-dsDNA antibodies, the subjectivity, low efficiency, low accuracy and low repeatability in its interpretation process limit its application in modern high-throughput and high-precision detection requirements. Summary of the Invention

[0005] In view of the above-mentioned deficiencies in the prior art, the present invention aims to provide a method for interpreting images of anti-double-stranded DNA antibodies by indirect immunofluorescence assay, so as to improve the accuracy and efficiency of anti-double-stranded DNA antibody detection and provide a more reliable basis for clinical diagnosis.

[0006] According to the embodiment of the present application, the first scheme is provided: obtaining a to-be-collected slide, the to-be-collected slide being obtained by processing a serum sample to be detected by an anti-double-stranded DNA indirect immunofluorescence method; performing image collection on the to-be-collected slide to obtain a collection image set corresponding to the to-be-collected slide; performing image processing on the collection image set to obtain a first target image set; performing image processing on the first target image set to obtain a cell region corresponding to each first target image in the first target image set; performing identification processing on the cell region corresponding to each first target image to obtain a sample report of the to-be-collected slide.

[0007] Further, as a more preferred embodiment of the present application, the identification processing on the cell region corresponding to each first target image to obtain a sample report of the to-be-collected slide comprises: performing image recognition on the cell region corresponding to each first target image to obtain a plurality of target cell regions meeting a preset first requirement; performing kinetochore recognition on the plurality of target cell regions to obtain a first kinetochore region in each target cell region; performing morphological feature recognition on the first kinetochore region in each target cell region to determine at least one second kinetochore region; performing color recognition on the at least one second kinetochore region to determine at least one target kinetochore region; and obtaining a sample report of the to-be-collected slide according to the at least one target kinetochore region.

[0008] Further, as a more preferred embodiment of the present application, the obtaining a sample report of the to-be-collected slide according to the at least one target kinetochore region comprises: obtaining a number of regions of the at least one target kinetochore region to obtain a first quantity; obtaining a total number of cell regions in the first target image set to obtain a second quantity; obtaining a target proportion according to the first quantity and the second quantity; judging whether the target proportion exceeds a preset proportion; if yes, generating a first sample report of the to-be-collected slide; or if no, generating a second sample report of the to-be-collected slide.

[0009] Further, as a more preferred embodiment of the present application, the morphological feature recognition on the first kinetochore region in each target cell region to determine at least one second kinetochore region comprises: morphological recognition on the first kinetochore region in each target cell region to determine the first kinetochore region meeting a preset second requirement as a to-be-recognized kinetochore region; feature processing on the target cell region corresponding to the to-be-recognized kinetochore region to obtain a plurality of first feature points; feature processing on the to-be-recognized kinetochore region to obtain a second feature point; and determination of the to-be-recognized kinetochore region meeting a preset third requirement as a second kinetochore region according to the plurality of first feature points and the second feature point, to obtain at least one second kinetochore region.

[0010] Further, as a more preferred embodiment of the present application, the color recognition on the at least one second kinetochore region to determine at least one target kinetochore region comprises: color recognition on the at least one second kinetochore region to obtain a first color of each second kinetochore region; screening out a first color meeting a preset color as a target color in the first color of each second kinetochore region to obtain at least one target color; and determination of the second kinetochore region corresponding to the at least one target color as a target kinetochore region, to obtain at least one target kinetochore region.

[0011] Further, as a more preferred embodiment of the present application, the image processing on the target image set to obtain a first target image set comprises: standardization processing on the target image set to obtain a second target image set; and image preprocessing on the second target image set to obtain the first target image set.

[0012] Further, as a more preferred embodiment of the present application, the image preprocessing on the second target image set to obtain a first target image set comprises: gray scale conversion processing on the second target image set to obtain a first preprocessed image set; denoising processing on the first preprocessed image set to obtain a second preprocessed image set; and contrast processing on the second preprocessed image set to obtain the first target image set.

[0013] Further, as a more preferred embodiment of the present application, the image processing on the first target image set to obtain a cell region corresponding to each first target image in the first target image set comprises: image recognition processing on each first target image in the first target image set to obtain a to-be-divided cell region of the each first target image; and image segmentation processing on the to-be-divided cell region of the each first target image to obtain a cell region corresponding to each first target image in the first target image set.

[0014] According to the embodiment of the present application, the second solution provided by the present application is an anti-double-stranded DNA antibody indirect immunofluorescence method picture interpretation device, characterized in that the device comprises an acquisition unit configured to acquire a to-be-acquired slide, the to-be-acquired slide being obtained by processing a serum sample to be detected by an anti-double-stranded DNA antibody indirect immunofluorescence method; an acquisition unit configured to acquire an image set corresponding to the to-be-acquired slide by performing image acquisition on the to-be-acquired slide; a processing unit configured to obtain a first target image set by performing image processing on the image set; the processing unit is further configured to obtain a cell region corresponding to each first target image in the first target image set by performing image processing on the first target image set; and the processing unit is further configured to obtain a sample report of the to-be-acquired slide by performing identification processing on the cell region corresponding to each first target image.

[0015] According to the embodiment of the present application, the third solution provided by the present application is a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the computer program is executed by the processor, and the program comprises instructions for the first solution.

[0016] According to the embodiment of the present application, the fourth solution provided by the present application is a computer storage medium storing one or more instructions, wherein the one or more instructions are adapted to be loaded and executed by a processor to perform the steps of the above-mentioned first solution and any possible implementation manner thereof.

[0017] The present application ensures the consistency and reliability of image quality by performing standardized image acquisition and processing on the to-be-acquired slide, further reduces human error, and improves the accuracy of detection; and through systematic image acquisition and processing steps, a large number of sample images can be quickly and efficiently acquired and processed, thereby improving the overall detection efficiency; after processing the images, the cell regions in each image can be accurately identified, thereby reducing the possibility of missed judgment and misjudgment, and the automatic identification reduces human intervention and reduces errors caused by subjective judgment; further, by performing identification processing on the cell regions of the first target image set, a sample report is automatically generated, thereby reducing the interpretation differences between different operators and improving the repeatability and reliability of the results. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments or background of the present application, the drawings needed to be used in the embodiments or background of the present application will be described below.

[0019] Figure 1 A flowchart of an anti-double-stranded DNA antibody indirect immunofluorescence method picture interpretation method provided by the embodiments of the present application; Figure 2 A target cell region structure schematic diagram provided for an embodiment of the present application; Figure 3 A structure schematic diagram of an anti-double-stranded DNA antibody indirect immunofluorescence method picture interpretation device provided for an embodiment of the present application; Figure 4 An internal structure diagram of a computer device in an embodiment. DETAILED DESCRIPTION

[0020] In order for those skilled in the art to better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0021] It should be noted that when an element is referred to as being "fixed to" or "disposed on" another component, it can be directly on the other component or indirectly disposed on the other component; when a component is referred to as being "connected to" another component, it can be directly connected to the other component or indirectly connected to the other component.

[0022] It should be understood that the terms "length", "width", "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the purpose of facilitating the description of the present application and simplifying the description, and do not indicate or imply that the device or component referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application.

[0023] In addition, the terms "first", "second" are only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the technical features indicated. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more of the features. In the description of the present application, the meaning of "multiple", "several" is two or more, unless otherwise explicitly and specifically limited.

[0024] It is to be understood that the structures, proportions, sizes, etc. shown in the drawings accompanying the present specification are merely intended to assist in understanding and reading the present specification and are not intended to limit the conditions under which the present application can be implemented. Therefore, any modification of the structure, change of the proportion relationship or adjustment of the size, which does not affect the effects and purposes that can be achieved by the present application, should still fall within the scope of the technical content disclosed by the present application.

[0025] The embodiments of the present application will be described below with reference to the accompanying drawings of the embodiments of the present application.

[0026] Please refer to Figure 1 , Figure 1 is a flowchart of a picture interpretation method of an indirect immunofluorescence method of an anti-double-stranded DNA antibody provided by the embodiments of the present application. The method can include: 101, obtaining a to-be-collected slide, the to-be-collected slide being obtained by processing a serum sample to be detected by an indirect immunofluorescence method of an anti-double-stranded DNA.

[0027] Among them, the to-be-collected slide is obtained by processing a serum sample to be detected by an indirect immunofluorescence method of an anti-double-stranded DNA, rather than an original serum. That is, obtaining a serum sample to be detected, processing by an indirect immunofluorescence method of an anti-double-stranded DNA, and obtaining a to-be-collected slide.

[0028] Specifically, the to-be-collected slide is a sample that has been processed by a series of specific experimental operations (i.e., an indirect immunofluorescence method of an anti-double-stranded DNA, IIF) and has formed a specific image.

[0029] Among them, the indirect immunofluorescence method is used to detect specific antibodies or antigens in the sample. The secondary antibody labeled by the fluorescent dye is combined to the specific antibodies in the sample, and the fluorescent signal is observed under the fluorescence microscope.

[0030] As can be seen, in the present embodiment, the to-be-collected slide meeting the interpretation method can be obtained by processing the original serum, and the accuracy of the present method is improved.

[0031] 102, image collection is performed on the to-be-collected slide to obtain a collection image set corresponding to the to-be-collected slide.

[0032] Among them, image collection can use a special imaging device (fluorescence microscope and its supporting digital camera / camera) to capture the fluorescent signal on the sample and convert it into a digital format image file.

[0033] In the image acquisition process, the magnification, exposure time, gain, and other parameters of the imaging device need to be controlled. For example, if the exposure time of the microscope is set to 500 ms and the gain is 1.5 times, the acquired image will have sufficient brightness and contrast, and the image distortion caused by overexposure will be avoided.

[0034] In the image acquisition process, the stability of the ambient light in the acquisition environment is maintained to avoid interference from external light. This helps to ensure the consistency and reliability of image acquisition.

[0035] For each sample, multiple images (e.g., 5-10) are acquired under different fields of view to ensure comprehensive coverage of the dynamic matrix information in the sample. This multi-view acquisition can improve the accuracy of detection and avoid missing important information.

[0036] The acquisition image set corresponding to the to-be-acquired slide contains multiple images (e.g., 5-10) from the same sample under different fields of view. These images collectively constitute a complete fluorescence performance record of the sample under the current IIF experimental conditions.

[0037] For example, suppose the laboratory is conducting an anti-double-stranded DNA antibody detection. The technicians need to acquire images of the processed to-be-acquired slide. First, they place the prepared slide sample on the stage of the fluorescence microscope. Then, the technicians adjust the magnification of the microscope to 400x to observe the details of the sample. They then set the exposure time to 500 ms and the gain to 1.5 times to ensure the quality of the images. When acquiring images, they close the curtains in the laboratory to maintain the stability of the ambient light and avoid interference from external light. To fully record the sample information, they acquire 10 images under different fields of view to ensure that the dynamic matrix information in the sample is fully captured.

[0038] As can be seen, the image acquisition process of the to-be-acquired slide in this embodiment can obtain an image set for the judgment method.

[0039] 103. Perform image processing on the acquisition image set to obtain a first target image set.

[0040] The image processing can include, but is not limited to, standardization processing and image preprocessing.

[0041] In an optional implementation, the image processing on the acquisition image set to obtain a first target image set includes: performing standardization processing on the target acquisition image set to obtain a second target image set; and performing image preprocessing on the second target image set to obtain the first target image set.

[0042] The standardization processing refers to converting the images into a unified and standardized form for subsequent processing. In this embodiment, the standardization can refer to cropping the useless background and unifying the image size. For example, the black background area at the edge of the image is cropped, and the image only retains the effective area containing sample information. Then, the resolution of the image is uniformly adjusted to 1024x1024 pixels, ensuring that all images have the same size and pixel density.

[0043] Specifically, in the process of cropping the useless background, the edge of the image (for example, the area with a certain pixel width from the edge of the image) is scanned, and the average pixel value of the area is detected. If the average pixel value is lower than a certain preset threshold (for example, lower than the gray value of 10), it is considered that the area is a black background. Then, the algorithm calculates the smallest rectangular area containing all non-black background pixels, and only retains the image content in the rectangular area, and discards the black background outside the rectangular area.

[0044] Specifically, the unification of the image size is realized by scaling operation. If the original image size is smaller than 1024x1024 (for example, 800x800 after cropping), it needs to be enlarged; if the original image size is larger than 1024x1024 (for example, 1200x1200 after cropping), it needs to be reduced. The scaling algorithm (such as bilinear interpolation, bicubic interpolation, etc.) calculates the value of each pixel in the new image according to the value of the original pixel, so as to maintain the clarity and details of the image as much as possible.

[0045] The image preprocessing can be gray scale conversion processing, noise removal processing, and contrast processing.

[0046] Specifically, the gray scale conversion is the process of converting a color image into a gray scale image. The gray scale image only contains brightness information and no color information, making the image processing simpler and more efficient; the noise removal is to reduce or eliminate the noise in the image through a specific algorithm to improve the image quality, and the commonly used methods include median filtering, mean filtering, etc.; the contrast enhancement is to adjust the brightness and contrast of the image, so that the difference between different regions in the image is more obvious, thereby improving the distinguishability of the image.

[0047] As can be seen, in this embodiment, the same specification standardization processing converts each image in the original image set into a second target image set containing only effective sample information and having a unified size; and through the image preprocessing, the images in the second target image set are converted into a gray scale first target image set which is single channel, has less noise, and has higher contrast between the target (moving base) and the background, thereby laying a good foundation for more accurately identifying the cell region and the moving base in the subsequent process.

[0048] In an alternative embodiment, the image preprocessing of the second target image set to obtain the first target image set comprises: performing a gray scale conversion processing on the second target image set to obtain a first preprocessed image set; performing a denoising processing on the first preprocessed image set to obtain a second preprocessed image set; and performing a contrast processing on the second preprocessed image set to obtain the first target image set.

[0049] In the implementation of the gray scale conversion processing, the RGB three-channel values of each pixel are combined to calculate a single gray scale value according to a preset conversion formula (for example, a common simple weighted average method: Gray = 0.299R + 0.587G + 0.114*B). For a fluorescent image, since it may mainly manifest as green fluorescence, the weight of the green channel may be relatively more important during conversion, but the standard conversion formula is usually sufficient. The calculated gray scale value is assigned to the corresponding pixel of a new image to generate an 8-bit gray scale image in a single channel. The converted gray scale image is stored to constitute the first preprocessed image set.

[0050] For example, assuming that there is an image showing a green bottle fly short film in the second target image set, the kinetoplast region presents bright green fluorescence. After the gray scale conversion, the image becomes a black and white image, the kinetoplast region presents a relatively bright white or light gray color due to strong fluorescence, and the surrounding background without fluorescence presents a relatively dark black or dark gray color.

[0051] In the implementation of the denoising processing, the image files in the first preprocessed image set are read, the filter kernel size is set to 3x3 pixels, for each pixel in the image (except for edge pixels, which have a specific method such as being ignored or filled), the system defines a 3x3 window centered on the pixel, collects the gray scale values of the 9 pixels in the window, arranges them in order from small to large (or from large to small), takes the 5th value (i.e., the median), replaces the original gray scale value of the center pixel of the window with the just calculated median, and repeats the above process until all pixels in the image are processed. The denoised image is stored to constitute the second preprocessed image set.

[0052] For example, assuming that there is a gray scale image in the first preprocessed image set, the kinetoplast region should present a uniform bright region, but it is mixed with some random black spots (salt and pepper noise). After applying 3x3 median filtering, these black spots are replaced by the median values (usually brighter values) of their surrounding pixels, thereby disappearing in the image, and the kinetoplast region becomes more uniform and clear.

[0053] During the contrast processing process, the image files in the second preprocessed image set are read. The number of pixels at each grayscale level (0-255) in the image is counted. The cumulative distribution function (CDF) for each grayscale level is calculated, which is the ratio of the number of pixels from 0 to that grayscale level to the total number of pixels. After normalization, the CDF typically ranges from 0 to 1. Based on the CDF, a mapping relationship is designed from the original grayscale level to the new grayscale level. Typically, the new grayscale level = (CDF * (L-1)), where L is the total number of grayscale levels (here, 256), and then rounded to the nearest integer. This mapping table ensures that densely populated grayscale areas in the original image are expanded to the wider new grayscale range. Each pixel in the image is iterated over, and the corresponding new grayscale value is found based on its original grayscale value and the mapping table. The new grayscale values ​​of all pixels are combined to generate a contrast-enhanced image, which is then stored to form the first target image set.

[0054] For example, suppose the second preprocessed image set contains a denoised grayscale image. The moving body region exhibits a certain brightness, but the contrast with the background is not ideal, and details are unclear. After applying histogram equalization, if the original image's pixels are primarily concentrated in the mid-grayscale range (for example, 100-150), the equalization algorithm will expand this concentrated range to a wider grayscale range (perhaps 50-200), while also enhancing detail in both dark areas (the background) and bright areas (the brightest part of the moving body). In the final image, the moving body will appear brighter and have clearer boundaries, while the background may become darker or more uniform, significantly improving overall contrast.

[0055] It can be seen that in this embodiment, through grayscale conversion, denoising, and contrast enhancement, the standardized original fluorescence image is gradually converted into a high-quality, single-channel grayscale image (first target image set) that is more suitable for computer automatic analysis, laying a solid foundation for the subsequent accurate identification of cells and kinetochores.

[0056] 104. Perform image processing on the first target image set to obtain a cell region corresponding to each first target image in the first target image set.

[0057] Specifically, performing image processing on the first target image set to obtain the cell area corresponding to each first target image in the first target image set includes: performing image recognition processing on each first target image in the first target image set to obtain the cell area to be segmented of each first target image; performing image segmentation processing on the cell area to be segmented of each first target image to obtain the cell area corresponding to each first target image in the first target image set.

[0058] Image recognition processing refers to the use of computer vision technology to identify specific objects or regions in an image. In this case, image recognition is used to detect and mark cell regions in the image, identifying the cell regions to be segmented.

[0059] Specifically, image recognition processing can use machine learning algorithms (such as convolutional neural networks (CNNs)) or traditional image processing methods (such as edge detection and template matching) to identify cell regions.

[0060] Image segmentation is the process of dividing an image into multiple meaningful parts or regions. The goal of image segmentation is to extract objects of interest (such as cell regions) for subsequent analysis and processing.

[0061] Specifically, image segmentation processing can use threshold segmentation, region growing, watershed algorithm or deep learning method (such as U-Net) for image segmentation.

[0062] The "cell region to be segmented" refers to the image region initially identified through image recognition processing as potentially containing intact Crithidia lucilia cells. This region may include the cell itself or a small portion of the surrounding background. It serves as a preliminary marker for this region, requiring further refinement.

[0063] For example, suppose we are processing a 1024x1024 pixel target image. First, the grayscale histogram of the image is calculated. It is found that the grayscale values ​​of most background pixels are concentrated below 50, while the grayscale values ​​of cells (especially the kinetochorium region) are concentrated above 150. A threshold is set, such as 100, and all connected regions with grayscale values ​​greater than 100 are found. Next, it filters out regions with an area smaller than 50 pixels or larger than 5000 pixels (assuming these sizes do not meet cell expectations), as well as regions with overly elongated or fragmented shapes. Ultimately, eight candidate regions may be identified in the image, their locations represented by their respective center coordinates and sizes (or bounding boxes). These eight regions are the cell regions to be segmented.

[0064] Furthermore, for example, for one of the eight cell regions to be segmented (assuming it is located in the upper left corner of the image and the bounding box is approximately 100x100 pixels), it first focuses on this 100x100 area. It may apply threshold segmentation again in this small area, such as calculating the average grayscale value in the area and setting a threshold relative to the average value (for example, pixels 20 above the average value belong to cells). Alternatively, it may apply Canny edge detection to find the cell edge and then fill the interior to form a cell area. Finally, it may use a small closing operation to fill in the small black spots (small holes) that may exist inside the cell. Finally, a binary mask that matches the exact outline of the cell is output. Repeat this process for the other seven regions.

[0065] It can be seen that in this embodiment, image recognition is responsible for quickly locating areas that may contain cells, greatly reducing the scope of subsequent processing; image segmentation is responsible for accurately separating these areas from the background, providing clear, well-defined input for the next step of feature extraction and interpretation specifically for cells (especially kinetoplasts), ensuring that subsequent analysis is based on real cells, rather than background noise or other interference.

[0066] 105. Perform identification processing on the cell region corresponding to each first target image to obtain a sample report of the slide to be collected.

[0067] The specific implementation process of identifying the cell region corresponding to each first target image can be referred to 201-204, which will not be repeated here.

[0068] The sample report for the slide to be collected can include sample identification information (such as sample number), test result (positive / negative), and possibly other information (such as interpretation parameters, links to sample images, etc.). The sample report can be connected to the laboratory information system (LIS) to achieve automated data upload.

[0069] Recognition processing refers to the process of extracting features and determining patterns within a single cell region. Recognition processing analyzes specific features within the cell to determine whether a fluorescent reaction caused by anti-dsDNA antibodies exists, further determining whether the sampled slide is positive or negative, and generating a sample report.

[0070] It can be seen that in this embodiment, by identifying and processing each cell region, an objective, rapid, and standardized judgment of the anti-dsDNA antibody test results is achieved, and a diagnostic report for clinical use is ultimately output, which solves the subjectivity and inefficiency of traditional manual interpretation.

[0071] The application ensures consistency and reliability of image quality, further reduces human error, and improves detection accuracy through standardized image acquisition and processing of the to-be-collected slide. Through systematic image acquisition and processing steps, a large number of sample images can be quickly and efficiently acquired and processed, and the overall detection efficiency is improved. After processing the image, the cell region in each image can be accurately identified, reducing the possibility of missed and misjudged, and the automatic identification reduces the human intervention and reduces the error caused by subjective judgment. Further, by identifying and processing the cell region of the first target image set, a sample report is automatically generated, reducing the interpretation difference between different operators, improving the repeatability and reliability of the results.

[0072] 201、In an optional implementation, the identifying and processing of the cell region corresponding to each first target image to obtain the sample report of the to-be-collected slide comprises: performing image recognition on the cell region corresponding to each first target image to obtain a plurality of target cell regions meeting a preset first requirement; performing kinetochore recognition on the plurality of target cell regions to obtain a first kinetochore region in each target cell region; performing morphological feature recognition on the first kinetochore region in each target cell region to determine at least one second kinetochore region; performing color recognition on the at least one second kinetochore region to determine at least one target kinetochore region; and obtaining the sample report of the to-be-collected slide according to the at least one target kinetochore region.

[0073] In the process, image recognition refers to preliminary quality evaluation and screening of a single cell region to determine whether the cell is suitable for subsequent kinetochore analysis.

[0074] The preset first requirement is a standard for screening cell regions, which is set in advance. The preset first requirement can be that the cell morphology is complete and not broken.

[0075] The target cell region refers to a cell region that is determined to meet the preset requirement and can be subjected to subsequent kinetochore analysis after the above image recognition and screening.

[0076] Specifically, in the cell region corresponding to each first target image, the area, perimeter, circularity (or compactness), and whether there is a breakage of the cell are calculated. For example, the ratio of the area to the perimeter of the cell is calculated, or a morphological operation (such as erosion) is used to detect whether the edge is continuous. If the area of the cell is too small (may be noise or debris), the morphology is extremely irregular (may be broken), or the edge has obvious breakage, the cell region will be marked as not meeting the requirements. Only the cell regions with complete morphology and appropriate size are retained to obtain the target cell regions. For example, in an image containing 50 segmented cells, the algorithm may determine that 45 of the cells have complete morphology, and thus 45 target cell regions are obtained, and the other 5 are excluded from subsequent analysis.

[0077] The motile base body recognition refers to locating and preliminarily circumscribing the position and approximate region of the motile base body in a single cell region.

[0078] The first motile base body region refers to marking the motile base body region in each target cell region.

[0079] Specifically, the inside of each target cell region is scanned to find a region with brightness (gray value) significantly higher than other regions of the cell (especially the background and cytoplasm). The motile base body is usually the brightest region because it binds fluorescent antibodies. Threshold segmentation, connected region analysis, and other methods are used to preliminarily identify the highest brightness connected region as the first motile base body region. For example, a gray threshold is set, and pixels above the threshold are considered part of the motile base body, and then these pixels are clustered into one or more regions.

[0080] The specific implementation description of obtaining the sample report of the slide to be collected according to the at least one target motile base body region can be referred to 202, which is not repeated here.

[0081] The specific implementation description of performing morphological feature recognition on the first motile base body region in each target cell region to determine at least one second motile base body region can be referred to 203, which is not repeated here.

[0082] The specific implementation description of performing color recognition on the at least one second motile base body region to determine at least one target motile base body region can be referred to 204, which is not repeated here.

[0083] As can be seen, in this embodiment, through the precise image processing and recognition steps, after obtaining a single cell region, fine recognition and analysis steps are performed to accurately locate and evaluate the motile base body in the cell from the original image to determine the presence or absence of anti-dsDNA antibodies and generate a final sample report, thereby improving the accuracy and interpretation efficiency of the method.

[0084] 202、In an optional embodiment, the sample report of the to-be-collected slide is obtained according to the at least one target dynamic matrix region, comprising: obtaining the number of the at least one target dynamic matrix region to obtain a first number; obtaining the total number of cell regions in the first target image set to obtain a second number; obtaining a target proportion according to the first number and the second number; judging whether the target proportion exceeds a preset proportion; if it exceeds, generating the sample report of the to-be-collected slide as a first sample report; or, if it does not exceed, generating the sample report of the to-be-collected slide as a second sample report.

[0085] Wherein, the first number refers to the total number of target dynamic matrix regions identified.

[0086] Wherein, the total number of cell regions refers to the total number of cell regions confirmed to be needed for subsequent analysis in the first target image set through image recognition and screening (meeting the preset first requirement, such as complete morphology). This is usually the total number of cells before dynamic matrix recognition, so the second number refers to the total number of cell regions obtained by statistics.

[0087] Wherein, the target proportion refers to the proportion of the number of target dynamic matrix regions (the first number) to the total number of cell regions (the second number). The target proportion is a key indicator for judging sample positivity.

[0088] Therefore, the target proportion = the first number / the second number.

[0089] Wherein, the preset proportion is a pre-set threshold value, usually determined based on clinical experience and laboratory standards. It represents the cell positivity rate threshold for determining sample positivity. For example, a common standard may be 20% or 30%. If the target proportion exceeds this threshold, it is considered that the anti-dsDNA antibody level in the sample is high enough to be determined as positive.

[0090] Wherein, the first sample report refers to a report containing positive detection result information. Specifically, if the target proportion > the preset proportion (for example, 25% is calculated, and 20% is preset), the sample is determined as positive. It will generate a first sample report and clearly mark "the detection result is positive" in the report. The report may also contain detailed information such as the number of positive cells, the total number of cells, and the target proportion.

[0091] Wherein, the second sample report refers to a report containing negative detection result information. Specifically, if the target proportion ≤ the preset proportion (for example, 11.11% is calculated, and 20% is preset), the sample is determined as negative. It will generate a second sample report and clearly mark "the detection result is negative" in the report. The report may also contain relevant statistical information.

[0092] It can be seen that in the embodiment, through quantitative analysis (statistical quantity, calculation ratio) and comparison of the preset threshold, complex image information is converted into simple and clear clinical diagnosis results (positive / negative), and is output in a standardized report form, which not only improves the consistency and accuracy of interpretation, but also greatly improves the detection efficiency.

[0093] 203、In an alternative embodiment, the morphological feature recognition of the first kinetoplast region in each target cell region determines at least one second kinetoplast region, comprising: morphological recognition of the first kinetoplast region in each target cell region, determining the first kinetoplast region meeting the preset second requirement as the to-be-recognized kinetoplast region; performing feature processing on the target cell region corresponding to the to-be-recognized kinetoplast region to obtain a plurality of first feature points; performing feature processing on the to-be-recognized kinetoplast region to obtain a second feature point; and determining the to-be-recognized kinetoplast region meeting the preset third requirement as the second kinetoplast region according to the plurality of first feature points and the second feature point, to obtain at least one second kinetoplast region.

[0094] Wherein, morphological recognition refers to analyzing the shape features of these regions, such as area, perimeter, circularity, aspect ratio, convex hull, etc.

[0095] Wherein, the preset second requirement is a screening standard set in the morphological recognition stage, which is used to preliminarily exclude regions whose shape obviously does not meet the kinetoplast feature. The preset second requirement can be that the kinetoplast is usually elliptical and the outline is relatively regular.

[0096] Wherein, the to-be-recognized kinetoplast region refers to those first kinetoplast regions that pass the preset second requirement screening. They are considered as "potential" kinetoplasts and need to be further analyzed for more detailed features.

[0097] Wherein, feature processing refers to extracting global or local feature points of the cell region. These feature points can be used to locate key structures of the cell, such as the center of the nucleus, the base of the flagellum attachment, etc.

[0098] Wherein, the first feature point refers to the key point extracted from the target cell region, which represents the overall structural feature of the cell. The first feature point can include the center point of the nucleus (N), the base center point of the flagellum and cell connection, etc.

[0099] Specifically, the cell region image is processed to locate the nucleus center (N) and the base of the flagellum. Locate the nucleus center (N): find the region with the highest fluorescence intensity or the largest area in the cell region (usually corresponding to the nucleus), and then calculate the centroid of the region as N(0, 0). Here, the origin is set to simplify subsequent calculations; locate the base of the flagellum center (B): the flagellum usually extends from the vicinity of the nucleus, and the starting point of the flagellum, i.e. the base center B(m, n), can be located by edge detection, line fitting or finding the fluorescent bright spot in a specific region of the cell edge.

[0100] wherein the second feature point refers to a key point extracted from the dynamic base region to be identified, which represents its own position and characteristics. The most core is the center point K(h, k) of the dynamic base region.

[0101] wherein the preset third requirement refers to a set of geometric rules for confirming the authenticity of the dynamic base, based on cell biology knowledge. It ensures that the position of the K point conforms to the typical distribution rule of the dynamic base in the cell.

[0102] wherein the preset third requirement can refer to Figure 2 for explanation, Figure 2 is a structural schematic diagram of the target cell region. Assuming that the nucleus center point is N(0, 0), the dynamic base center point is K(h, k), and the intersection of the flagellum and the base is B(m, n), then the preset third requirement is: Three points are not collinear: ; the dynamic base is located between the nucleus and the base of the flagellum: the coordinates satisfy (arranged along the x-axis); the dynamic base is close to the flagellum: equivalent to ; in terms of geometric meaning, represents the distance between point and point , represents the distance between point and the origin , and the original inequality represents that the distance from point to point is less than the distance from point to the origin.

[0103] If the dynamic base region to be identified satisfies all three preset third requirements, it is confirmed as a second dynamic base region.

[0104] wherein the second dynamic base region refers to a dynamic base region to be identified that meets the requirements in terms of shape and position satisfying the preset third requirements. The second dynamic base region is considered to be a highly reliable positive signal generated by the combination of anti-dsDNA antibody.

[0105] It can be seen that in this embodiment, the most likely to represent the dynamic matrix combined with anti-dsDNA antibodies is accurately screened from the preliminary fluorescence signal through morphological and geometric feature analysis. The preset second requirement ensures the reasonableness of the shape, and the preset third requirement utilizes the relatively fixed spatial position relationship of the dynamic matrix in the cell, greatly improving the accuracy and specificity of identification, effectively eliminating background noise and false positive signals.

[0106] 204、In an alternative embodiment, the color identification of the at least one second dynamic matrix region to determine at least one target dynamic matrix region comprises: color identification of the at least one second dynamic matrix region to obtain the first color of each second dynamic matrix region; in the first color of each second dynamic matrix region, the first color meeting the preset color is selected as the target color to obtain at least one target color; the second dynamic matrix region corresponding to the at least one target color is determined as the target dynamic matrix region to obtain at least one target dynamic matrix region.

[0107] Wherein, color identification refers to analyzing and identifying color information in an image through image processing technology to determine the color characteristics of a specific region.

[0108] Wherein, the first color refers to the average gray value, the maximum gray value, the minimum gray value presented by each second dynamic matrix region, or a value representing the overall brightness characteristics of the region. This value reflects the fluorescence intensity of the region.

[0109] Wherein, the preset color is a gray value (or intensity) range preset according to the typical performance of anti-dsDNA antibody positive signal in indirect immunofluorescence experiment. Positive signal usually shows strong fluorescence, corresponding to higher gray value in gray image. For example, the preset color may be defined as the region with gray value greater than 200 is considered as strong fluorescence region.

[0110] Wherein, the target color refers to those first colors that are screened out and meet the preset color standard. They represent regions with high enough brightness, which may be the real positive signal.

[0111] Wherein, the target dynamic matrix region refers to the second dynamic matrix region whose first color is screened as the target color. The target dynamic matrix region not only passes the morphological and geometric position screening, but also reaches the standard of positive signal in fluorescence intensity (color / brightness).

[0112] It can be seen that in the embodiment, by setting a brightness threshold (preset color), the positive dynamic matrix actually emitting strong fluorescence is distinguished from the negative or weak positive dynamic matrix which has correct position and shape but insufficient fluorescence intensity, the accuracy of interpretation is improved, and only the dynamic matrix meeting all the standards (shape, position and intensity) is finally determined as a positive signal, thereby providing a reliable basis for subsequent quantity statistics and result determination.

[0113] Based on the description of the picture interpretation method embodiment of the anti-double-stranded DNA antibody indirect immunofluorescence method, the embodiment of the application further discloses an anti-double-stranded DNA antibody indirect immunofluorescence method picture interpretation device, as shown in the accompanying drawings. Figure 3 The anti-double-stranded DNA antibody indirect immunofluorescence method picture interpretation device 300 comprises: An acquisition unit 301 is configured to acquire a to-be-acquired slide, wherein the to-be-acquired slide is obtained by processing a serum sample to be detected by an anti-double-stranded DNA indirect immunofluorescence method; An acquisition unit 302 is configured to acquire an image of the to-be-acquired slide to obtain a set of acquisition images corresponding to the to-be-acquired slide; A processing unit is configured to perform image processing on the set of acquisition images to obtain a first target image set; The processing unit is further configured to perform image processing on the first target image set to obtain a cell region corresponding to each first target image in the first target image set; The processing unit is further configured to perform identification processing on the cell region corresponding to each first target image to obtain a sample report of the to-be-acquired slide.

[0114] The application ensures the consistency and reliability of image quality by performing standardized image acquisition and processing on the to-be-acquired slide, further reduces human error, and improves the accuracy of detection. Through systematic image acquisition and processing steps, a large number of sample images can be quickly and efficiently acquired and processed, and the overall detection efficiency is improved. After processing the images, the cell regions in each image can be accurately identified, the possibility of missed judgment and misjudgment is reduced, and the automatic identification reduces human intervention and reduces the error caused by subjective judgment. Further, by performing identification processing on the cell regions of the first target image set, a sample report is automatically generated, the interpretation difference between different operators is reduced, and the repeatability and reliability of the results are improved.

[0115] The embodiment of the present application further provides a computer storage medium (Memory). The computer storage medium is a memory device in an electronic device, and is used for storing programs and data. It can be understood that the computer storage medium can include a built-in storage medium in the electronic device, and of course can include an expansion storage medium supported by the electronic device. The computer storage medium provides a storage space, and the storage space stores an operating system of the electronic device. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and the instructions can be one or more computer programs (including program codes). It should be noted that the computer storage medium can be a high-speed RAM memory, or a non-volatile memory such as at least one disk memory; optionally, the computer storage medium can be at least one computer storage medium located away from the processor.

[0116] In one embodiment, one or more instructions stored in the computer storage medium can be loaded and executed by the processor to implement the corresponding steps in the above-described embodiments; in a specific implementation, one or more instructions in the computer storage medium can be loaded and executed by the processor to implement any step of the method in the above-described embodiments, and details are not described herein again. Figure 1 and / or Figure 2

[0117] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described apparatus and modules can refer to the corresponding process in the foregoing method embodiments, and details are not described herein again.

[0118] In several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method can be implemented in other ways. For example, the division of the module is only a logical function division, and in actual implementation, another division mode can be used, for example, a plurality of modules or components can be combined or integrated into another system, or some features can be ignored or not executed. The coupling or direct coupling or communication connection between the displayed or discussed modules can be indirect coupling or communication connection between the modules through some interfaces, devices or modules, and can be electrical, mechanical or other forms.

[0119] The modules described as separate components can or can not be physically separated, and the components displayed as modules can or can not be physical modules, that is, can be located in one place, or can be distributed on a plurality of network modules. According to actual needs, part or all of the modules can be selected to achieve the purpose of the embodiment scheme.

[0120] Figure 4 An internal structure diagram of a computer device in an embodiment is shown. The computer device can be a terminal. As shown in FIG. 1, the computer device includes an application processor (Application Processor, AP), a display unit, a storage unit, a communication unit, a sensor unit and a power supply unit.​Figure 4 As shown in the figure, the computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the computer device has a storage operating system, and can also have a computer program which is executed by the processor to enable the processor to implement the above-mentioned anti-double-stranded DNA antibody indirect immunofluorescence method picture reading method. The internal memory can also store a computer program which is executed by the processor to enable the processor to execute the above-mentioned anti-double-stranded DNA antibody indirect immunofluorescence method picture reading method. Those skilled in the art can understand that, Figure 4 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the equipment to which the scheme of the present application is applied. The specific equipment can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0121] A computer readable storage medium stores a computer program, which is executed by a processor to enable the processor to execute the steps of the anti-double-stranded DNA antibody indirect immunofluorescence method picture reading method in any of the above embodiments.

[0122] A computer device includes a memory and a processor, the memory stores a computer program, and the computer program is executed by the processor to enable the processor to execute the steps of the anti-double-stranded DNA antibody indirect immunofluorescence method picture reading method in any of the above embodiments.

[0123] In the above embodiments, all or part of the embodiments can be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions according to the embodiments of the present application are generated. The computer can be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions can be stored in or transmitted from a computer readable storage medium. The computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center through a wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) manner. The computer readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. integrated with one or more available media sets. The available medium can be a read-only memory (ROM), or a random access memory (RAM), or a magnetic medium, such as a floppy disk, a hard disk, a magnetic tape, a magnetic disk, or an optical medium, such as a digital versatile disc (DVD), or a semiconductor medium, such as a solid state disk (SSD), etc.

[0124] The above description of disclosed embodiments enables one skilled in the art to make or use the application. Various modifications to these embodiments will be obvious to those skilled in the art, and the generic principles defined herein can be applied to other embodiments without departing from the spirit or scope of the application. Therefore, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for interpreting images of an anti-double-stranded DNA antibody indirect immunofluorescence assay, characterized in that: The method comprises: Obtaining a glass slide to be collected, wherein the glass slide to be collected is obtained by treating the serum sample to be tested with an anti-double-stranded DNA indirect immunofluorescence assay; Performing image acquisition on the glass slide to be acquired to obtain an acquired image set corresponding to the glass slide to be acquired; performing image processing on the acquired image set to obtain a first target image set; performing image processing on the first target image set to obtain a cell region corresponding to each first target image in the first target image set; The cell region corresponding to each first target image is identified and processed to obtain a sample report of the slide to be collected.

2. The method according to claim 1, characterized in that The identifying and processing the cell region corresponding to each first target image to obtain a sample report of the slide to be collected includes: Performing image recognition on the cell region corresponding to each of the first target images to obtain a plurality of target cell regions that meet a preset first requirement; performing kinetochoroid identification on the plurality of target cell regions to obtain a first kinetochoroid region in each target cell region; performing morphological feature recognition on the first kinetoclase region in each target cell region to determine at least one second kinetoclase region; Performing color recognition on the at least one second dynamic matrix region to determine at least one target dynamic matrix region; A sample report of the slide to be collected is obtained according to the at least one target dynamic matrix area.

3. The method according to claim 2, characterized in that Obtaining a sample report of the slide to be collected based on the at least one target dynamic matrix area includes: Obtaining the number of regions of the at least one target moving substrate region to obtain a first quantity; Obtaining the total number of cell regions in the first target image set to obtain a second number; Obtaining a target proportion based on the first quantity and the second quantity; Determining whether the target proportion exceeds a preset proportion; If it exceeds, then the sample report of the slide to be collected is generated as the first sample report; or, If not, a sample report of the slide to be collected is generated as a second sample report.

4. The method according to claim 2, characterized in that The performing morphological feature recognition on the first kinetoplast region in each target cell region to determine at least one second kinetoplast region comprises: performing morphological recognition on the first kinetochoroid region in each target cell region, and determining the first kinetochoroid region that meets the preset second requirement as the kinetochoroid region to be identified; Performing feature processing on the target cell region corresponding to the kinetochoroid region to be identified to obtain a plurality of first feature points; Performing feature processing on the moving base region to be identified to obtain a second feature point; According to the plurality of first feature points and the second feature points, a moving base region to be identified that meets a preset third requirement is determined as a second moving base region, and at least one second moving base region is obtained.

5. The method according to claim 2, characterized in that The performing color recognition on the at least one second moving base region to determine at least one target moving base region includes: Performing color recognition on the at least one second moving base region to obtain a first color of each second moving base region; Screening out the first colors that meet the preset color from the first colors of each second moving base region as a target color, and obtaining at least one target color; The second dynamic matrix region corresponding to the at least one target color is determined as a target dynamic matrix region, thereby obtaining at least one target dynamic matrix region.

6. The method according to claim 1, characterized in that The performing image processing on the acquired image set to obtain a first target image set includes: performing standardization processing on the acquired image set to obtain a second target image set; Perform image preprocessing on the second target image set to obtain a first target image set.

7. The method according to claim 6, characterized in that The performing image preprocessing on the second target image set to obtain the first target image set includes: performing grayscale conversion processing on the second target image set to obtain a first preprocessed image set; performing denoising processing on the first preprocessed image set to obtain a second preprocessed image set; Perform contrast processing on the second preprocessed image set to obtain a first target image set.

8. The method according to claim 1, characterized in that The performing image processing on the first target image set to obtain a cell region corresponding to each first target image in the first target image set includes: performing image recognition processing on each first target image in the first target image set to obtain a cell region to be segmented in each first target image; Perform image segmentation processing on the cell region to be segmented in each of the first target images to obtain the cell region corresponding to each first target image in the first target image set.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the anti-double-stranded DNA antibody indirect immunofluorescence image interpretation method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that A computer program is stored, and when the computer program is executed by a processor, the processor is caused to perform the steps of the anti-double-stranded DNA antibody indirect immunofluorescence image interpretation method according to any one of claims 1 to 7.

Images