Blood vessel image processing method and device, computer device and storage medium

Abstract

The application relates to a blood vessel image processing method and device, computer equipment and a storage medium. The method comprises the following steps: determining a plurality of blood vessel cross-section images of a to-be-processed blood vessel image; inputting the plurality of blood vessel cross-section images into a first marking model respectively, and obtaining a plurality of marking images corresponding to the plurality of blood vessel cross-section images; and reconstructing the plurality of marking images according to blood vessel wall marks in each marking image, and obtaining a three-dimensional blood vessel wall image. By analyzing the three-dimensional blood vessel wall image provided by the application, the related information of a plaque region in the three-dimensional blood vessel wall image can be determined more accurately.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular to a method, apparatus, computer device, readable storage medium, and program product for processing vascular images. Background Technology

[0002] Arterial plaque rupture is a major cause of acute embolism in blood vessels or organs, and the nature of the plaque plays an important role in the occurrence, development, and prognosis of cardiovascular and cerebrovascular diseases. Therefore, three-dimensional reconstruction of plaques and corresponding blood vessels can facilitate diagnosis and subsequent treatment for doctors.

[0003] Traditional techniques typically involve reconstructing the blood vessel lumen to obtain images, which medical professionals then analyze to determine plaque-related information. However, the plaque-related information determined from these traditional techniques is inaccurate. Summary of the Invention

[0004] Therefore, it is necessary to provide a vascular image processing method, apparatus, computer equipment, readable storage medium, and program product to address the aforementioned technical problems.

[0005] In a first aspect, one embodiment of this application provides a method for processing blood vessel images, including:

[0006] Identify multiple cross-sectional images of the blood vessels to be processed;

[0007] Multiple blood vessel cross-sectional images are input into the first labeling model to obtain multiple labeled images corresponding to the multiple blood vessel cross-sectional images; the labeled images include blood vessel wall labels;

[0008] Based on the blood vessel wall markers in each marked image, multiple marked images are reconstructed to obtain a three-dimensional blood vessel wall image.

[0009] In one embodiment, multiple cross-sectional images of blood vessels in the blood vessel image to be processed are determined, including:

[0010] Acquire the image of the blood vessel to be processed, and determine the centerline of the blood vessel based on the image.

[0011] Multiple cross-sectional images of blood vessels are reconstructed based on the images of the blood vessels to be processed and the centerline of the blood vessels.

[0012] In one embodiment, determining the vessel centerline based on the vessel image to be processed includes:

[0013] The blood vessel image to be processed is segmented to obtain the blood vessel image;

[0014] Skeletonization processing is performed on the blood vessel image to obtain the blood vessel centerline.

[0015] In one embodiment, determining the vessel centerline based on the vessel image to be processed includes:

[0016] The input parameters of the second labeling model are determined based on the blood vessel image to be processed. The input parameters are then input into the second labeling model to obtain the blood vessel type labeling image. The blood vessel type labeling image includes different types of blood vessel labels.

[0017] Skeletonization extraction was performed on the blood vessel type calibration image to obtain the blood vessel centerline.

[0018] In one embodiment, determining the input parameters of the second labeling model based on the blood vessel image to be processed includes:

[0019] Use the vascular image to be processed and / or the vascular image as input parameters.

[0020] In one embodiment, the three-dimensional blood vessel wall image includes different types of blood vessel markers.

[0021] In one embodiment, the three-dimensional blood vessel wall image includes information about the blood vessel wall region.

[0022] In one embodiment, acquiring the blood vessel image to be processed includes:

[0023] An initial blood vessel image is obtained, and grayscale normalization is performed on the initial blood vessel image to obtain the blood vessel image to be processed.

[0024] Secondly, one embodiment of this application provides a vascular image processing apparatus, the apparatus comprising:

[0025] The first determining module is used to determine multiple cross-sectional images of blood vessels in the blood vessel image to be processed;

[0026] The second determining module is used to input multiple blood vessel cross-sectional images into the first labeling model to obtain multiple labeling images corresponding to the multiple blood vessel cross-sectional images; the labeling images include blood vessel wall labels.

[0027] The reconstruction module is used to reconstruct multiple labeled images based on the blood vessel wall markers in each labeled image to obtain a three-dimensional blood vessel wall image.

[0028] Thirdly, one embodiment of this application provides a computer device including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of the method provided in the above embodiments.

[0029] Fourthly, one embodiment of this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method provided in the above embodiments.

[0030] Fifthly, one embodiment of this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the method provided in the above embodiments.

[0031] This application provides a method, apparatus, computer device, readable storage medium, and program product for processing vascular images. The method involves determining multiple cross-sectional images of a vascular image to be processed; inputting these multiple cross-sectional images into a first labeling model to obtain multiple labeled images corresponding to the cross-sectional images; and reconstructing the multiple labeled images based on vascular wall labels in each labeled image to obtain a three-dimensional vascular wall image. The vascular image processing method provided in this embodiment can obtain a three-dimensional vascular wall image including the vascular wall region. Since the plaque region is located inside the vascular wall region, compared with the traditional technique of determining plaque-related information by analyzing vascular lumen images, the three-dimensional vascular wall image obtained by this embodiment can more intuitively and comprehensively display the plaque region. By analyzing the plaque region, relevant information about the plaque region can be determined more accurately. Attached Figure Description

[0032] To more clearly illustrate the technical solutions in the embodiments of this application or the conventional technology, the drawings used in the description of the embodiments or the conventional technology will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0033] Figure 1 An application environment diagram of a blood vessel image processing method provided in one embodiment;

[0034] Figure 2 A flowchart illustrating the steps of a vascular image processing method provided in one embodiment;

[0035] Figure 3 A schematic diagram of a three-dimensional blood vessel wall image reconstruction process provided in one embodiment;

[0036] Figure 4 A flowchart illustrating the steps of a vascular image processing method provided in another embodiment;

[0037] Figure 5 A flowchart illustrating the steps of a vascular image processing method provided in another embodiment;

[0038] Figure 6 A flowchart illustrating the steps of a vascular image processing method provided in another embodiment;

[0039] Figure 7A flowchart illustrating the steps of a vascular image processing method provided in another embodiment;

[0040] Figure 8 A schematic diagram of a blood vessel image and a blood vessel centerline provided for one embodiment;

[0041] Figure 9 A flowchart illustrating the steps of a vascular image processing method provided in another embodiment;

[0042] Figure 10 A schematic diagram of a vascular image processing apparatus provided in one embodiment;

[0043] Figure 11 This is a schematic diagram of the structure of a computer device provided in one embodiment of this application. Detailed Implementation

[0044] To make the above-mentioned objectives, features, and advantages of this application more apparent and understandable, the specific embodiments of this application are described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of this application. However, this application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of this application. Therefore, this application is not limited to the specific embodiments disclosed below.

[0045] The serial numbers assigned to components in this article, such as "first" and "second", are used only to distinguish the objects being described and have no sequential or technical meaning.

[0046] The vascular image processing method provided in this application embodiment can be applied to, for example... Figure 1 The application environment shown includes a terminal 100 and a medical scanning device 200. The terminal 100 can communicate with the medical scanning device 200 via a network. The terminal 100 can be, but is not limited to, various personal computers, laptops, and tablets. The medical scanning device 200 can be, but is not limited to, MR (Magnetic Resonance), CT (Computed Tomography), and PET (Positron Emission Computed Tomography)-CT devices.

[0047] In one embodiment, such as Figure 2 As shown, this application provides a method for processing blood vessel images, which can be applied to... Figure 1 Taking the terminal in the example, the explanation includes the following steps:

[0048] Step 200: Determine multiple cross-sectional images of the blood vessels to be processed.

[0049] The vascular image to be processed refers to a three-dimensional image obtained using medical scanning equipment that requires processing. Different medical scanning equipment yields different vascular images to be processed; similarly, the same medical scanning equipment using different scanning methods can also produce different vascular images to be processed. This embodiment does not limit the type of vascular image to be processed, as long as its function can be achieved. Specifically, when the medical scanning equipment is a magnetic resonance imaging (MRI) device, different scanning methods can yield different vascular images to be processed, such as images obtained using TOF-MRA (Time-of-Flight Magnetic Resonance Angiography), images obtained using CE-MRA (Contrast-Enhanced Magnetic Resonance Angiography), and images obtained by subtracting CE-MRA images.

[0050] A vascular cross-sectional image is a three-dimensional cross-sectional view of the vascular image to be processed. Multiple vascular cross-sectional images of the vascular image to be processed can be directly stored in the terminal's memory, and can be retrieved directly from the memory when needed. This embodiment does not limit the number of vascular cross-sectional images or the specific method of acquiring them, as long as the function can be achieved.

[0051] Step 210: Input multiple blood vessel cross-sectional images into the first labeling model to obtain multiple labeling images corresponding to the multiple blood vessel cross-sectional images; the labeling images include blood vessel wall labels.

[0052] After obtaining multiple vascular cross-sectional images, the terminal inputs each vascular cross-sectional image into a pre-trained first labeling model. The first labeling model processes the images, specifically labeling the vessel wall in each vascular cross-sectional image, to obtain a labeled image corresponding to each vascular cross-sectional image. Optionally, the first labeling model can also be used to label plaques in each vascular cross-sectional image, thus the resulting labeled image may also include plaque labels.

[0053] The first labeling model can be obtained by the terminal training a convolutional neural network based on vascular cross-section image samples and storing it in the terminal's memory; alternatively, it can be obtained by the terminal training a convolutional neural network based on vascular cross-section image samples, as well as vascular wall markers and plaque markers in each vascular cross-section image within the vascular cross-section image samples, and storing it in the terminal's memory. The vascular cross-section image samples include multiple vascular cross-section images, and the type of each vascular cross-section image can be the same as the multiple vascular cross-section images of the vascular image to be processed.

[0054] Step 220: Reconstruct multiple labeled images based on the blood vessel wall markers in each labeled image to obtain a three-dimensional blood vessel wall image.

[0055] After obtaining multiple labeled images corresponding to multiple cross-sectional images of blood vessels, the terminal reconstructs these images based on the blood vessel wall markers in each labeled image to obtain a three-dimensional blood vessel wall image. The three-dimensional blood vessel wall image includes the blood vessel wall region; that is, the blood vessel wall region can be displayed in the obtained three-dimensional blood vessel wall image. When the labeled images also include plaque markers, the reconstructed three-dimensional blood vessel wall image also includes plaque regions; that is, plaque regions can also be displayed in the obtained three-dimensional blood vessel wall image. Furthermore, the three-dimensional blood vessel wall image can also display both blood vessel wall regions and plaque regions in combination. This embodiment does not limit the specific method for reconstructing multiple labeled images, as long as a three-dimensional blood vessel wall image can be obtained.

[0056] In an optional embodiment, when reconstructing multiple labeled images, the terminal can first perform intra-layer reconstruction on each labeled image based on the blood vessel wall markers and plaque markers in each labeled image, thereby obtaining each reconstructed labeled image, such as... Figure 3 As shown in Figure a. After obtaining each reconstructed labeled image, inter-layer interpolation reconstruction is performed between each pair of reconstructed labeled images, as shown in Figure a. Figure 3 As shown in Figure b, a three-dimensional image of the blood vessel wall can ultimately be obtained, as follows. Figure 3 As shown in Figure c.

[0057] The vascular image processing method provided in this embodiment determines multiple vascular cross-sectional images of the vascular image to be processed; inputs the multiple vascular cross-sectional images into a first labeling model to obtain multiple labeling images corresponding to the multiple vascular cross-sectional images; and reconstructs the multiple labeling images based on the vascular wall label and plaque label in each labeling image to obtain a three-dimensional vascular wall image. The vascular image processing method provided in this embodiment can obtain a three-dimensional vascular wall image including a vascular wall region and a plaque region. The plaque region is located inside the vascular wall region. Compared with traditional techniques that determine plaque-related information by analyzing vascular lumen images, the three-dimensional vascular wall image obtained by this embodiment can more intuitively and comprehensively display the plaque region. By analyzing the plaque region, relevant information of the plaque region can be determined more accurately.

[0058] In one embodiment, such as Figure 4 As shown, a possible implementation method for determining multiple cross-sectional images of blood vessels in a blood vessel image to be processed includes the following steps:

[0059] Step 400: Obtain the image of the blood vessel to be processed, and determine the center line of the blood vessel based on the image of the blood vessel to be processed.

[0060] The description of the vascular image to be processed can be found in the specific description in the above embodiments, and will not be repeated here. The vascular image to be processed can be directly stored in the terminal's memory, and the terminal can retrieve it directly from the memory when needed.

[0061] After acquiring the image of the blood vessel to be processed, the terminal processes the blood vessels in the image to obtain the centerline of the blood vessels. This embodiment does not limit the specific method for extracting the centerline of the blood vessels in the image to be processed, as long as the function can be achieved.

[0062] Step 410: Reconstruct multiple cross-sectional images of blood vessels based on the image of the blood vessel to be processed and the centerline of the blood vessel.

[0063] After obtaining the blood vessel centerline, the terminal can obtain multiple cross-sectional images of the blood vessel image to be processed based on the blood vessel centerline and the blood vessel image to be processed.

[0064] Specifically, such as Figure 5 As shown, one possible method for reconstructing multiple cross-sectional images of blood vessels based on the image of the blood vessel to be processed and the blood vessel centerline includes the following steps:

[0065] Step 500: Determine multiple cross-sectional points based on the blood vessel centerline.

[0066] After obtaining the blood vessel centerline, the terminal randomly selects multiple points on the blood vessel centerline, which are recorded as cross-section points. The distance between adjacent cross-section points can be the same or different.

[0067] Step 510: For each cross-sectional point, determine the adjacent pixels of that cross-sectional point based on the blood vessel image to be processed and the cross-sectional point.

[0068] For each cross-sectional point on the blood vessel centerline, the terminal determines the adjacent pixels of the cross-sectional point in the blood vessel image in a direction perpendicular to the blood vessel centerline. This embodiment does not limit the number of adjacent pixels determined for each cross-sectional point, as long as the function can be achieved.

[0069] In an optional embodiment, the terminal may first select multiple pixels adjacent to the cross-section point in the direction perpendicular to the center line of the blood vessel in the image to be processed, and denoted as the first adjacent pixel. Then, it may select multiple pixels adjacent to the first adjacent pixel in the direction perpendicular to the center line of the blood vessel in the image to be processed, and denoted as the second adjacent pixel. The first adjacent pixel and the second adjacent pixel are collectively referred to as the adjacent pixels of the cross-section point.

[0070] Step 520: Perform pixel interpolation on the adjacent pixels of the cross-section point based on the vertical direction of the blood vessel centerline to obtain the blood vessel cross-section image corresponding to the cross-section point.

[0071] The terminal obtains the adjacent pixels of each cross-section point, and performs pixel interpolation processing on the adjacent pixels of each cross-section point in the direction perpendicular to the blood vessel centerline to obtain the blood vessel cross-section image corresponding to that cross-section point. Thus, multiple blood vessel cross-section images corresponding to multiple cross-section points in the blood vessel image to be processed can be obtained.

[0072] This embodiment uses the vessel centerline corresponding to the vessel image to be processed and the method of reconstructing multiple vessel cross-sectional images from the vessel image to be processed. It is simple, easy to understand, and easy to implement.

[0073] In one embodiment, such as Figure 6 As shown, this relates to a possible method for determining the centerline of a blood vessel based on a blood vessel image to be processed, the steps of which include:

[0074] Step 600: Perform blood vessel segmentation processing on the blood vessel image to be processed to obtain the blood vessel image.

[0075] After obtaining the blood vessel image to be processed, the terminal performs segmentation processing to separate the blood vessel regions in the image, thereby obtaining the blood vessel image. This embodiment does not limit the specific method of blood vessel segmentation processing, as long as it can achieve its function.

[0076] In an optional embodiment, the terminal can input the blood vessel image to be processed into a pre-trained segmentation model, through which the blood vessel image can be obtained. The segmentation model can be obtained by the terminal training a neural network based on blood vessel image samples of the same type as the blood vessel image to be processed.

[0077] Step 610: Perform skeletonization processing on the blood vessel image to obtain the blood vessel centerline.

[0078] After obtaining the vascular image, the terminal performs skeletonization processing on the vascular image, that is, it uses a shape erosion operation to remove the boundary of the vascular image, which can obtain the center line of the vascular image.

[0079] In an optional embodiment, the terminal can process the vascular image based on any one of the distance transformation method, path planning method, and tracking method to obtain the vascular centerline.

[0080] This embodiment obtains a vascular image by segmenting the vascular image to be processed, and then performs skeletonization processing on the vascular image to obtain the vascular centerline. This method of determining the vascular centerline is simple, convenient, and easy to implement.

[0081] In one embodiment, such as Figure 7 As shown, this relates to a possible method for determining the centrality of blood vessels based on a blood vessel image to be processed, the steps of which include:

[0082] Step 700: Determine the input parameters of the second labeling model based on the blood vessel image to be processed, and input the input parameters into the second labeling model to obtain the blood vessel type labeling image; the blood vessel type labeling image includes different types of blood vessel labels.

[0083] After obtaining the blood vessel image to be processed, the terminal determines the input parameters for the second labeling model based on the image. This embodiment does not limit the specific method for determining the input parameters of the second labeling model.

[0084] After receiving the input parameters, the terminal inputs them into a pre-trained second labeling model. Based on the second labeling model, a blood vessel type labeling image including different types of blood vessel labels can be obtained. In other words, the second labeling model can label different types of blood vessels in the blood vessel image to be processed.

[0085] The second labeling model can be trained by the terminal using input samples that are identical to the input parameters. The description of the input samples can be found in the specific description of the input parameters, which will not be repeated here.

[0086] Step 710: Extract the skeleton from the blood vessel type calibration image to obtain the blood vessel centerline.

[0087] After obtaining the blood vessel type calibration image, the terminal performs skeletonization processing on it, that is, it uses a shape erosion operation to remove the boundaries of the blood vessel image, thus obtaining the blood vessel centerline. Since the blood vessel type calibration image contains different types of blood vessel markers, the obtained blood vessel centerline also includes different types of blood vessel markers.

[0088] The blood vessel image obtained by segmenting the blood vessel image to be processed is as follows: Figure 8 As shown in Figure a, the centerline of the blood vessel obtained by skeletonization of the blood vessel image is as follows: Figure 8 As shown in Figure b.

[0089] In an optional embodiment, the terminal can process the blood vessel type calibration image based on any one of the distance transformation method, path planning method, and tracking method to obtain the blood vessel centerline.

[0090] In this embodiment, by inputting the input parameters determined according to the blood vessel image to be processed into the second labeling model, a blood vessel type labeling image can be obtained. By processing the blood vessel type labeling image, the resulting blood vessel centerline can also include different types of blood vessel labels, which facilitates subsequent applications.

[0091] In one embodiment, one possible implementation involves determining input parameters of a second labeling model based on a blood vessel image to be processed, the method comprising:

[0092] Use the vascular image to be processed and / or the vascular image as input parameters.

[0093] When the terminal determines the input parameters of the second labeling model based on the blood vessel image to be processed, it can directly use the blood vessel image to be processed as the input parameter of the second labeling model; it can also use the blood vessel image obtained after blood vessel segmentation processing of the blood vessel image to be processed as the input parameter of the second labeling model; or it can use both the blood vessel image to be processed and the blood vessel image as input parameters, that is, the second labeling model includes two input channels, which are used to input the blood vessel image to be processed and the blood vessel image respectively.

[0094] In this embodiment, multiple types of input parameters are proposed, allowing users to choose according to their specific application scenarios, thus improving the applicability of the vascular image processing method. Furthermore, when the input parameters are the vascular image to be processed and the vascular image itself, the input channel of the vascular image can focus on vascular information while eliminating interference from related backgrounds. This allows for a more accurate and faster acquisition of the vascular type calibration image based on the input parameters, thereby improving the efficiency and accuracy of the vascular image processing method.

[0095] In one embodiment, the three-dimensional blood vessel wall image includes different types of blood vessel markers.

[0096] When the terminal extracts the vessel centerline from the vessel type calibration image using skeletonization, this centerline includes markers for different types of vessels. Therefore, the 3D vessel wall image obtained from the reconstructed vessel cross-section image based on this centerline also includes these different types of vessel markers. This allows users to easily obtain the 3D vessel wall image corresponding to a target vessel type based on the different types of vessel markers in the 3D vessel wall image. The target vessel type can be a single vessel or any combination of multiple vessel types. In other words, when the terminal includes a display, it can display 3D vessel wall images corresponding to all vessel types as a whole, or display 3D vessel wall images corresponding to a single vessel type, or 3D vessel wall images corresponding to any combination of multiple vessel types, depending on actual needs.

[0097] In this embodiment, the three-dimensional blood vessel wall image includes different types of blood vessel markers, which makes it easier for users to display the target blood vessel segment according to actual needs, thereby making the blood vessel image processing method more applicable.

[0098] In one embodiment, the three-dimensional blood vessel wall image includes information about the blood vessel wall region. This information includes the size of the blood vessel wall region, the stenosis rate of the blood vessel wall, and the grayscale value of the blood vessel wall region. When the three-dimensional blood vessel wall image includes plaque regions, it also includes information about the plaque regions. This information includes the size of the plaque regions and the vulnerability probability of the plaque regions. This embodiment does not limit the specific content of the information about the blood vessel wall region and the plaque regions.

[0099] In this embodiment, by displaying information about the blood vessel wall region and plaque region in a three-dimensional blood vessel wall image, users can intuitively obtain information about the plaque region and blood vessel wall region through the three-dimensional blood vessel wall image.

[0100] In one embodiment, one possible implementation method involving acquiring a blood vessel image to be processed includes:

[0101] An initial blood vessel image is obtained, and grayscale normalization is performed on the initial blood vessel image to obtain the blood vessel image to be processed.

[0102] The initial vascular image can refer to a three-dimensional image obtained using medical scanning equipment. After obtaining the initial vascular image, the terminal performs grayscale normalization processing on it, that is, processes the grayscale values ​​in the initial vascular image to bring them within a preset, fixed grayscale range, thus obtaining the vascular image to be processed. Specifically, the terminal obtains a fixed grayscale region preset by the user, and adaptively selects a portion of the grayscale region (mainly referring to the area where the vascular region is located in the initial vascular image) from the grayscale distribution range of the initial vascular image for processing, bringing it within the fixed grayscale range. In a specific embodiment, the terminal selects more than 95% of the grayscale range from the grayscale distribution range of the initial vascular image for processing, and this portion of the grayscale range is mapped to the grayscale range of 0-200.

[0103] In this embodiment, by performing grayscale normalization on the initial vascular image, the differences in grayscale values ​​between different vascular images to be processed can be avoided, which would lead to inaccurate output results obtained using the first labeling model and the second labeling model. This can improve the accuracy of determining the vascular centerline and thus improve the accuracy of the vascular image processing method.

[0104] like Figure 9 As shown, in one embodiment, a blood vessel image processing method is proposed, the steps of which include:

[0105] Step 900: Obtain the image of the blood vessel to be processed;

[0106] Step 910: Perform blood vessel segmentation on the blood vessel image to be processed to obtain a blood vessel image; perform skeletonization on the blood vessel image to obtain the blood vessel centerline;

[0107] Step 920: Determine the input parameters of the second labeling model based on the blood vessel image to be processed, input the input parameters into the second labeling model to obtain the blood vessel type labeling image; extract the skeleton from the blood vessel type labeling image to obtain the blood vessel centerline;

[0108] Step 930: Reconstruct multiple cross-sectional images of blood vessels based on the image of the blood vessel to be processed and the centerline of the blood vessel;

[0109] Step 940: Input multiple blood vessel cross-sectional images into the first labeling model to obtain multiple labeled images corresponding to the multiple blood vessel cross-sectional images;

[0110] Step 950: Reconstruct multiple labeled images based on the blood vessel wall markers in each labeled image to obtain a three-dimensional blood vessel wall image.

[0111] It should be understood that although the steps in the flowchart are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order constraint on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the diagram may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0112] Based on the same inventive concept, this application also provides a vascular image processing apparatus for implementing the aforementioned vascular image processing method. The solution provided by this apparatus is similar to the implementation described in the above method; therefore, the specific limitations in one or more embodiments of the vascular image processing apparatus provided below can be found in the limitations of the vascular image processing method described above, and will not be repeated here.

[0113] Please see Figure 10 One embodiment of this application provides a vascular image processing apparatus 10, which includes a first determining module 11, a second determining module 12, and a reconstruction module 13.

[0114] The first determining module 11 is used to determine multiple vascular interface images of the vascular image to be processed;

[0115] The second determining module 12 is used to input multiple vascular interface images into the first labeling model respectively to obtain multiple labeling images corresponding to the multiple vascular interface images; the labeling images include vascular wall labels.

[0116] The reconstruction module 13 is used to reconstruct multiple labeled images based on the blood vessel wall markings in each labeled image to obtain a three-dimensional blood vessel wall image.

[0117] In one embodiment, the first determining module 12 includes an acquisition unit and a reconstruction unit. The acquisition unit acquires a blood vessel image to be processed and determines the blood vessel centerline based on the blood vessel image; the reconstruction unit reconstructs multiple cross-sectional images of the blood vessel based on the blood vessel image to be processed and the blood vessel centerline.

[0118] In one embodiment, the acquisition unit is specifically used to perform blood vessel segmentation processing on the blood vessel image to be processed to obtain a blood vessel image; and to perform skeletonization processing on the blood vessel image to obtain the blood vessel centerline.

[0119] In one embodiment, the acquisition unit is further configured to determine the input parameters of the second labeling model based on the blood vessel image to be processed, input the input parameters into the second labeling model to obtain a blood vessel type labeling image; the blood vessel type labeling image includes different types of blood vessel labels; and perform skeletonization extraction on the blood vessel type labeling image to obtain the blood vessel centerline.

[0120] In one embodiment, the acquisition unit is further configured to use the vascular image to be processed and / or the vascular image as input parameters.

[0121] In one embodiment, the three-dimensional blood vessel wall image includes different types of blood vessel markers.

[0122] In one embodiment, the three-dimensional blood vessel wall image includes information about the blood vessel wall region.

[0123] In one embodiment, the acquisition unit is further configured to acquire an initial vascular image, perform grayscale normalization processing on the initial vascular image, and obtain a vascular image to be processed.

[0124] Each module in the aforementioned vascular 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 corresponding operations of each module.

[0125] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 11As shown, the computer device includes a processor, memory, communication interface, display screen, and input devices 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 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 communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a vascular image processing method. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0126] Those skilled in the art will understand that Figure 11 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.

[0127] 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:

[0128] Identify multiple cross-sectional images of the blood vessels to be processed;

[0129] Multiple blood vessel cross-sectional images are input into the first labeling model to obtain multiple labeled images corresponding to the multiple blood vessel cross-sectional images; the labeled images include blood vessel wall labels;

[0130] Based on the blood vessel wall markers in each marked image, multiple marked images are reconstructed to obtain a three-dimensional blood vessel wall image.

[0131] In one embodiment, when the processor executes the computer program, it further performs the following steps: acquiring a blood vessel image to be processed; determining the blood vessel centerline based on the blood vessel image to be processed; and reconstructing multiple blood vessel cross-sectional images based on the blood vessel image to be processed and the blood vessel centerline.

[0132] In one embodiment, when the processor executes the computer program, it further performs the following steps: performing blood vessel segmentation processing on the blood vessel image to be processed to obtain a blood vessel image; and performing skeletonization processing on the blood vessel image to obtain the blood vessel centerline.

[0133] In one embodiment, when the processor executes the computer program, it further performs the following steps: determining the input parameters of the second labeling model based on the blood vessel image to be processed, inputting the input parameters into the second labeling model to obtain a blood vessel type calibration image; the blood vessel type calibration image includes different types of blood vessel labels; and extracting the skeleton from the blood vessel type calibration image to obtain the blood vessel centerline.

[0134] In one embodiment, the processor, when executing the computer program, further performs the following steps: taking the vascular image to be processed and / or the vascular image as input parameters.

[0135] In one embodiment, the three-dimensional blood vessel wall image includes different types of blood vessel markers.

[0136] In one embodiment, the three-dimensional blood vessel wall image includes information about the blood vessel wall region.

[0137] In one embodiment, when the processor executes the computer program, it further performs the following steps: acquiring an initial vascular image, performing grayscale normalization processing on the initial vascular image, and obtaining a vascular image to be processed.

[0138] 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:

[0139] Identify multiple cross-sectional images of the blood vessels to be processed;

[0140] Multiple blood vessel cross-sectional images are input into the first labeling model to obtain multiple labeled images corresponding to the multiple blood vessel cross-sectional images; the labeled images include blood vessel wall labels;

[0141] Based on the blood vessel wall markers in each marked image, multiple marked images are reconstructed to obtain a three-dimensional blood vessel wall image.

[0142] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: acquiring a blood vessel image to be processed; determining the blood vessel centerline based on the blood vessel image to be processed; and reconstructing multiple blood vessel cross-sectional images based on the blood vessel image to be processed and the blood vessel centerline.

[0143] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: performing blood vessel segmentation processing on the blood vessel image to be processed to obtain a blood vessel image; and performing skeletonization processing on the blood vessel image to obtain the blood vessel centerline.

[0144] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: determining the input parameters of the second labeling model based on the blood vessel image to be processed, inputting the input parameters into the second labeling model to obtain a blood vessel type labeling image; the blood vessel type labeling image includes different types of blood vessel labels; and extracting the skeleton from the blood vessel type labeling image to obtain the blood vessel centerline.

[0145] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: taking the vascular image to be processed and / or the vascular image as input parameters.

[0146] In one embodiment, the three-dimensional blood vessel wall image includes different types of blood vessel markers.

[0147] In one embodiment, the three-dimensional blood vessel wall image includes information about the blood vessel wall region.

[0148] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: acquiring an initial vascular image, performing grayscale normalization on the initial vascular image, and obtaining a vascular image to be processed.

[0149] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0150] Identify multiple cross-sectional images of the blood vessels to be processed;

[0151] Multiple blood vessel cross-sectional images are input into the first labeling model to obtain multiple labeled images corresponding to the multiple blood vessel cross-sectional images; the labeled images include blood vessel wall labels;

[0152] Based on the blood vessel wall markers in each marked image, multiple marked images are reconstructed to obtain a three-dimensional blood vessel wall image.

[0153] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: acquiring a blood vessel image to be processed; determining the blood vessel centerline based on the blood vessel image to be processed; and reconstructing multiple blood vessel cross-sectional images based on the blood vessel image to be processed and the blood vessel centerline.

[0154] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: performing blood vessel segmentation processing on the blood vessel image to be processed to obtain a blood vessel image; and performing skeletonization processing on the blood vessel image to obtain the blood vessel centerline.

[0155] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: determining the input parameters of the second labeling model based on the blood vessel image to be processed, inputting the input parameters into the second labeling model to obtain a blood vessel type labeling image; the blood vessel type labeling image includes different types of blood vessel labels; and extracting the skeleton from the blood vessel type labeling image to obtain the blood vessel centerline.

[0156] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: taking the vascular image to be processed and / or the vascular image as input parameters.

[0157] In one embodiment, the three-dimensional blood vessel wall image includes different types of blood vessel markers.

[0158] In one embodiment, the three-dimensional blood vessel wall image includes information about the blood vessel wall region.

[0159] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: acquiring an initial vascular image, performing grayscale normalization on the initial vascular image, and obtaining a vascular image to be processed.

[0160] 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 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, etc., and are not limited to these.

[0161] 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.

[0162] 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 patent application should be determined by the appended claims.

Claims

1. A method for processing blood vessel images, characterized in that, include: Identify multiple cross-sectional images of the blood vessels to be processed; The multiple cross-sectional images of blood vessels are respectively input into a first labeling model to obtain multiple labeled images corresponding to the multiple cross-sectional images of blood vessels; the labeled images include blood vessel wall labels and plaque labels; Based on the vessel wall markers and plaque markers in each of the marked images, the plurality of marked images are reconstructed to obtain a three-dimensional vessel wall image; The step of reconstructing the plurality of labeled images based on the vessel wall markers and plaque markers in each labeled image to obtain a three-dimensional vessel wall image includes: Intra-layer reconstruction is performed on each of the labeled images based on the vessel wall markers and plaque markers in each labeled image to obtain each reconstructed labeled image; Interlayer interpolation reconstruction is performed between each pair of the reconstructed labeled images to obtain the three-dimensional blood vessel wall image.

2. The vascular image processing method according to claim 1, characterized in that, The determination of multiple vascular cross-sectional images of the vascular image to be processed includes: Acquire the image of the blood vessel to be processed, and determine the centerline of the blood vessel based on the image of the blood vessel to be processed; The multiple cross-sectional images of blood vessels are reconstructed based on the blood vessel image to be processed and the blood vessel centerline.

3. The vascular image processing method according to claim 2, characterized in that, Determining the vessel centerline based on the vessel image to be processed includes: The blood vessel image to be processed is segmented to obtain a blood vessel image; The blood vessel image is processed into a skeleton to obtain the center line of the blood vessel.

4. The vascular image processing method according to claim 2, characterized in that, Determining the vessel centerline based on the vessel image to be processed includes: The input parameters of the second labeling model are determined based on the blood vessel image to be processed, and the input parameters are input into the second labeling model to obtain a blood vessel type labeling image; the blood vessel type labeling image includes different types of blood vessel labels; The skeletonization extraction of the blood vessel type calibration image is performed to obtain the center line of the blood vessel.

5. The vascular image processing method according to claim 4, characterized in that, The step of determining the input parameters of the second labeling model based on the blood vessel image to be processed includes: The blood vessel image to be processed and / or the blood vessel image are used as the input parameters.

6. The vascular image processing method according to claim 4, characterized in that, The three-dimensional blood vessel wall image includes the different types of blood vessel markers.

7. The vascular image processing method according to claim 1, characterized in that, The three-dimensional blood vessel wall image includes information about the blood vessel wall region.

8. A vascular image processing device, characterized in that, The apparatus, used in the vascular image processing method as described in any one of claims 1-7, comprises: The first determining module is used to determine multiple cross-sectional images of blood vessels in the blood vessel image to be processed; The second determining module is used to input the plurality of blood vessel cross-sectional images into the first labeling model respectively to obtain a plurality of labeling images corresponding to the plurality of blood vessel cross-sectional images; the labeling images include blood vessel wall labels; The reconstruction module is used to reconstruct the plurality of labeled images based on the blood vessel wall markers in each labeled image to obtain a three-dimensional blood vessel wall 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 7.

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 7.

Images