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

Abstract

This application relates to a chromosome image processing method, apparatus, computer device, computer-readable storage medium, and computer program product. The method includes: acquiring an input image containing multiple chromosomes in metaphase; extracting individual chromosome images from the input image; performing feature analysis on each chromosome based on its image to obtain multidimensional chromosome feature information; and determining the phase quality assessment result of the input image based on statistical information corresponding to the multidimensional chromosome feature information. This method can improve the accuracy of chromosome image processing.

Description

Technical Field

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

[0002] With the rapid development of biomedical imaging technology, chromosome microscopic image analysis has become an important tool in genetic research and clinical diagnosis. In cytogenetics, metaphase chromosomes are widely used for karyotype analysis due to their highly condensed and morphologically stable characteristics.

[0003] Traditional methods require manual observation of chromosome images under a microscope, relying on experience to judge the quality of mitotic phases, and then selecting samples that meet the analysis criteria. However, manual assessment is highly subjective, inefficient, and susceptible to fatigue, especially when processing large numbers of samples, making it difficult to guarantee the consistency of assessment results. Therefore, traditional chromosome image processing methods suffer from inaccuracies. Summary of the Invention

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

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

[0006] Obtain an input image containing multiple chromosomes in the metaphase phase;

[0007] Extract the chromosome image of each chromosome from the input image;

[0008] Based on the images of each chromosome, feature analysis is performed on each chromosome to obtain the multidimensional chromosomal feature information of each chromosome.

[0009] Based on the statistical information corresponding to the multidimensional feature information of each chromosome, the quality assessment result of the splitting phase of the input image is determined.

[0010] In one embodiment, determining the cleavage phase quality assessment result of the input image based on the statistical information corresponding to the multidimensional feature information of each chromosome includes:

[0011] By fusing the multidimensional feature information of each chromosome according to a set dimension, the statistical information of the input image under the set dimension is obtained.

[0012] Based on the statistical information, the quality assessment result of the split phase of the input image is determined.

[0013] In one embodiment, the number of defined dimensions is multiple; determining the split phase quality assessment result of the input image based on the statistical information includes:

[0014] Obtain the weights corresponding to each of the defined dimensions;

[0015] The statistical information under each of the defined dimensions is weighted according to the weights described above to obtain weighted statistical information.

[0016] Based on the weighted statistical information, the split phase quality assessment result of the input image is determined.

[0017] In one embodiment, the defined dimension includes a monomer morphology dimension; the chromosome multidimensional feature information includes chromosome area and chromosome length;

[0018] The step of fusing the multidimensional feature information of each chromosome according to a set dimension to obtain the statistical information of the input image under the set dimension includes:

[0019] For each chromosome, determine the ratio of the chromosome area to an area threshold, and the difference between the chromosome length and a length threshold.

[0020] Based on the area ratio and the length difference, the chromosome score in the monomorphic dimension is determined; the chromosome score is positively correlated with the area ratio and negatively correlated with the length difference.

[0021] By statistically analyzing the chromosome scores of each chromosome in the monomer morphology dimension, statistical information of the input image in the monomer morphology dimension is obtained.

[0022] In one embodiment, the defined dimension includes a spatial distribution dimension; the chromosome multidimensional feature information includes chromosome centroid coordinates;

[0023] The step of fusing the multidimensional feature information of each chromosome according to a set dimension to obtain the statistical information of the input image under the set dimension includes:

[0024] Obtain the image center coordinates of the input image;

[0025] For each chromosome, determine the distance between the chromosome centroid coordinates and the image center coordinates;

[0026] Based on the average distance of each of the aforementioned distances, the statistical information of the input image under the spatial distribution dimension is determined; the average distance is positively correlated with the statistical information under the spatial distribution dimension.

[0027] In one embodiment, the defined dimension includes an overlapping dimension;

[0028] The step of fusing the multidimensional feature information of each chromosome according to a set dimension to obtain the statistical information of the input image under the set dimension includes:

[0029] For each chromosome, determine the Euclidean distance between the chromosome's centroid coordinates and the centroid coordinates of other chromosomes.

[0030] The Euclidean distance with the smallest distance among all the Euclidean distances is selected as the nearest neighbor distance of the chromosome.

[0031] Based on the average nearest neighbor distance of each nearest neighbor distance, the statistical information of the input image under the overlapping dimension is determined; the average nearest neighbor distance is positively correlated with the statistical information under the overlapping dimension.

[0032] Secondly, this application also provides a chromosome image processing apparatus, comprising:

[0033] The input image acquisition module is used to acquire an input image containing multiple chromosomes in the metaphase phase.

[0034] A chromosome image extraction module is used to extract individual chromosome images from the input image;

[0035] The feature analysis module is used to perform feature analysis on each chromosome based on the chromosome images to obtain the multidimensional chromosome feature information of each chromosome.

[0036] The quality assessment result determination module is used to determine the quality assessment result of the splitting phase of the input image based on the statistical information corresponding to the multidimensional feature information of each chromosome.

[0037] 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.

[0038] 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.

[0039] 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.

[0040] The aforementioned chromosome image processing methods, apparatus, computer equipment, computer-readable storage media, and computer program products achieve precise localization and isolation of individual chromosomes by automatically extracting independent images of each chromosome from the input image, avoiding feature confusion caused by chromosome overlap or background interference in traditional methods. Multidimensional feature analysis based on each chromosome image can quantitatively describe the morphological and structural information of chromosomes, providing objective data support for assessing the quality of mitotic phases. Further statistical analysis of the multidimensional feature information of each chromosome allows for a comprehensive evaluation of the overall quality of mitotic phases, replacing subjective human judgment. Ultimately, this method significantly improves the accuracy of mitotic phase screening through automated feature extraction and quantitative evaluation. Attached Figure Description

[0041] 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.

[0042] Figure 1 This is an application environment diagram of a chromosome image processing method in one embodiment;

[0043] Figure 2 This is a flowchart illustrating a chromosome image processing method in one embodiment;

[0044] Figure 3 This is a high-quality sample image of the chromosome image quality assessment process in one embodiment;

[0045] Figure 4 This is a low-quality sample image of the chromosome image quality assessment process in one embodiment;

[0046] Figure 5 This is a flowchart illustrating the chromosome image processing method in another embodiment;

[0047] Figure 6 This is a structural block diagram of a chromosome image processing device in one embodiment;

[0048] Figure 7 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0049] 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.

[0050] The chromosome image processing method provided in this application embodiment can be applied to, for example... Figure 1 In 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 on a cloud or other network server. 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 image processing, server 102 acquires an input image containing multiple chromosomes in metaphase from image acquisition device 104; extracts individual chromosome images from the input image; performs feature analysis on each chromosome based on its image to obtain multidimensional chromosome feature information; and determines the phase quality assessment result of the input image based on the statistical information corresponding to the multidimensional chromosome feature information.

[0051] In one exemplary embodiment, such as Figure 2 As shown, a chromosome image processing method is provided, which can be applied to... Figure 1 Taking server 102 as an example, the explanation includes the following steps S202 to S208. Wherein:

[0052] Step S202: Obtain an input image containing multiple chromosomes in the metaphase phase.

[0053] Metaphase is a specific stage in the process of cell mitosis. During this stage, chromosomes align on the equatorial plate in the center of the cell, exhibiting a relatively stable morphology and clear structure, making them easy to observe and analyze. It is a crucial period for studying chromosome morphology, structure, and number. Input images, which are images containing chromosomes in metaphase and captured by imaging equipment such as microscopes, serve as the foundational data source for subsequent processing and analysis.

[0054] Specifically, this step requires the use of appropriate microscopy equipment, such as an optical microscope or an electron microscope, to photograph cells in metaphase of mitosis. During the imaging process, it is crucial to ensure that the microscope parameters are set appropriately, such as magnification and focal length, to guarantee a clear image that clearly shows the multiple chromosomes in metaphase within the cell. Simultaneously, it is essential to maintain the stability of the imaging environment to avoid image blurring due to factors such as vibration. The input image can be a single image or a sequence of consecutively captured images.

[0055] Step S204: Extract the chromosome images of each chromosome from the input image.

[0056] Chromosome images, in particular, refer to the portion of an input image containing only a single chromosome, separated from an input image that contains multiple chromosomes. They remove interference from other chromosomes and the background, focusing more on the features of a single chromosome.

[0057] Specifically, the input image first needs to be preprocessed, such as through image enhancement to improve contrast and clarity, making chromosome boundaries more distinct. Then, image segmentation algorithms, such as threshold-based, edge-detection-based, or region-growing-based methods, are used to separate each chromosome from the overall image. During segmentation, the parameters of the segmentation algorithm may need to be adjusted based on the chromosome's morphology, color, and other features to ensure accurate extraction of each chromosome's image. For chromosomes that are stuck together, special processing techniques, such as morphological processing and watershed algorithms, are required to separate them and extract their individual images. For example, chromosome images can be segmented using a deep learning segmentation model. This deep learning segmentation model can be trained using a labeled chromosome segmentation dataset. Segmentation models could be, for example, Mask R-CNN (Mask Region-based Convolutional Neural Network) or SAM (Segment Anything Model).

[0058] Step S206: Based on the images of each chromosome, perform feature analysis on each chromosome to obtain the multidimensional chromosomal feature information of each chromosome.

[0059] Feature analysis is the process of extracting and quantifying various features of chromosomes. By analyzing the morphology, structure, texture, and other characteristics of chromosomes, detailed information about them can be obtained. Multidimensional chromosome feature information refers to a collection of information describing chromosome features from multiple dimensions. These dimensions may include average chromosome compactness, average area and border compactness, relative length fraction, overlapping area, convex hull defects, average Euclidean distance, average Euclidean distance between average center points, Hopkins statistic, nearest neighbor distance, average confidence level, confidence level standard deviation, chromosome number, chromosome category entropy, chromosome category number deviation, autosomal category and number entropy, length, width, centromere position, arm ratio, banding pattern, etc.

[0060] Specifically, for each extracted chromosome image, feature analysis is performed from multiple aspects. In terms of morphological features, the length and width of the chromosome are measured, the position of the centromere is determined, and the length ratio of the long arm to the short arm is calculated. Regarding structural features, the banding patterns of the chromosome are observed; different chromosomes have different banding distributions, and these bands can be displayed using specific staining methods. Analysis of the banding can further identify the chromosome type and structural characteristics. In terms of texture features, the grayscale distribution and texture roughness of the chromosome image are analyzed. These texture features can reflect the differences in the surface structure and composition of the chromosome. Through these multi-faceted analyses, the obtained feature information is organized and quantified to form multi-dimensional feature information for each chromosome.

[0061] Step S208: Based on the statistical information corresponding to the multidimensional feature information of each chromosome, determine the quality assessment result of the splitting phase of the input image.

[0062] The statistical information refers to the results obtained after statistical analysis of the multidimensional feature information of each chromosome, such as feature evaluation scores, mean, standard deviation, coefficient of variation, etc. This statistical information can reflect the quality, overall distribution, and dispersion of chromosome features. The cell division phase quality assessment result is a comprehensive evaluation of the quality of cell division phases in the input image based on the statistical analysis of the multidimensional feature information of chromosomes, used to determine whether the cell division phase is suitable for subsequent research and analysis.

[0063] Optionally, the server can directly perform statistical analysis on the multidimensional feature information of each chromosome, calculating the statistics for each feature, such as the mean and standard deviation of all chromosome lengths, and the coefficient of variation of centromere position distribution. Then, corresponding evaluation criteria are formulated based on this statistical information. For example, a small standard deviation in chromosome length indicates relatively consistent chromosome length and good mitotic quality; a large coefficient of variation in centromere position distribution may indicate abnormal chromosome structure or problems in the mitotic process, resulting in poor mitotic quality. By combining the statistical information of each feature and the evaluation criteria, the quality of the mitotic phase in the input image is comprehensively evaluated, yielding a complete assessment result that determines whether the mitotic phase is high-quality, acceptable, or unacceptable, providing a reference for subsequent research and analysis.

[0064] Optionally, the server can also fuse multidimensional feature information of each chromosome according to a set dimension to obtain statistical information of the input image under the set dimension, and determine the cleavage phase quality assessment result of the input image based on the statistical information. Furthermore, the number of set dimensions can be one or more.

[0065] The aforementioned chromosome image processing method achieves precise localization and isolation of individual chromosomes by automatically extracting independent images of each chromosome from the input image, avoiding feature confusion caused by chromosome overlap or background interference in traditional methods. Multidimensional feature analysis based on each chromosome image quantifies the morphological and structural information of the chromosomes, providing objective data support for assessing the quality of mitotic phases. Further statistical analysis of the multidimensional feature information of each chromosome allows for a comprehensive evaluation of the overall quality of mitotic phases, replacing subjective human judgment. Ultimately, this method significantly improves the accuracy of mitotic phase selection through automated feature extraction and quantitative evaluation.

[0066] In an exemplary embodiment, the quality assessment result of the splitting phase of the input image is determined based on the statistical information corresponding to the multidimensional feature information of each chromosome, including: fusing the multidimensional feature information of each chromosome according to a set dimension to obtain the statistical information of the input image under the set dimension; and determining the quality assessment result of the splitting phase of the input image based on the statistical information.

[0067] In this context, "defined dimensions" refers to specific directions or categories used for classifying and comprehensively analyzing the multidimensional features of chromosomes. Examples include monomer morphology dimensions, spatial distribution dimensions, and overlap dimensions, reflecting the characteristics and quality of chromosomes from different perspectives. For instance, the monomer morphology dimension focuses on the morphological features of the chromosome itself, the spatial distribution dimension focuses on the positional distribution of chromosomes in an image, and the overlap dimension focuses on whether chromosomes overlap. Statistical information, such as the mean and standard deviation, is the result of statistical analysis of the multidimensional features of each chromosome within the defined dimensions. It describes the overall distribution and dispersion of chromosome features within that dimension.

[0068] Specifically, the process begins by clearly defining dimensions, which are key directions for classifying chromosome features from multiple perspectives. Next, the multidimensional feature information of each chromosome is integrated according to these defined dimensions; for example, features related to chromosome morphology are grouped into the morphological dimension, and features related to spatial location into the spatial dimension. Statistical analysis is then performed on the feature information under each defined dimension, calculating statistical measures such as mean and standard deviation. These statistical measures reflect the overall situation of chromosome features under that dimension. Finally, based on this statistical information and according to pre-defined evaluation criteria, the quality assessment result of the splitting phase of the input image is determined. For example, if the statistical information falls within a certain range, it is judged as a high-quality splitting phase; if it exceeds the range, it is judged as unqualified.

[0069] In this embodiment, by setting dimensions to fuse feature information and evaluating it, the characteristics of chromosomes can be comprehensively considered from multiple perspectives, making the assessment results of mitotic phase quality more comprehensive and accurate, and avoiding the one-sidedness of single feature assessment.

[0070] In an exemplary embodiment, the number of dimensions is set to multiple; the split phase quality assessment result of the input image is determined based on statistical information, including: obtaining the weights corresponding to each set dimension; weighting the statistical information under each set dimension according to each weight to obtain weighted statistical information; and determining the split phase quality assessment result of the input image based on the weighted statistical information.

[0071] In multi-dimensional comprehensive evaluation, the weight is a numerical value assigned to each set dimension to represent the importance of that dimension in the overall evaluation. The larger the weight, the greater the impact of that dimension on the evaluation result.

[0072] Specifically, since there are multiple dimensions, their importance in assessing the quality of the splitting phase may vary. Therefore, it is necessary to first obtain the weight corresponding to each dimension. The weights can be determined based on actual needs and experience; for example, dimensions with a greater impact on splitting phase quality can be assigned larger weights. Then, the statistical information under each dimension is weighted according to these weights. For example, the statistics of the morphological dimension are multiplied by their weights, the statistics of the spatial dimension are multiplied by their weights, and so on. Finally, all the weighted statistics are summed to obtain the weighted statistical information. Based on the weighted statistical information, the splitting phase quality assessment result of the input image is determined according to the evaluation criteria. The weighted statistical information better reflects the degree of influence of each dimension on the overall quality.

[0073] In this embodiment, the differences in importance of different set dimensions in the evaluation are considered, and the evaluation results are made more reasonable through weighted processing, which can more accurately reflect the actual quality of the split phase of the input image.

[0074] In an exemplary embodiment, the defined dimension includes a single-cell morphology dimension; the chromosome multidimensional feature information includes chromosome area and chromosome length; the multidimensional feature information of each chromosome is fused according to the defined dimension to obtain the statistical information of the input image under the defined dimension, including: for each chromosome, determining the area ratio of the chromosome area to an area threshold and the length difference between the chromosome length and a length threshold; based on the area ratio and length difference, determining the chromosome score under the single-cell morphology dimension; the chromosome score is positively correlated with the area ratio and negatively correlated with the length difference; by statistically analyzing the chromosome scores of each chromosome under the single-cell morphology dimension, the statistical information of the input image under the single-cell morphology dimension is obtained.

[0075] The area threshold is a pre-set standard value used to measure whether the chromosome area is normal. For example, in this embodiment, the area threshold is the area of ​​the bounding rectangle of the chromosome. By comparing it with the actual chromosome area, it can be determined whether the chromosome meets the requirements in terms of area. The length threshold is a pre-set standard value used to measure whether the chromosome length is normal. By comparing it with the actual chromosome length, it can be determined whether the chromosome meets the requirements in terms of length. The chromosome score is a quantitative value determined based on the area ratio and length difference of the chromosome in the monomorphic dimension. It is used to represent the degree of conformity of the chromosome in terms of morphology. The larger the area ratio, the straighter the chromosome shape, and the higher the chromosome score; while the smaller the length difference, the closer the chromosome length is to the standard chromosome length, and the higher the chromosome score, indicating that the chromosome morphology is closer to the normal standard. The score is positively correlated with the area ratio and negatively correlated with the length difference. It is a set dimension for observing and evaluating chromosomes from the perspective of the morphological characteristics of the individual chromosome itself. It focuses on the morphological attributes of the chromosome itself, such as shape and size, and uses these attributes to measure whether the chromosome conforms to the normal standard in terms of morphology, thereby reflecting the quality of the chromosome in the mitotic phase.

[0076] Specifically, in the single-chromosome morphology dimension, chromosome area and length are used as the main feature information. For each chromosome, the difference between its area and a preset area threshold, and the difference between its length and a preset length threshold are calculated first. The score for that chromosome in the single-chromosome morphology dimension is determined based on these two differences. A larger area ratio and a smaller length difference indicate that the chromosome morphology better conforms to the standard, resulting in a higher score. That is, the score is positively correlated with the area ratio and negatively correlated with the length difference. Finally, the scores of all chromosomes in the single-chromosome morphology dimension are statistically analyzed, such as calculating the average score, highest score, and lowest score. These statistics constitute the statistical information of the input image in the single-chromosome morphology dimension.

[0077] In this embodiment, the morphology of chromosome monomorphism is quantitatively evaluated based on two key morphological features: chromosome area and length. This can accurately reflect the morphological quality of chromosomes and provide an important basis for evaluating the quality of mitotic phases.

[0078] In an exemplary embodiment, the defined dimension includes a spatial distribution dimension; the chromosome multidimensional feature information includes chromosome centroid coordinates; the multidimensional feature information of each chromosome is fused according to the defined dimension to obtain the statistical information of the input image under the defined dimension, including: obtaining the image center coordinates of the input image; for each chromosome, determining the distance between the chromosome centroid coordinates and the image center coordinates; determining the statistical information of the input image under the spatial distribution dimension based on the average distance of each distance; the average distance is positively correlated with the statistical information under the spatial distribution dimension.

[0079] The chromosome centroid coordinates are the coordinates of the chromosome's center of gravity in the image, used to describe the chromosome's position in space. The image center coordinates are the coordinates of the center of the input image, serving as a reference point for judging the spatial distribution of chromosomes. The spatial distribution dimension mainly focuses on the spatial position and arrangement of chromosomes in the cell division image. It considers the relative positional relationships between chromosomes and their distances from reference points such as the image center, using this information to assess whether the spatial distribution of chromosomes is uniform and reasonable, thus reflecting the quality of the mitotic phase.

[0080] Specifically, in the spatial distribution dimension, the centroid coordinates of chromosomes are used as feature information. First, the center coordinates of the input image are determined as a reference point for judging the spatial distribution of chromosomes. Then, for each chromosome, the distance between its centroid coordinates and the image center coordinates is calculated. This distance reflects the degree of deviation of the chromosome in the image. Finally, the average distance between all chromosomes and the image center coordinates is calculated. The smaller the average distance, the more concentrated the chromosome distribution in the image, the more likely it is to overlap, and the worse the spatial distribution quality. That is, the average distance is positively correlated with the statistical information in the spatial distribution dimension. This average distance is the statistical information of the input image in the spatial distribution dimension.

[0081] In this embodiment, by calculating the distance between the chromosome and the center of the image and statistically analyzing the average distance, the spatial distribution of chromosomes in the image can be intuitively reflected, which helps to assess whether the distribution of chromosomes in the mitotic phase is uniform and reasonable.

[0082] In an exemplary embodiment, the defined dimension includes an overlap dimension; the multidimensional feature information of each chromosome is fused according to the defined dimension to obtain the statistical information of the input image under the defined dimension, including: for each chromosome, determining the Euclidean distances corresponding to the centroid coordinates of the chromosome and the centroid coordinates of other chromosomes respectively; selecting the Euclidean distance with the smallest distance from each Euclidean distance as the nearest neighbor distance of the chromosome; determining the statistical information of the input image under the overlap dimension based on the average nearest neighbor distance of each nearest neighbor distance; the average nearest neighbor distance is positively correlated with the statistical information under the overlap dimension.

[0083] Euclidean distance is the straight-line distance between two points in two-dimensional or multi-dimensional space, used to measure the spatial proximity of chromosomes. Nearest neighbor distance is the smallest Euclidean distance among the centroid coordinates of all other chromosomes for a given chromosome. It reflects the distance of that chromosome to its nearest neighbor and is used to assess chromosome overlap. A smaller distance indicates a greater likelihood of overlap or proximity, thus resulting in a lower chromosome score. In other words, the average nearest neighbor distance is positively correlated with the statistical information under the overlap dimension. The overlap dimension is a defined dimension used to assess the degree of overlap between chromosomes. By analyzing indicators such as the nearest neighbor distance between chromosomes, it determines whether chromosomes overlap in the image and the severity of the overlap. The degree of overlap affects the assessment of the quality of the mitotic phase.

[0084] Specifically, in the overlap dimension, chromosome centroid coordinates are also used as feature information. For each chromosome, the Euclidean distance between its centroid coordinates and those of all other chromosomes in the image is calculated. These distances reflect the spatial proximity of that chromosome to other chromosomes. The smallest of these Euclidean distances is selected as the nearest neighbor distance for that chromosome, i.e., the distance to the nearest chromosome. Then, the average nearest neighbor distance of all chromosomes is calculated. The smaller the average nearest neighbor distance, the greater the probability of chromosomes overlapping or being close to each other, and the more severe the overlap. That is, the average nearest neighbor distance is positively correlated with the statistical information in the overlap dimension. This average nearest neighbor distance is the statistical information of the input image in the overlap dimension.

[0085] In this embodiment, by calculating the nearest neighbor distance and the average nearest neighbor distance, the overlap between chromosomes can be effectively assessed, providing important information about the degree of chromosome overlap for judging the quality of the mitotic phase, and helping to more accurately assess the quality of the mitotic phase.

[0086] For example, the above method achieves a closed-loop technology of deep segmentation, feature extraction, and automatic quality assessment. For instance, as... Figure 3 As shown, through this technical solution, Figure 3 The chromosomes in the middle are relatively long, well-shaped (long and with minimal curvature), well-dispersed (no overlap), and the bands are clear and easily distinguishable, making them suitable for analysis. The scoring system gives a high automatic score (0.86 points); while Figure 4 The chromosomes are short, generally dispersed, and the bands are unclear, making them difficult to identify and impossible to distinguish whether there are structural abnormalities in the chromosomes. They are not suitable for analysis, and the scoring system gives a low automatic score (0.47 points).

[0087] In a specific embodiment, such as Figure 5 As shown, a chromosome image processing method is also provided, including:

[0088] Step S501: Obtain an input image containing multiple chromosomes in metaphase.

[0089] Step S502: Extract the chromosome image of each chromosome from the input image;

[0090] Step S503: Based on the images of each chromosome, perform feature analysis on each chromosome to obtain the multidimensional feature information of each chromosome.

[0091] Step S504: For each chromosome, determine the ratio of the chromosome area to the area threshold, and the length difference between the chromosome length and the length threshold.

[0092] Step S505: Based on the area ratio and length difference, determine the chromosome score in the monomorphic dimension;

[0093] Among them, chromosome score is positively correlated with area ratio, and chromosome score is negatively correlated with length difference;

[0094] Step S506: By statistically analyzing the chromosome scores of each chromosome in the monomer morphology dimension, the statistical information of the input image in the monomer morphology dimension is obtained.

[0095] Step S507: Obtain the coordinates of the image center of the input image;

[0096] Step S508: For each chromosome, determine the distance between the chromosome centroid coordinates and the image center coordinates;

[0097] Step S509: Based on the average distance of each distance, determine the statistical information of the input image in the spatial distribution dimension;

[0098] Among them, the average distance is positively correlated with the statistical information under the spatial distribution dimension;

[0099] Step S510: For each chromosome, determine the Euclidean distance between the chromosome centroid coordinates and the other centroid coordinates of other chromosomes.

[0100] Step S511: Select the Euclidean distance with the smallest distance from all Euclidean distances as the nearest neighbor distance of the chromosome;

[0101] Step S512: Based on the average nearest neighbor distance of each nearest neighbor distance, determine the statistical information of the input image in the overlapping dimension;

[0102] Among them, the average nearest neighbor distance is positively correlated with the statistical information under the overlapping dimension; the number of dimensions is set to be multiple.

[0103] Step S513: Obtain the weights corresponding to each set dimension;

[0104] Step S514: Weight the statistical information under each set dimension according to each weight to obtain weighted statistical information;

[0105] Step S515: Based on weighted statistical information, determine the quality assessment result of the split phase of the input image.

[0106] 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.

[0107] Based on the same inventive concept, this application also provides a chromosome image processing apparatus for implementing the chromosome image processing 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 image processing apparatus embodiments provided below can be found in the limitations of the chromosome image processing method described above, and will not be repeated here.

[0108] In one exemplary embodiment, such as Figure 6 As shown, a chromosome image processing device 600 is provided, including: an input image acquisition module 602, a chromosome image extraction module 604, a feature analysis module 606, and a quality assessment result determination module 608, wherein:

[0109] The input image acquisition module 602 is used to acquire an input image containing multiple chromosomes in the metaphase phase.

[0110] The chromosome image extraction module 604 is used to extract the individual chromosome images of each chromosome from the input image;

[0111] The feature analysis module 606 is used to perform feature analysis on each chromosome based on each chromosome image to obtain the multidimensional feature information of each chromosome.

[0112] The quality assessment result determination module 608 is used to determine the quality assessment result of the splitting phase of the input image based on the statistical information corresponding to the multidimensional feature information of each chromosome.

[0113] In one exemplary embodiment, the quality assessment result determination module 608 includes:

[0114] The statistical information determination unit is used to fuse the multidimensional feature information of each chromosome according to a set dimension to obtain the statistical information of the input image under the set dimension.

[0115] The quality assessment result determination unit is used to determine the quality assessment result of the split phase of the input image based on statistical information.

[0116] In one exemplary embodiment, the number of dimensions is set to multiple. In this embodiment, the quality assessment result determination unit is specifically used for:

[0117] Obtain the weights corresponding to each defined dimension;

[0118] The statistical information under each set dimension is weighted according to each weight to obtain weighted statistical information;

[0119] Based on weighted statistical information, the quality assessment result of the split phase of the input image is determined.

[0120] In an exemplary embodiment, the defined dimension includes a monomer morphology dimension; the chromosome multidimensional feature information includes chromosome area and chromosome length. In this embodiment, the statistical information determination unit is further configured to:

[0121] For each chromosome, determine the ratio of the chromosome area to an area threshold, and the difference between the chromosome length and a length threshold.

[0122] Chromosome scores in the monomorphic dimension were determined based on area ratio and length difference; chromosome scores were positively correlated with area ratio and negatively correlated with length difference.

[0123] By statistically analyzing the chromosome scores of each chromosome in the monomer morphology dimension, we can obtain statistical information about the input image in the monomer morphology dimension.

[0124] In an exemplary embodiment, the defined dimension includes a spatial distribution dimension; the chromosome multidimensional feature information includes chromosome centroid coordinates. In this embodiment, the statistical information determination unit is further configured to:

[0125] Obtain the coordinates of the center of the input image;

[0126] For each chromosome, determine the distance between the chromosome's centroid coordinates and the image center coordinates;

[0127] Based on the average distance of each distance, the statistical information of the input image in the spatial distribution dimension is determined; the average distance is positively correlated with the statistical information in the spatial distribution dimension.

[0128] In one exemplary embodiment, the defined dimension includes an overlapping dimension. In this embodiment, the statistical information determination unit is further configured to:

[0129] For each chromosome, determine the Euclidean distances between the chromosome's centroid coordinates and the centroid coordinates of other chromosomes.

[0130] Select the Euclidean distance with the smallest distance from all Euclidean distances as the nearest neighbor distance of the chromosome;

[0131] Based on the average nearest neighbor distance of each nearest neighbor, the statistical information of the input image in the overlapping dimension is determined; the average nearest neighbor distance is positively correlated with the statistical information in the overlapping dimension.

[0132] Each module in the aforementioned chromosome image processing 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.

[0133] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 7 As 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 image processing 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.

[0134] Those skilled in the art will understand that Figure 7 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.

[0135] 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.

[0136] 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.

[0137] 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.

[0138] 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.

[0139] 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.

[0140] 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.

[0141] 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 image processing method, characterized in that, The method includes: Obtain an input image containing multiple chromosomes in metaphase; Extract the chromosome image of each chromosome from the input image; Based on the images of each chromosome, feature analysis is performed on each chromosome to obtain the multidimensional chromosomal feature information of each chromosome. Based on the statistical information corresponding to the multidimensional feature information of each chromosome, the quality assessment result of the splitting phase of the input image is determined.

2. The method according to claim 1, characterized in that, The determination of the cleavage phase quality assessment result of the input image based on the statistical information corresponding to the multidimensional feature information of each chromosome includes: By fusing the multidimensional feature information of each chromosome according to a set dimension, the statistical information of the input image under the set dimension is obtained. Based on the statistical information, the split phase quality assessment result of the input image is determined.

3. The method according to claim 2, characterized in that, The number of defined dimensions is multiple; the determination of the split phase quality assessment result of the input image based on the statistical information includes: Obtain the weights corresponding to each of the defined dimensions; The statistical information under each of the defined dimensions is weighted according to the weights described above to obtain weighted statistical information. Based on the weighted statistical information, the split phase quality assessment result of the input image is determined.

4. The method according to claim 2, characterized in that, The defined dimension includes the monomer morphology dimension; the chromosome multidimensional feature information includes chromosome area and chromosome length; The step of fusing the multidimensional feature information of each chromosome according to a set dimension to obtain the statistical information of the input image under the set dimension includes: For each chromosome, determine the ratio of the chromosome area to an area threshold and the length difference between the chromosome length and a length threshold. Based on the area ratio and the length difference, the chromosome score in the monomorphic dimension is determined; the chromosome score is positively correlated with the area ratio and negatively correlated with the length difference. By statistically analyzing the chromosome scores of each chromosome in the monomer morphology dimension, statistical information of the input image in the monomer morphology dimension is obtained.

5. The method according to claim 2, characterized in that, The defined dimension includes the spatial distribution dimension; the chromosome multidimensional feature information includes the chromosome centroid coordinates. The step of fusing the multidimensional feature information of each chromosome according to a set dimension to obtain the statistical information of the input image under the set dimension includes: Obtain the image center coordinates of the input image; For each chromosome, determine the distance between the chromosome centroid coordinates and the image center coordinates; Based on the average distance of each of the aforementioned distances, the statistical information of the input image under the spatial distribution dimension is determined; the average distance is positively correlated with the statistical information under the spatial distribution dimension.

6. The method according to claim 5, characterized in that, The defined dimensions include overlapping dimensions; The step of fusing the multidimensional feature information of each chromosome according to a set dimension to obtain the statistical information of the input image under the set dimension includes: For each chromosome, determine the Euclidean distance between the chromosome's centroid coordinates and the centroid coordinates of other chromosomes. The Euclidean distance with the smallest distance among all the Euclidean distances is selected as the nearest neighbor distance of the chromosome. Based on the average nearest neighbor distance of each nearest neighbor distance, the statistical information of the input image under the overlapping dimension is determined; the average nearest neighbor distance is positively correlated with the statistical information under the overlapping dimension.

7. A chromosome image processing device, characterized in that, The device includes: The input image acquisition module is used to acquire an input image containing multiple chromosomes in the metaphase phase. A chromosome image extraction module is used to extract individual chromosome images from the input image; The feature analysis module is used to perform feature analysis on each chromosome based on the chromosome images to obtain the multidimensional chromosome feature information of each chromosome. The quality assessment result determination module is used to determine the quality assessment result of the splitting phase of the input image based on the statistical information corresponding to the multidimensional feature information of each chromosome.

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