Method, device, electronic equipment and program product for analyzing protein co-localization in fluorescence images

Abstract

The application discloses a kind of analysis methods, device, electronic equipment and program product to protein co-localization in fluorescence image.The method comprises: determining target region of interest from the fluorescence image of tissue sample, and the target fluorescence image is obtained by cutting target region of interest;According to the preset pixel threshold, separate the protein staining pixel in target fluorescence image, to obtain the sample distribution area of each protein staining pixel;The target fluorescence image is divided into multiple protein sample blocks;For a single protein sample block, the position distribution information of each protein staining pixel in the protein sample block is determined respectively, and the distance and overlap data between various protein staining pixels are determined according to the position distribution information of different protein staining pixels;According to the distance and overlap data between various protein staining pixels in target fluorescence image, determine the protein co-localization analysis result in target fluorescence image.To this end, the comprehensive, automatic, unbiased analysis of protein co-localization signal is realized.

Description

Technical Field

[0001] This invention relates to the fields of life sciences, medical imaging, and other technologies, and in particular to an analytical method, apparatus, electronic device, and program product for colocalizing proteins in fluorescence images. Background Technology

[0002] Fluorescent images are images obtained by utilizing the property of fluorescent substances to emit light under specific excitation light, and are widely used in life sciences, medical imaging, and materials science.

[0003] In related technologies, existing immunofluorescence quantitative methods typically rely on manually selecting two representative co-labeled sites at specific locations for quantitative analysis. This approach suffers from human selection bias and a limited number of sampling points, failing to comprehensively and objectively reflect the spatial distribution and intensity characteristics of co-localization of the two proteins in the fluorescence image, thus resulting in insufficient representativeness and accuracy of the analytical results. Furthermore, existing methods struggle to perform automated, high-throughput differential analysis and comparison of co-localization signals of the two proteins in fluorescence images at a holistic level, especially when using fluorescence images acquired using ordinary optical microscopes, lacking efficient and precise intelligent analysis tools. Summary of the Invention

[0004] This invention provides a method, apparatus, electronic device, and program product for analyzing protein colocalization in fluorescence images, to solve the technical problem that existing technologies struggle to objectively and accurately quantify the colocalization of different proteins when processing fluorescence images.

[0005] According to one aspect of the present invention, a method for analyzing protein colocalization in fluorescence images is provided, the method comprising:

[0006] A target region of interest is determined from the fluorescence image of a tissue sample, and the target region of interest is cropped to obtain a target fluorescence image; wherein, the target fluorescence image includes the target fluorescence image in the control group fluorescence image and the target fluorescence image in the experimental group fluorescence image; the target fluorescence image includes at least two protein staining pixels corresponding to the two protein samples;

[0007] Based on a preset pixel threshold, protein staining pixel images corresponding to different protein samples are separated from the target fluorescence image, and the protein sample distribution region corresponding to each protein staining pixel image is determined; wherein, the preset pixel threshold includes the pixel threshold corresponding to each protein staining pixel;

[0008] Based on the horizontal coordinate of the target fluorescence image, the target fluorescence image is divided into multiple protein sample blocks according to a preset pixel spacing;

[0009] For a single protein sample block, the positional distribution information of each type of protein staining pixel within the protein sample block is determined, and the distance and overlap data between various types of protein staining pixels are determined based on the positional distribution information of different protein staining pixels;

[0010] Based on the distance and overlap data between the various protein-stained pixels in the target fluorescence image, the protein colocalization analysis results in the target fluorescence image are determined.

[0011] According to another aspect of the present invention, an analytical apparatus for co-localization of proteins in fluorescence images is provided, the apparatus comprising:

[0012] The target image acquisition module is used to determine the target region of interest from the fluorescence image of the tissue sample, and crop the target region of interest to obtain the target fluorescence image; wherein, the target fluorescence image includes the target fluorescence image in the control group fluorescence image and the target fluorescence image in the experimental group fluorescence image; the target fluorescence image includes at least two protein staining pixels corresponding to the two protein samples;

[0013] The protein sample separation module is used to separate protein staining pixel images corresponding to different protein samples from the target fluorescence image according to a preset pixel threshold, and to determine the protein sample distribution region corresponding to each protein staining pixel image; wherein, the preset pixel threshold includes the pixel threshold corresponding to each protein staining pixel;

[0014] The protein sample block separation module is used to divide the target fluorescence image into multiple protein sample blocks based on the horizontal coordinate of the target fluorescence image and according to a preset pixel spacing.

[0015] The protein sample distance determination module is used to determine the position distribution information of each type of protein staining pixel within a single protein sample block, and to determine the distance and overlap data between various types of protein staining pixels based on the position distribution information of different protein staining pixels.

[0016] The colocalization analysis module is used to determine the protein colocalization analysis results in the target fluorescence image based on the distance and overlap data between various protein-stained pixels in the target fluorescence image.

[0017] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0018] One or more processors; and a memory communicatively connected to at least one of the processors; wherein the memory stores a computer program executable by at least one of the processors, which, when executed by one or more of the processors, causes one or more of the processors to implement the method for analyzing protein colocalization in fluorescence images as described in the embodiments of the present invention.

[0019] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the method for analyzing protein colocalization in fluorescence images as described in the embodiments of the present invention.

[0020] According to another aspect of the present invention, embodiments of this disclosure also provide a computer program product, including a computer program that, when executed by a processor, implements the method for analyzing protein colocalization in fluorescence images as described in embodiments of the present invention.

[0021] The technical solution of this invention involves determining a target region of interest from a fluorescence image of a tissue sample and cropping the target region of interest to obtain a target fluorescence image. The target fluorescence image includes target fluorescence images from a control group fluorescence image and target fluorescence images from an experimental group fluorescence image. The target fluorescence image includes at least two protein staining pixels corresponding to different protein samples. Based on a preset pixel threshold, protein staining pixel images corresponding to different protein samples are separated from the target fluorescence image, and a corresponding protein sample distribution region is determined for each protein staining pixel image. The preset pixel threshold includes a pixel threshold corresponding to each protein staining pixel. Using the horizontal coordinate of the target fluorescence image as a reference, the target fluorescence image is divided into multiple protein sample blocks according to a preset pixel spacing. For a single protein sample block, the positional distribution information of each protein staining pixel within the protein sample block is determined, and the distance and overlap data between various protein staining pixels are determined based on the positional distribution information of different protein staining pixels. Based on the distance and overlap data between various protein staining pixels in the target fluorescence image, the protein co-localization analysis results in the target fluorescence image are determined. By adopting the above-mentioned operational steps, the technical problem of difficulty in comprehensively, objectively, accurately, and automatically quantifying colocalization signals of different proteins in existing analytical methods is solved. It also enables comprehensive, automatic, and unbiased integrated analysis of colocalization signals of different types of proteins in fluorescence images, thereby realizing the differential quantitative comparison of colocalization signals of large-scale different protein samples and effectively improving the interpretability of biological tissues.

[0022] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a flowchart of a method for analyzing protein colocalization in fluorescence images according to Embodiment 1 of the present invention;

[0025] Figure 2 This is a flowchart of a method for analyzing protein colocalization in fluorescence images according to Embodiment 2 of the present invention;

[0026] Figure 3 This is a flowchart of a method for analyzing protein colocalization in fluorescence images according to Embodiment 3 of the present invention;

[0027] Figure 4 This is a schematic diagram of the structure of an analytical device for co-localization of proteins in fluorescence images according to Embodiment 4 of the present invention;

[0028] Figure 5 This is a schematic diagram of the structure of an electronic device that implements the method for analyzing protein co-localization in fluorescence images provided in Embodiment 5 of the present invention. Detailed Implementation

[0029] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0030] It should be noted that the terms "target fluorescence image in the control group fluorescence image," "target fluorescence image in the experimental group fluorescence image," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0031] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0032] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0033] It is understood that the data involved in this technical solution (including but not limited to the data itself, the acquisition or use of the data) shall comply with the requirements of relevant laws, regulations and related provisions.

[0034] Example 1

[0035] Figure 1 This invention provides a flowchart of a method for analyzing protein co-localization in fluorescence images, as described in Embodiment 1. This embodiment is applicable to scenarios involving the analysis and detection of fluorescence images in fields such as life sciences, medical imaging, and materials science. The method can be executed by an analysis device for protein co-localization in fluorescence images. This device can be implemented in hardware and / or software, optionally through electronic devices such as PCs or servers. Figure 1 As shown, the method may specifically include:

[0036] S110. Determine the target region of interest from the fluorescence image of the tissue sample, and crop the target region of interest to obtain the target fluorescence image.

[0037] The target region of interest can be understood as the image region to be analyzed subsequently; typically, the stained pixels in this image region are displayed relatively well. The target fluorescence image includes the target fluorescence image in the control group fluorescence image and the target fluorescence image in the experimental group fluorescence image; the target fluorescence image includes at least the protein stained pixels corresponding to the two protein samples.

[0038] Specifically, edge extraction algorithms can be used to detect and identify the fluorescence images of tissue samples, pinpointing the precise boundaries of each region. Then, based on the fluorescence signal intensity within each region, the region with the highest fluorescence signal intensity is identified as the target region of interest (ROI). Finally, the selected ROI is cropped to obtain the target fluorescence image. This technical solution, by employing an automatic detection algorithm to determine the ROI, improves the continuity and naturalness of the cropped ROI, ensuring that the stained pixels within the cropped ROI are clearly visible, facilitating subsequent identification and analysis of the stained pixels.

[0039] S120. Based on a preset pixel threshold, separate the protein staining pixel images corresponding to different protein samples from the target fluorescence image, and determine the protein sample distribution area corresponding to each protein staining pixel image.

[0040] The preset pixel threshold can include the pixel threshold corresponding to each type of protein staining pixel. The preset pixel thresholds corresponding to different protein staining pixels can be the same or different. The protein staining pixel image can be understood as a staining image composed of pixels corresponding to protein samples of the same color. The protein sample distribution area can be understood as the staining image obtained after post-processing optimization.

[0041] Specifically, for any protein staining pixel of a protein sample, the pixel threshold corresponding to that protein staining pixel is iterated against each pixel in the target fluorescence image to obtain a protein staining pixel image for that protein sample. Similarly, the pixel threshold corresponding to each color of protein staining pixel is iterated against each pixel in the target fluorescence image to separate protein staining pixel images corresponding to multiple protein samples from the target fluorescence image. Furthermore, post-processing such as anomaly labeling and deletion, and hole filling are performed on each protein staining pixel image to obtain the sample distribution region of each protein staining pixel image. This technical solution uses a preset pixel threshold to segment different types of protein staining pixels, thereby obtaining completely independent and non-interfering distribution regions of various protein staining pixels.

[0042] In one embodiment, the step of separating protein staining pixel images corresponding to different protein samples from the target fluorescence image according to a preset pixel threshold, and determining the corresponding protein sample distribution region according to each protein staining pixel image, includes: extracting staining pixel channels corresponding to each protein staining pixel from the target fluorescence image; for any staining pixel channel, setting pixels smaller than the pixel threshold in the staining pixel channel as background pixels, and setting pixels greater than or equal to the pixel threshold as candidate marker pixels, so as to generate a protein staining pixel image based on the background pixels and the candidate marker pixels; and using an eight-neighbor connectivity rule to identify each pixel in the protein staining pixel image to obtain the sample distribution region of each protein sample; wherein, the eight-neighbor connectivity rule is any pixel adjacent to each other in the horizontal, vertical, and diagonal directions.

[0043] In this context, the staining pixel channel can be understood as a grayscale image specific to a particular protein sample dye. The pixel threshold can be understood as a pre-set critical grayscale value used to filter candidate marker pixels for different colored dyes. Candidate marker pixels can be understood as potentially valid pixels, indicating that the target protein sample is clustered at that pixel location. The protein staining pixel image can be understood as the final binarized or masked output, clearly showing the distribution area of ​​the protein sample.

[0044] Specifically, firstly, the extraction algorithm in OpenCV can be used to extract the staining pixel channels corresponding to each protein staining pixel from the target fluorescence image, thereby decomposing the multi-channel target fluorescence image into multiple staining pixel channels containing one color. Then, for any staining pixel channel, noise pixels in the staining pixel channel can be removed based on a preset noise pixel threshold. Next, based on the preset pixel threshold, pixels smaller than the noise threshold in the noise-removed staining pixel channel are set as background pixels, and pixels greater than or equal to the pixel threshold are set as candidate marker pixels, thus generating a protein staining pixel image for that protein sample based on the background pixels and candidate marker pixels. Similarly, the protein staining pixel images corresponding to each protein sample can be obtained using the above method. Finally, the eight-neighborhood connectivity rule is used to traverse each pixel in each protein staining pixel image, dividing interconnected pixels into the same connected region, and deleting connected regions that do not meet the minimum area requirement, to extract the remaining connected regions as the sample distribution areas for the protein samples. By employing the above processing, this technical solution can avoid interference from pixels stained with different colors, count the number of pixels contained in each connected region, and delete isolated connected regions with an area smaller than a preset area threshold, thereby reducing scattered noise and interference from non-target fluorescence signals to obtain a clearer and more distinct protein sample distribution area.

[0045] In one embodiment, the step of using the eight-neighbor connectivity rule to identify each pixel in the protein staining pixel image to obtain the sample distribution region of each protein sample includes: for each pixel in the protein staining pixel image, taking all pixels adjacent to the pixel as neighboring pixels; counting the number of pixels contained in each connected region, and deleting isolated connected regions with an area smaller than a preset area threshold to obtain the sample distribution region of each protein sample.

[0046] Adjacent pixels can be understood as pixels that are adjacent to the currently selected pixel. Isolated connected regions can be understood as regions composed of multiple adjacent pixels, but the pixel area of ​​this region is relatively small.

[0047] Specifically, for each pixel in the protein-stained pixel image, all adjacent pixels are obtained as neighboring pixels, and each pixel is scanned to determine whether each neighboring pixel is a pre-labeled pixel. Then, pixels belonging to the same label are merged into the same connected region, and the number of pixels contained in each connected region is counted to remove isolated connected regions with an area smaller than a preset area threshold, thereby obtaining the sample distribution area for each protein sample. This technical solution, by using the eight-neighborhood connectivity rule to process pixels, can effectively ensure the accuracy of area calculation and improve the computational efficiency of subsequent protein co-localization analysis.

[0048] In another embodiment, the step of separating protein staining pixel images corresponding to different protein samples from the target fluorescence image according to a preset pixel threshold, and determining the protein sample distribution region corresponding to each protein staining pixel image, further includes: inputting the target fluorescence image into a deep segmentation network to obtain the sample distribution region of each protein sample.

[0049] Specifically, deep segmentation networks can include at least one of the following: U-Net network, CNN network, etc.

[0050] Furthermore, the target fluorescence image can be input into a deep segmentation network. By processing the target fluorescence image using the deep segmentation network, the sample distribution area of ​​each protein sample can be directly extracted, thereby determining the sample distribution area more efficiently.

[0051] S130. Using the horizontal coordinate of the target fluorescence image as a reference, the target fluorescence image is divided into multiple protein sample blocks according to a preset pixel spacing.

[0052] Specifically, the longer side of the target fluorescence image can be used as the horizontal axis, and the horizontal axis can be used as the basis for dividing the sample blocks. By following the preset pixel spacing, such as 15 pixels, the target fluorescence image can be divided into multiple long strip-shaped protein sample blocks in the order of the horizontal axis.

[0053] Furthermore, if the horizontal coordinate length of the last protein sample block is less than 15 pixels, it can be deleted or padded to ensure the integrity of the protein sample block division, so as to facilitate the comparison of subsequent protein sample blocks.

[0054] S140. For a single protein sample block, determine the positional distribution information of each type of protein staining pixel within the protein sample block, and determine the distance and overlap data between various types of protein staining pixels based on the positional distribution information of different protein staining pixels.

[0055] The distance data includes centroid distance, absolute distance, relative distance, and standard deviation distance. Overlap data can be understood as the number of protein-stained pixels of different colors that overlap at the same pixel coordinates.

[0056] Specifically, for a single protein sample block, the positional distribution information of different protein-stained pixels within that block is determined based on their horizontal and vertical coordinates. Furthermore, by comparing the positional distribution information of different protein-stained pixels pixel-by-pixel within each protein sample block, the distances and overlap data between various protein-stained pixels are determined. This technical solution, by comparing the coordinates of different protein-stained pixels across multiple protein sample blocks, enables refined and automated overlap calculations, achieving efficient computation for different protein samples.

[0057] S150. Based on the distance and overlap data between the various protein staining pixels in the target fluorescence image, determine the protein colocalization analysis results in the target fluorescence image.

[0058] The results of protein colocalization analysis can include whether different proteins are spatially clustered in the same region, the degree of overlap, the distribution pattern in the spatial distribution, and whether there is an interaction.

[0059] Specifically, the distance and overlap data between various protein-stained pixels in the target fluorescence images of different experimental groups can be plotted and quantitatively analyzed to determine the protein co-localization analysis results in the target fluorescence images of the experimental groups.

[0060] In one embodiment, determining the protein colocalization analysis results in the target fluorescence image based on the distance and overlap data between various protein-stained pixels in the target fluorescence image includes: statistically verifying the distance and overlap data between various protein-stained pixels in the target fluorescence images of the control group and the experimental group to determine the protein colocalization analysis results in the target fluorescence image.

[0061] Specifically, the distance and overlap data between various protein-stained pixels in the target fluorescence images of the control and experimental groups can be organized and grouped to obtain data columns for the control and experimental groups. Statistical pre-validation of the two data columns is then performed to determine whether parametric or non-parametric validation should be used. Finally, statistical software (such as GraphPad Prism, SPSS, or Python's SciPy library) is used to calculate the similarity probability values ​​between the control and experimental data columns to determine the protein co-localization analysis results in the target fluorescence images of the experimental group. This technical solution, by employing the above method, can effectively eliminate subjective bias and capture subtle changes between data points, effectively improving the reliability of statistical validation.

[0062] The technical solution of this invention involves determining a target region of interest from a fluorescence image of a tissue sample and cropping the target region of interest to obtain a target fluorescence image. The target fluorescence image includes target fluorescence images from a control group fluorescence image and target fluorescence images from an experimental group fluorescence image. The target fluorescence image includes at least two protein staining pixels corresponding to different protein samples. Based on a preset pixel threshold, protein staining pixel images corresponding to different protein samples are separated from the target fluorescence image, and a corresponding protein sample distribution region is determined for each protein staining pixel image. The preset pixel threshold includes a pixel threshold corresponding to each protein staining pixel. Using the horizontal coordinate of the target fluorescence image as a reference, the target fluorescence image is divided into multiple protein sample blocks according to a preset pixel spacing. For a single protein sample block, the positional distribution information of each protein staining pixel within the protein sample block is determined, and the distance and overlap data between various protein staining pixels are determined based on the positional distribution information of different protein staining pixels. Based on the distance and overlap data between various protein staining pixels in the target fluorescence image, the protein co-localization analysis results in the target fluorescence image are determined. By adopting the above-mentioned operational steps, the technical problem of difficulty in comprehensively, objectively, accurately, and automatically quantifying colocalization signals of different proteins in existing analytical methods is solved. It also enables comprehensive, automatic, and unbiased integrated analysis of colocalization signals of different types of proteins in fluorescence images, thereby realizing the differential quantitative comparison of colocalization signals of large-scale different protein samples and effectively improving the interpretability of biological tissues.

[0063] Example 2

[0064] Figure 2 This is a flowchart of a method for analyzing protein co-localization in fluorescence images provided in Embodiment 2 of the present invention. This embodiment refines the technical solution of "determining the positional distribution information of each protein-stained pixel within the protein sample block, and determining the distance and overlap data between various protein-stained pixels based on the positional distribution information of different protein-stained pixels" based on the above embodiments. Detailed implementation can be found in the description of this embodiment. Technical features that are the same as or similar to those in the foregoing embodiments will not be repeated here. Figure 2 As shown, the method may specifically include:

[0065] S210. Determine the target region of interest from the fluorescence image of the tissue sample, and crop the target region of interest to obtain the target fluorescence image.

[0066] S220. Based on a preset pixel threshold, separate the protein staining pixel images corresponding to different protein samples from the target fluorescence image, and determine the protein sample distribution area corresponding to each protein staining pixel image.

[0067] S230. Using the horizontal coordinate of the target fluorescence image as a reference, the target fluorescence image is divided into multiple protein sample blocks according to a preset pixel spacing.

[0068] S240. For a single protein staining pixel, determine the candidate marker pixel corresponding to each horizontal coordinate in the protein sample block, and record the vertical coordinate of each candidate marker pixel, so as to determine the position distribution information of the protein staining pixel based on the horizontal and vertical coordinates of multiple candidate marker pixels.

[0069] Specifically, for a single protein staining pixel, within a protein sample block, candidate marker pixels are identified at each horizontal coordinate, and the corresponding vertical coordinate of each candidate marker pixel is recorded. By statistically analyzing the coordinate values ​​of each candidate marker pixel at each horizontal and vertical coordinate, the positional distribution information of a protein staining pixel within a protein sample block is obtained. Furthermore, the horizontal and vertical coordinates of different protein staining pixels within multiple protein sample blocks are statistically analyzed to determine the positional distribution information of each type of protein staining pixel. This technical solution, by determining the positional distribution information of protein staining pixels, facilitates subsequent overlap comparison calculations of protein staining pixels, ensuring the accuracy of the overlap ratio calculation in subsequent fluorescence images.

[0070] S250. For the same protein sample block, based on the positional distribution information of various protein staining pixels, determine the centroid coordinates of each protein staining pixel in the protein sample block, so as to determine the centroid distance between various protein staining pixels based on the multiple centroid coordinates.

[0071] Specifically, for the same protein sample block, based on the positional distribution information of various protein staining pixels within that block—that is, the coordinate values ​​of each protein staining pixel within the protein sample block—the centroid coordinates of each protein staining pixel are determined. Then, the centroid coordinates corresponding to different protein staining pixels are calculated to determine the centroid distance between various protein staining pixels. By calculating the centroid coordinates of different protein staining pixels, the proximity distance between different protein staining pixels in the two-dimensional plane can be effectively assessed, thus aiding in the evaluation of the overlap state of different protein staining pixels in planar space.

[0072] S260. The positional distribution information of the various protein staining pixels is compared pixel by pixel to determine the number of overlaps of the various protein staining pixels in the protein sample block.

[0073] Specifically, the positional distribution information of various protein staining pixels can be compared pixel by pixel. By identifying overlapping pixels of different protein staining pixels at the same coordinates, the overlap of various protein staining pixels in the protein sample block can be determined based on multiple overlapping pixels. By using pixel-by-pixel comparison, the accuracy of calculating the overlap ratio of different protein staining pixels in the target fluorescence image can be ensured.

[0074] S270. Based on the distance and overlap data between the various protein staining pixels in the target fluorescence image, determine the protein colocalization analysis results in the target fluorescence image.

[0075] The technical solution of this invention involves determining the candidate marker pixels corresponding to each horizontal coordinate within a protein sample block for each protein staining pixel, and recording the vertical coordinate of each candidate marker pixel. Based on the horizontal and vertical coordinates of multiple candidate marker pixels, the positional distribution information of the protein staining pixels is determined. For the same protein sample block, based on the positional distribution information of multiple protein staining pixels, the centroid coordinates of each type of protein staining pixel within the protein sample block are determined. The centroid distances between various protein staining pixels are then determined based on these centroid coordinates. Finally, the positional distribution information of multiple protein staining pixels is compared pixel-by-pixel to determine the overlap of multiple protein staining pixels within the protein sample block. This improves the accuracy of calculating the distance data and overlap ratio between protein staining pixels, enabling multidimensional protein co-localization analysis.

[0076] Example 3

[0077] Figure 3 This is a schematic flowchart illustrating the analysis of protein co-localization in fluorescence images according to Embodiment 3 of the present invention. The technical solution of this embodiment refines the implementation method of protein co-localization based on the above embodiments. For parts not described in detail in this method embodiment, please refer to the above embodiments. See details... Figure 3 As shown, the method may include the following steps:

[0078] S301. Target Region Extraction. Obtain an original image containing two types of protein fluorescent labels, where red fluorescent labels represent postsynaptic membrane protein sample regions, and green fluorescent labels represent presynaptic membrane protein sample regions. Select a target region from the original fluorescence image that simultaneously contains both red and green fluorescent signals, and crop the original fluorescence image according to the boundaries of the target region to obtain the target fluorescence image for subsequent analysis.

[0079] S302. Fluorescence signal separation and noise removal. The red and green fluorescence channels of the target fluorescence image are extracted separately, and thresholding is applied to each channel. For any channel, pixels with a value less than 20 are set as background pixels, while pixels with a value greater than or equal to 20 are retained as candidate marker pixels for the corresponding protein sample. This generates binary masks for red and green protein samples, respectively.

[0080] For any abnormal marked areas in the image, the operator manually selects the abnormal areas and excludes the entire selected area from subsequent analysis.

[0081] For the binary mask after excluding abnormal regions, the eight-neighbor connectivity rule is used to identify connected regions, that is, the eight pixels adjacent to the center pixel in the horizontal, vertical and diagonal directions are considered as neighboring pixels. The number of pixels contained in each connected region is counted, and isolated connected regions with an area smaller than a preset area threshold are deleted, thereby reducing the interference of scattered noise and non-target fluorescence signals, and obtaining the processed red protein sample region and green protein sample region.

[0082] S303. Protein Sample Block Division and Location Distribution Statistics. Using the horizontal axis of the target fluorescence image as the division benchmark, the target fluorescence image is divided into several consecutive protein sample blocks with a width of 15 pixels each. If the width of the target fluorescence image is not divisible by 15 pixels, the remaining pixels in the horizontal axis direction are divided into the last block.

[0083] For the k-th block, read the red and green marker pixels corresponding to each horizontal coordinate within the block, and record the vertical coordinates of the corresponding marker pixels. Merge all red marker pixels detected within the block into a single red protein sample region, and merge all green marker pixels detected within the block into a single green protein sample region.

[0084] The minimum, maximum, and all ordinate sets of the labeled pixels are calculated separately to determine the vertical positional distribution range of the two protein samples within the block. The minimum and maximum ordinates are only used to characterize the positional distribution range; subsequent region overlap calculations are based on the actual set of detected labeled pixels.

[0085] S304. Centroid Distance and Region Overlap Ratio Calculation. When both red and green marker pixels exist in the k-th block, all marker pixels of the same color in the block are considered as points with the same mass, and the longitudinal centroids of the red protein sample region and the green protein sample region are calculated respectively.

[0086] The overlap ratio of the regions is further calculated based on the intersection and union of the red protein sample regions and the green protein sample regions. When a block is missing red marker pixels, green marker pixels, or both marker pixels, the absolute distance of the centroid of the block is recorded as invalid because it is impossible to obtain the two vertical centroids at the same time, and the overlap ratio of the region is recorded as 0.

[0087] S305. Comparison and verification between the control group and the experimental group. Statistical data for both groups were calculated and quantitatively analyzed by plotting graphs to assess the significance of the co-localization signals of the two proteins.

[0088] This invention proposes an intelligent analysis method for co-labeling two proteins in fluorescence images, which can realize comprehensive, automatic and unbiased analysis of co-localization signals of two proteins in fluorescence images, thereby enabling differential quantitative comparison and processing large-scale protein samples.

[0089] Example 4

[0090] Figure 4 This is a schematic diagram of the structure of an analytical device for co-localization of proteins in fluorescence images provided in Embodiment 4 of the present invention. Figure 4 As shown, the device includes: a target image acquisition module 401, a protein sample separation module 402, a protein sample block separation module 403, a protein sample distance determination module 404, and a colocalization analysis module 405.

[0091] The target image acquisition module 401 is used to determine a target region of interest from the fluorescence image of the tissue sample and crop the target region of interest to obtain a target fluorescence image; wherein the target fluorescence image includes target fluorescence images in the control group fluorescence image and target fluorescence images in the experimental group fluorescence image; the target fluorescence image includes at least two protein staining pixels corresponding to two protein samples; the protein sample separation module 402 is used to separate protein staining pixel images corresponding to different protein samples from the target fluorescence image according to a preset pixel threshold, and determine the corresponding protein sample distribution region according to each protein staining pixel image; wherein the preset pixel threshold includes each protein staining pixel The corresponding pixel threshold; protein sample block separation module 403, used to divide the target fluorescence image into multiple protein sample blocks based on the horizontal coordinate of the target fluorescence image and according to a preset pixel spacing; protein sample distance determination module 404, used to determine the position distribution information of each type of protein staining pixel in the protein sample block for a single protein sample block, and to determine the distance and overlap data between various types of protein staining pixels based on the position distribution information of different types of protein staining pixels; colocalization analysis module 405, used to determine the protein colocalization analysis results in the target fluorescence image based on the distance and overlap data between various types of protein staining pixels in the target fluorescence image.

[0092] In this embodiment of the invention, the target image acquisition module 401 determines a target region of interest from the fluorescence image of a tissue sample and crops the target region of interest to obtain a target fluorescence image. The target fluorescence image includes target fluorescence images from the control group fluorescence image and target fluorescence images from the experimental group fluorescence image. The target fluorescence image includes at least two protein staining pixels corresponding to two protein samples. The protein sample separation module 402 separates protein staining pixel images corresponding to different protein samples from the target fluorescence image according to a preset pixel threshold, and determines the corresponding protein sample distribution region based on each protein staining pixel image. The preset pixel threshold includes the distribution region of each protein sample. The process involves several steps: a pixel threshold corresponding to the protein staining pixels; a protein sample block separation module 403 dividing the target fluorescence image into multiple protein sample blocks based on the horizontal coordinate of the target fluorescence image and a preset pixel spacing; a protein sample distance determination module 404 determining the positional distribution information of each type of protein staining pixel within a single protein sample block, and determining the distance and overlap data between various protein staining pixels based on their positional distribution information; and a colocalization analysis module 405 determining the protein colocalization analysis results in the target fluorescence image based on the distance and overlap data between various protein staining pixels in the target fluorescence image. By employing these steps, the technical problem of comprehensively, objectively, accurately, and automatically quantifying different protein colocalization signals in existing analysis methods is solved. This allows for comprehensive, automatic, and unbiased integrated analysis of different types of protein colocalization signals in fluorescence images, thereby achieving differential quantitative comparison of colocalization signals from large-scale different protein samples and effectively improving the interpretability of biological tissues.

[0093] Based on the above-mentioned optional technical solutions, the protein sample separation module 402 may optionally include: a pixel channel extraction unit, a protein staining pixel image generation unit, and a sample distribution region determination unit. Specifically, the pixel channel extraction unit is used to extract staining pixel channels corresponding to each type of protein staining pixel from the target fluorescence image; the protein staining pixel image generation unit is used to, for any staining pixel channel, set pixels smaller than a pixel threshold as background pixels and pixels greater than or equal to the pixel threshold as candidate marker pixels, to generate a protein staining pixel image based on the background pixels and the candidate marker pixels; the sample distribution region determination unit is used to identify each pixel in the protein staining pixel image using an eight-neighbor connectivity rule to obtain the sample distribution region of each type of protein sample; wherein the eight-neighbor connectivity rule refers to any pixel being adjacent to other pixels in the horizontal, vertical, and diagonal directions.

[0094] Based on the above-mentioned optional technical solutions, optionally, the sample distribution region determination unit may include: an adjacent pixel determination subunit and a sample distribution region determination subunit. The adjacent pixel determination subunit is used to, for each pixel in the target protein staining pixel image, identify all pixels adjacent to that pixel as adjacent pixels; the sample distribution region determination subunit is used to count the number of pixels contained in each connected region and delete isolated connected regions with an area smaller than a preset area threshold, to obtain the sample distribution region for each protein sample.

[0095] Based on the above-mentioned optional technical solutions, the protein sample separation module 402 may optionally include a segmentation network generation unit. The segmentation network generation unit is used to input the target fluorescence image into a deep segmentation network to obtain the sample distribution region of each protein sample.

[0096] Based on the above-mentioned optional technical solutions, the protein sample distance determination module 404 may optionally include: a pixel coordinate determination unit, a centroid distance determination unit, and an overlap quantity determination unit. The distance includes the centroid distance; the pixel coordinate determination unit is used to determine, for each type of protein staining pixel, the corresponding candidate marker pixel at each horizontal coordinate within the protein sample block, and record the vertical coordinate of each candidate marker pixel, so as to determine the positional distribution information of the protein staining pixel based on the horizontal and vertical coordinates of multiple candidate marker pixels; the centroid distance determination unit is used to determine, for the same protein sample block, the centroid coordinates of each type of protein staining pixel in the protein sample block based on the positional distribution information of multiple types of protein staining pixels, so as to determine the centroid distance between various types of protein staining pixels based on the multiple centroid coordinates; the overlap quantity determination unit is used to compare the positional distribution information of multiple types of protein staining pixels pixel by pixel to determine the overlap quantity of multiple types of protein staining pixels in the protein sample block.

[0097] Based on the above-mentioned optional technical solutions, the colocalization analysis module 405 may optionally include a comparison verification unit. The comparison verification unit is used to statistically verify the distance and overlap data between various protein-stained pixels in the target fluorescence images of the control group and the experimental group to determine the protein colocalization analysis results in the target fluorescence images.

[0098] Based on the above-mentioned optional technical solutions, the distance may optionally include absolute distance, relative distance, and standard deviation distance.

[0099] The analytical apparatus for protein colocalization in fluorescence images provided in this embodiment of the invention can execute the analytical method for protein colocalization in fluorescence images provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0100] Example 5

[0101] Figure 5 A schematic diagram of an electronic device 10, which can be used to implement embodiments of the present invention, is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0102] like Figure 5 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0103] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0104] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as a method for analyzing protein co-localization in fluorescence images.

[0105] In some embodiments, a method for analyzing protein colocalization in fluorescence images may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or mounted on electronic device 10 via read-only memory (ROM) 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the method for analyzing protein colocalization in fluorescence images described above may be performed. Alternatively, in other embodiments, processor 11 may be configured by any other suitable means (e.g., by means of firmware) to perform a method for analyzing protein colocalization in fluorescence images.

[0106] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0107] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0108] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0109] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0110] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0111] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0112] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication unit 19, or installed from storage unit 18, or installed from ROM 12. When the computer program is executed by processor 11, it performs the functions defined in the methods of the embodiments of the present invention.

[0113] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0114] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for analyzing protein co-localization in fluorescence images, characterized in that, include: A target region of interest is determined from the fluorescence image of a tissue sample, and the target region of interest is cropped to obtain a target fluorescence image; wherein, the target fluorescence image includes the target fluorescence image in the control group fluorescence image and the target fluorescence image in the experimental group fluorescence image; the target fluorescence image includes at least two protein staining pixels corresponding to the two protein samples; Based on a preset pixel threshold, protein staining pixel images corresponding to different protein samples are separated from the target fluorescence image, and the protein sample distribution region corresponding to each protein staining pixel image is determined; wherein, the preset pixel threshold includes the pixel threshold corresponding to each protein staining pixel; Based on the horizontal coordinate of the target fluorescence image, the target fluorescence image is divided into multiple protein sample blocks according to a preset pixel spacing; For a single protein sample block, the positional distribution information of each type of protein staining pixel within the protein sample block is determined, and the distance and overlap data between various types of protein staining pixels are determined based on the positional distribution information of different protein staining pixels; Based on the distance and overlap data between the various protein-stained pixels in the target fluorescence image, the protein colocalization analysis results in the target fluorescence image are determined.

2. The method for analyzing protein co-localization in fluorescence images according to claim 1, characterized in that, The step of separating protein staining pixel images corresponding to different protein samples from the target fluorescence image according to a preset pixel threshold, and determining the corresponding protein sample distribution region according to each protein staining pixel image, includes: Extract the staining pixel channel corresponding to each protein staining pixel from the target fluorescence image; For any stained pixel channel, pixels in the stained pixel channel that are less than the pixel threshold are set as background pixels, and pixels that are greater than or equal to the pixel threshold are set as candidate marker pixels, so as to generate a protein stained pixel image based on the background pixels and the candidate marker pixels; The eight-neighbor connectivity rule is used to identify each pixel in the protein staining pixel image to obtain the sample distribution area of ​​each protein sample; wherein, the eight-neighbor connectivity rule is any pixel that is adjacent to the pixel in the horizontal, vertical and diagonal directions.

3. The method for analyzing protein co-localization in fluorescence images according to claim 2, characterized in that, The process of identifying each pixel in the protein-stained pixel image using the eight-neighbor connectivity rule to obtain the sample distribution region for each protein sample includes: For each pixel in the protein-stained pixel image, all pixels adjacent to that pixel are considered as neighboring pixels; The number of pixels contained in each connected region is counted, and isolated connected regions with an area smaller than a preset area threshold are deleted to obtain the sample distribution area for each protein sample.

4. The method for analyzing protein co-localization in fluorescence images according to claim 1, characterized in that, The step of separating protein staining pixel images corresponding to different protein samples from the target fluorescence image according to a preset pixel threshold, and determining the corresponding protein sample distribution region according to each protein staining pixel image, further includes: The target fluorescence image is input into a deep segmentation network to obtain the sample distribution region for each protein sample.

5. The method for analyzing protein co-localization in fluorescence images according to claim 1, characterized in that, The distance includes the centroid distance; determining the positional distribution information of each protein staining pixel within the protein sample block, and determining the distance and overlap data between various protein staining pixels based on the positional distribution information of different protein staining pixels, includes: For a single protein staining pixel, the candidate marker pixel corresponding to each horizontal coordinate in the protein sample block is determined one by one, and the vertical coordinate of each candidate marker pixel is recorded respectively, so as to determine the position distribution information of the protein staining pixel based on the horizontal and vertical coordinates of multiple candidate marker pixels; For the same protein sample block, based on the positional distribution information of various protein staining pixels, the centroid coordinates of each protein staining pixel in the protein sample block are determined, so as to determine the centroid distance between various protein staining pixels based on the multiple centroid coordinates. The positional distribution information of the various protein staining pixels is compared pixel by pixel to determine the number of overlaps of the various protein staining pixels in the protein sample block.

6. The method for analyzing protein co-localization in fluorescence images according to claim 1, characterized in that, The step of determining the protein colocalization analysis results in the target fluorescence image based on the distance and overlap data between various protein-stained pixels in the target fluorescence image includes: The distances and overlaps between various protein-stained pixels in the target fluorescence images of the control group and the experimental group were statistically verified to determine the protein colocalization analysis results in the target fluorescence images.

7. The method for analyzing protein co-localization in fluorescence images according to claim 1, characterized in that, The distance also includes absolute distance, relative distance, and standard deviation distance.

8. An analytical device for protein co-localization in fluorescence images, characterized in that, The device includes: The target image acquisition module is used to determine the target region of interest from the fluorescence image of the tissue sample, and crop the target region of interest to obtain the target fluorescence image; wherein, the target fluorescence image includes the target fluorescence image in the control group fluorescence image and the target fluorescence image in the experimental group fluorescence image; the target fluorescence image includes at least two protein staining pixels corresponding to the two protein samples; The protein sample separation module is used to separate protein staining pixel images corresponding to different protein samples from the target fluorescence image according to a preset pixel threshold, and to determine the protein sample distribution region corresponding to each protein staining pixel image; wherein, the preset pixel threshold includes the pixel threshold corresponding to each protein staining pixel; The protein sample block separation module is used to divide the target fluorescence image into multiple protein sample blocks based on the horizontal coordinate of the target fluorescence image and according to a preset pixel spacing. The protein sample distance determination module is used to determine the position distribution information of each type of protein staining pixel within a single protein sample block, and to determine the distance and overlap data between various types of protein staining pixels based on the position distribution information of different protein staining pixels. The colocalization analysis module is used to determine the protein colocalization analysis results in the target fluorescence image based on the distance and overlap data between various protein-stained pixels in the target fluorescence image.

9. An electronic device, characterized in that, The electronic device includes: One or more processors; Storage device for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the analytical method for colocalization of proteins in fluorescence images as described in any one of claims 1-7.

10. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the method for analyzing protein colocalization in fluorescence images as described in any one of claims 1-7.

Images