Chromosome segmentation method, device, computer equipment, readable storage medium and program product

Abstract

The application relates to a chromosome segmentation method and device, computer equipment, a readable storage medium and a program product. The method comprises the following steps: acquiring a sampling image of a chromosome; the sampling image comprises a plurality of image regions; for each image region, based on the gray value of each pixel point in the image region, determining the texture attribute of the image region and the gray statistical value matched with the texture attribute; based on the gray statistical value, determining a segmentation threshold value used for background segmentation of the image region; performing background segmentation on the image region according to the segmentation threshold value to obtain a foreground image in the image region; and splicing the foreground images of the respective image regions to obtain a chromosome image of the chromosome. The method can improve the chromosome segmentation accuracy.

Description

Technical Field

[0001] This application relates to the field of biomedical imaging technology, and in particular to a chromosome segmentation method, apparatus, computer equipment, readable storage medium, and program product. Background Technology

[0002] With the development of biomedical imaging technology, chromosome karyotype analysis has become increasingly important in fields such as genetic disease diagnosis and tumor research. Currently, chromosome microscopy can achieve high-resolution chromosome sample acquisition. However, due to factors such as uneven illumination, interference from cell debris, and the complex morphology of chromosomes, the contrast between chromosome regions and the background in the original sampled images is low, and the noise distribution in different regions varies significantly.

[0003] Traditional methods typically employ global thresholding or edge detection techniques with fixed parameters. These methods assume a uniform grayscale distribution across the image, leading to the potential for chromosome breakage or background remnants in low-contrast local areas. This is particularly problematic when dealing with curled or overlapping chromosomes, where the segmentation accuracy falls short of clinical diagnostic requirements. Therefore, traditional methods for chromosome segmentation suffer from low accuracy. Summary of the Invention

[0004] Therefore, it is necessary to provide a chromosome segmentation method, apparatus, computer equipment, readable storage medium, and program product that can improve the accuracy of chromosome segmentation in response to the above-mentioned technical problems.

[0005] In a first aspect, this application provides a chromosome segmentation method, including:

[0006] Acquire sampled images of chromosomes; the sampled images include multiple image regions;

[0007] For each image region, based on the grayscale value of each pixel in the image region, the texture attribute of the image region and the grayscale statistical value matching the texture attribute are determined;

[0008] Based on the grayscale statistical values, a segmentation threshold for background segmentation of the image region is determined;

[0009] The image region is segmented according to the segmentation threshold to obtain the foreground image in the image region;

[0010] By stitching together the foreground images of each of the aforementioned image regions, a chromosome image is obtained.

[0011] In one embodiment, the method further includes:

[0012] Obtain a pre-trained labeled chromosome detection model;

[0013] The chromosome image is detected based on the labeled chromosome detection model to obtain detection results for the labeled chromosome.

[0014] In one embodiment, determining the texture attributes of the image region based on the grayscale values ​​of each pixel in the image region includes:

[0015] By statistically analyzing the gray values ​​of each pixel in the image region, the gray standard deviation and gray histogram of the image region are determined.

[0016] The texture attributes of the image region are determined based on the gray-level histogram and the gray-level standard deviation.

[0017] In one embodiment, the grayscale statistics include the grayscale peak value of the grayscale histogram; determining the segmentation threshold for background segmentation of the image region based on the grayscale statistics includes:

[0018] If the contrast of the image region indicated by the texture attribute is greater than or equal to the contrast threshold, the gray value peak of the gray value histogram is obtained.

[0019] Based on the grayscale peak value, a segmentation threshold for background segmentation of the image region is determined.

[0020] In one embodiment, the grayscale statistics include the average grayscale value of the image region; determining the segmentation threshold for background segmentation of the image region based on the grayscale statistics includes:

[0021] If the texture attribute indicates that the contrast of the image region is less than the contrast threshold, the average gray value of the image region is determined based on each of the gray values.

[0022] Based on the mean gray level and the standard deviation of gray level, a segmentation threshold for background segmentation of the image region is determined.

[0023] In one embodiment, the grayscale statistics include the average grayscale value of the image region; determining the segmentation threshold for background segmentation of the image region based on the grayscale statistics includes:

[0024] If the texture attribute indicates that the sharpness of the image region is greater than or equal to the sharpness threshold, obtain the upper limit of the preset standard deviation set for the grayscale standard deviation;

[0025] Based on the mean gray level, the standard deviation of gray level, and the preset upper limit of standard deviation, a segmentation threshold for background segmentation of the image region is determined.

[0026] Secondly, this application also provides a chromosome segmentation device, comprising:

[0027] A sampling image acquisition module is used to acquire sampling images of chromosomes; the sampling image includes multiple image regions;

[0028] The texture attribute determination module is used to determine the texture attribute of each image region and the grayscale statistical value matching the texture attribute based on the grayscale value of each pixel in the image region for each image region.

[0029] The segmentation threshold determination module is used to determine a segmentation threshold for background segmentation of the image region based on the grayscale statistical value.

[0030] The background segmentation module is used to segment the image region according to the segmentation threshold to obtain the foreground image in the image region;

[0031] The foreground image stitching module is used to stitch together the foreground images of each of the image regions to obtain the chromosome image.

[0032] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the method described above.

[0033] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the steps of the method described above.

[0034] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the steps of the method described above.

[0035] The aforementioned chromosome segmentation method, apparatus, computer equipment, readable storage medium, and program products, through a region-based processing strategy, independently calculate texture attributes and grayscale statistics for each image region, enabling them to adapt to contrast differences in different regions and effectively solve segmentation errors caused by uneven illumination. The grayscale statistical method based on texture feature matching can accurately characterize the periodic structure of chromosome stripes, thereby determining a more reasonable segmentation threshold and avoiding misidentification of cell fragments as chromosome regions. The foreground image extracted through threshold segmentation retains the complete morphological contour of the chromosome, reducing the chromosome breakage problem caused by edge blurring in traditional methods. The final stitched foreground image combines the high-precision segmentation results of each region, significantly improving the accuracy of chromosome segmentation. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a diagram illustrating the application environment of the chromosome segmentation method in one embodiment;

[0038] Figure 2 This is a flowchart illustrating a chromosome segmentation method in one embodiment;

[0039] Figure 3 This is a schematic diagram of the detection results for labeled chromosome detection in one embodiment;

[0040] Figure 4 This is a flowchart illustrating the chromosome segmentation method in another embodiment;

[0041] Figure 5 This is a structural block diagram of a chromosome segmentation device in one embodiment;

[0042] Figure 6 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0044] The chromosome segmentation method provided in this application can be applied to, for example... Figure 1In the application environment shown, server 102 communicates with image acquisition device 104 via a network. A data storage system can store the data that server 102 needs to process. The data storage system can be integrated onto server 102, or it can be located in the cloud or on other network servers. Server 102 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. Image acquisition device 104 is a high-precision optical device specifically designed to capture details of chromosome samples under a microscope. Its core function is to convert the microscopic chromosome structure into a digital image. Specifically, during chromosome segmentation, server 102 acquires sampled images of the chromosome; the sampled images include multiple image regions; for each image region, based on the grayscale values ​​of each pixel in the image region, the texture attributes of the image region and the grayscale statistical values ​​matching the texture attributes are determined; based on the grayscale statistical values, a segmentation threshold for background segmentation of the image region is determined; the image region is segmented according to the segmentation threshold to obtain the foreground image in the image region; the foreground images of each image region are stitched together to obtain the chromosome image.

[0045] In one exemplary embodiment, such as Figure 2 As shown, a chromosome segmentation method is provided, which can be applied to... Figure 1 Taking server 102 as an example, the explanation includes the following steps S202 to S210. Wherein:

[0046] Step S202: Obtain a sampled image of the chromosome.

[0047] The sampled image is a digital image of a chromosome sample captured by image acquisition equipment such as a microscope. It contains complete information about the chromosome and its background and includes multiple image regions. The image region is formed by dividing the sampled image into multiple local regions (such as grid-like segmentation), and each region is processed independently to accommodate uneven lighting or noise differences.

[0048] Specifically, image acquisition requires a microscopic imaging device. First, the chromosome sample (such as a stained preparation of metaphase cells) is placed on the microscope stage, and the objective lens magnification is adjusted to make the chromosome morphology clearly visible. The microscope must be equipped with a stable light source system to ensure uniform illumination of the sample. During image acquisition, the camera captures complete information about the chromosomes and their background, generating a sampled image containing multiple chromosome instances. To accommodate subsequent regional processing, the sampled image is automatically divided into multiple non-overlapping image regions, each covering a portion of the chromosome or background. The core of this step is to ensure image clarity and reasonable regional division, providing fundamental data for subsequent processing.

[0049] Step S204: For each image region, based on the grayscale value of each pixel in the image region, determine the texture attribute of the image region and the grayscale statistical value that matches the texture attribute.

[0050] Among them, pixel grayscale value is the brightness value of each pixel (0-255), reflecting the strength of the signal of the chromosome or background. Texture attributes describe the spatial characteristics of grayscale distribution in an image region, such as stripe thickness and directionality, and are used to distinguish chromosomes from the background. Grayscale statistics are the statistical characteristics of grayscale values ​​in a region, such as mean, variance, and histogram peak value, reflecting the overall brightness and contrast.

[0051] Specifically, the server needs to extract texture features and grayscale distribution information for each segmented image region. First, it reads the grayscale values ​​of all pixels within the region, generating a grayscale value list. By analyzing the spatial arrangement pattern of the grayscale values, texture attributes are calculated to quantify the difference between the region and the background. Simultaneously, the overall characteristics of the grayscale values ​​are statistically analyzed, including the mean, standard deviation, and peak distribution of the grayscale histogram. These statistical values, along with the texture attributes, constitute a region feature description, providing a basis for the adaptive determination of the subsequent segmentation threshold.

[0052] Optionally, the server can determine the texture attributes of an image region using a pre-trained texture attribute determination model, or it can determine the gray-level standard deviation and gray-level histogram of the image region by statistically analyzing the gray-level values ​​of each pixel in the image region, and then determine the texture attributes of the image region based on the gray-level histogram and gray-level standard deviation.

[0053] Step S206: Based on grayscale statistics, determine the segmentation threshold used for background segmentation of the image region.

[0054] The segmentation threshold is a grayscale threshold used to distinguish between the foreground (chromosome) and the background. Pixel values ​​higher than the threshold are considered foreground, and those lower are considered background.

[0055] Specifically, the server can determine the segmentation threshold for background segmentation of the image region based on the region state represented by the texture attributes and the corresponding grayscale statistical values.

[0056] Optionally, if the contrast of the image region represented by the texture attribute is greater than or equal to a contrast threshold, the server can obtain the gray-level peak value of the gray-level histogram and determine the segmentation threshold for background segmentation of the image region based on the gray-level peak value. Alternatively, if the contrast of the image region represented by the texture attribute is less than the contrast threshold, the server can determine the gray-level mean value of the image region based on each gray-level value and determine the segmentation threshold for background segmentation of the image region based on the gray-level mean value and the gray-level standard deviation.

[0057] Step S208: Perform background segmentation on the image region according to the segmentation threshold to obtain the foreground image in the image region.

[0058] Background segmentation is used to divide the image region into two parts: the foreground (chromosome) and the background, and to remove irrelevant signals.

[0059] Specifically, background segmentation is achieved through binarization. The server compares the grayscale values ​​of all pixels within a region with a threshold; pixels above the threshold are set to 1 (foreground, i.e., chromosome), and those below the threshold are set to 0 (background). To optimize the segmentation results, morphological processing is performed: opening operations are used to eliminate small-area noise, and closing operations are used to fill breaks or holes inside the chromosome. Finally, the foreground boundaries are smoothed to reduce jagged artifacts, resulting in a clean and continuous chromosome foreground image.

[0060] Step S210: stitch together the foreground images of each image region to obtain the chromosome image.

[0061] Specifically, stitching involves combining the foreground images obtained from processing each image region into a complete image according to their original positions. First, the coordinates of each foreground image in the global image are determined based on the row and column indices used during region division. If there is overlap between regions, a weighted average or maximum value fusion method is used to eliminate seams, ensuring a natural stitching transition. Finally, the integrity of the stitched chromosome is checked, and any potential local breaks are repaired. The resulting chromosome image covers the entire sample region, providing highly complete visualization data for subsequent karyotype analysis or disease diagnosis.

[0062] The aforementioned chromosome segmentation method employs a region-based processing strategy, independently calculating texture attributes and grayscale statistics for each image region. This adapts to contrast differences in different regions, effectively addressing segmentation errors caused by uneven illumination. The grayscale statistical method based on texture feature matching accurately characterizes the periodic structure of chromosome stripes, thereby determining a more reasonable segmentation threshold and avoiding misidentification of cell fragments as chromosome regions. The foreground image extracted through threshold segmentation preserves the complete morphological outline of the chromosome, reducing chromosome breakage issues caused by edge blurring in traditional methods. The final stitched foreground image combines high-precision segmentation results from each region, significantly improving the accuracy of chromosome segmentation. It is understood that the method in this embodiment can also use a semantic segmentation model to remove background, impurity cell nuclei, and other regions to obtain a chromosome image.

[0063] In one exemplary embodiment, the chromosome segmentation method further includes: acquiring a pre-trained labeled chromosome detection model; and detecting chromosome images based on the labeled chromosome detection model to obtain detection results for the labeled chromosomes.

[0064] The labeled chromosome detection model is an algorithm trained using deep learning (such as convolutional neural networks) that can identify the location and type of specific markers (such as abnormal structures, missing segments, or specific gene loci) in chromosome images. The detection result is the labeled chromosome information output by the model, including marker type, location coordinates, and confidence score.

[0065] Specifically, after stitching the chromosome images, further detection is needed to identify the presence of marker chromosomes. First, the server loads a pre-trained marker chromosome detection model from the database. This model is typically trained on a large amount of labeled data containing both normal and abnormal chromosome samples, and possesses the ability to recognize specific markers. The stitched chromosome image is then input into the marker chromosome detection model, which analyzes image features pixel-by-pixel, such as texture, edges, and color distribution. Afterward, the model locates areas where markers may exist and outputs the detection results. For example, the detection result might indicate a deletion of the long arm of chromosome 5 or successful fluorescent probe labeling. These results can help doctors quickly determine whether a sample contains genetic abnormalities. Figure 3 The image shown is the result of successful detection of the marked chromosome.

[0066] In this embodiment, automated model detection replaces manual microscope observation, significantly improving the efficiency and accuracy of labeled chromosome identification and reducing the risk of missed or misdiagnosis.

[0067] In an exemplary embodiment, determining the texture attributes of an image region based on the grayscale values ​​of each pixel in the image region includes: determining the grayscale standard deviation and grayscale histogram of the image region by statistically analyzing the grayscale values ​​of each pixel in the image region; and determining the texture attributes of the image region based on the grayscale histogram and grayscale standard deviation.

[0068] Among them, the grayscale standard deviation reflects the dispersion of pixel grayscale values ​​in an image region; the larger the standard deviation, the higher the contrast. The grayscale histogram is a distribution curve of the number of pixels at each grayscale level in an image region, with the peak representing the dominant grayscale level. Texture attributes describe the spatial characteristics of the grayscale distribution in an image region and are used to distinguish chromosomes from the background.

[0069] Specifically, for each image region, the grayscale values ​​of all pixels are first counted, the grayscale standard deviation is calculated, and a grayscale histogram is generated. The grayscale standard deviation measures contrast, while the grayscale histogram displays the grayscale distribution. If the grayscale histogram exhibits a bimodal distribution (one peak for the chromosome and one for the background) and a large standard deviation, it indicates high contrast in the region, and the texture attribute may be characterized by "clear stripes," meaning the contrast of the image region is greater than or equal to the contrast threshold. If the histogram is unimodal and the standard deviation is small, the contrast is low, and the texture attribute may be characterized by "uniform blur," meaning the contrast of the image region is less than the contrast threshold. By combining the grayscale standard deviation and the shape of the grayscale histogram, the texture features of the region can be quantified, providing a basis for subsequent segmentation threshold selection.

[0070] In this embodiment, the texture of the region is accurately quantified by combining grayscale standard deviation and histogram analysis, avoiding misjudgment caused by a single indicator and improving the adaptability of background segmentation.

[0071] In an exemplary embodiment, the grayscale statistics include the grayscale peak value of the grayscale histogram; based on the grayscale statistics, determining a segmentation threshold for background segmentation of the image region includes: obtaining the grayscale peak value of the grayscale histogram when the contrast of the image region represented by the texture attribute is greater than or equal to a contrast threshold; and determining the segmentation threshold for background segmentation of the image region based on the grayscale peak value.

[0072] The contrast threshold is a pre-set value used to determine whether the contrast of a region is high enough. For example, with a contrast threshold of 30, if the grayscale standard deviation is greater than 30, then the contrast of the image region represented by the texture attribute is considered to be greater than or equal to the contrast threshold. The grayscale peak is the highest point in the grayscale histogram, representing the dominant grayscale level of the region. For example, the grayscale peak can refer to the typical brightness of a chromosome or background.

[0073] Specifically, when the region texture attribute shows a contrast greater than or equal to the contrast threshold, it indicates a significant difference in grayscale between the chromosome and the background. In this case, directly analyze the grayscale histogram to locate the highest peak. For example, the valley value between the two peaks can be directly taken. The grayscale histogram has peaks at grayscale values ​​of 80 and 180, with the lowest point at 120. Therefore, the obtained segmentation threshold is 120. Alternatively, the Otsu algorithm can be used to automatically find the valley, which is more suitable for cases with obvious bimodalities.

[0074] In this embodiment, the threshold is determined directly using the grayscale peak value in the high-contrast region, which simplifies the calculation process and ensures clear segmentation boundaries while reducing noise interference.

[0075] In an exemplary embodiment, the grayscale statistics include the grayscale mean of the image region; based on the grayscale statistics, determining a segmentation threshold for background segmentation of the image region includes: determining the grayscale mean of the image region according to each grayscale value when the texture attribute indicates that the contrast of the image region is less than the contrast threshold; and determining the segmentation threshold for background segmentation of the image region based on the grayscale mean and the grayscale standard deviation.

[0076] Among them, the grayscale mean is the average grayscale value of all pixels in the image area, reflecting the overall brightness.

[0077] Specifically, when the contrast of a region's texture attribute is less than the contrast threshold, it indicates that the gray levels of the chromosome and the background are similar. This could be due to uneven lighting or blurred chromosome edges. The threshold needs to be determined by combining the gray-level mean and standard deviation. First, calculate the gray-level mean. Set the segmentation threshold to "gray-level mean - k × gray-level standard deviation" (k is an empirical coefficient, usually taken as 0.2~0.5) to differentiate the foreground from the background. For example: gray-level mean = 100, gray-level standard deviation = 20, k = 0.3, threshold = 100 - 6 = 94.

[0078] In this embodiment, the threshold is dynamically adjusted by the grayscale mean and standard deviation in the low-contrast region to avoid over-segmentation or under-segmentation caused by similar grayscale values, thereby improving robustness.

[0079] In an exemplary embodiment, the grayscale statistics include the grayscale mean of the image region; based on the grayscale statistics, a segmentation threshold for background segmentation of the image region is determined, including: when the texture attribute indicates that the sharpness of the image region is greater than or equal to the sharpness threshold, obtaining a preset upper limit of the standard deviation set for the grayscale standard deviation; and determining the segmentation threshold for background segmentation of the image region based on the grayscale mean, the grayscale standard deviation, and the preset upper limit of the standard deviation.

[0080] The sharpness threshold is a preset value used to determine whether the texture of a region is sharp. The preset standard deviation upper limit is the maximum allowed standard deviation value used to limit the influence of noise (e.g., set to 20).

[0081] Specifically, when the clarity of the region's texture attribute is greater than or equal to the clarity threshold, it indicates that the grayscale standard deviation is moderate. For example, the grayscale standard deviation may be between 30 and 80, and the grayscale histogram may have multiple small or wide peaks, meaning that local details exist. In this embodiment, when the standard deviation is large, the threshold deviates more from the mean (preserving details); when the standard deviation is small, the threshold is close to the mean (avoiding noise). In this case, the segmentation threshold = grayscale mean × (1 - k × (1 - Std / R)) (R is usually taken as 128, and k is taken as 0.2~0.5). For example: if the mean = 150, the grayscale standard deviation Std = 60, the maximum standard deviation of R preset = 128, and k = 0.3, then the segmentation threshold = 150 × (1 - 0.3 × (1 - 60 / 128)) ≈ 150 × 0.85 ≈ 127.5.

[0082] In this embodiment, the upper limit of the standard deviation is used to constrain the threshold in the high-resolution region to balance detail preservation and noise suppression, and to avoid loss of chromosome information due to excessive smoothing.

[0083] For example, if the texture attributes do not meet the above conditions, the segmentation threshold can be obtained by adding or subtracting a fixed value from the grayscale mean. For example: segmentation threshold = grayscale mean ± 30 (adjust the ± value according to the actual effect) or directly take the grayscale mean.

[0084] In a specific embodiment, such as Figure 4 As shown, a chromosome segmentation method is also provided, including:

[0085] Step S401: Obtain a sampled image of the chromosome;

[0086] The sampled image includes multiple image regions;

[0087] Step S402: For each image region, determine the gray standard deviation and gray histogram of the image region by statistically analyzing the gray values ​​of each pixel in the image region.

[0088] Step S403: Determine the texture attributes of the image region based on the gray-level histogram and gray-level standard deviation;

[0089] Step S404: If the contrast of the image region represented by the texture attribute is greater than or equal to the contrast threshold, obtain the gray value peak of the gray value histogram.

[0090] Step S405: Determine the segmentation threshold for background segmentation of the image region based on the grayscale peak value;

[0091] Step S406: When the contrast of the image region represented by the texture attribute is less than the contrast threshold, determine the average gray value of the image region based on each gray value.

[0092] Step S407: Determine the segmentation threshold for background segmentation of the image region based on the gray mean and gray standard deviation;

[0093] Step S408: If the sharpness of the image region represented by the texture attribute is greater than or equal to the sharpness threshold, obtain the preset upper limit of the standard deviation set for the grayscale standard deviation.

[0094] Step S409: Based on the gray mean, gray standard deviation and preset upper limit of standard deviation, determine the segmentation threshold used for background segmentation of the image region;

[0095] Step S410: Perform background segmentation on the image region according to the segmentation threshold to obtain the foreground image in the image region;

[0096] Step S411: stitch together the foreground images of each image region to obtain the chromosome image;

[0097] Step S412: Obtain the pre-trained labeled chromosome detection model;

[0098] Step S413: Detect chromosome images based on the labeled chromosome detection model to obtain detection results for labeled chromosomes.

[0099] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0100] Based on the same inventive concept, this application also provides a chromosome segmentation apparatus for implementing the chromosome segmentation method described above. The solution provided by this apparatus is similar to the implementation described in the above method; therefore, the specific limitations in one or more chromosome segmentation apparatus embodiments provided below can be found in the limitations of the chromosome segmentation method described above, and will not be repeated here.

[0101] In one exemplary embodiment, such as Figure 5As shown, a chromosome segmentation device 500 is provided, including: a sampling image acquisition module 502, a texture attribute determination module 504, a segmentation threshold determination module 506, a background segmentation module 508, and a foreground image stitching module 510, wherein:

[0102] The sampling image acquisition module 502 is used to acquire a sampling image of the chromosome; the sampling image includes multiple image regions;

[0103] The texture attribute determination module 504 is used to determine the texture attribute of each image region and the gray-scale statistical value that matches the texture attribute based on the gray-scale value of each pixel in the image region for each image region.

[0104] The segmentation threshold determination module 506 is used to determine the segmentation threshold for background segmentation of the image region based on grayscale statistical values.

[0105] Background segmentation module 508 is used to segment the background of the image region according to the segmentation threshold to obtain the foreground image in the image region;

[0106] The foreground image stitching module 510 is used to stitch together the foreground images of each image region to obtain a chromosome image.

[0107] In one exemplary embodiment, the chromosome segmentation device 500 further includes a labeled chromosome detection module, specifically used for:

[0108] Obtain a pre-trained labeled chromosome detection model;

[0109] Chromosome images are detected using a labeled chromosome detection model to obtain detection results for labeled chromosomes.

[0110] In an exemplary embodiment, the texture attribute determination module 504 is specifically used for:

[0111] By statistically analyzing the gray values ​​of each pixel in an image region, the gray standard deviation and gray histogram of the image region are determined.

[0112] The texture attributes of image regions are determined based on gray-level histograms and gray-level standard deviations.

[0113] In one exemplary embodiment, the grayscale statistics include the grayscale peak values ​​of the grayscale histogram. In this embodiment, the segmentation threshold determination module 506 is further configured to:

[0114] When the contrast of the image region represented by the texture attribute is greater than or equal to the contrast threshold, obtain the gray value peak of the gray value histogram.

[0115] Based on the grayscale peak value, determine the segmentation threshold used for background segmentation of the image region.

[0116] In an exemplary embodiment, the grayscale statistics include the average grayscale value of the image region. In this embodiment, the segmentation threshold determination module 506 is further configured to:

[0117] When the contrast of an image region represented by the texture attribute is less than the contrast threshold, the average gray value of the image region is determined based on each gray value.

[0118] Based on the gray mean and gray standard deviation, a segmentation threshold is determined for background segmentation of image regions.

[0119] In an exemplary embodiment, the segmentation threshold determination module 506 is further configured to:

[0120] If the sharpness of the image region represented by the texture attribute is greater than or equal to the sharpness threshold, obtain the upper limit of the preset standard deviation set for the grayscale standard deviation;

[0121] Based on the gray mean, gray standard deviation, and a preset upper limit of standard deviation, a segmentation threshold is determined for background segmentation of the image region.

[0122] Each module in the aforementioned chromosome segmentation device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0123] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 6As shown, the computer device includes a processor, memory, input / output interfaces, a communication interface, a display unit, and an input device. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When executed by the processor, the computer program implements a chromosome segmentation method. The display unit is used to form a visually visible image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0124] Those skilled in the art will understand that Figure 6 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0125] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described above.

[0126] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0127] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the method described above.

[0128] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0129] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0130] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0131] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A chromosome segmentation method, characterized in that, The method includes: Acquire sampled images of chromosomes; the sampled images include multiple image regions; For each image region, based on the grayscale value of each pixel in the image region, the texture attribute of the image region and the grayscale statistical value matching the texture attribute are determined; Based on the grayscale statistical values, a segmentation threshold for background segmentation of the image region is determined; The image region is segmented according to the segmentation threshold to obtain the foreground image in the image region; By stitching together the foreground images of each of the aforementioned image regions, a chromosome image is obtained.

2. The method according to claim 1, characterized in that, The method further includes: Obtain a pre-trained labeled chromosome detection model; The chromosome image is detected based on the labeled chromosome detection model to obtain detection results for the labeled chromosome.

3. The method according to claim 1, characterized in that, Determining the texture attributes of the image region based on the grayscale values ​​of each pixel in the image region includes: By statistically analyzing the gray values ​​of each pixel in the image region, the gray standard deviation and gray histogram of the image region are determined. The texture attributes of the image region are determined based on the gray-level histogram and the gray-level standard deviation.

4. The method according to claim 3, characterized in that, The gray-level statistical values ​​include the gray-level peak values ​​of the gray-level histogram; determining the segmentation threshold for background segmentation of the image region based on the gray-level statistical values ​​includes: If the contrast of the image region indicated by the texture attribute is greater than or equal to the contrast threshold, the gray value peak of the gray value histogram is obtained. Based on the grayscale peak value, a segmentation threshold for background segmentation of the image region is determined.

5. The method according to claim 4, characterized in that, The grayscale statistical value includes the average grayscale value of the image region; the step of determining a segmentation threshold for background segmentation of the image region based on the grayscale statistical value includes: If the texture attribute indicates that the contrast of the image region is less than the contrast threshold, the average gray value of the image region is determined based on each of the gray values. Based on the mean gray level and the standard deviation of gray level, a segmentation threshold for background segmentation of the image region is determined.

6. The method according to claim 5, characterized in that, The step of determining the segmentation threshold for background segmentation of the image region based on the grayscale statistical values ​​includes: If the texture attribute indicates that the sharpness of the image region is greater than or equal to the sharpness threshold, obtain the upper limit of the preset standard deviation set for the grayscale standard deviation; Based on the mean gray level, the standard deviation of gray level, and the preset upper limit of standard deviation, a segmentation threshold for background segmentation of the image region is determined.

7. A chromosome splitting device, characterized in that, The device includes: A sampling image acquisition module is used to acquire sampling images of chromosomes; the sampling image includes multiple image regions; The texture attribute determination module is used to determine the texture attribute of each image region and the grayscale statistical value matching the texture attribute based on the grayscale value of each pixel in the image region for each image region. The segmentation threshold determination module is used to determine a segmentation threshold for background segmentation of the image region based on the grayscale statistical value. The background segmentation module is used to segment the image region according to the segmentation threshold to obtain the foreground image in the image region; The foreground image stitching module is used to stitch together the foreground images of each of the image regions to obtain the chromosome image.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images