Fluorescence crosstalk correction method, device and equipment

Abstract

The embodiment of the invention provides a fluorescence crosstalk correction method, device and equipment. The method comprises the following steps: acquiring a four-channel image set of a to-be-detected sample; determining the concentration corresponding to the to-be-detected sample according to the image set and a predetermined first concentration threshold, a predetermined second concentration threshold and a predetermined third concentration threshold; determining a target channel corresponding to the to-be-tested sample according to the image set; and performing crosstalk correction according to predetermined crosstalk logic, the target channel corresponding to the to-be-detected sample, the concentration corresponding to the to-be-detected sample and the predetermined correction coefficient of each channel under each concentration. The method provided by the embodiment of the invention can simply and conveniently identify the target channel and has high-precision four-channel crosstalk correction capability.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of molecular biotechnology, and specifically to a fluorescence crosstalk correction method, device and equipment. Background Art

[0002] Digital polymerase chain reaction (PCR) is an emerging technology for quantitative and precise detection of nucleic acids. The diluted nucleic acid template is distributed into a large number of different reaction units, so that each reaction unit has one or no nucleic acid; then detectable fluorescence is added during PCR amplification; when the amplification is completed, the number of fluorescence occurrences in each reaction unit is collected using statistical methods to quantitatively detect the nucleic acid concentration in the sample.

[0003] Compared with previous second-generation sequencing, gene chips and quantitative PCR, digital PCR has the advantages of high sensitivity and absolute quantification. The sensitivity of digital PCR can reach a single nucleic acid molecule and the lowest detection line is 0.001%. At the same time, there is no need for standard curves and reference genes to compare to determine the amount of nucleic acid. The starting amount of the target molecule can be absolutely quantified, and the number of deoxyribonucleic acid (DNA) molecules can be directly obtained. Based on the above advantages, digital PCR technology has been widely used in tumor liquid biopsy, targeted therapy companion diagnosis, non-invasive prenatal testing, pathogenic microorganisms (viruses, bacteria, fungi, etc.) detection, genetic disease diagnosis, drug genome testing, gene expression analysis, environmental testing and molecular ecology and other research directions.

[0004] Fluorescent probes are mainly composed of three parts: fluorescent groups, linking groups and recognition groups. Among them, the linking group is used to connect the recognition group and the fluorescent group. After the recognition group binds to the analyte, the linking group transmits the recognition information to the fluorescent group, thereby generating a change in the fluorescent signal. Multi-channel nucleic acid detection technology requires that the digital PCR reaction system can be compatible with more fluorescent probes / dyes. However, due to the small wavelength difference of commonly used nucleic acid fluorescent probes or the intersection of wavelength ranges, fluorescence crosstalk will be caused during fluorescence detection. At present, in the four-channel detection process, the target channel will crosstalk with other channels, which will affect the judgment of the positive and negative nature of the molecule. Therefore, there is an urgent need for a fluorescence crosstalk correction method for digital PCR four-channel detection. Summary of the invention

[0005] The embodiments of the present invention provide a fluorescence crosstalk correction method, device and equipment to solve the problem of crosstalk in the existing four-channel detection process.

[0006] In a first aspect, an embodiment of the present invention provides a fluorescence crosstalk correction method, comprising:

[0007] Acquire a four-channel image set of the sample to be tested, the image set including a FAM channel image, a HEX channel image, a ROX channel image and a CY-5 channel image;

[0008] Determine the concentration corresponding to the sample to be tested according to the image set and a predetermined first concentration threshold, a second concentration threshold, and a third concentration threshold;

[0009] Determine the target channel corresponding to the sample to be tested according to the image set;

[0010] Crosstalk correction is performed according to a predetermined crosstalk logic, a target channel corresponding to the sample to be tested, a concentration corresponding to the sample to be tested, and a predetermined correction coefficient of each channel at each concentration.

[0011] In one embodiment, the crosstalk logic includes:

[0012] If the FAM channel is the target channel, then the HEX channel has major crosstalk at the first concentration;

[0013] If the HEX channel is the target channel, there is major crosstalk in the ROX channel at the first and second concentrations;

[0014] If the ROX channel is the target channel, then the CY-5 channel has major crosstalk at the first concentration.

[0015] In one embodiment, crosstalk correction is performed according to a predetermined crosstalk logic, a target channel corresponding to the sample to be tested, a concentration corresponding to the sample to be tested, and a predetermined correction coefficient of each channel at each concentration, including:

[0016] When the concentration corresponding to the sample to be tested is the first concentration and the target channel corresponding to the sample to be tested is the FAM channel, a correction matrix is ​​determined according to the correction coefficient of the FAM channel at the first concentration and the FAM channel image, and a correction image of the HEX channel is obtained by subtracting the HEX channel image from the correction matrix;

[0017] When the concentration corresponding to the sample to be tested is the first concentration and the target channel corresponding to the sample to be tested is the HEX channel, a correction matrix is ​​determined according to the correction coefficient of the HEX channel at the first concentration and the HEX channel image, and a correction image of the ROX channel is obtained by subtracting the ROX channel image from the correction matrix;

[0018] When the concentration corresponding to the sample to be tested is the second concentration and the target channel corresponding to the sample to be tested is the HEX channel, a correction matrix is ​​determined according to the correction coefficient of the HEX channel at the second concentration and the HEX channel image, and a correction image of the ROX channel is obtained by subtracting the ROX channel image from the correction matrix;

[0019] When the concentration corresponding to the sample to be tested is the first concentration and the target channel corresponding to the sample to be tested is the ROX channel, the correction matrix is ​​determined according to the correction coefficient of the ROX channel at the first concentration and the ROX channel image, and the CY-5 channel image is subtracted from the correction matrix to obtain the correction image of the CY-5 channel.

[0020] In one embodiment, determining a target channel corresponding to a sample to be tested according to an image set includes:

[0021] Determine the information entropy of each channel image in the image set;

[0022] The channel corresponding to the channel image with the largest information entropy is determined as the target channel corresponding to the sample to be tested.

[0023] In one embodiment, the correction coefficients for each channel at each concentration are determined according to the following steps:

[0024] Acquire multiple images of the target channel and its corresponding main crosstalk channel at each concentration;

[0025] For each image of the target channel, locate the position of the droplet and calculate the droplet gray value based on the droplet center according to the preset radius and calculate the global droplet gray value to obtain the target channel droplet gray value;

[0026] For each image of the main crosstalk channel, locate the position of the droplet and calculate the droplet gray value based on the center of the droplet according to the preset radius and calculate the global droplet gray value to obtain the droplet gray value of the main crosstalk channel;

[0027] The ratio of the average grayscale value of the droplets in the main crosstalk channel to the average grayscale value of the droplets in the target channel is determined as the correction coefficient of the target channel to the main crosstalk channel at this concentration.

[0028] In one embodiment, determining the concentration corresponding to the sample to be tested according to the image set and the predetermined first concentration threshold, second concentration threshold, and third concentration threshold includes:

[0029] The number of pixel values ​​greater than the first concentration threshold in each channel image in the image set is obtained respectively, and the number of channels whose number is greater than the preset threshold is counted, and when the number of channels is 1, the concentration corresponding to the sample to be tested is determined to be the first concentration;

[0030] Respectively obtain the number of pixel values ​​greater than the second concentration threshold in each channel image in the image set, and count the number of channels greater than the preset threshold, and when the number of channels is 1, determine that the concentration corresponding to the sample to be tested is the second concentration;

[0031] The number of pixel values ​​greater than the third concentration threshold in each channel image in the image set is obtained respectively, and the number of channels greater than the preset threshold is counted, and when the number of channels is 1, it is determined that the concentration corresponding to the sample to be tested is the third concentration.

[0032] In one embodiment, the first concentration threshold, the second concentration threshold and the third concentration threshold are determined according to the following steps:

[0033] Multiple channel images at each concentration are obtained respectively, and the gray value histogram of each channel image is determined. The average of the highest and second highest gray values ​​in the histogram is used as the segmentation threshold of the channel image. The average value of the segmentation thresholds of the multiple channel images at each concentration is determined as the concentration threshold corresponding to the concentration.

[0034] In one embodiment, before determining the concentration corresponding to the sample to be tested, the method further includes: performing denoising processing on each channel image in the image set in sequence, and the denoising processing is performed according to the following steps:

[0035] Grayscale the image to be processed to obtain a single-channel grayscale image;

[0036] Perform mean filtering on the grayscale image to obtain a mean filtered image;

[0037] The data types of the grayscale image and the mean filter image are converted into floating-point types, and a floating-point noise image is obtained according to the floating-point grayscale image and the floating-point mean filter image;

[0038] The pixel values ​​in the noise image that are less than 0 are truncated and the data type of the noise image is converted to an unsigned integer;

[0039] Perform threshold segmentation on the noise image to separate the noise pixels and obtain a binary segmented image;

[0040] The segmented image is sequentially subjected to morphological erosion, morphological opening and morphological dilation operations to obtain the corresponding morphological image, and the segmented image and its corresponding morphological image are subjected to XOR operation to obtain the noise mask image;

[0041] Perform an AND operation on the mean filter image and the noise mask image to obtain a filtered noise region image;

[0042] Determine a non-noise mask image according to the noise mask image, and use the non-noise mask image to obtain a non-noise region image from the grayscale image;

[0043] The filtered noise area image and the non-noise area image are added and fused.

[0044] In a second aspect, an embodiment of the present invention provides a fluorescence crosstalk correction device, comprising:

[0045] An acquisition module is used to obtain a four-channel image set of the sample to be tested, the image set including a FAM channel image, a HEX channel image, a ROX channel image and a CY-5 channel image;

[0046] An analysis module, used to determine the concentration corresponding to the sample to be tested according to the image set and a predetermined first concentration threshold, a second concentration threshold, and a third concentration threshold;

[0047] A processing module, used to determine the target channel corresponding to the sample to be tested according to the image set;

[0048] The correction module is used to perform crosstalk correction according to the predetermined crosstalk logic, the target channel corresponding to the sample to be tested, the concentration corresponding to the sample to be tested, and the correction coefficient of each channel at each predetermined concentration.

[0049] In a third aspect, an embodiment of the present invention provides an electronic device, including:

[0050] at least one processor and memory;

[0051] Memory stores computer-executable instructions;

[0052] At least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor performs the fluorescence crosstalk correction method as described in any one of the first aspects.

[0053] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer execution instructions, and when the computer execution instructions are executed by a processor, they are used to implement the fluorescence crosstalk correction method as described in any one of the first aspects.

[0054] The fluorescence crosstalk correction method, device and equipment provided by the embodiment of the present invention obtain a four-channel image set of the sample to be tested; and determine the concentration corresponding to the sample to be tested according to the image set and the predetermined first concentration threshold, second concentration threshold and third concentration threshold; and determine the target channel corresponding to the sample to be tested according to the image set; finally, according to the predetermined crosstalk logic, the target channel corresponding to the sample to be tested, the concentration corresponding to the sample to be tested and the correction coefficient of each channel at each predetermined concentration, crosstalk correction is performed, which can simply and conveniently identify the target channel and the crosstalk channel, and has a high-precision four-channel crosstalk correction capability. In the current industry standard of 10%, the method provided by the present application can reduce the fluorescence crosstalk intensity to less than 1%. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0056] Figure 1 A flow chart of a fluorescence crosstalk correction method provided by an embodiment of the present invention;

[0057] Figure 2 A schematic diagram of the structure of a fluorescence crosstalk correction device provided by an embodiment of the present invention;

[0058] Figure 3 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention.

[0059] The above drawings have shown clear embodiments of the present invention, which will be described in more detail below. These drawings and text descriptions are not intended to limit the scope of the present invention in any way, but to illustrate the concept of the present invention for those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0060] The present invention is further described in detail below by specific embodiments in conjunction with the accompanying drawings. Wherein similar elements in different embodiments adopt associated similar element numbers. In the following embodiments, many detailed descriptions are for making the present application better understood. However, those skilled in the art can easily recognize that some features can be omitted in different situations, or can be replaced by other elements, materials, methods. In some cases, some operations related to the present application are not shown or described in the specification, this is to avoid the core part of the present application being overwhelmed by too much description, and for those skilled in the art, it is not necessary to describe these related operations in detail, and they can fully understand the related operations according to the description in the specification and the general technical knowledge in the art.

[0061] In addition, the features, operations or characteristics described in the specification can be combined in any appropriate manner to form various implementations. At the same time, the steps or actions in the method description can also be interchanged or adjusted in a manner that is obvious to those skilled in the art. Therefore, the various sequences in the specification and the drawings are only for the purpose of clearly describing a certain embodiment and are not meant to be a required sequence, unless otherwise specified that a certain sequence must be followed.

[0062] The serial numbers of the components in this document, such as "first", "second", etc., are only used to distinguish the objects described and do not have any order or technical meaning. The "connection" and "coupling" mentioned in this application, unless otherwise specified, include direct and indirect connections (couplings).

[0063] The crosstalk problem in the four-channel detection process is affected by the molecular concentration, camera exposure time and temperature, etc., resulting in different target fluorescence intensity and crosstalk intensity. Due to different concentrations, the crosstalk situation of the target channel to the non-target channel is different (then its correction coefficient should also be adjusted accordingly according to the crosstalk situation), so the concentration is of great significance for the correction of fluorescence crosstalk in the four-channel detection. First, the concentration involved in this application is explained. The numerical values ​​of the first concentration (i.e., high concentration), the second concentration (i.e., medium concentration) and the third concentration (i.e., low concentration) in this application meet the following quantitative relationship: the first concentration> the second concentration> the third concentration, and the first concentration is two to four times the second concentration, and the second concentration is two to four times the third concentration. The first concentration threshold is the threshold under the first concentration, the second concentration threshold is the threshold under the second concentration, and the third concentration threshold is the threshold under the third concentration. It can be understood that the first concentration threshold> the second concentration threshold> the third concentration threshold.

[0064] Figure 1 Flow chart of a fluorescence crosstalk correction method provided by one embodiment of the present invention. Figure 1 As shown, the fluorescence crosstalk correction method provided in this embodiment may include:

[0065] S101 , obtaining a four-channel image set of a sample to be tested, the image set including a FAM channel image, a HEX channel image, a ROX channel image and a CY-5 channel image.

[0066] In this embodiment, the existing image acquisition method of the digital PCR reaction system can be used to obtain image data of the sample to be tested. For example, the laser emitted by the excitation light source can be passed through a converging lens and a filter to illuminate and excite the sample to be tested to produce fluorescence, and fluorescence imaging is used. In this embodiment, images of four channels, FAM, HEX, ROX and CY-5, are collected. Therefore, the four-channel image set of the sample to be tested obtained in this embodiment can specifically include: FAM channel image, HEX channel image, ROX channel image and CY-5 channel image.

[0067] S102 : Determine the concentration corresponding to the sample to be tested according to the image set and a predetermined first concentration threshold, a second concentration threshold, and a third concentration threshold.

[0068] Accurately determining the concentration corresponding to the sample to be tested is a prerequisite for high-precision correction of fluorescence crosstalk. In an optional implementation, after obtaining the first concentration threshold, the second concentration threshold, and the third concentration threshold, the obtained thresholds can be used to perform a threshold segmentation operation on the four-channel grayscale image to determine the concentration corresponding to the sample to be tested. Specifically, the following operation process can be followed:

[0069] First, the first concentration threshold (high concentration threshold) is used to perform threshold segmentation on the four different channel images, the grayscale values ​​greater than the first concentration threshold are set to 255, and the other grayscale values ​​are set to 0, and then the number of non-zero grayscale values ​​in each channel image is counted to obtain the number of non-zero grayscale values ​​in the FAM channel image under the first concentration threshold (Count_Pixel_FAM1), the number of non-zero grayscale values ​​in the HEX channel image under the first concentration threshold (Count_Pixel_HEX1), the number of non-zero grayscale values ​​in the ROX channel image under the first concentration threshold (Count_Pixel_ROX1), and the number of non-zero grayscale values ​​in the CY-5 channel image under the first concentration threshold (Count_Pixel_CY-5-1), and then the number of channels that satisfy the number of non-zero pixel values ​​greater than 900000 is counted (for example, when Count_Pixel_FAM1>900000, the number of channels is increased by 1). If the number of channels is greater than 1, it is a multi-target channel and no fluorescence correction is performed. If the number of channels is 1, the target channel is determined to be in a high concentration state, that is, the concentration corresponding to the sample to be tested is the first concentration, and the concentration judgment process ends. If the number of channels is 0, it is determined that the target channel is not in a high concentration state, and it is necessary to continue to judge whether other concentration conditions are met.

[0070] If it is not a high concentration state, the second concentration threshold (medium concentration threshold) is used to perform threshold segmentation on the four different channel images based on the above idea, and the grayscale values ​​greater than the second concentration threshold are set to 255, and the other grayscale values ​​are set to 0, and then the number of non-zero grayscale values ​​in each channel image is counted to obtain the number of non-zero grayscale values ​​in the FAM channel image under the second concentration threshold (Count_Pixel_FAM2), the number of non-zero grayscale values ​​in the HEX channel image under the second concentration threshold (Count_Pixel_HEX2), the number of non-zero grayscale values ​​in the ROX channel image under the second concentration threshold (Count_Pixel_ROX2) and the number of non-zero grayscale values ​​in the CY-5 channel image under the second concentration threshold (Count_Pixel_CY-5-2), and then the number of channels that meet the non-zero pixel value number greater than 900000 is counted. If the number of channels is greater than 1, it is a multi-target channel without fluorescence correction processing. If the number of channels is 1, it is determined that the target channel is in a medium concentration state, that is, the concentration corresponding to the sample to be tested is the second concentration, and the concentration judgment process ends. If the number of channels is 0, it is determined that the target channel is not in a medium concentration state, and it is necessary to continue to determine whether other concentrations meet the conditions.

[0071] If it is not in the medium concentration state, the third concentration threshold (low concentration threshold) is used to perform threshold segmentation on the four different channel images based on the above idea, and the grayscale values ​​greater than the third concentration threshold are set to 255, and the other grayscale values ​​are set to 0, and then the number of non-zero grayscale values ​​in each channel image is counted to obtain the number of non-zero grayscale values ​​in the FAM channel image under the third concentration threshold (Count_Pixel_FAM3), the number of non-zero grayscale values ​​in the HEX channel image under the third concentration threshold (Count_Pixel_HEX3), the number of non-zero grayscale values ​​in the ROX channel image under the third concentration threshold (Count_Pixel_ROX3) and the number of non-zero grayscale values ​​in the CY-5 channel image under the third concentration threshold (Count_Pixel_CY-5-3), and then the number of channels that meet the non-zero pixel value number greater than 900000 is counted. If the number of channels is greater than 1, it is a multi-target channel without fluorescence correction processing. If the number of channels is 1, it is determined that the target channel is in a low concentration state, that is, the concentration corresponding to the sample to be tested is the third concentration, and the concentration judgment process ends.

[0072] That is to say, in an optional implementation, determining the concentration corresponding to the sample to be tested according to the image set and the predetermined first concentration threshold, second concentration threshold, and third concentration threshold may specifically include:

[0073] The number of pixel values ​​greater than the first concentration threshold in each channel image in the image set is obtained respectively, and the number of channels whose number is greater than the preset threshold is counted, and when the number of channels is 1, the concentration corresponding to the sample to be tested is determined to be the first concentration;

[0074] Respectively obtain the number of pixel values ​​greater than the second concentration threshold in each channel image in the image set, and count the number of channels greater than the preset threshold, and when the number of channels is 1, determine that the concentration corresponding to the sample to be tested is the second concentration;

[0075] The number of pixel values ​​greater than the third concentration threshold in each channel image in the image set is obtained respectively, and the number of channels greater than the preset threshold is counted, and when the number of channels is 1, it is determined that the concentration corresponding to the sample to be tested is the third concentration.

[0076] It should be noted that when making concentration judgments, the order of the first concentration, the second concentration, and the third concentration must be followed, and the order cannot be adjusted. When there is only one concentration that satisfies a single target, a single-target multi-channel fluorescence crosstalk correction is performed. The methods provided in this application are all performed on single-target multi-channel images.

[0077] S103, determining the target channel corresponding to the sample to be tested according to the image set.

[0078] It is understandable that different target channels will have different crosstalk to other channels. Therefore, accurately determining the target channel corresponding to the sample to be tested is also a prerequisite for high-precision correction of fluorescence crosstalk. When judging the target channel of a four-channel grayscale image, the following process can be followed: first, the grayscale value distribution in the image is counted, which represents the distribution of the same pixel value, that is, the number of pixels with the same pixel value is counted; then, the probability distribution of each pixel is obtained based on the number of pixels counted, that is, the proportion of the number of pixels with the corresponding value to the total number of pixels; finally, the information entropy of each channel image is obtained using the following formula:

[0079]

[0080] Among them, H represents the information entropy of the channel image, p i Represents the probability of a pixel with pixel value i.

[0081] Information entropy reflects the richness of image information. In the case of four channels, the target channel has more information than the non-target channel. Therefore, by comparing the information entropy of the four channel images and determining the channel with the largest information entropy as the target channel, the target channel corresponding to the sample to be tested can be accurately determined.

[0082] That is to say, in an optional implementation, determining the target channel corresponding to the sample to be tested based on the image set may specifically include: determining the information entropy of each channel image in the image set; and determining the channel corresponding to the channel image with the largest information entropy as the target channel corresponding to the sample to be tested.

[0083] S104, performing crosstalk correction according to a predetermined crosstalk logic, a target channel corresponding to the sample to be tested, a concentration corresponding to the sample to be tested, and a predetermined correction coefficient of each channel at each concentration.

[0084] The crosstalk logic between the four channels (FAM, HEX, ROX, and CY-5) is shown in Table 1.

[0085] Table 1

[0086] index 0 1 2 3 aisle FAM HEX ROX CY-5 Case 1 Main target Main crosstalk (high) Weak secondary crosstalk No crosstalk Case 2 No crosstalk Main target Main crosstalk (high, medium) Weak secondary crosstalk Case 3 No crosstalk No crosstalk Main target Main crosstalk (high) Case 4 No crosstalk No crosstalk No crosstalk Main target

[0087] As shown in Table 1, its internal crosstalk logic is as follows: 1. FAM is the main target and its main crosstalk is HEX (crosstalk only exists at high concentrations). At the same time, there may be weak signals in other channels, which are secondary crosstalk (they may not be detected by the segmentation network, but the image still has extremely weak crosstalk noise); 2. HEX is the main target and its main crosstalk is ROX (crosstalk only exists at high and medium concentrations), and the secondary crosstalk is CY-5 (caused by the main crosstalk). At the same time, there may be weak secondary crosstalk in other channels; 3. ROX is the main target and its main crosstalk is CY-5 (crosstalk only exists at high concentrations); 4. If CY5 is the main target, there is no crosstalk in other channels.

[0088] That is to say, in an optional embodiment, the crosstalk logic may specifically include: if the FAM channel is the target channel, then the HEX channel has a main crosstalk at a first concentration; if the HEX channel is the target channel, then the ROX channel has a main crosstalk at a first concentration and a second concentration; if the ROX channel is the target channel, then the CY-5 channel has a main crosstalk at the first concentration.

[0089] The correction coefficient of each channel at each concentration reflects the effect of the target channel on the crosstalk channel. For example, the correction coefficients obtained at different concentrations are shown in Table 2.

[0090] Table 2

[0091] FAM HEX ROX CY5 High concentration 0.01 0.05 0.036 0 Medium concentration 0 0.063 0 0 Low concentration 0 0 0 0

[0092] 0.01 in Table 2 represents the correction coefficient of the FAM channel for the main crosstalk channel when the FAM channel is the target channel at the first concentration (high concentration), and the others are similar and will not be described in detail here.

[0093] Based on the above crosstalk logic and correction coefficient, crosstalk correction can be performed in the following steps:

[0094] If the concentration is high and the target channel index is 0 (FAM), the correction coefficient is used to correct the fluorescence crosstalk of the HEX channel. As shown in Table 2, the correction coefficient of FAM 0.01 is multiplied by the gray value of the FAM channel to obtain the correction matrix, and then the HEX and correction matrix are subtracted to obtain the HEX correction image. All possible weak secondary crosstalks are corrected by the constant matrix, and the gray value of the constant matrix size is set between 5-15 and selected according to the specific situation.

[0095] If the concentration is high and the target channel index is 1 (HEX), the correction coefficient is used to correct the fluorescence crosstalk of the ROX channel. As shown in Table 2, the correction coefficient of HEX 0.05 is multiplied by the gray value of the HEX channel to obtain the correction matrix, and then the ROX and correction matrix are subtracted to obtain the ROX correction image. All possible weak secondary crosstalks are corrected by the constant matrix, and the gray value of the constant matrix size is set to between 5 and 15 according to the specific situation.

[0096] If the concentration is medium and the target channel index is 1 (HEX), the correction coefficient is used to correct the fluorescence crosstalk of the ROX channel. As shown in Table 2, the correction coefficient of HEX 0.063 is multiplied by the gray value of the HEX channel to obtain the correction matrix, and then the ROX and correction matrix are subtracted to obtain the ROX correction image. All possible weak crosstalks are corrected by the constant matrix, and the gray value of the constant matrix size is set between 5-15 and selected according to the specific situation.

[0097] If the concentration is high and the target channel index is 2 (ROX), the correction coefficient is used to correct the fluorescence crosstalk of the CY-5 channel. As shown in Table 2, the correction coefficient of ROX 0.036 is multiplied by the gray value of the ROX channel to obtain the correction matrix, and then the correction image of CY-5 is obtained by subtracting CY-5 from the correction matrix.

[0098] If the target channel index is 3 (CY-5), no processing is performed.

[0099] That is to say, in an optional implementation, crosstalk correction is performed according to a predetermined crosstalk logic, a target channel corresponding to the sample to be tested, a concentration corresponding to the sample to be tested, and a correction coefficient of each channel at each predetermined concentration, which may specifically include:

[0100] When the concentration corresponding to the sample to be tested is the first concentration and the target channel corresponding to the sample to be tested is the FAM channel, a correction matrix is ​​determined according to the correction coefficient of the FAM channel at the first concentration and the FAM channel image, and a correction image of the HEX channel is obtained by subtracting the HEX channel image from the correction matrix;

[0101] When the concentration corresponding to the sample to be tested is the first concentration and the target channel corresponding to the sample to be tested is the HEX channel, a correction matrix is ​​determined according to the correction coefficient of the HEX channel at the first concentration and the HEX channel image, and a correction image of the ROX channel is obtained by subtracting the ROX channel image from the correction matrix;

[0102] When the concentration corresponding to the sample to be tested is the second concentration and the target channel corresponding to the sample to be tested is the HEX channel, a correction matrix is ​​determined according to the correction coefficient of the HEX channel at the second concentration and the HEX channel image, and a correction image of the ROX channel is obtained by subtracting the ROX channel image from the correction matrix;

[0103] When the concentration corresponding to the sample to be tested is the first concentration and the target channel corresponding to the sample to be tested is the ROX channel, the correction matrix is ​​determined according to the correction coefficient of the ROX channel at the first concentration and the ROX channel image, and the CY-5 channel image is subtracted from the correction matrix to obtain the correction image of the CY-5 channel.

[0104] The fluorescence crosstalk correction method provided in this embodiment obtains a four-channel image set of the sample to be tested; determines the concentration corresponding to the sample to be tested according to the image set and the predetermined first concentration threshold, second concentration threshold and third concentration threshold; and determines the target channel corresponding to the sample to be tested according to the image set; finally, crosstalk correction is performed according to the predetermined crosstalk logic, the target channel corresponding to the sample to be tested, the concentration corresponding to the sample to be tested and the predetermined correction coefficient of each channel at each concentration, which can simply and conveniently identify the target channel and the crosstalk channel, and has a high-precision four-channel crosstalk correction capability. In the current industry standard of 10%, the method provided in this application can reduce the fluorescence crosstalk intensity to less than 1%.

[0105] The accuracy of the correction coefficient directly determines the precision of the fluorescence crosstalk correction. The following will use a specific example to illustrate how to determine the correction coefficient of each channel at each concentration.

[0106] Acquire multiple (such as 4) four-channel image sets at high concentration, where the target channel is the FAM channel and the main crosstalk channel is the HEX channel, then there will be 4 FAM channel images and 4 HEX channel images. For each FAM channel image, the u-net segmentation algorithm can be used to locate the position of each droplet and calculate the droplet gray value based on the droplet center with a preset radius (for example, 10) and count the global droplet gray value to obtain the FAM channel global droplet gray value; for each HEX channel image, the u-net segmentation algorithm can be used to locate the position of each droplet and calculate the droplet gray value based on the droplet center with a preset radius (for example, 10) and count the global droplet gray value to obtain the HEX channel global droplet gray value. The FAM channel global droplet gray value and the HEX global channel droplet gray value at high concentration are shown in Table 3.

[0107] Table 3

[0108]

[0109] The average grayscale value of the FAM channel droplets at high concentrations in Table 3 is taken to obtain the average grayscale value of the FAM channel droplets at high concentrations. The average grayscale value of the HEX channel droplets at high concentrations in Table 3 is taken to obtain the average grayscale value of the HEX channel droplets at high concentrations. This is the correction factor of the FAM channel to the HEX channel at high concentration.

[0110] Acquire multiple four-channel image sets at high concentration, where the target channel is the HEX channel and the main crosstalk channel is the ROX channel. According to the above method, the correction coefficient of the HEX channel to the ROX channel at high concentration can be obtained. By analogy, the correction coefficient of the ROX channel to the CY-5 channel at high concentration and the correction coefficient of the HEX channel to the ROX channel at medium concentration can also be obtained.

[0111] On the basis of the above embodiments, in the fluorescence crosstalk correction method provided in this embodiment, the correction coefficient of each channel at each concentration can be determined specifically according to the following steps:

[0112] Acquire multiple images of the target channel and its corresponding main crosstalk channel at each concentration;

[0113] For each image of the target channel, locate the position of the droplet and calculate the droplet gray value based on the droplet center according to the preset radius and calculate the global droplet gray value to obtain the target channel droplet gray value;

[0114] For each image of the main crosstalk channel, locate the position of the droplet and calculate the droplet gray value based on the center of the droplet according to the preset radius and calculate the global droplet gray value to obtain the droplet gray value of the main crosstalk channel;

[0115] The ratio of the average grayscale value of the droplets in the main crosstalk channel to the average grayscale value of the droplets in the target channel is determined as the correction coefficient of the target channel to the main crosstalk channel at this concentration.

[0116] On the basis of the above embodiments, in the fluorescence crosstalk correction method provided in this embodiment, the first concentration threshold, the second concentration threshold and the third concentration threshold can be determined according to the following steps: respectively obtain multiple channel images at each concentration, determine the gray value histogram of each channel image, and use the average of the highest and second highest gray values ​​in the histogram as the segmentation threshold of the channel image, and respectively determine the average value of the segmentation thresholds of the multiple channel images at each concentration as the concentration threshold corresponding to the concentration.

[0117] Specifically, multiple channel images (grayscale images) at a first concentration (high concentration) can be obtained, and grayscale value histogram analysis can be performed on the multiple grayscale images to find the grayscale values ​​of the two highest peaks (the two peaks represent the grayscale value concentration area of ​​the background area and the main droplet pixel area, which can be recorded as gray and gray respectively). background and gray droplet ), then the threshold for threshold segmentation can be obtained according to the grayscale values ​​of the two peaks. Specifically, the formula thresh = (gray background +gray droplet) / 2 determines the segmentation threshold thresh of each channel image, and further calculates the average value of multiple segmentation thresholds to obtain the first concentration threshold. Obtain multiple channel images (grayscale images) at the second concentration (medium concentration), and the second concentration threshold can be obtained according to the above method; obtain multiple channel images (grayscale images) at the third concentration (low concentration), and the third concentration threshold can be obtained according to the above method.

[0118] It is understandable that the image will inevitably be affected by noise during the acquisition process, so there is a certain amount of Gaussian white noise and salt and pepper noise in the acquired image. The presence of noise may affect the correction of fluorescence crosstalk in the image, so it is necessary to perform denoising on the acquired image to eliminate the above-mentioned adverse effects. In an optional implementation, before determining the concentration corresponding to the sample to be tested, each channel image in the image set is denoised in turn, and the denoising process can be specifically performed according to the following steps:

[0119] Step A0, grayscale the image to be processed to obtain a single-channel grayscale image. For example, an RGB three-channel image can be converted into a single-channel grayscale image to reduce the amount of image data, thereby reducing the complexity of computer operation.

[0120] Step A1, perform mean filtering on the grayscale image to obtain a mean filtered image, which can suppress noise and improve analysis reliability.

[0121] Step A2, converting the data types of the grayscale image and the mean filter image into floating point types, and obtaining a floating point noise image according to the floating point grayscale image and the floating point mean filter image. Specifically, the noise image can be obtained by subtracting the mean filter image from the original image, thereby separating the noise image.

[0122] Step A3, truncate the pixel values ​​less than 0 in the noise image and convert the data type of the noise image into an unsigned integer, for example, it can be converted into uint8 type, which is converted into uint8 to save space and limit the grayscale value to between 0-255.

[0123] Step A4, threshold segmentation is performed on the noise image to separate the noise pixels and obtain a binary segmented image. Specifically, the segmentation threshold is set to 18, if the gray value is greater than 18, the gray value is converted to 255, if the gray value is less than 18, the gray value is converted to 0, and finally a binary segmented image thresh_img can be obtained. After the noise is separated by threshold segmentation, the gray value of the noise is increased to the maximum.

[0124] Step A5, perform morphological erosion, morphological opening and morphological dilation operations on the segmented image in sequence to obtain the corresponding morphological image, perform XOR operation on the segmented image and its corresponding morphological image to obtain the noise mask image. Specifically, perform morphological erosion operation on the segmented image, set the morphological structure element to 3x 3 structure size and obtain the morphological image imgA; then perform morphological opening and morphological dilation operations on imgA in sequence, set the structure element to 3x 3 structure size to obtain the morphological image imgB; then multiply imgB by 255 and perform XOR operation with thresh_img to obtain the noise mask image. Through a series of operations on the image, the noise mask image is finally obtained.

[0125] Step A6, performing an AND operation on the mean filter image and the noise mask image to obtain a filtered noise region image, thereby avoiding the noise information in the original image.

[0126] Step A7, determine the non-noise mask image according to the noise mask image, and use the non-noise mask image to obtain the non-noise area image from the grayscale image. Specifically, subtract the noise mask image from 255 to obtain image imgC, and perform an AND operation between imgC and the original grayscale image to obtain the non-noise area image.

[0127] Step A8, adding and fusing the filtered noise area image and the non-noise area image, so as to obtain a denoised image.

[0128] The method provided in this embodiment, based on the above embodiment, further separates the noise area and the non-noise area for denoising, which can remove noise losslessly without damaging the droplet information. The noise mask is extracted based on morphological operations and the global noise of the image is removed. At the same time, this method solves the disadvantages of image blurring and loss of target key information in the denoising process of traditional algorithms.

[0129] In summary, the present application provides a correction coefficient-based fluorescence crosstalk correction method, which is not only fast and accurate, but also can easily and conveniently identify targets and crosstalk channels, and has high-precision multi-channel crosstalk correction capabilities. Based on morphological operations, the noise mask is extracted and the global noise of the droplet image is removed, solving the problems of image blur and loss of key target information in the denoising process of the traditional algorithm; based on the multi-data statistics of the double peaks, the threshold parameters under different concentrations are determined and the channel concentration judgment is completed based on the concentration threshold division results; based on the logic from high concentration to low concentration, the logical judgment of whether the single target condition is met is completed; based on the grayscale distribution, the information entropy is calculated and the target channel is judged based on the information entropy; based on the segmentation algorithm, the position of each droplet is located and the droplet gray value is calculated based on the droplet center with a radius of 10 and the global droplet average value is counted, and the crosstalk correction coefficients of different channels under different concentrations are obtained using the experimental parameter table of multiple comparisons of different concentrations and channels; based on the target channel, concentration and crosstalk logic, four-channel crosstalk correction is performed. This application reduces the intensity of fluorescence crosstalk to less than 1 percent, while the current industry standard is 10 percent.

[0130] Figure 2 FIG. 1 is a schematic diagram of the structure of a fluorescence crosstalk correction device provided by an embodiment of the present invention. Figure 2 As shown, the fluorescence crosstalk correction device 20 provided in this embodiment may include: a collection module 201 , an analysis module 202 , a processing module 203 and a correction module 204 .

[0131] The acquisition module 201 is used to obtain a four-channel image set of the sample to be tested, wherein the image set includes a FAM channel image, a HEX channel image, a ROX channel image and a CY-5 channel image;

[0132] An analysis module 202 is used to determine the concentration corresponding to the sample to be tested according to the image set and a predetermined first concentration threshold, a second concentration threshold, and a third concentration threshold;

[0133] The processing module 203 is used to determine the target channel corresponding to the sample to be tested according to the image set;

[0134] The correction module 204 is used to perform crosstalk correction according to a predetermined crosstalk logic, a target channel corresponding to the sample to be tested, a concentration corresponding to the sample to be tested, and a predetermined correction coefficient of each channel at each concentration.

[0135] The device of this embodiment can be used to perform Figure 1 The technical solution of the method embodiment shown has similar implementation principles and technical effects, which will not be repeated here.

[0136] In an optional implementation manner, the crosstalk logic may specifically include:

[0137] If the FAM channel is the target channel, then the HEX channel has major crosstalk at the first concentration;

[0138] If the HEX channel is the target channel, there is major crosstalk in the ROX channel at the first and second concentrations;

[0139] If the ROX channel is the target channel, then the CY-5 channel has major crosstalk at the first concentration.

[0140] In an optional implementation, the correction module 204 is used to perform crosstalk correction according to a predetermined crosstalk logic, a target channel corresponding to the sample to be tested, a concentration corresponding to the sample to be tested, and a correction coefficient of each channel at each predetermined concentration, which may specifically include:

[0141] When the concentration corresponding to the sample to be tested is the first concentration and the target channel corresponding to the sample to be tested is the FAM channel, a correction matrix is ​​determined according to the correction coefficient of the FAM channel at the first concentration and the FAM channel image, and a correction image of the HEX channel is obtained by subtracting the HEX channel image from the correction matrix;

[0142] When the concentration corresponding to the sample to be tested is the first concentration and the target channel corresponding to the sample to be tested is the HEX channel, a correction matrix is ​​determined according to the correction coefficient of the HEX channel at the first concentration and the HEX channel image, and a correction image of the ROX channel is obtained by subtracting the ROX channel image from the correction matrix;

[0143] When the concentration corresponding to the sample to be tested is the second concentration and the target channel corresponding to the sample to be tested is the HEX channel, a correction matrix is ​​determined according to the correction coefficient of the HEX channel at the second concentration and the HEX channel image, and a correction image of the ROX channel is obtained by subtracting the ROX channel image from the correction matrix;

[0144] When the concentration corresponding to the sample to be tested is the first concentration and the target channel corresponding to the sample to be tested is the ROX channel, the correction matrix is ​​determined according to the correction coefficient of the ROX channel at the first concentration and the ROX channel image, and the CY-5 channel image is subtracted from the correction matrix to obtain the correction image of the CY-5 channel.

[0145] In an optional implementation, the processing module 203 is used to determine the target channel corresponding to the sample to be tested according to the image set, which may specifically include:

[0146] Determine the information entropy of each channel image in the image set;

[0147] The channel corresponding to the channel image with the largest information entropy is determined as the target channel corresponding to the sample to be tested.

[0148] In one embodiment, the correction coefficients for each channel at each concentration are determined according to the following steps:

[0149] Acquire multiple images of the target channel and its corresponding main crosstalk channel at each concentration;

[0150] For each image of the target channel, locate the position of the droplet and calculate the droplet gray value based on the droplet center according to the preset radius and calculate the global droplet gray value to obtain the target channel droplet gray value;

[0151] For each image of the main crosstalk channel, locate the position of the droplet and calculate the droplet gray value based on the center of the droplet according to the preset radius and calculate the global droplet gray value to obtain the droplet gray value of the main crosstalk channel;

[0152] The ratio of the average grayscale value of the droplets in the main crosstalk channel to the average grayscale value of the droplets in the target channel is determined as the correction coefficient of the target channel to the main crosstalk channel at this concentration.

[0153] In an optional implementation, the analysis module 202 is used to determine the concentration corresponding to the sample to be tested according to the image set and the predetermined first concentration threshold, second concentration threshold, and third concentration threshold, which may specifically include:

[0154] The number of pixel values ​​greater than the first concentration threshold in each channel image in the image set is obtained respectively, and the number of channels whose number is greater than the preset threshold is counted, and when the number of channels is 1, the concentration corresponding to the sample to be tested is determined to be the first concentration;

[0155] Respectively obtain the number of pixel values ​​greater than the second concentration threshold in each channel image in the image set, and count the number of channels greater than the preset threshold, and when the number of channels is 1, determine that the concentration corresponding to the sample to be tested is the second concentration;

[0156] The number of pixel values ​​greater than the third concentration threshold in each channel image in the image set is obtained respectively, and the number of channels greater than the preset threshold is counted, and when the number of channels is 1, it is determined that the concentration corresponding to the sample to be tested is the third concentration.

[0157] In an optional implementation manner, the first concentration threshold, the second concentration threshold and the third concentration threshold may be determined according to the following steps:

[0158] Multiple channel images at each concentration are obtained respectively, and the gray value histogram of each channel image is determined. The average of the highest and second highest gray values ​​in the histogram is used as the segmentation threshold of the channel image. The average value of the segmentation thresholds of the multiple channel images at each concentration is determined as the concentration threshold corresponding to the concentration.

[0159] In an optional embodiment, the fluorescence crosstalk correction device 20 may further include a denoising module (not shown in the figure), which is used to perform denoising on each channel image in the image set in turn before determining the concentration corresponding to the sample to be tested, and is specifically used to perform denoising according to the following steps:

[0160] Grayscale the image to be processed to obtain a single-channel grayscale image;

[0161] Perform mean filtering on the grayscale image to obtain a mean filtered image;

[0162] The data types of the grayscale image and the mean filter image are converted into floating-point types, and a floating-point noise image is obtained according to the floating-point grayscale image and the floating-point mean filter image;

[0163] The pixel values ​​in the noise image that are less than 0 are truncated and the data type of the noise image is converted to an unsigned integer;

[0164] Perform threshold segmentation on the noise image to separate the noise pixels and obtain a binary segmented image;

[0165] The segmented image is sequentially subjected to morphological erosion, morphological opening and morphological dilation operations to obtain the corresponding morphological image, and the segmented image and its corresponding morphological image are subjected to XOR operation to obtain the noise mask image;

[0166] Perform an AND operation on the mean filter image and the noise mask image to obtain a filtered noise region image;

[0167] Determine a non-noise mask image according to the noise mask image, and use the non-noise mask image to obtain a non-noise region image from the grayscale image;

[0168] The filtered noise area image and the non-noise area image are added and fused.

[0169] The present invention also provides an electronic device, see Figure 3 As shown, the embodiment of the present invention is only Figure 3 This is just an example for explanation, and it does not mean that the present invention is limited to this. Figure 3 This is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. Figure 3 As shown, the electronic device 30 provided in this embodiment may include: a memory 301, a processor 302 and a bus 303. The bus 303 is used to realize the connection between various components.

[0170] The memory 301 stores a computer program, and when the computer program is executed by the processor 302 , the technical solution of any of the above method embodiments can be implemented.

[0171] The memory 301 and the processor 302 are directly or indirectly electrically connected to each other to achieve data transmission or interaction. For example, these elements can be electrically connected to each other through one or more communication buses or signal lines, such as through bus 303. The memory 301 stores a computer program for implementing the fluorescence crosstalk correction method, including at least one software function module that can be stored in the memory 301 in the form of software or firmware. The processor 302 executes various functional applications and data processing by running the software program and module stored in the memory 301.

[0172] The memory 301 may be, but is not limited to, a random access memory (RAM), a read only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electric erasable programmable read-only memory (EEPROM), etc. The memory 301 is used to store programs, and the processor 302 executes the programs after receiving the execution instruction. Furthermore, the software programs and modules in the above-mentioned memory 301 may also include an operating system, which may include various software components and / or drivers for managing system tasks (such as memory management, storage device control, power management, etc.), and may communicate with various hardware or software components to provide an operating environment for other software components.

[0173] The processor 302 may be an integrated circuit chip with signal processing capability. The processor 302 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc. The methods, steps, and logic diagrams disclosed in the embodiments of the present invention may be implemented or executed. The general-purpose processor may be a microprocessor or any conventional processor. It is understood that Figure 3 The structure is only for illustration and may also include Figure 3 More or fewer components as shown, or with Figure 3 Different configurations shown. Figure 3 Each component shown in the figure may be implemented by hardware and / or software.

[0174] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. The computer program is executed by a processor to implement the technical solution of any of the above method embodiments.

[0175] The various embodiments in the present disclosure are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0176] The protection scope of the present disclosure is not limited to the above-mentioned embodiments. Obviously, those skilled in the art can make various changes and modifications to the present disclosure without departing from the scope and spirit of the present disclosure. If these changes and modifications fall within the scope of the claims of the present disclosure and their equivalents, the intention of the present disclosure also includes these changes and modifications.

Claims

1. A fluorescence crosstalk correction method, characterized in that: include: Acquire a four-channel image set of the sample to be tested, wherein the image set includes a FAM channel image, a HEX channel image, a ROX channel image, and a CY-5 channel image; Determine the concentration corresponding to the sample to be tested according to the image set and a predetermined first concentration threshold, a second concentration threshold, and a third concentration threshold; Determine the target channel corresponding to the sample to be tested according to the image set; Crosstalk correction is performed according to a predetermined crosstalk logic, a target channel corresponding to the sample to be tested, a concentration corresponding to the sample to be tested, and a predetermined correction coefficient of each channel at each concentration.

2. The method according to claim 1, characterized in that The crosstalk logic includes: If the FAM channel is the target channel, then the HEX channel has major crosstalk at the first concentration; If the HEX channel is the target channel, there is major crosstalk in the ROX channel at the first and second concentrations; If the ROX channel is the target channel, then the CY-5 channel has major crosstalk at the first concentration.

3. The method according to claim 2, characterized in that The crosstalk correction is performed according to a predetermined crosstalk logic, a target channel corresponding to the sample to be tested, a concentration corresponding to the sample to be tested, and a predetermined correction coefficient of each channel at each concentration, including: When the concentration corresponding to the sample to be tested is the first concentration and the target channel corresponding to the sample to be tested is the FAM channel, a correction matrix is ​​determined according to the correction coefficient of the FAM channel at the first concentration and the FAM channel image, and a correction image of the HEX channel is obtained by subtracting the HEX channel image from the correction matrix; When the concentration corresponding to the sample to be tested is the first concentration and the target channel corresponding to the sample to be tested is the HEX channel, a correction matrix is ​​determined according to the correction coefficient of the HEX channel at the first concentration and the HEX channel image, and a ROX channel image is subtracted from the correction matrix to obtain a corrected image of the ROX channel; When the concentration corresponding to the sample to be tested is the second concentration and the target channel corresponding to the sample to be tested is the HEX channel, a correction matrix is ​​determined according to the correction coefficient of the HEX channel at the second concentration and the HEX channel image, and a correction image of the ROX channel is obtained by subtracting the ROX channel image from the correction matrix; When the concentration corresponding to the sample to be tested is the first concentration and the target channel corresponding to the sample to be tested is the ROX channel, a correction matrix is ​​determined according to the correction coefficient of the ROX channel at the first concentration and the ROX channel image, and the CY-5 channel image is subtracted from the correction matrix to obtain a correction image of the CY-5 channel.

4. The method according to claim 1, characterized in that: The step of determining the target channel corresponding to the sample to be tested according to the image set includes: Determining the information entropy of each channel image in the image set; The channel corresponding to the channel image with the largest information entropy is determined as the target channel corresponding to the sample to be tested.

5. The method according to claim 1, characterized in that The correction coefficient of each channel at each concentration is determined according to the following steps: Acquire multiple images of the target channel and its corresponding main crosstalk channel at each concentration; For each image of the target channel, locate the position of the droplet and calculate the droplet gray value based on the droplet center according to the preset radius and calculate the global droplet gray value to obtain the target channel droplet gray value; For each image of the main crosstalk channel, locate the position of the droplet and calculate the droplet gray value based on the center of the droplet according to the preset radius and calculate the global droplet gray value to obtain the droplet gray value of the main crosstalk channel; The ratio of the average grayscale value of the droplets in the main crosstalk channel to the average grayscale value of the droplets in the target channel is determined as the correction coefficient of the target channel to the main crosstalk channel at this concentration.

6. The method according to claim 1, characterized in that The step of determining the concentration corresponding to the sample to be tested according to the image set and a predetermined first concentration threshold, a second concentration threshold, and a third concentration threshold comprises: Respectively obtaining the number of pixel values ​​greater than the first concentration threshold in each channel image in the image set, and counting the number of channels whose number is greater than a preset threshold, and determining that the concentration corresponding to the sample to be tested is the first concentration when the number of channels is 1; Respectively obtaining the number of pixel values ​​greater than the second concentration threshold in each channel image in the image set, and counting the number of channels whose number is greater than a preset threshold, and determining that the concentration corresponding to the sample to be tested is the second concentration when the number of channels is 1; The number of pixel values ​​greater than the third concentration threshold in each channel image in the image set is obtained respectively, and the number of channels whose number is greater than the preset threshold is counted, and when the number of channels is 1, it is determined that the concentration corresponding to the sample to be tested is the third concentration.

7. The method according to claim 1, characterized in that The first concentration threshold, the second concentration threshold and the third concentration threshold are determined according to the following steps: Multiple channel images at each concentration are obtained respectively, and the gray value histogram of each channel image is determined. The average of the highest and second highest gray values ​​in the histogram is used as the segmentation threshold of the channel image. The average value of the segmentation thresholds of the multiple channel images at each concentration is determined as the concentration threshold corresponding to the concentration.

8. The method according to any one of claims 1 to 7, characterized in that: Before determining the concentration corresponding to the sample to be tested, the method further includes: performing denoising processing on each channel image in the image set in sequence, and the denoising processing is performed according to the following steps: Grayscale the image to be processed to obtain a single-channel grayscale image; Performing mean filtering on the grayscale image to obtain a mean filtered image; Converting the data types of the grayscale image and the mean filter image into floating-point types, and obtaining a floating-point noise image according to the floating-point grayscale image and the floating-point mean filter image; Performing truncation processing on pixel values ​​less than 0 in the noise image and converting the data type of the noise image into an unsigned integer; Performing threshold segmentation on the noise image to separate noise pixels and obtain a binary segmented image; Performing a morphological erosion operation, a morphological opening operation, and a morphological dilation operation on the segmented image in sequence to obtain a corresponding morphological image, and performing an XOR operation on the segmented image and its corresponding morphological image to obtain a noise mask image; Performing an AND operation on the mean filter image and the noise mask image to obtain a filtered noise region image; Determine a non-noise mask image according to the noise mask image, and use the non-noise mask image to acquire a non-noise region image from the grayscale image; The filtered noise area image and the non-noise area image are added and fused.

9. A fluorescence crosstalk correction device, characterized in that: include: An acquisition module is used to obtain a four-channel image set of the sample to be tested, wherein the image set includes a FAM channel image, a HEX channel image, a ROX channel image and a CY-5 channel image; An analysis module, used to determine the concentration corresponding to the sample to be tested according to the image set and a predetermined first concentration threshold, a second concentration threshold, and a third concentration threshold; A processing module, used to determine the target channel corresponding to the sample to be tested according to the image set; The correction module is used to perform crosstalk correction according to a predetermined crosstalk logic, a target channel corresponding to the sample to be tested, a concentration corresponding to the sample to be tested, and a predetermined correction coefficient of each channel at each concentration.

10. An electronic device, characterized in that: include: at least one processor and memory; The memory stores computer-executable instructions; The at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor performs the fluorescence crosstalk correction method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the fluorescence crosstalk correction method according to any one of claims 1 to 8.

Images