Tagging methods, apparatus, computer equipment, and storage media

Abstract

This application relates to a labeling processing method, apparatus, computer device, and storage medium. The method includes: acquiring an image segmentation result of a target image, wherein the image segmentation result includes a first initial label; calculating the similarity between the image segmentation result and a preset target tissue segmentation result; and establishing a first association relationship between the first initial label and the target tissue label based on the similarity. This first association relationship indicates that the target tissue in the target image can be identified based on the first initial label. This enables the identification of the tissue meaning represented by the first initial label contained in the image segmentation result, allowing medical application software to process each other's image segmentation results, thus improving the universality of the image segmentation results.

Description

Technical Field

[0001] This application relates to the field of marker recognition technology, and in particular to a marker processing method, apparatus, computer equipment, and storage medium. Background Technology

[0002] Image segmentation is a prerequisite for image understanding and computer vision, and one of the fundamental techniques for image processing and analysis. Image segmentation divides an image into regions with different semantic information and labels each region with corresponding semantic tags. Image segmentation has been applied in many fields, such as in the medical field, for segmenting magnetic resonance imaging (MRI) images and blood cell images.

[0003] In traditional techniques, different medical application software uses different image segmentation standards. Therefore, when processing image segmentation results obtained by other medical application software, each application software may not be able to identify the meaning of the markings in the image segmentation results, such as tissue meaning or lesion meaning. This results in medical application software being unable to process each other's image segmentation results. Summary of the Invention

[0004] Therefore, it is necessary to provide a labeling processing method, apparatus, computer equipment, and storage medium that can improve the universality of image segmentation results in response to the above-mentioned technical problems.

[0005] A tagging processing method, the method comprising:

[0006] Obtain the image segmentation result of the target image, wherein the image segmentation result includes a first initial label;

[0007] Calculate the similarity between the image segmentation result and the preset target tissue segmentation result;

[0008] Based on the similarity, a first association relationship is established between the first initial marker and the target tissue marker.

[0009] In one embodiment, obtaining the image segmentation result of the target image includes:

[0010] Obtain the initial image segmentation result of the target image;

[0011] The initial image segmentation result is registered with the preset target tissue segmentation result to determine the registration parameters;

[0012] According to the registration parameters, the resolution of the initial image segmentation result is adjusted to obtain the adjusted image segmentation result, wherein the resolution of the adjusted image segmentation result is the same as the resolution of the preset tissue segmentation result;

[0013] The adjusted image segmentation result is determined as the image segmentation result of the target image.

[0014] In one embodiment, calculating the similarity between the image segmentation result and the preset target tissue segmentation result includes:

[0015] Calculate the intersection of the first initial label contained in the image segmentation result and the target tissue label contained in the preset target tissue segmentation result;

[0016] Calculate the union of the first initial label contained in the image segmentation result and the target tissue label contained in the preset target tissue segmentation result;

[0017] Based on the intersection and the union, the similarity between the image segmentation result and the preset target tissue segmentation result is determined.

[0018] In one embodiment, the method further includes:

[0019] If the image region corresponding to the first initial marker contains the second initial marker, and the volume of the image region corresponding to the second initial marker is less than the volume threshold, then a second association relationship between the second initial marker and the lesion marker is established, wherein the image region corresponding to the first initial marker is the region occupied by the image segmentation result corresponding to the first initial marker in the target image, and the image region corresponding to the second initial marker is the region occupied by the image segmentation result corresponding to the second initial marker in the target image.

[0020] In one embodiment, the method further includes:

[0021] Based on the first initial marker and the second initial marker, a target replacement marker is generated, wherein the target replacement marker is used to characterize the attribution relationship between the target tissue and the lesion;

[0022] Replace the first initial tag and the second initial tag with the target replacement tag.

[0023] In one embodiment, after replacing the first initial marker and the second initial marker with the target replacement marker, the method further includes:

[0024] The target replacement tag is converted into the first initial tag and the second initial tag.

[0025] In one embodiment, obtaining the image segmentation result of the target image includes:

[0026] Obtain the initial image segmentation result of the target image;

[0027] Locate a third initial marker in the target image located in the image edge region, and establish a third association relationship between the third initial marker and the background marker, wherein the third association relationship is used to indicate that the background in the target image can be identified based on the third initial marker;

[0028] The image segmentation results other than those corresponding to the third initial label in the initial image segmentation results are determined as the image segmentation results of the target image.

[0029] A marking processing apparatus, the apparatus comprising:

[0030] The segmentation result acquisition module is used to acquire the image segmentation result of the target image, wherein the image segmentation result includes a first initial label;

[0031] The similarity calculation module is used to calculate the similarity between the image segmentation result and the preset target tissue segmentation result;

[0032] The association establishment module is used to establish a first association between the first initial tag and the target organization tag based on the similarity.

[0033] A computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program performing the following steps:

[0034] Obtain the image segmentation result of the target image, wherein the image segmentation result includes a first initial label;

[0035] Calculate the similarity between the image segmentation result and the preset target tissue segmentation result;

[0036] Based on the similarity, a first association relationship is established between the first initial marker and the target tissue marker.

[0037] A computer-readable storage medium having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0038] Obtain the image segmentation result of the target image, wherein the image segmentation result includes a first initial label;

[0039] Calculate the similarity between the image segmentation result and the preset target tissue segmentation result;

[0040] Based on the similarity, a first association relationship is established between the first initial marker and the target tissue marker.

[0041] The aforementioned labeling processing method, apparatus, computer equipment, and storage medium establish a first association relationship between a first initial label contained in the image segmentation result and a preset target tissue segmentation result by calculating the similarity between the image segmentation result of the target image and the target tissue label. This first association relationship indicates that the target tissue in the target image can be identified based on the first initial label. This achieves the identification of the tissue meaning represented by the first initial label contained in the image segmentation result, enabling medical application software to process each other's image segmentation results and improving the universality of image segmentation results. Attached Figure Description

[0042] Figure 1 This is a flowchart illustrating a marking process in one embodiment;

[0043] Figure 2 This is a flowchart illustrating the process of calculating the similarity between the initial image segmentation result of the target image and the preset target tissue segmentation result in one embodiment.

[0044] Figure 3 This is a structural block diagram of a marking processing device in one embodiment;

[0045] Figure 4 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0046] 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.

[0047] In one embodiment, such as Figure 1 As shown, a tagging processing method is provided. This embodiment illustrates the application of this method to a tagging processing device, which can be a terminal, such as, but not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices; it can also be a server, such as, but not limited to, a standalone server and a server cluster composed of multiple servers; or it can be a system including a terminal and a server, implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0048] Step S102: Obtain the image segmentation result of the target image.

[0049] The image segmentation result includes a first initial label. This first initial label has an unknown meaning, and its meaning needs to be identified.

[0050] Specifically, the labeling processing device acquires the image segmentation results obtained by other image segmentation devices from segmenting the target image.

[0051] Step S104: Calculate the similarity between the image segmentation result and the preset target tissue segmentation result.

[0052] Specifically, the labeling processing device calculates the similarity between the image segmentation result and the preset target tissue segmentation result. Optionally, the labeling processing device calculates the DICE coefficient between the image segmentation result and the preset target tissue segmentation result as the similarity between the two. When the target tissue segmentation result consists of segmentation results for multiple tissues, the labeling processing device calculates the DICE coefficient between the image segmentation result and the segmentation result of each tissue separately.

[0053] Optionally, the target tissue segmentation result can be a target tissue model. This model includes morphological information of multiple typical tissues. It can be labeled using multiple independent segmentation results, each containing morphological information of typical tissues. Taking the liver model as an example, this model is a segmentation result with a resolution of 512*512*300. This model contains various types of labels with known meanings; specifically, 0 is used to label the background, 1 to label the liver tissue, 2 to label the left lobe of the liver tissue, and 3 to label the right lobe of the liver tissue. Optionally, the target tissue model can include multiple sub-models, each corresponding to a label recognition method. For example, sub-model 0 is used to identify the background, and sub-model 1 is used to identify the liver tissue. The target tissue model can be updated as needed.

[0054] Optionally, the target tissue segmentation result can be the historical tissue segmentation result. A correspondence between pixels and labels in historical tissue images is established based on historical data. Methods for establishing this correspondence can include similarity analysis, deep learning, manual establishment, etc.

[0055] Step S106: Based on the similarity, establish the first association relationship between the first initial marker and the target tissue marker.

[0056] Specifically, if the similarity meets a preset similarity condition, the labeling processing device determines that the tissue corresponding to the first initial label and the tissue corresponding to the target tissue label are the same tissue, and therefore establishes a first association between the first initial label and the target tissue label. This first association is used to indicate that the target tissue in the target image can be identified based on the first initial label. Optionally, if the similarity is greater than or equal to a similarity threshold, the labeling processing device establishes the first association between the first initial label and the target tissue label. Otherwise, the labeling processing device ends the recognition of the image segmentation result.

[0057] In the aforementioned labeling method, a first association relationship is established between the first initial label contained in the image segmentation result and the preset target tissue segmentation result by calculating the similarity between the image segmentation result and the target tissue label. This first association relationship indicates that the target tissue in the target image can be identified based on the first initial label. This achieves the identification of the tissue meaning represented by the first initial label contained in the image segmentation result, enabling medical application software to process each other's image segmentation results and improving the universality of the image segmentation results.

[0058] In one embodiment, such as Figure 2 The diagram illustrates one possible implementation of step S102, "obtaining the image segmentation result of the target image." Based on the above embodiment, step S102 can be specifically implemented through the following steps:

[0059] Step S1022: Obtain the initial image segmentation result of the target image;

[0060] Step S1024: Register the initial image segmentation result with the preset target tissue segmentation result to determine the registration parameters;

[0061] Step S1026: Adjust the resolution of the initial image segmentation result according to the registration parameters to obtain the adjusted image segmentation result;

[0062] Step S1028: The adjusted image segmentation result is determined as the image segmentation result of the target image.

[0063] The resolution of the adjusted image segmentation result is the same as the resolution of the preset tissue segmentation result.

[0064] Specifically, the labeling processing device acquires the initial image segmentation result of the target image. Then, the device registers the initial image segmentation result with a preset target tissue segmentation result using bounding boxes to determine registration parameters. Registration can employ rigid registration, non-rigid registration, deep learning, or other methods. These registration parameters reflect the registration relationship between the initial image segmentation result and the preset target tissue segmentation result. Next, the labeling processing device adjusts the resolution of the initial image segmentation result according to the registration parameters, obtaining the adjusted image segmentation result. Finally, the device identifies the adjusted image segmentation result as the final image segmentation result for the target image.

[0065] In this embodiment, by registering and adjusting the resolution of the image segmentation results, the geometric structure of the image segmentation results and the target tissue segmentation results are at the same scale, which helps to improve the accuracy of subsequent similarity calculations and thus improve the accuracy of image segmentation result recognition.

[0066] In one embodiment, one possible implementation of step S104, "calculating the similarity between the image segmentation result and the preset target tissue segmentation result," is as follows: Based on the above embodiment, step S104 can be specifically implemented through the following steps:

[0067] Step S1042: Calculate the intersection of the first initial label contained in the image segmentation result and the target tissue label contained in the preset target tissue segmentation result;

[0068] Step S1044: Calculate the union of the first initial label contained in the image segmentation result and the target tissue label contained in the preset target tissue segmentation result;

[0069] Step S1046: Determine the similarity between the image segmentation result and the preset target tissue segmentation result based on the intersection and union.

[0070] Specifically, the labeling processing device first calculates the intersection between the first initial label contained in the image segmentation result and the target tissue label contained in the preset target tissue segmentation result. Then, it calculates the union between the first initial label contained in the image segmentation result and the target tissue label contained in the preset target tissue segmentation result. Finally, it calculates the ratio of the intersection to the union to obtain the Dice coefficient, i.e., Dice = intersection / union. The Dice coefficient is then used as the similarity between the image segmentation result and the preset target tissue segmentation result.

[0071] In this embodiment, the Dice coefficient is used to characterize the similarity between the image segmentation result and the target tissue segmentation result, which is beneficial to improving the accuracy of image segmentation result recognition.

[0072] In other embodiments, instead of Dice coefficients, similarity can be obtained using intersection / first initial label or intersection / target organization label. This saves the need for union calculation, reduces computation, and improves the efficiency of image segmentation result recognition.

[0073] Furthermore, both the image segmentation result and the preset target tissue segmentation result have geometric shapes. Therefore, other methods for calculating geometric similarity can also be used to calculate the similarity between the image segmentation result and the preset target tissue segmentation result. For example, the shape fit between the image segmentation result and the preset target tissue segmentation result can also be used as the similarity between the two. The shape fit can be calculated using variance analysis or deep learning methods.

[0074] In one embodiment, the method further includes the following steps:

[0075] Step S112: If the image region corresponding to the first initial marker contains the second initial marker, and the volume of the image region corresponding to the second initial marker is less than the volume threshold, then a second association relationship between the second initial marker and the lesion marker is established.

[0076] Wherein, the image region corresponding to the first initial label is the region occupied by the image segmentation result corresponding to the first initial label in the target image, and the image region corresponding to the second initial label is the region occupied by the image segmentation result corresponding to the second initial label in the target image.

[0077] Specifically, lesions in tissue are typically completely contained within the tissue. Marked regions that are contained within the tissue and are relatively small in size can be identified as lesions or sub-tissues. Based on this, for the identification of image segmentation results containing lesions, the labeling processing device first determines whether the image region corresponding to the first initial label contains the second initial label. If the image region corresponding to the first initial label contains the second initial label, the labeling processing device determines whether the volume of the image region corresponding to the second initial label is less than a volume threshold. If the volume of the image region corresponding to the second initial label is less than the volume threshold, a second association relationship is established between the second initial label and the lesion label. This second association relationship indicates that a lesion in the target image can be identified based on the second initial label. Otherwise, the labeling processing device terminates the identification of the image segmentation results.

[0078] Furthermore, the first initial markers constitute multiple marker groups and are distributed in different regions of the target image. Some marker groups contain a marker group consisting of a second initial marker, and some marker groups contain multiple marker groups consisting of second initial markers, indicating that the tissue includes different lesions and / or sub-tissues.

[0079] In this embodiment, the lesion markers in the image segmentation results of the target image can be accurately identified based on the positional relationship of the image segmentation results.

[0080] In one embodiment, the method further includes the following steps:

[0081] Step S122: Generate a target replacement marker based on the first initial marker and the second initial marker;

[0082] Step S124: Replace the first initial marker and the second initial marker with the target replacement marker.

[0083] Among them, the target replacement marker is used to characterize the attribution relationship between the target tissue and the lesion.

[0084] Specifically, the labeling processing device interprets the labels in the image segmentation results of the target image based on the first association between the first initial label and the target tissue label, and the second association between the second initial label and the lesion label. To better process the image segmentation results of the target image, a segmentation standard protocol can be used to convert the labels in the image segmentation results of the target image into label values ​​that conform to the segmentation standard protocol, thereby generating target replacement labels. Then, the labeling processing device replaces the first initial label and the second initial label with the target replacement labels.

[0085] For example, taking medical images as an example, the segmentation standard protocol stipulates that the segmentation result of a pixel includes tissue bit markers and lesion bit markers. The tissue bit marker is used to identify the tissue to which the pixel belongs, for example: 00 represents the background, 01 represents lung tissue, 02 represents the liver, etc.; the lesion bit marker is used to identify the type of lesion in the pixel, for example: 00 represents normal tissue, 01 represents a nodule, 02 represents a tumor, etc.

[0086] Medical image segmentation results can all be converted into a standard format that conforms to the above segmentation standard protocol. For example, please refer to Table 1; the standard format segmentation result contains the following four pixels:

[0087]

[0088]

[0089] Table 1

[0090] Based on the segmentation standard protocol, the segmentation results for the above four pixel locations are: background, normal lung tissue, lung nodules, and liver tumors, respectively.

[0091] In this embodiment, replacing the markers in the image segmentation result of the target image with target replacement markers can enrich the meaning of the marker values, reflect the subordinate relationship between markers, and facilitate the processing of image segmentation results by the marker processing device.

[0092] In one embodiment, the method further includes the following steps:

[0093] Step S126: Convert the target replacement marker into a first initial marker and a second initial marker.

[0094] Specifically, the tag processing device can also convert the target replacement tag into a first initial tag and a second initial tag according to the segmentation standard protocol.

[0095] In one embodiment, one possible implementation of step S102, "obtaining the image segmentation result of the target image," is as follows: Based on the above embodiment, step S102 can be specifically implemented through the following steps:

[0096] Step S102a: Obtain the initial image segmentation result of the target image;

[0097] Step S102b: Locate the third initial marker in the edge region of the target image and establish a third association relationship between the third initial marker and the background marker;

[0098] Step S102c: The image segmentation results other than the image segmentation results corresponding to the third initial label in the initial image segmentation results are determined as the image segmentation results of the target image.

[0099] The third association is used to indicate that the background in the target image can be identified based on the third initial label.

[0100] Specifically, the labeling processing device acquires the initial image segmentation result of the target image, and then counts the third initial labels located in the image edge regions of the target image. Since most edge labels are generally background labels, the labeling processing device establishes a third association between the third initial labels and background labels. In this way, the labeling processing device can identify the background region in the target image based on the third initial labels, that is, the meaning of the third initial label is background. Finally, the labeling processing device determines the image segmentation results of the initial image segmentation results other than those corresponding to the third initial labels as the image segmentation results of the target image.

[0101] In this embodiment, the background markers are first determined based on the positional distribution of the image segmentation results, and then the results of the non-background markers are determined as the image segmentation results of the target image. This helps to improve the recognition accuracy of the target tissue markers.

[0102] In one embodiment, the server obtains the initial image segmentation result of the target image; locates the third initial marker located in the image edge region of the target image, and establishes a third association relationship between the third initial marker and the background marker; determines the image segmentation results other than those corresponding to the third initial marker in the initial image segmentation result as the filtered image segmentation results; registers the filtered image segmentation results with the preset target tissue segmentation results to determine the registration parameters; adjusts the resolution of the initial image segmentation results according to the registration parameters to obtain the adjusted image segmentation results; and determines the adjusted image segmentation results as the image segmentation results of the target image.

[0103] It should be understood that, although Figure 1-2 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 1-2At least some of the steps in the process may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but may be executed at different times. The execution order of these steps or stages is not necessarily sequential, but may be executed in turn or alternately with other steps or at least some of the steps or stages in other steps.

[0104] In one embodiment, such as Figure 3 As shown, a marking processing apparatus is provided, comprising:

[0105] The segmentation result acquisition module 202 is used to acquire the image segmentation result of the target image, wherein the image segmentation result includes a first initial label;

[0106] Similarity calculation module 204 is used to calculate the similarity between the image segmentation result and the preset target tissue segmentation result;

[0107] The association establishment module 206 is used to establish a first association between the first initial tag and the target organization tag based on similarity.

[0108] In the aforementioned labeling processing device, a first association relationship is established between a first initial label contained in the image segmentation result and a preset target tissue segmentation result by calculating the similarity between the image segmentation result and the target tissue label. This first association relationship indicates that the target tissue in the target image can be identified based on the first initial label. This achieves the identification of the tissue meaning represented by the first initial label contained in the image segmentation result, enabling medical application software to process each other's image segmentation results and improving the universality of the image segmentation results.

[0109] In one embodiment, the segmentation result acquisition module 202 is specifically used to acquire the initial image segmentation result of the target image; register the initial image segmentation result with the preset target tissue segmentation result to determine the registration parameters; adjust the resolution of the initial image segmentation result according to the registration parameters to obtain the adjusted image segmentation result, wherein the resolution of the adjusted image segmentation result is the same as the resolution of the preset tissue segmentation result; and determine the adjusted image segmentation result as the image segmentation result of the target image.

[0110] In one embodiment, the similarity calculation module 204 is specifically used to calculate the intersection of the first initial label contained in the image segmentation result and the target tissue label contained in the preset target tissue segmentation result; calculate the union of the first initial label contained in the image segmentation result and the target tissue label contained in the preset target tissue segmentation result; and determine the similarity between the image segmentation result and the preset target tissue segmentation result based on the intersection and the union.

[0111] In one embodiment, the device further includes:

[0112] The association establishment module is used to establish a second association between the second initial label and the lesion label if the image region corresponding to the first initial label contains the second initial label and the volume of the image region corresponding to the second initial label is less than the volume threshold. The image region corresponding to the first initial label is the region occupied by the image segmentation result corresponding to the first initial label in the target image, and the image region corresponding to the second initial label is the region occupied by the image segmentation result corresponding to the second initial label in the target image.

[0113] In one embodiment, the device further includes:

[0114] The marker generation module is used to generate target replacement markers based on a first initial marker and a second initial marker, wherein the target replacement markers are used to characterize the attribution relationship between target tissue and lesion;

[0115] The tag replacement module is used to replace the first initial tag and the second initial tag with the target replacement tag.

[0116] In one embodiment, the device further includes:

[0117] The tag conversion module is used to convert the target replacement tag into a first initial tag and a second initial tag.

[0118] In one embodiment, the segmentation result acquisition module 202 is specifically used to acquire the initial image segmentation result of the target image; locate the third initial marker located in the image edge region of the target image; establish a third association relationship between the third initial marker and the background marker, wherein the third association relationship is used to indicate that the background in the target image can be identified according to the third initial marker; and determine the image segmentation result of the target image other than the image segmentation result corresponding to the third initial marker in the initial image segmentation result as the image segmentation result of the target image.

[0119] For specific limitations regarding the tag processing device, please refer to the limitations regarding the tag processing method above, which will not be repeated here. Each module in the aforementioned tag processing device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in hardware or independently of the processor in a computer device, or stored in software in the memory of a computer device, so that the processor can call and execute the operations corresponding to each module.

[0120] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 4As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a tagging method.

[0121] Those skilled in the art will understand that Figure 4 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.

[0122] 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 perform the following steps:

[0123] Obtain the image segmentation result of the target image, wherein the image segmentation result contains a first initial label;

[0124] Calculate the similarity between the image segmentation result and the preset target tissue segmentation result;

[0125] Based on similarity, establish the first association between the first initial tag and the target organization tag.

[0126] In the aforementioned computer device, by calculating the similarity between the image segmentation result of the target image and the preset target tissue segmentation result, a first association relationship is established between the first initial label contained in the image segmentation result and the target tissue label. This first association relationship can indicate that the target tissue in the target image can be identified based on the first initial label. This achieves the identification of the tissue meaning represented by the first initial label contained in the image segmentation result, enabling medical application software to process each other's image segmentation results, thus improving the universality of the image segmentation results.

[0127] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0128] Obtain the initial image segmentation result of the target image; register the initial image segmentation result with the preset target tissue segmentation result to determine the registration parameters; adjust the resolution of the initial image segmentation result according to the registration parameters to obtain the adjusted image segmentation result, wherein the resolution of the adjusted image segmentation result is the same as the resolution of the preset tissue segmentation result; determine the adjusted image segmentation result as the image segmentation result of the target image.

[0129] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0130] Calculate the intersection of the first initial label contained in the image segmentation result and the target tissue label contained in the preset target tissue segmentation result; calculate the union of the first initial label contained in the image segmentation result and the target tissue label contained in the preset target tissue segmentation result; determine the similarity between the image segmentation result and the preset target tissue segmentation result based on the intersection and union.

[0131] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0132] If the image region corresponding to the first initial label contains the second initial label, and the volume of the image region corresponding to the second initial label is less than the volume threshold, then a second association relationship between the second initial label and the lesion label is established, wherein the image region corresponding to the first initial label is the region occupied by the image segmentation result corresponding to the first initial label in the target image, and the image region corresponding to the second initial label is the region occupied by the image segmentation result corresponding to the second initial label in the target image.

[0133] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0134] Based on the first initial marker and the second initial marker, a target replacement marker is generated, wherein the target replacement marker is used to characterize the attribution relationship between the target tissue and the lesion; the first initial marker and the second initial marker are replaced by the target replacement marker.

[0135] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0136] Transform the target replacement tag into a first initial tag and a second initial tag.

[0137] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0138] Obtain the initial image segmentation result of the target image; locate the third initial marker in the image edge region of the target image, and establish the third association relationship between the third initial marker and the background marker, wherein the third association relationship is used to indicate that the background in the target image can be identified based on the third initial marker; determine the image segmentation result of the target image other than the image segmentation result corresponding to the third initial marker in the initial image segmentation result.

[0139] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0140] Obtain the image segmentation result of the target image, wherein the image segmentation result contains a first initial label;

[0141] Calculate the similarity between the image segmentation result and the preset target tissue segmentation result;

[0142] Based on similarity, establish the first association between the first initial tag and the target organization tag.

[0143] In the aforementioned computer-readable storage medium, by calculating the similarity between the image segmentation result of the target image and the preset target tissue segmentation result, a first association relationship is established between the first initial label contained in the image segmentation result and the target tissue label. This first association relationship can indicate that the target tissue in the target image can be identified based on the first initial label. This achieves the identification of the tissue meaning represented by the first initial label contained in the image segmentation result, enabling medical application software to process each other's image segmentation results, thus improving the universality of the image segmentation results.

[0144] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:

[0145] Obtain the initial image segmentation result of the target image; register the initial image segmentation result with the preset target tissue segmentation result to determine the registration parameters; adjust the resolution of the initial image segmentation result according to the registration parameters to obtain the adjusted image segmentation result, wherein the resolution of the adjusted image segmentation result is the same as the resolution of the preset tissue segmentation result; determine the adjusted image segmentation result as the image segmentation result of the target image.

[0146] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:

[0147] Calculate the intersection of the first initial label contained in the image segmentation result and the target tissue label contained in the preset target tissue segmentation result; calculate the union of the first initial label contained in the image segmentation result and the target tissue label contained in the preset target tissue segmentation result; determine the similarity between the image segmentation result and the preset target tissue segmentation result based on the intersection and union.

[0148] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:

[0149] If the image region corresponding to the first initial label contains the second initial label, and the volume of the image region corresponding to the second initial label is less than the volume threshold, then a second association relationship between the second initial label and the lesion label is established, wherein the image region corresponding to the first initial label is the region occupied by the image segmentation result corresponding to the first initial label in the target image, and the image region corresponding to the second initial label is the region occupied by the image segmentation result corresponding to the second initial label in the target image.

[0150] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:

[0151] Based on the first initial marker and the second initial marker, a target replacement marker is generated, wherein the target replacement marker is used to characterize the attribution relationship between the target tissue and the lesion; the first initial marker and the second initial marker are replaced by the target replacement marker.

[0152] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:

[0153] Transform the target replacement tag into a first initial tag and a second initial tag.

[0154] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:

[0155] Obtain the initial image segmentation result of the target image; locate the third initial marker in the image edge region of the target image, and establish the third association relationship between the third initial marker and the background marker, wherein the third association relationship is used to indicate that the background in the target image can be identified based on the third initial marker; determine the image segmentation result of the target image other than the image segmentation result corresponding to the third initial marker in the initial image segmentation result.

[0156] 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 methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0157] 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 specification.

[0158] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. 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 patent application should be determined by the appended claims.

Claims

1. A marking processing method, applied to a marking processing device, characterized in that, The method includes: Obtaining image segmentation results of a target image includes: obtaining image segmentation results obtained by segmenting the target image using other image segmentation devices; wherein, the image segmentation results contain a first initial label with unknown meaning; Calculate the similarity between the image segmentation result and the preset target tissue segmentation result; Based on the similarity, a first association relationship is established between the first initial marker and the target tissue marker; If the image region corresponding to the first initial marker contains the second initial marker, and the volume of the image region corresponding to the second initial marker is less than the volume threshold, then a second association relationship between the second initial marker and the lesion marker is established; Based on the first and second association relationships, and in conjunction with the segmentation standard protocol, the labels of the image segmentation results of the target image are converted into label values ​​that conform to the segmentation standard protocol.

2. The method according to claim 1, characterized in that, The process of obtaining the image segmentation result of the target image includes: Obtain the initial image segmentation result of the target image; The initial image segmentation result is registered with the preset target tissue segmentation result to determine the registration parameters; Based on the registration parameters, the resolution of the initial image segmentation result is adjusted to obtain the adjusted image segmentation result; The adjusted image segmentation result is determined as the image segmentation result of the target image.

3. The method according to claim 1 or 2, characterized in that, The calculation of the similarity between the image segmentation result and the preset target tissue segmentation result includes: Calculate the intersection of the first initial label contained in the image segmentation result and the target tissue label contained in the preset target tissue segmentation result; Calculate the union of the first initial label contained in the image segmentation result and the target tissue label contained in the preset target tissue segmentation result; Based on the intersection and the union, the similarity between the image segmentation result and the preset target tissue segmentation result is determined.

4. The method according to claim 1, characterized in that, The method further includes: Based on the first initial marker and the second initial marker, a target replacement marker is generated, wherein the target replacement marker is used to characterize the attribution relationship between the target tissue and the lesion; Replace the first initial tag and the second initial tag with the target replacement tag.

5. The method according to claim 4, characterized in that, After replacing the first initial marker and the second initial marker with the target replacement marker, the method further includes: The target replacement tag is converted into the first initial tag and the second initial tag.

6. The method according to claim 1, characterized in that, The process of obtaining the image segmentation result of the target image includes: Obtain the initial image segmentation result of the target image; Locate the third initial marker located in the edge region of the target image, and establish a third association relationship between the third initial marker and the background marker; The image segmentation results other than those corresponding to the third initial label in the initial image segmentation results are determined as the image segmentation results of the target image.

7. A marking processing apparatus, applied to a marking processing device, characterized in that, The device includes: The segmentation result acquisition module is used to acquire the image segmentation result of the target image, including: acquiring the image segmentation result obtained by segmenting the target image by other image segmentation devices; wherein, the image segmentation result contains a first initial label with unknown meaning; The similarity calculation module is used to calculate the similarity between the image segmentation result and the preset target tissue segmentation result; The association establishment module is used to establish a first association between the first initial tag and the target organization tag based on the similarity. The association establishment module is used to establish a second association between the second initial marker and the lesion marker if the image region corresponding to the first initial marker contains the second initial marker and the volume of the image region corresponding to the second initial marker is less than a volume threshold. The device includes modules for converting the image segmentation result labels of the target image into label values ​​that conform to the segmentation standard protocol, based on the first association relationship and the second association relationship and in conjunction with the segmentation standard protocol.

8. The apparatus according to claim 7, characterized in that, The segmentation result acquisition module is used for: Obtain the initial image segmentation result of the target image; The initial image segmentation result is registered with the preset target tissue segmentation result to determine the registration parameters; Based on the registration parameters, the resolution of the initial image segmentation result is adjusted to obtain the adjusted image segmentation result; The adjusted image segmentation result is determined as the image segmentation result of the target image.

9. 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.

10. 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.

Images