A fluorescence image detection method, system, electronic device and medium

Abstract

The application discloses a fluorescence image detection method and system, electronic equipment and medium, comprising: acquiring a fluorescence image corresponding to a to-be-detected sample, the fluorescence image comprising a microwell array fluorescence image containing coded microspheres and a droplet fluorescence image; based on a local multi-threshold segmentation identification method, performing multiple digital detection information extraction on the microwell array fluorescence image with the coded microspheres as carriers and based on the microwell array to obtain first fluorescence detection information; performing multiple digital detection information extraction on the droplet fluorescence image with the coded microspheres as carriers and based on the droplet to obtain second fluorescence detection information; and obtaining fluorescence detection information corresponding to the to-be-detected sample through the first fluorescence detection information and the second fluorescence detection information. The application can not only realize accurate identification of the position information of the droplet and the microwell array reaction unit in the fluorescence image, but also accurately identify the single-microsphere wrapping and the multi-microsphere wrapping states of the reaction unit, thereby realizing higher data statistical accuracy.

Description

Technical Field

[0001] This invention relates primarily to the field of image processing technology, and in particular to a fluorescence image detection method, system, electronic device, and medium. Background Technology

[0002] Disease is a process of quantitative change leading to qualitative change, and effective diagnosis is the first step in treating disease. High-throughput digital diagnostic technologies with high sensitivity, high specificity, and the ability to simultaneously analyze multiple biological factors are considered reliable methods that promise to significantly improve diagnostic capabilities.

[0003] Digital detection methods are single-molecule detection methods that can effectively reduce detection errors and improve detection sensitivity. The detection solution containing the target molecule is repeatedly diluted and dispersed in thousands of picoliter microreaction units, theoretically ensuring that each unit contains at most one target molecule. Each unit contains a detection substrate and a luminescent enzyme group. When the probe detects the target molecule, a specific reaction catalyzes the enzyme substrate to emit light, and the unit state changes to "1," a "positive" signal state. Units that do not undergo a specific reaction show "0," a "negative" signal state. By reading the states of these units, the initial molecular concentration of the reactant can be calculated.

[0004] Microdroplets and microarrays are two commonly used methods for isolating reaction units in digital single-molecule detection. Microdroplet-based detection methods, utilizing specialized microfluidic chips with T-structures and flow focusing structures, can generate hundreds of thousands of uniformly sized, well-sealed, and well-dispersed microfluidic reaction units in a short time. Microarrays, on the other hand, construct tens of thousands of physically isolated, neatly arranged micro-wells capable of holding microliter-level reaction solutions through chemical and physical etching. Each reaction unit is filled with coded microspheres of 5-10 micrometer diameter coupled with specific detection probes, achieving detection limits (LODs) as low as 10 fen-liters. Both methods can be used for the detection of multiplex digital nucleic acid and protein biomarkers.

[0005] However, traditional image processing methods in digital detection processes are not suitable for the complex imaging environments of high-throughput, ultra-sensitive multiplex digital nucleic acid and protein detection. Therefore, there is an urgent need for a fluorescence image detection method that can effectively detect reactive units in fluorescence images. Summary of the Invention

[0006] The purpose of this application is to provide a fluorescence image detection method, system, electronic device, and medium that effectively detects reaction units in fluorescence images while improving image detection accuracy.

[0007] In a first aspect, this application provides a fluorescence image detection method, comprising: acquiring a fluorescence image corresponding to a sample to be detected, the fluorescence image including a fluorescence image of a micropore array containing coded microspheres and a droplet fluorescence image; extracting multiple digital detection information based on the micropore array using coded microspheres as carriers from the micropore array fluorescence image to obtain first fluorescence detection information; extracting multiple digital detection information based on the droplet fluorescence image using coded microspheres as carriers to obtain second fluorescence detection information; and obtaining fluorescence detection information corresponding to the sample to be detected through the first fluorescence detection information and the second fluorescence detection information.

[0008] This application enables the acquisition of fluorescence images corresponding to the sample to be tested. These fluorescence images include fluorescence images of a micropore array containing coded microspheres and fluorescence images of droplets. Then, based on a local multi-threshold segmentation and recognition method, multiple digital detection information is extracted from the micropore array fluorescence image using coded microspheres as carriers, yielding first fluorescence detection information. Simultaneously, multiple digital detection information is extracted from the droplet fluorescence image using coded microspheres as carriers, yielding second fluorescence detection information. Finally, the fluorescence detection information corresponding to the sample to be tested is obtained through the first and second fluorescence detection information. This application achieves accurate identification and analysis of potential reaction units in the preprocessed image.

[0009] In one implementation of the first aspect, obtaining the fluorescence image corresponding to the sample to be tested includes: detecting the sample to be tested using a fluorescence microscope to obtain the fluorescence image corresponding to the sample to be tested.

[0010] In this application, a fluorescence microscope is used to acquire the fluorescence image of the sample to be tested. The microscope mainly includes an excitation light source (including but not limited to LED, xenon lamp, laser, etc.), a filter (to retain light of a specific wavelength, including but not limited to 488nm, 645nm, etc.), a sample stage and a sample to be photographed (including but not limited to microfluidic chips made of PB, PDMS, PMMA materials, which contain fluorescently encoded microspheres and emulsion droplets encapsulating the reaction substrate), an objective lens (a focusing lens, including but not limited to x4 and x10 magnification), and an image sensor (the working principle of the image sensor includes but is not limited to CCD (charge-coupled device) and CMOS (complementary metal-oxide-semiconductor)). This application captures and acquires image data signals (including but not limited to bright field, CY5, and dark field images of RITC / FITC fluorescence channels) of the microfluidic chip sample according to the set parameters, and transmits the image signals to the memory of the computer processing system via a data cable for the next step of signal processing.

[0011] In one implementation of the first aspect, based on a local multi-threshold segmentation and recognition method, multiple digital detection information is extracted from the micro-pore array fluorescence image using coded microspheres as carriers to obtain first fluorescence detection information. This includes: preprocessing the micro-pore array fluorescence image to obtain a preprocessed micro-pore array fluorescence image; identifying the first coded microsphere information corresponding to the preprocessed micro-pore array fluorescence image using the local multi-threshold segmentation and recognition method, wherein the first coded microsphere information includes the coded microsphere position, coded microsphere boundary, and fluorescence intensity signal; and obtaining the first fluorescence detection information based on the first coded microsphere information.

[0012] This application achieves accurate identification and analysis of potential reactive units (microspheres, microdroplets) in preprocessed images through the Localized Area Multi-threshold Segmentation (LAMS) method.

[0013] In one implementation of the first aspect, the micro-pore array fluorescence image is preprocessed to obtain a preprocessed micro-pore array fluorescence image, including: performing full-image fluorescence intensity equalization, median filtering, Gaussian filtering, morphological opening operation, and structure dilation on the micro-pore array fluorescence image.

[0014] In this application, mathematical morphology employs two fundamental operations: erosion and dilation. These operations, when combined, form opening and closing operations. Morphological opening involves erosion followed by dilation, while closing involves dilation followed by erosion. Structural dilation enlarges the target region, merging background points in contact with the target into the target object, thus expanding the target boundary outwards. This serves to fill voids within the target region and eliminate small particle noise contained within it. This application further improves image detection accuracy by preprocessing the micro-aperture fluorescence image.

[0015] In one implementation of the first aspect, the first coded microsphere information corresponding to the preprocessed micropore array fluorescence image is identified by a local multi-threshold segmentation and recognition method, including: when performing the nth small-region recognition on the preprocessed micropore array fluorescence image, acquiring a grayscale image corresponding to the preprocessed micropore array fluorescence image for the nth small-region recognition, where the iteration number n is a positive integer; performing binarization processing on the grayscale image for the nth small-region recognition to obtain the nth binarized image; and extracting small-region information from the nth binarized image to obtain the first... The nth recognition region information; when the recognition region information meets the preset screening conditions, the nth recognition region is used as a reaction unit in the preprocessed micro-well array fluorescence image; the pixel value of the nth recognition region in the nth binarized image is set to the first value to obtain a grayscale image for the (n+1)th small region recognition; when all reaction units in the preprocessed micro-well array fluorescence image have undergone binarized recognition screening, the recognition region information of each recognition region is used as the first coded microsphere information corresponding to the preprocessed micro-well array fluorescence image.

[0016] In one implementation of the first aspect, when the identified region information meets the preset screening conditions, the nth identified region is used as a reaction unit in the preprocessed micro-well array fluorescence image. The preset screening conditions include region area screening conditions, eccentricity screening conditions, roundness screening conditions, and convex hull ratio screening conditions. The region area screening condition is: the area of ​​the identified region is greater than the minimum allowable area minValidArea and less than the maximum allowable area maxValidArea; where the minimum allowable area minValidArea = D. index ×0.75×R real 2 ×π, maximum allowed area maxValidArea=D index ×1.25×R real 2 ×π, R real D is the actual average radius of the microsphere. index The attenuation coefficient is... baseThresh is the initial binarization threshold, and curThresh is the threshold value. n Here, `curThresh` represents the actual threshold in the nth binarization process, and `BitDepth` represents the image bit depth. The eccentricity selection condition is: when the actual threshold `curThresh` in the nth binarization process... n When the eccentricity of the recognition region is less than the preset binarization threshold, the allowed range is: validEcc∈[0.75,2]; when curThresh is not less than the preset binarization threshold, the allowed range is: validEcc∈[0.75×D]. index[2]; The roundness filtering condition is: Roundness∈[0.75,2], and the roundness of the recognition region is Roundness=4×Π×Area / (perimeter). 2 Where perimeter is the perimeter of the identified region and area is the area of ​​the identified region; the convex hull rate filtering condition is: the convex hull rate of the identified region is greater than the convex hull rate threshold.

[0017] In one implementation of the first aspect, the first coded microsphere information corresponding to the preprocessed micropore array fluorescence image is identified by a local multi-threshold segmentation and recognition method. The method further includes: when the identification region information does not meet preset screening conditions, obtaining a binarized threshold curThresh for the (n+1)th small region identification. n+1 The binarization threshold curThresh for the (n+1)th small region identification is used. n+1 = baseThresh + (n+1) × stepValue, where baseThresh is the initial binarization threshold and stepValue is the step gradient value.

[0018] In one implementation of the first aspect, the droplet fluorescence image is subjected to multiple digital detection information extraction based on the droplet using coded microspheres as carriers to obtain second fluorescence detection information. The droplet fluorescence image includes a coded signal image with coded microspheres and a report signal image with probes in the droplet. The method includes: performing region extraction on the coded signal image with coded microspheres and the report signal image with probes in the droplet based on a local multi-threshold segmentation and recognition method to obtain droplet fluorescence image information.

[0019] In one implementation of the first aspect, the fluorescence image detection method further includes: determining the state of the droplet-encapsulated microspheres; the droplet fluorescence image information includes coded microsphere fluorescence information and droplet information; based on the coded microsphere fluorescence information, a corrected microsphere centroid position parameter, denoted as B(x1, y1), is extracted from the microsphere structure identified from the coded signal image containing the coded microsphere; based on the droplet information, a corrected center coordinate position parameter, denoted as D(x0, y0), is obtained from the report signal image containing the probe in the droplet; when the corrected microsphere centroid position parameter satisfies... At that time, it was determined that the identified microsphere structure was encapsulated by the droplet, R drop Let θ be the radius of a single droplet, d0 and h0 be the coordinate correction coefficients, and θ be the distance coefficient.

[0020] In one implementation of the first aspect, the droplet fluorescence image information includes the number of coded microspheres encapsulated by the droplet, and the determination of the state of the microspheres encapsulated by the droplet, and further includes: when the state of the microspheres encapsulated by the droplet is determined to be an empty droplet, encapsulating a single microsphere, or encapsulating multiple microspheres based on the number of coded microspheres encapsulated by the droplet.

[0021] In the process of digital droplet image detection, this application realizes the segmentation, recognition, and parameter extraction of a single droplet in a fluorescence image. In addition, it also realizes the discrimination function of the state of droplet encapsulating coded microspheres (no microspheres encapsulated, encapsulating a single coded microsphere, or encapsulating two or more coded microspheres), further improving the accuracy of microsphere statistical results.

[0022] Secondly, this application provides a fluorescence image detection system, including an image acquisition module for acquiring a fluorescence image corresponding to a sample to be detected, the fluorescence image including a fluorescence image of a micropore array containing coded microspheres and a droplet fluorescence image; a first image detection module for extracting multiple digital detection information based on the micropore array using coded microspheres as carriers and the micropore array based on a local multi-threshold segmentation and recognition method to obtain first fluorescence detection information; a second image detection module for extracting multiple digital detection information based on the droplet fluorescence image using coded microspheres as carriers and the droplet based on the droplet to obtain second fluorescence detection information; and an information acquisition module for obtaining fluorescence detection information corresponding to the sample to be detected through the first fluorescence detection information and the second fluorescence detection information.

[0023] Thirdly, this application provides an electronic device, comprising: a memory storing multiple instructions; and a processor loading instructions from the memory to perform steps as described in any of the fluorescence image detection methods above.

[0024] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by an electronic device, implements the steps in any of the above-described fluorescence image detection methods.

[0025] This application utilizes a fluorescence image detection method to identify the position and boundaries of coded microspheres encapsulated in droplets and suspended in the reaction solution of a microporous array, as well as extract signals such as fluorescence intensity. This application solves the problem of accurately identifying high-density reaction unit (coded microspheres, droplets) signals in fluorescence images during high-throughput multiplex digital detection, and the problem of accurately interpreting the state of droplet-encapsulated microspheres (empty droplets, single-microsphere encapsulation, multiple-microsphere encapsulation) during multiplex digital droplet detection using coded microspheres as carriers. This application not only achieves accurate identification of the positional information of droplets and microporous array reaction units in fluorescence images, but also accurately distinguishes the single-microsphere encapsulation and multiple-microsphere encapsulation states of reaction units, achieving higher data statistical accuracy. This application is also applicable to the analysis of reaction chamber image signals captured by CMOD or CCD camera units, and does not rely on reagents such as fluorescein for auxiliary imaging, reducing experimental complexity. Attached Figure Description

[0026] Figure 1 This is a schematic diagram illustrating an application scenario according to an embodiment of this application.

[0027] Figure 2a This is a schematic flowchart of a fluorescence image detection method according to an embodiment of this application.

[0028] Figure 2b This is a schematic diagram comparing the encoded signal image and the reported signal image according to an embodiment of this application.

[0029] Figure 3 This is a schematic diagram of the structure of a fluorescence image detection system according to an embodiment of this application.

[0030] Figure 4 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0031] The following specific examples illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be noted that, unless otherwise specified, the following embodiments and features in the embodiments can be combined with each other.

[0032] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this application. Therefore, the drawings only show the components related to this application and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0033] For droplets and microarrays, which correspond to two types of reaction units, achieving a stable and rapid signal capture method is a prerequisite for high-throughput, accurate digital nucleic acid and protein detection. Considering manufacturing costs and detection efficiency, methods based on CMOS or CCD photosensitive devices for imaging and computer image processing are effective ways to identify each reaction unit and read fluorescence signals, and are currently the mainstream development direction for experimental research and commercial equipment.

[0034] In digital detection processes using microdroplets as reaction units, smaller droplet units are needed to accommodate more reaction units while ensuring reaction efficiency. At the same time, the droplets are arranged more closely together, allowing more droplets to be spread out in the collection chamber.

[0035] On solid-phase chips using micropore arrays as reaction units, the number of usable reaction units is limited. To accommodate more analyte, the reaction chamber loading needs to be maximized to 100%, requiring the use of objectives with lower magnification to capture more reaction units within a single image. To achieve imaging of a large number of reaction units in a short time, the objective magnification of the fluorescence microscope should be as low as possible. This results in poor image quality, with individual reaction units occupying smaller pixel sizes, making them more difficult to identify. Furthermore, due to the use of multiple detection probes, the fluorescence signal intensity differences between reaction units can be significant.

[0036] When a large number of reaction units are close together in an image, the difficulty of image recognition increases significantly: a single image contains a large number of droplets, and uneven lighting during the shooting process makes it difficult to identify individual reaction units; the close proximity of reaction units makes crosstalk of fluorescence signals between units more severe, which will seriously interfere with the accuracy of subsequent data and further affect the quality of data calculation.

[0037] In droplet images, closely adjacent droplets appear to be stuck together due to the mutual compression of their surfaces, making it difficult to completely distinguish and identify the signals of different fluorescent units.

[0038] In micropore array images, the fluorescence intensity distribution of the encoded microspheres is very large, resulting in a low signal-to-noise ratio for some microspheres. Furthermore, microspheres with weak fluorescence signals are affected by the scattered light from surrounding microspheres with high fluorescence signals, making their morphology difficult to distinguish.

[0039] Experiments have shown that traditional image processing methods in digital detection processes are not suitable for the complex imaging environments of high-throughput, ultra-sensitive multiplex digital nucleic acid and protein detection. Furthermore, there is a lack of publicly available literature and data addressing this issue to achieve stable high-throughput digital reaction unit identification. Therefore, this application proposes a fluorescence image detection method, system, electronic device, and medium that can stably identify high-throughput multiplex digital nucleic acid and protein detection images and obtain reliable results.

[0040] The following embodiments of this application provide a fluorescence image detection method, system, electronic device, and medium. Specifically, the fluorescence image detection system can be integrated into an electronic device, such as a terminal or server. The terminal can be a mobile phone, tablet computer, smart Bluetooth device, laptop computer, or personal computer (PC). The server can be a single server or a server cluster consisting of multiple servers.

[0041] In some embodiments, the fluorescence image detection system may also be integrated into multiple electronic devices, such as multiple servers, with multiple servers implementing the fluorescence image detection method of this application.

[0042] In some embodiments, the server may also be implemented as a terminal.

[0043] For example, Figure 1 This is a schematic diagram of an application scenario according to an embodiment of this application. The application scenario may include a fluorescence microscope 11, a storage terminal 12, a server 13, etc. The fluorescence microscope 11 may include a sample to be tested, an objective lens, a light source, a spectrometer, a filter, an image sensor, etc. The storage terminal 12 can be used to store fluorescence images corresponding to the sample to be tested. The fluorescence microscope 11, the storage terminal 12, and the server 13 are interconnected and will not be described further here.

[0044] Server 13 may include a processor and memory. Server 13 may include an image data reading module: calling relevant program interfaces / functions (such as the imread() function in Matlab, the cv2.imread() function in OpenCV) to transfer image data from storage media (hard disk, including but not limited to SSD solid-state drives and HDD mechanical hard disks) to the computer's memory media in a certain reading order.

[0045] Server 13 may also include an image recognition module for fluorescently encoded microspheres, which uses a series of image processing methods to identify the position and boundaries of the encoded microspheres encapsulated in the droplet and suspended in the microporous array reaction liquid, as well as extract fluorescence intensity signals. That is, server 13 can acquire fluorescence images corresponding to the sample to be tested, including fluorescence images of the microporous array containing the encoded microspheres and fluorescence images of the droplets. Based on a local multi-threshold segmentation and recognition method, the fluorescence images of the microporous array are subjected to multiple digital detection information extraction based on the microporous array with the encoded microspheres as carriers to obtain first fluorescence detection information; the fluorescence images of the droplets are subjected to multiple digital detection information extraction based on the droplets with the encoded microspheres as carriers to obtain second fluorescence detection information; and the fluorescence detection information corresponding to the sample to be tested is obtained through the first fluorescence detection information and the second fluorescence detection information.

[0046] The technical solutions in the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0047] like Figure 2a As shown, with server 13 as the execution entity, this embodiment provides a fluorescence image detection method, including steps S210 to S240, as follows:

[0048] S210, Server 13 acquires the fluorescence image corresponding to the sample to be tested. The fluorescence image includes a micropore array fluorescence image containing coded microspheres and a droplet fluorescence image.

[0049] In one embodiment, obtaining a fluorescence image corresponding to the sample to be tested includes: detecting the sample to be tested using a fluorescence microscope to obtain a fluorescence image corresponding to the sample to be tested.

[0050] In this application, a fluorescence microscope is used to acquire the fluorescence image of the sample to be tested. The microscope mainly includes an excitation light source (including but not limited to LED, xenon lamp, laser, etc.), a filter (to retain light of a specific wavelength, including but not limited to 488nm, 645nm, etc.), a sample stage and a sample to be photographed (including but not limited to microfluidic chips made of PB, PDMS, PMMA materials, which contain fluorescently encoded microspheres and emulsion droplets encapsulating the reaction substrate), an objective lens (a focusing lens, including but not limited to x4 and x10 magnification), and an image sensor (the working principle of the image sensor includes but is not limited to CCD (charge-coupled device) and CMOS (complementary metal-oxide-semiconductor)). This application captures and acquires image data signals (including but not limited to bright field, CY5, and dark field images of RITC / FITC fluorescence channels) of the microfluidic chip sample according to the set parameters, and transmits the image signals to the memory of the computer processing system via a data cable for the next step of signal processing.

[0051] S220. Based on the local multi-threshold segmentation and recognition method, the fluorescence image of the micropore array is subjected to multiple digital detection information extraction based on the micropore array with coded microspheres as carriers to obtain the first fluorescence detection information.

[0052] In one embodiment, based on a local multi-threshold segmentation and recognition method, multiple digital detection information is extracted from the micropore array fluorescence image using coded microspheres as carriers and based on the micropore array to obtain first fluorescence detection information. This includes: preprocessing the micropore array fluorescence image to obtain a preprocessed micropore array fluorescence image; identifying the first coded microsphere information corresponding to the preprocessed micropore array fluorescence image using the local multi-threshold segmentation and recognition method, the first coded microsphere information including the coded microsphere position, coded microsphere boundary, and fluorescence intensity signal; and obtaining the first fluorescence detection information based on the first coded microsphere information.

[0053] In this embodiment, the microsphere identification process for extracting multiple digital detection information based on the micropore array using coded microspheres as carriers from micropore array fluorescence images can include three parts: image preprocessing → local multi-threshold segmentation and identification method → ​​data aggregation. The LAMS method is used to identify microspheres in fluorescence images of coded microspheres with high fill rate. Data aggregation integrates the information of droplets identified from different channel fluorescence images (structural variables recording data such as the position, boundary region, roundness, and eccentricity of individual microspheres) and removes duplicate data (structures).

[0054] This application proposes a Localized Area Multi-threshold Segmentation (LAMS) method. The LAMS method enables the identification of microspheres in fluorescence images of high-fill-rate encoded microspheres, and achieves accurate identification and analysis of potential reaction units in the preprocessed image.

[0055] In one embodiment, the micropore array fluorescence image is preprocessed to obtain a preprocessed micropore array fluorescence image, including: performing full-image fluorescence intensity equalization, median filtering, Gaussian filtering, morphological opening operation, and structure dilation on the micropore array fluorescence image.

[0056] In this application, mathematical morphology employs two fundamental operations: erosion and dilation. These operations, when combined, form opening and closing operations. Morphological opening involves erosion followed by dilation, while closing involves dilation followed by erosion. Structural dilation enlarges the target region, merging background points in contact with the target into the target object, thus expanding the target boundary outwards. This serves to fill voids within the target region and eliminate small particle noise contained within it. This application further improves image detection accuracy by preprocessing the micro-aperture fluorescence image.

[0057] Specifically, the image preprocessing process may include: full-image fluorescence intensity equalization, median filtering (template size: 5 pixels), Gaussian filtering (parameter σ = 0.25), morphological opening operation (disk circular structure with radius 3), and structure dilation (disk circular structure with radius 3).

[0058] In one embodiment, the first coded microsphere information corresponding to the preprocessed microporous array fluorescence image is identified by a local multi-threshold segmentation and recognition method, including: when performing the nth small region recognition on the preprocessed microporous array fluorescence image, obtaining a grayscale image corresponding to the preprocessed microporous array fluorescence image for the nth small region recognition, where the iteration number n is a positive integer; performing binarization processing on the grayscale image for the nth small region recognition to obtain the nth binarized image; extracting small region information from the nth binarized image to obtain the recognition region information of the nth recognition region; when the recognition region information meets the preset screening conditions, using the nth recognition region as a reaction unit in the preprocessed microporous array fluorescence image; setting the pixel value of the nth recognition region in the nth binarized image to a first value to obtain a grayscale image for the (n+1)th small region recognition; when all reaction units in the preprocessed microporous array fluorescence image have undergone binarization recognition screening processing, using the recognition region information of each recognition region as the first coded microsphere information corresponding to the preprocessed microporous array fluorescence image.

[0059] In one embodiment, when the identified region information meets preset screening conditions, the nth identified region is used as a reaction unit in the preprocessed micro-well array fluorescence image. The preset screening conditions include region area screening conditions, eccentricity screening conditions, roundness screening conditions, and convex hull ratio screening conditions. The region area screening condition is: the area of ​​the identified region is greater than the minimum allowable area minValidArea and less than the maximum allowable area maxValidArea; where the minimum allowable area minValidArea = D index ×0.75×R real 2 ×π, maximum allowed area maxValidArea=D index ×1.25×R real 2 ×π, R real D is the actual average radius of the microsphere. index The attenuation coefficient is... baseThresh is the initial binarization threshold, and curThresh is the threshold value. nHere, `curThresh` represents the actual threshold in the nth binarization process, and `BitDepth` represents the image bit depth. The eccentricity selection condition is: when the actual threshold `curThresh` in the nth binarization process... n When the eccentricity of the recognition region is less than the preset binarization threshold, the allowed range is: validEcc∈[0.75,2]; when curThresh is not less than the preset binarization threshold, the allowed range is: validEcc∈[0.75×D]. index [2]; The roundness filtering condition is: Roundness∈[0.75,2], and the roundness of the recognition region is Roundness=4×Π×Area / (perimeter). 2 Where perimeter is the perimeter of the identified region and area is the area of ​​the identified region; the convex hull rate filtering condition is: the convex hull rate of the identified region is greater than the convex hull rate threshold.

[0060] In one embodiment, the method of identifying the first coded microsphere information corresponding to the preprocessed micropore array fluorescence image using a local multi-threshold segmentation and recognition method further includes: when the identification region information does not meet the preset screening conditions, obtaining the binarized threshold curThresh for the (n+1)th small region identification. n+1 The binarization threshold curThresh for the (n+1)th small region identification is used. n+1 = baseThresh + (n+1) × stepValue, where baseThresh is the initial binarization threshold and stepValue is the step gradient value.

[0061] In this embodiment, during the effective region selection process, the effective region after binarization gradually decreases as the binarization threshold curThresh increases. Therefore, this application proposes a dynamic selection equation based on the binarization threshold curThresh: First, the attenuation coefficient Dindex of pixel area, convexity-to-area ratio, eccentricity, and roundness is calculated. Then, based on the attenuation coefficient Dindex, the selection conditions for potential reaction unit regions under different binarization thresholds curThresh can be determined: region area selection condition, eccentricity selection condition, roundness selection condition, and convex hull ratio selection condition. Small recognition regions that do not meet the above selection conditions are input into the next round of recognition. For each sub-region, the binarization threshold for the next round is set as curThresh = baseThresh + (n + 1) × stepValue. Then, the binarized sub-region information is extracted and selected using the above selection conditions.

[0062] The image recognition process ends when no information about any small region is identified after binarization, or when the number of iterations StepCount equals the maximum number of iterations MaxStepCount.

[0063] Specifically, in this embodiment, the structure variable `validRegions` is first initialized to store the information of the filtered valid reaction unit regions. Then, the initial binarization threshold `baseThresh`, the step gradient value `stepValue`, and the maximum number of iterations `maxStepCount` are set. For the nth (n = 1, 2, 3, ..., maxStepCount) small region recognition, the input grayscale image is binarized multiple times, and the binarization threshold is set as `curThresh = baseThresh + n × stepValue`. That is, as the number of iterations within a small region increases, the binarization threshold `curThresh` will also increase accordingly. In this embodiment, binarization refers to the fact that the value of a pixel in a grayscale image is between 0 and 65535 (16-bit depth). After binarization, the value of a pixel in the binarized image has only two values: 0 and 1, where 0 represents the background and 1 represents the valid signal.

[0064] In this embodiment, after each binarization, small region information is extracted from the obtained binarized image (in the binarized image, 0 signal is the background and 1 signal is the effective pixel point). The pixel area, minimum convex area ratio, eccentricity (major axis to minor axis ratio), roundness, centroid coordinates, and bounding box coordinates of the recognition region are calculated respectively. All data results can be stored in computer memory as a list of structures.

[0065] Next, the obtained recognition region information (structure list) is filtered and eliminated according to the following method. Recognition regions that meet all the following filtering conditions are considered reaction units in the image, and their data is saved to the structure list variable `validRegions`. Then, using the region's BoundingBox information, the pixel values ​​of the corresponding region in the grayscale image are set to 0 using an assignment method, meaning only image information that does not meet the current filtering conditions is retained.

[0066] In this embodiment, during the recognition process of a single binarized image, regions that do not meet the current filtering criteria will proceed to the next round of recognition: the recognition count variable StepCount is incremented by 1, that is, the value of curThresh is increased to curThresh = baseThresh + (n + 1) × stepValue. The input image is then re-binarized, and small region information extraction and filtering processes are performed within the image.

[0067] When the number of iterations (n) for a certain region equals the set maximum number of iterations (maxStepCount), the program ends the recognition process and returns a list of summarized validRegions structures. Small regions in the binarized image below this threshold (curThresh = baseThresh + maxStepCount × stepValue) that still do not meet the following filtering conditions will not be further recognized. After all potential response units in the image have undergone binarization recognition and filtering, the processing system returns validRegions variable information.

[0068] S230. Extract the second fluorescence detection information from the droplet fluorescence image using coded microspheres as carriers and droplets as the basis.

[0069] like Figure 2b As shown, the left side is the encoded signal image of the coded microspheres, and the right side is the report signal image of the probe in the droplet. In one embodiment, the droplet fluorescence image is subjected to multiple digital detection information extraction based on the droplet, using the coded microspheres as a carrier, to obtain second fluorescence detection information. The droplet fluorescence image includes the encoded signal image of the coded microspheres and the report signal image of the probe in the droplet, including: based on a local multi-threshold segmentation and recognition method, performing region extraction on the encoded signal image of the coded microspheres and the report signal image of the probe in the droplet respectively to obtain droplet fluorescence image information.

[0070] In one embodiment, the fluorescence image detection method further includes: determining the state of the droplet-encapsulated microspheres; the droplet fluorescence image information includes coded microsphere fluorescence information and droplet information; based on the coded microsphere fluorescence information, a corrected microsphere centroid position parameter, denoted as B(x1, y1), is extracted from the microsphere structure identified from the coded signal image containing the coded microsphere; based on the droplet information, a corrected center coordinate position parameter, denoted as D(x0, y0), is obtained from the report signal image containing the probe in the droplet; when the corrected microsphere centroid position parameter satisfies... At that time, it was determined that the identified microsphere structure was encapsulated by the droplet, R drop Let d0 and h0 be the radius of a single droplet, respectively, and θ be the coordinate correction coefficients. d0 and h0 are the coordinate correction coefficients to reduce image shift caused by filters with different cutoff frequencies during shooting. θ is the distance coefficient, which can be adjusted according to the actual droplet size and uniformity.

[0071] In this embodiment, for a single coded microsphere, a reliable calculation method (small area distance extraction + Euclidean distance determination) is used to determine the microsphere and the droplet in which it is located.

[0072] In one embodiment, the droplet fluorescence image information includes the number of coded microspheres encapsulated by the droplet. Determining the state of the microspheres encapsulated by the droplet further includes: when determining the state of the microspheres encapsulated by the droplet based on the number of coded microspheres encapsulated by the droplet, the droplet is an empty droplet, encapsulates a single microsphere, or encapsulates multiple microspheres.

[0073] In the process of digital droplet image detection, this application realizes the segmentation, recognition, and parameter extraction of a single droplet in a fluorescence image. In addition, it also realizes the discrimination function of the state of droplet encapsulating coded microspheres (no microspheres encapsulated, encapsulating a single coded microsphere, or encapsulating two or more coded microspheres), further improving the accuracy of microsphere statistical results.

[0074] S240. Obtain the fluorescence detection information corresponding to the sample to be tested through the first fluorescence detection information and the second fluorescence detection information.

[0075] Currently, there are no commercially available devices capable of detecting multiplex digital droplets based on coded microspheres, nor can coded microspheres be used to detect target molecules. Although Quanterix's SiMOA technology (products such as QH-1) can achieve multiplex digital detection based on coded microspheres using microporous arrays as carriers, according to publicly available data (literature, patent descriptions), it does not yet possess accurate detection capabilities at high coding multiplicity.

[0076] This application enables the acquisition of fluorescence images corresponding to the sample to be tested. These fluorescence images include fluorescence images of a micropore array containing coded microspheres and fluorescence images of droplets. Then, based on a local multi-threshold segmentation and recognition method, multiple digital detection information is extracted from the micropore array fluorescence image using coded microspheres as carriers, yielding first fluorescence detection information. Simultaneously, multiple digital detection information is extracted from the droplet fluorescence image using coded microspheres as carriers, yielding second fluorescence detection information. Finally, the fluorescence detection information corresponding to the sample to be tested is obtained through the first and second fluorescence detection information. This application achieves accurate identification and analysis of potential reaction units in the preprocessed image.

[0077] This application utilizes a fluorescence image detection method to identify the position and boundaries of coded microspheres encapsulated in droplets and suspended in the reaction solution of a microporous array, as well as extract signals such as fluorescence intensity. This application solves the problem of accurately identifying high-density reaction unit (coded microspheres, droplets) signals in fluorescence images during high-throughput multiplex digital detection, and the problem of accurately interpreting the state of droplet-encapsulated microspheres (empty droplets, single-microsphere encapsulation, multiple-microsphere encapsulation) during multiplex digital droplet detection using coded microspheres as carriers. This application not only achieves accurate identification of the positional information of droplets and microporous array reaction units in fluorescence images, but also accurately distinguishes the single-microsphere encapsulation and multiple-microsphere encapsulation states of reaction units, achieving higher data statistical accuracy. This application is also applicable to the analysis of reaction chamber image signals captured by CMOD or CCD camera units, and does not rely on reagents such as fluorescein for auxiliary imaging, reducing experimental complexity.

[0078] The protection scope of the fluorescence image detection method in this application is not limited to the execution order of the steps listed in this embodiment. Any solution implemented by adding, deleting, or replacing steps in the prior art based on the principles of this application is included within the protection scope of this application.

[0079] This application also provides a fluorescence image detection system, which can implement the fluorescence image detection method of this application. However, the implementation device of the fluorescence image detection method of this application includes, but is not limited to, the structure of the fluorescence image detection system listed in this embodiment. All structural modifications and substitutions of the prior art made in accordance with the principles of this application are included within the protection scope of this application.

[0080] like Figure 3 As shown, this application also provides a fluorescence image detection system, including an image acquisition module 310, a first image detection module 320, a second image detection module 330, and an information acquisition module 340. The image acquisition module 310 is configured to acquire a fluorescence image corresponding to the sample to be detected, the fluorescence image including a fluorescence image of a micropore array containing coded microspheres and a droplet fluorescence image; the first image detection module 320 is configured to extract multiple digital detection information based on the micropore array using coded microspheres as a carrier from the micropore array fluorescence image using a local multi-threshold segmentation and recognition method to obtain first fluorescence detection information; the second image detection module 330 is configured to extract multiple digital detection information based on the droplet using coded microspheres as a carrier from the droplet fluorescence image to obtain second fluorescence detection information; the information acquisition module 340 is configured to obtain fluorescence detection information corresponding to the sample to be detected through the first and second fluorescence detection information.

[0081] This application enables the acquisition of fluorescence images corresponding to the sample to be tested. These fluorescence images include fluorescence images of a micropore array containing coded microspheres and fluorescence images of droplets. Then, based on a local multi-threshold segmentation and recognition method, multiple digital detection information is extracted from the micropore array fluorescence image using coded microspheres as carriers, yielding first fluorescence detection information. Simultaneously, multiple digital detection information is extracted from the droplet fluorescence image using coded microspheres as carriers, yielding second fluorescence detection information. Finally, the fluorescence detection information corresponding to the sample to be tested is obtained through the first and second fluorescence detection information. This application achieves accurate identification and analysis of potential reaction units in the preprocessed image.

[0082] In one embodiment, the image acquisition module 310 includes an image acquisition submodule, which is configured to: detect the sample to be detected using a fluorescence microscope to obtain a fluorescence image corresponding to the sample to be detected.

[0083] In this application, a fluorescence microscope is used to acquire the fluorescence image of the sample to be tested. The microscope mainly includes an excitation light source (including but not limited to LED, xenon lamp, laser, etc.), a filter (to retain light of a specific wavelength, including but not limited to 488nm, 645nm, etc.), a sample stage and a sample to be photographed (including but not limited to microfluidic chips made of PB, PDMS, PMMA materials, which contain fluorescently encoded microspheres and emulsion droplets encapsulating the reaction substrate), an objective lens (a focusing lens, including but not limited to x4 and x10 magnification), and an image sensor (the working principle of the image sensor includes but is not limited to CCD (charge-coupled device) and CMOS (complementary metal-oxide-semiconductor)). This application captures and acquires image data signals (including but not limited to bright field, CY5, and dark field images of RITC / FITC fluorescence channels) of the microfluidic chip sample according to the set parameters, and transmits the image signals to the memory of the computer processing system via a data cable for the next step of signal processing.

[0084] In one embodiment, the first image detection module 320 includes a first image detection submodule, which is configured to: preprocess the micro-pore array fluorescence image to obtain a preprocessed micro-pore array fluorescence image; identify the first coded microsphere information corresponding to the preprocessed micro-pore array fluorescence image using a local multi-threshold segmentation and recognition method, the first coded microsphere information including the coded microsphere position, the coded microsphere boundary and the fluorescence intensity signal; and obtain first fluorescence detection information based on the first coded microsphere information.

[0085] This application proposes a Localized Area Multi-threshold Segmentation (LAMS) method. The LAMS method enables the identification of microspheres in fluorescence images of high-fill-rate encoded microspheres, and achieves accurate identification and analysis of potential reaction units in the preprocessed image.

[0086] In one embodiment, the first image detection submodule includes a preprocessing module configured to perform full-image fluorescence intensity equalization, median filtering, Gaussian filtering, morphological opening operation, and structure dilation on the micropore array fluorescence image.

[0087] In this application, mathematical morphology employs two fundamental operations: erosion and dilation. These operations, when combined, form opening and closing operations. Morphological opening involves erosion followed by dilation, while closing involves dilation followed by erosion. Structural dilation enlarges the target region, merging background points in contact with the target into the target object, thus expanding the target boundary outwards. This serves to fill voids within the target region and eliminate small particle noise contained within it. This application further improves image detection accuracy by preprocessing the micro-aperture fluorescence image.

[0088] In one embodiment, the first image detection submodule includes a recognition module, which is configured to: when performing the nth small-region recognition on the preprocessed micro-well array fluorescence image, acquire a grayscale image corresponding to the preprocessed micro-well array fluorescence image for the nth small-region recognition, where the iteration number n is a positive integer; perform binarization processing on the grayscale image for the nth small-region recognition to obtain the nth binarized image; extract small-region information from the nth binarized image to obtain the recognition region information of the nth recognition region; when the recognition region information meets the preset screening conditions, use the nth recognition region as a reaction unit in the preprocessed micro-well array fluorescence image; set the pixel value of the nth recognition region in the nth binarized image to a first value to obtain a grayscale image for the (n+1)th small-region recognition; when all reaction units in the preprocessed micro-well array fluorescence image have undergone binarization recognition screening processing, use the recognition region information of each recognition region as the first coded microsphere information corresponding to the preprocessed micro-well array fluorescence image.

[0089] In one embodiment, when the identified region information meets preset screening conditions, the nth identified region is used as a reaction unit in the preprocessed micro-well array fluorescence image. The preset screening conditions include region area screening conditions, eccentricity screening conditions, roundness screening conditions, and convex hull ratio screening conditions. The region area screening condition is: the area of ​​the identified region is greater than the minimum allowable area minValidArea and less than the maximum allowable area maxValidArea; where the minimum allowable area minValidArea = D. index ×0.75×R real 2 ×π, maximum allowed area maxValidArea=D index ×1.25×R real 2 ×π, R real D is the actual average radius of the microsphere. index The attenuation coefficient is... baseThresh is the initial binarization threshold, and curThresh is the threshold value. n for The actual threshold in the nth binarization process, where BitDepth is the image bit depth; the eccentricity selection condition is: when the actual threshold in the nth binarization process curThresh... n When the eccentricity of the recognition region is less than the preset binarization threshold, the allowed range is: validEcc∈[0.75,2]; when curThresh is not less than the preset binarization threshold, the allowed range is: validEcc∈[0.75×D]. index [2]; The roundness filtering condition is: Roundness∈[0.75,2], and the roundness of the recognition region is Roundness=4×Π×Area / (perimeter). 2 Where perimeter is the perimeter of the identified region and area is the area of ​​the identified region; the convex hull rate filtering condition is: the convex hull rate of the identified region is greater than the convex hull rate threshold.

[0090] In one embodiment, the identification module further includes an identification submodule, which is configured to: when the identification region information does not meet the preset filtering conditions, obtain the binarization threshold curThresh for performing the (n+1)th small region identification. n+1 The binarization threshold curThresh for the (n+1)th small region identification is used. n+1 = baseThresh + (n+1) × stepValue, where baseThresh is the initial binarization threshold and stepValue is the step gradient value.

[0091] In one embodiment, the droplet fluorescence image includes an encoded signal image of the encoded microsphere and a report signal image of the probe in the droplet. The second image detection module includes a second image detection submodule, which is configured to: extract regions from the encoded signal image of the encoded microsphere and the report signal image of the probe in the droplet based on a local multi-threshold segmentation and recognition method to obtain droplet fluorescence image information.

[0092] In one embodiment, the fluorescence image detection system further includes a first discrimination module, configured to: determine the state of the droplet-encapsulated microspheres; the droplet fluorescence image information includes coded microsphere fluorescence information and droplet information; based on the coded microsphere fluorescence information, extract the corrected microsphere centroid position parameter, denoted as B(x1, y1), from the microsphere structure identified from the coded signal image containing the coded microsphere; based on the droplet information, obtain the corrected center coordinate position parameter of a single droplet structure, denoted as D(x0, y0), from the report signal image containing the probe in the droplet; when the corrected microsphere centroid position parameter satisfies At that time, it was determined that the identified microsphere structure was encapsulated by the droplet, R drop Let θ be the radius of a single droplet, d0 and h0 be the coordinate correction coefficients, and θ be the distance coefficient.

[0093] In one embodiment, the droplet fluorescence image information includes the number of coded microspheres encapsulated by the droplet, and the first discrimination module includes a first discrimination submodule, which is configured to: determine the state of the droplet encapsulating microspheres as an empty droplet, encapsulating a single microsphere, or encapsulating multiple microspheres based on the number of coded microspheres encapsulated by the droplet.

[0094] In the process of digital droplet image detection, this application realizes the segmentation, recognition, and parameter extraction of a single droplet in a fluorescence image. In addition, it also realizes the discrimination function of the state of droplet encapsulating coded microspheres (no microspheres encapsulated, encapsulating a single coded microsphere, or encapsulating two or more coded microspheres), further improving the accuracy of microsphere statistical results.

[0095] In practice, the above modules can be implemented as independent entities or combined in any way to be implemented as the same or several entities. For the specific implementation of the above modules, please refer to the previous method implementation examples, which will not be repeated here.

[0096] As described above, this application utilizes a fluorescence image detection method to identify the position and boundaries of coded microspheres encapsulated in droplets and suspended in the microporous array reaction solution, as well as extract signals such as fluorescence intensity. This application solves the problem of accurately identifying high-density reaction unit (coded microspheres, droplets) signals in fluorescence images during high-throughput multiplex digital detection, and the problem of accurately interpreting the state of droplet-encapsulated microspheres (empty droplets, single-microsphere encapsulation, multiple-microsphere encapsulation) during multiplex digital droplet detection using coded microspheres as carriers. This application not only achieves accurate identification of the positional information of droplets and microporous array reaction units in fluorescence images, but also accurately distinguishes the single-microsphere encapsulation and multiple-microsphere encapsulation states of reaction units, achieving higher data statistical accuracy. This application is also applicable to the analysis of reaction chamber image signals captured by CMOD or CCD camera units, and does not rely on reagents such as fluorescein for auxiliary imaging, reducing experimental complexity.

[0097] In the embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, or methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of modules / units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or units may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection of apparatuses or modules or units may be electrical, mechanical, or other forms.

[0098] The modules / units described as separate components may or may not be physically separate. The components shown as modules / units may or may not be physical modules; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules / units can be selected to achieve the objectives of the embodiments of this application, depending on actual needs. For example, the functional modules / units in the various embodiments of this application may be integrated into one processing module, or each module / unit may exist physically separately, or two or more modules / units may be integrated into one module / unit.

[0099] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0100] This application also provides an electronic device, which can be a terminal, a server, or other similar device. The terminal can be a mobile phone, tablet computer, smart Bluetooth device, laptop computer, personal computer, etc.; the server can be a single server or a server cluster composed of multiple servers, etc.

[0101] In some embodiments, the fluorescence image detection system provided in this application can also be integrated into multiple electronic devices. For example, the fluorescence image detection system can be integrated into multiple servers, and the fluorescence image detection method of this application can be implemented by multiple servers.

[0102] In this embodiment, a server will be used as an example for detailed description. For example, ... Figure 4 As shown, it illustrates a schematic diagram of the server structure involved in an embodiment of this application. Specifically:

[0103] The server may include components such as a processor 410 with one or more processing cores, a memory 420 with one or more computer-readable storage media, a power supply 430, an input module 440, and a communication module 450. Those skilled in the art will understand that... Figure 4 The server architecture shown does not constitute a limitation on the server and may include more or fewer components than illustrated, or combine certain components, or have different component arrangements. Wherein:

[0104] The processor 410 is the control center of the server, connecting various parts of the server through various interfaces and lines. It performs various server functions and processes data by running or executing software programs and / or modules stored in the memory 420, and by calling data stored in the memory 420, thereby providing overall monitoring of the server. In some embodiments, the processor 410 may include one or more processing cores; in some embodiments, the processor 410 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user interface, and applications, and the modem processor mainly handles wireless communication. It is understood that the modem processor may also not be integrated into the processor 410.

[0105] The memory 420 can be used to store software programs and modules. The processor 410 executes various functional applications and data processing by running the software programs and modules stored in the memory 420. The memory 420 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the server, etc. In addition, the memory 420 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 420 may also include a memory controller to provide the processor 410 with access to the memory 420.

[0106] The server also includes a power supply 430 that supplies power to the various components. In some embodiments, the power supply 430 can be logically connected to the processor 410 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. The power supply 430 may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0107] The server may also include an input module 440, which can be used to receive input numeric or character information, and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

[0108] The server may also include a communication module 450. In some embodiments, the communication module 450 may include a wireless module, through which the server can perform short-range wireless transmission, thereby providing users with wireless broadband internet access. For example, the communication module 450 can be used to help users send and receive emails, browse web pages, and access streaming media.

[0109] Although not shown, the server may also include a display unit, etc., which will not be described in detail here. Specifically, in this embodiment, the processor 410 in the server loads the executable files corresponding to the processes of one or more applications into the memory 420 according to the following instructions, and the processor 410 runs the applications stored in the memory 420, thereby realizing various functions of the fluorescence image detection system.

[0110] In some embodiments, this application also provides a computer-readable storage medium. Those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing a processor, and the program can be stored in a computer-readable storage medium. The storage medium is a non-transitory medium, such as random access memory, read-only memory, flash memory, hard disk, solid-state drive, magnetic tape, floppy disk, optical disk, and any combination thereof. The above storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., digital video disc (DVD)), or a semiconductor medium (e.g., solid-state disk (SSD)).

[0111] This application embodiment may also provide a computer program product, which includes one or more computer instructions. When the computer instructions are loaded and executed on a computing device, all or part of the flow or function according to the embodiments of this application is generated. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, or data center to another website, computer, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means.

[0112] When the computer program product is executed by a computer, the computer performs the method described in the foregoing method embodiments. The computer program product may be a software installation package; when the foregoing method is required, the computer program product can be downloaded and executed on the computer.

[0113] The descriptions of the processes or structures corresponding to the above figures each have their own emphasis. For parts of a process or structure that are not described in detail, please refer to the relevant descriptions of other processes or structures.

[0114] The above embodiments are merely illustrative of the principles and effects of this application and are not intended to limit this application. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in this application should still be covered by the claims of this application.

Claims

1. A fluorescence image detection method, characterized in that, The method includes: Acquire fluorescence images corresponding to the sample to be tested, including fluorescence images of a microporous array containing coded microspheres and fluorescence images of droplets; Based on a local multi-threshold segmentation and recognition method, multiple digital detection information is extracted from the micropore array fluorescence image using coded microspheres as carriers to obtain first fluorescence detection information. This includes: preprocessing the micropore array fluorescence image to obtain a preprocessed micropore array fluorescence image; identifying the first coded microsphere information corresponding to the preprocessed micropore array fluorescence image using the local multi-threshold segmentation and recognition method; the first coded microsphere information includes the coded microsphere position, coded microsphere boundary, and fluorescence intensity signal; and obtaining the first fluorescence detection information based on the first coded microsphere information. The droplet fluorescence image is subjected to multiple digital detection information extraction based on the droplet using coded microspheres as carriers to obtain second fluorescence detection information; the droplet fluorescence image includes the coded signal image with coded microspheres and the report signal image with probes in the droplet. Based on the local multi-threshold segmentation and recognition method, the coded signal image with coded microspheres and the report signal image with probes in the droplet are extracted into regions to obtain droplet fluorescence image information; The fluorescence detection information corresponding to the sample to be tested is obtained by using the first fluorescence detection information and the second fluorescence detection information.

2. The method according to claim 1, characterized in that, The acquisition of the fluorescence image corresponding to the sample to be tested includes: The sample to be tested is detected by a fluorescence microscope to obtain a fluorescence image of the sample.

3. The method according to claim 1 or 2, characterized in that, The step of preprocessing the micropore array fluorescence image to obtain a preprocessed micropore array fluorescence image includes: The fluorescence image of the micropore array is subjected to full-image fluorescence intensity equalization, median filtering, Gaussian filtering, morphological opening operation, and structure dilation.

4. The method according to claim 1 or 2, characterized in that, The step of identifying the first coded microsphere information corresponding to the preprocessed micropore array fluorescence image using the local multi-threshold segmentation recognition method includes: When performing the nth small region identification on the preprocessed micro-pore array fluorescence image, a grayscale image corresponding to the preprocessed micro-pore array fluorescence image is obtained for the nth small region identification, and the iteration number n is a positive integer; The grayscale image used for the nth small region recognition is binarized to obtain the nth binarized image; Small region information is extracted from the nth binarized image to obtain the recognition region information of the nth recognition region; When the identification region information meets the preset screening conditions, the nth identification region is used as a reaction unit in the preprocessed micro-pore array fluorescence image; The pixel values ​​of the nth recognition region in the nth binarized image are set to the first value to obtain a grayscale image for the (n+1)th small region recognition. When all reaction units in the preprocessed microporous array fluorescence image have undergone binarization recognition and screening, the recognition area information of each recognition region is used as the first coded microsphere information corresponding to the preprocessed microporous array fluorescence image.

5. The method according to claim 4, characterized in that, When the identification region information meets the preset screening conditions, the nth identification region is used as a reaction unit in the preprocessed micro-well array fluorescence image. The preset screening conditions include region area screening conditions, eccentricity screening conditions, roundness screening conditions, and convex hull ratio screening conditions. The area screening condition is: the area of ​​the identified region is greater than the minimum allowed area. And smaller than the maximum allowable area Wherein, the minimum allowable area The maximum allowable area , for , The attenuation coefficient is... , The initial binarization threshold is used. for The actual threshold in the nth binarization process, where BitDepth is the image bit depth; The centrifugation rate screening condition is: when the actual threshold is reached during the nth binarization process... When the eccentricity of the recognition region is less than the preset binarization threshold, the allowable range is as follows: When curThresh is not less than the preset binarization threshold, the allowed eccentricity range of the recognition region is: ; The roundness filtering condition is: Roundness ∈ [0.75, 2], and the roundness of the recognition region is Roundness = 4 × ×Area / (perimeter) 2 Where perimeter is the perimeter of the identified region and area is the area of ​​the identified region; The convex hull rate screening condition is: the convex hull rate of the identified region is greater than the convex hull rate threshold.

6. The method according to claim 4, characterized in that, The step of identifying the first coded microsphere information corresponding to the preprocessed micropore array fluorescence image through the local multi-threshold segmentation recognition method further includes: When the identified region information does not meet the preset filtering conditions, the binarization threshold curThresh for performing the (n+1)th small region identification is obtained. n+1 The binarization threshold curThresh used for the (n+1)th small region identification is... n+1 = baseThresh + (n+1) × stepValue, where baseThresh is the initial binarization threshold and stepValue is the step gradient value.

7. The method according to claim 1, characterized in that, The method further includes: determining the state of the droplets encapsulating the microspheres; The droplet fluorescence image information includes coded microsphere fluorescence information and droplet information. Based on the coded microsphere fluorescence information, the corrected microsphere centroid position parameter is extracted from the microsphere structure identified from the coded signal image of the coded microsphere, and denoted as B(x1, y1). Based on the droplet information, the corrected center coordinate position parameters of a single droplet structure are obtained from the reported signal image of the probe in the droplet, denoted as D(x0, y0). When the corrected microsphere centroid position parameter satisfies At that time, it was determined that the identified microsphere structure was encapsulated by the droplet. Let be the radius of a single droplet, and d0 and h0 be the correction factors for the coordinates, respectively. This is the distance coefficient.

8. The method according to claim 7, characterized in that, The droplet fluorescence image information includes the number of coded microspheres encapsulated by the droplet, and the determination of the state of the microspheres encapsulated by the droplet further includes: The state of the droplet-encapsulated microspheres is determined based on the number of microspheres encapsulated in the droplet, whether it is an empty droplet, encapsulates a single microsphere, or encapsulates multiple microspheres.

9. A fluorescence image detection system, characterized in that, The system includes: The image acquisition module is used to acquire the fluorescence image corresponding to the sample to be detected. The fluorescence image includes a fluorescence image of a micropore array containing coded microspheres and a fluorescence image of droplets. The first image detection module is used to extract multiple digital detection information based on the micropore array fluorescence image using coded microspheres as carriers and the micropore array as a basis, based on a local multi-threshold segmentation and recognition method, to obtain first fluorescence detection information. This includes: preprocessing the micropore array fluorescence image to obtain a preprocessed micropore array fluorescence image; identifying the first coded microsphere information corresponding to the preprocessed micropore array fluorescence image using the local multi-threshold segmentation and recognition method; the first coded microsphere information includes the position of the coded microsphere, the boundary of the coded microsphere, and the fluorescence intensity signal; and obtaining the first fluorescence detection information based on the first coded microsphere information. The second image detection module is used to extract multiple digital detection information based on the droplet using coded microspheres as carriers from the droplet fluorescence image to obtain second fluorescence detection information. The droplet fluorescence image includes a coded signal image with coded microspheres and a report signal image with probes in the droplet. Based on the local multi-threshold segmentation and recognition method, the coded signal image with coded microspheres and the report signal image with probes in the droplet are extracted into regions to obtain droplet fluorescence image information. The information acquisition module is used to obtain the fluorescence detection information corresponding to the sample to be tested through the first fluorescence detection information and the second fluorescence detection information.

10. An electronic device, characterized in that, The electronic device includes: The memory stores multiple instructions; A processor that loads instructions from the memory to perform the steps of the method as described in any one of claims 1 to 8.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by an electronic device, it performs the steps of the method described in any one of claims 1 to 8.

Images