Vessel key point position determination method and device, computer device, and storage medium

Abstract

The application relates to the computer technical field, in particular to a blood vessel key point position determination method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring a segmented mask image of a blood vessel; determining a region of interest layer of a to-be-determined blood vessel key point corresponding to the blood vessel from the segmented mask image; and determining a key point coordinate of the blood vessel key point according to a blood vessel connected domain in the region of interest layer. The method can improve the intelligent level of data processing and the accuracy of the key point position.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a method, apparatus, computer device, and storage medium for determining the location of key points in blood vessels. Background Technology

[0002] In medical imaging research, doctors can use vascular images to analyze and process vascular function. For example, using aortic images, doctors can observe the shape characteristics of the aorta and assess its functional status, thereby enabling early quantitative diagnosis and risk assessment of various diseases. Since the vascular centerline plays a crucial role in quantitative analysis such as vascular topological representation, vascular diameter, and volume calculation, its extraction is often a vital step in acquiring vascular morphology and conducting quantitative analysis. The extraction of the vascular centerline typically requires determining several seed points, or key points, as initial conditions. The accuracy of these initial conditions directly affects the accuracy of the extracted vascular centerline.

[0003] Traditionally, doctors typically need to manually input several seed points, i.e., input key vascular points manually. However, manual input of key vascular points makes the data processing less intelligent. Furthermore, manual input relies on the doctor's experience and judgment, and different doctors have different criteria for judgment. For less experienced doctors, this results in low accuracy in determining the location of key points. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, apparatus, computer equipment, and storage medium for determining the location of key points in blood vessels, which can improve the level of intelligence in data processing and enhance the accuracy of key point locations, in order to address the aforementioned technical problems.

[0005] A method for determining the location of key points in a blood vessel, the method comprising:

[0006] Obtain the segmentation mask image of the blood vessels;

[0007] Determine the region of interest (ROI) layer from the segmented mask image to identify the key points of the corresponding blood vessel;

[0008] Based on the connected regions of blood vessels in the region of interest layer, determine the key point coordinates of the blood vessel key points.

[0009] In one embodiment, determining the region of interest layer for key points of the corresponding blood vessel from the segmented mask image includes:

[0010] Obtain the scanning direction of the segmentation mask image;

[0011] Based on the scanning direction, the region of interest (ROI) layer for the key points of the corresponding blood vessel to be determined is determined from the segmented mask image.

[0012] In one embodiment, obtaining the scanning direction of the segmentation mask image includes:

[0013] Determine the maximum connected blood vessel region in each mask layer of the segmented mask image;

[0014] Based on each largest connected vessel region, a reference surface mask layer is determined from the segmented mask image;

[0015] The scanning direction of the segmentation mask image is determined based on the reference surface mask layer.

[0016] In one embodiment, determining the key point coordinates of key points of blood vessels based on the connected components of blood vessels in the region of interest layer includes:

[0017] Obtain the morphological parameters of the blood vessel connected regions in each mask layer of the region of interest layer;

[0018] Based on various morphological parameters, the target mask layer is determined from the region of interest layer;

[0019] Based on the connected components of the blood vessels in the target mask layer, determine the key point coordinates of the blood vessel key points.

[0020] In one embodiment, the morphological parameters of the vascular connected domains include at least one of the following: the number of vascular connected domains, their location coordinates, and the area of ​​the region.

[0021] In one embodiment, determining a target mask layer from a region of interest layer based on various morphological parameters includes:

[0022] The initial mask layer is determined from the region of interest layer based on the number of blood vessel connected regions in each mask layer;

[0023] The target mask layer is determined from the initial mask layer based on the area and / or coordinate position of the blood vessel connected regions in the initial mask layer.

[0024] In one embodiment, determining a target mask layer from the initial mask layer based on the area and / or coordinate position of the vascular connected regions in the initial mask layer includes:

[0025] The candidate mask layer is determined from the initial mask image based on the area of ​​the blood vessel connected regions in the initial mask layer;

[0026] Determine the positional distance between the blood vessel connected region and the reference plane based on the coordinate position of the blood vessel connected region in the candidate mask layer;

[0027] The candidate mask layer corresponding to the nearest connected vascular region is determined, and it becomes the target mask layer for the corresponding ascending aortic key point.

[0028] In one embodiment, determining the key point coordinates of blood vessel key points based on the blood vessel connected components in the target mask layer includes:

[0029] Based on the coordinates of the vascular connected region closest to the body surface in the target mask image, determine the key point coordinates of the ascending aorta.

[0030] In one embodiment, determining a target mask layer from the initial mask layer based on the area and / or coordinate position of the vascular connected regions in the initial mask layer includes:

[0031] The largest connected vessel region in each initial mask layer is determined based on the area of ​​the connected vessel region in the initial mask layer.

[0032] Based on the area and / or coordinates of the largest connected vessel region in the initial mask layer, and the area and / or coordinates of each non-largest connected vessel region, the initial mask layer that satisfies the area ratio condition and / or position distance condition is determined as the target mask layer for the corresponding head branch vessel key point.

[0033] In one embodiment, determining the key point coordinates of blood vessel key points based on the target mask layer includes:

[0034] Based on the coordinates of the largest connected vessel region in the target mask layer, determine the key point coordinates of the first head branch vessel key points;

[0035] Based on the coordinates of the non-maximum connected vessel regions in the target mask layer that satisfy the area ratio condition and / or position distance condition, determine the key point coordinates of the second head branch vessel key points and the third head branch vessel key points.

[0036] In one embodiment, determining a target mask layer from a region of interest layer based on various morphological parameters includes:

[0037] Based on the coordinates of the blood vessel connected regions in each mask layer of the region of interest layer, determine the shortest distances between the blood vessel connected regions in each mask layer and the reference surface.

[0038] Determine the nearest distance from the shortest distances, and determine the mask layer corresponding to the nearest distance as the target mask layer for the key points of the lower limb arteries.

[0039] In one embodiment, determining the key point coordinates of blood vessel key points based on the blood vessel connected components of the target mask layer includes:

[0040] Based on the coordinates of the blood vessel connected regions in the target mask layer, determine the first target connected region closest to the body surface;

[0041] Based on the coordinate position of the first target connected region, determine the key point position of the first lower limb artery.

[0042] Based on the positional distance between non-target connected components and target connected components in the target mask layer, determine the candidate connected components that meet the preset distance;

[0043] Determine the second target connected component that is closest to the body surface from the candidate connected components;

[0044] Based on the coordinates of the second target connected region, determine the key point locations of the second lower limb artery.

[0045] A device for determining the location of key points in a blood vessel, the device comprising:

[0046] The segmentation mask image acquisition module is used to acquire segmentation mask images of blood vessels;

[0047] The Region of Interest (ROI) layer determination module is used to determine the ROI layer of key points of the corresponding blood vessel from the segmented mask image.

[0048] The key point coordinate determination module is used to determine the key point coordinates of blood vessel key points based on the blood vessel connected domains in the region of interest layer.

[0049] A computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the method described in any of the above embodiments.

[0050] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any of the above embodiments.

[0051] The aforementioned method, apparatus, computer equipment, and storage medium for determining the location of key blood vessel points acquire a segmented mask image of the blood vessel, then determine the region of interest (ROI) layer for the corresponding key blood vessel points from the segmented mask image, and further determine the key point coordinates of the key blood vessel points based on the connected blood vessels within the ROI layer. Therefore, compared to manually inputting the coordinates of key blood vessel points, this application's solution improves the intelligence level of key blood vessel point location determination. Furthermore, by determining the ROI layer based on the segmented mask image, and then determining the key point coordinates of the key blood vessel points based on the connected blood vessels within the ROI layer, compared to relying on manual experience, the ROI layer and the connected blood vessels within the ROI layer can be accurately determined, thereby improving the accuracy of key blood vessel point coordinate determination. Attached Figure Description

[0052] Figure 1 This is an application scenario diagram of a method for determining the location of key blood vessel points in one embodiment;

[0053] Figure 2 This is a flowchart illustrating a method for determining the location of key points in a blood vessel in one embodiment;

[0054] Figure 3 This is a schematic diagram of the aorta in one embodiment;

[0055] Figure 4 This is a schematic diagram of the vascular connectivity region between the ascending aorta and the branch vessels in the top of the head in one embodiment;

[0056] Figure 5 This is a schematic diagram showing the location of key points in the aorta in one embodiment;

[0057] Figure 6 This is a schematic diagram showing the location of key points in the ascending aorta in one embodiment;

[0058] Figure 7 This is a schematic diagram showing the location of key points of the head branch vessels in one embodiment;

[0059] Figure 8 This is a schematic diagram showing the location of key points in the lower limb arteries in one embodiment;

[0060] Figure 9 A structural block diagram of a blood vessel key point location determination device in one embodiment;

[0061] Figure 10 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0062] 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.

[0063] The method for determining the location of key points in blood vessels provided in this application can be applied to, for example... Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. Terminal 102 can generate instructions based on user triggers and request the generation of key point coordinates. Server 104 can acquire a segmentation mask image of the blood vessel based on the instructions, and determine the region of interest (ROI) layer for the key points of the corresponding blood vessel from the segmentation mask image. Furthermore, server 104 can determine the key point coordinates of the blood vessel key points based on the connected components of the blood vessel in the ROI layer. Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices, and server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers.

[0064] In one embodiment, such as Figure 2 As shown, a method for determining the location of key points in a blood vessel is provided, which can be applied to... Figure 1 Taking the server in the example, the following steps may be included:

[0065] Step S202: Obtain the segmentation mask image of the blood vessels.

[0066] Wherein, "blood vessel" refers to the blood vessel whose location is to be determined. It can refer to blood vessels within human body tissues, such as the aorta, or blood vessels connecting abdominal organs. This application does not limit this. For example, the aorta... Figure 3 As shown, the following explanation uses the aorta as an example.

[0067] A segmentation mask image refers to three-dimensional volumetric data of a cross-section of the aorta, which may include multiple mask layers. A segmentation mask image can be represented as (X, Y, Z), where X and Y represent the size of each mask layer, and Z represents the dimension in the layer direction. For example, if the segmentation mask image has 100 layers, then Z corresponds to 100.

[0068] In this embodiment, the image pixels in the mask layer are represented by binary segmentation, such that the area corresponding to pixel value 0 represents non-blood vessels, and the area corresponding to pixel value 1 represents blood vessels.

[0069] In this embodiment, the server can acquire the scan data of the object under test through computed tomography (CT), and after image reconstruction based on the scan data, obtain the segmentation mask image of the corresponding aortic vessels.

[0070] Step S204: Determine the region of interest layer of the key points of the corresponding blood vessel from the segmentation mask image.

[0071] As will be understood by those skilled in the art, the key vascular points to be determined refer to the critical points used in extracting the vascular skeleton line. For example, taking the aorta as an example, the key vascular points to be determined include the ascending aorta key point, two lower limb vascular key points, and three head branch vascular key points. This is merely an illustrative example; the key vascular points to be determined correspond to the corresponding blood vessels, and will not be elaborated further here.

[0072] A region of interest (ROI) layer refers to the layer area of ​​a mask layer in a segmented mask image that corresponds to the approximate location of key points of blood vessels to be determined. A ROI layer can include multiple mask layers.

[0073] In this embodiment, the server can determine the region of interest layer corresponding to each key point of the blood vessel to be determined based on the acquired segmentation mask image and prior knowledge combined with anatomical information. For example, the region of interest layer corresponding to each key point of the blood vessel to be determined can be determined from the segmentation mask image by combining the scanning direction of the scanning data corresponding to the segmentation mask image.

[0074] Step S206: Determine the key point coordinates of the blood vessel key points based on the blood vessel connected domain in the region of interest layer.

[0075] In this context, a blood vessel connected region refers to a connected area formed by pixels with a value of 1 in a mask layer. For example... Figure 4 In (a), 401 and 402 both refer to the vascular connectivity domain.

[0076] In this embodiment, the server can analyze the determined region of interest (ROI) layers corresponding to each key blood vessel point to be determined, locate the mask layer where each key blood vessel point is located based on the blood vessel connectivity of each mask layer in the determined ROI layer, and thus determine the coordinate position of each key blood vessel point.

[0077] In this embodiment, the regions of interest (ROI) layers corresponding to each key point of the blood vessel to be determined are not the same. The server can perform parallel analysis and processing on each key point of the blood vessel to be determined based on the ROI layer of the key point of the blood vessel to be determined, so as to obtain the key point coordinates of the key points of the blood vessel in parallel.

[0078] In this embodiment, after determining the key point coordinates of each blood vessel, the server can mark the corresponding location on the blood vessel. Continuing with the aorta as an example, after determining six key points—including the ascending aorta, lower limb arteries, and head branch vessels—the server can display the markings as shown below. Figure 5 As shown in the figure. Among them, 501-503 represent key points of the head branch vessels, 504 represents key points of the ascending aorta, and 505 and 506 represent key points of the lower limb arteries.

[0079] In this embodiment, after determining the key point coordinates of the key points of the blood vessel, the server can extract the vascular skeleton line and perform subsequent processing based on the determined key point coordinates of the blood vessel, such as extracting the vascular skeleton line (e.g., the vascular center line) through skeletonization and refinement processing.

[0080] In the aforementioned method for determining the location of key blood vessel points, a segmentation mask image of the blood vessel is acquired. Then, a region of interest (ROI) layer is determined from the segmentation mask image for the corresponding key blood vessel points to be determined. Furthermore, the key point coordinates are determined based on the connected components of the blood vessels within the ROI layer. Therefore, compared to manually inputting the coordinates of key blood vessel points, this solution improves the intelligence level of key blood vessel point location determination. Moreover, by determining the ROI layer based on the segmentation mask image, and then determining the key point coordinates based on the connected components of the blood vessels within the ROI layer, compared to relying on manual experience, the ROI layer and the connected components of the blood vessels within the ROI layer can be accurately determined, thereby improving the accuracy of key blood vessel point coordinate determination.

[0081] In one embodiment, determining the region of interest (ROI) layer of the key points of the corresponding blood vessel from the segmentation mask image may include: obtaining the scanning direction of the segmentation mask image; and determining the ROI layer of the key points of the corresponding blood vessel from the segmentation mask image based on the scanning direction.

[0082] The scanning direction refers to the direction in which the object to be scanned is scanned using scanning equipment such as CT. For example, when scanning the human body, the scan can be performed from the head to the lower limbs or from the lower limbs to the head.

[0083] In this embodiment, the server can analyze the blood vessel connected components of each mask layer in the segmented mask image, and then determine the scanning direction of the segmented mask image based on the determined blood vessel connected components and the layer index of each mask layer in the segmented mask image, i.e., the Z-direction dimension of the mask layer. Specifically, if scanning from the head to the lower limbs, the layer index of each mask layer increases layer by layer from the head to the lower limbs; if scanning from the lower limbs to the head, the layer index of each mask layer decreases layer by layer from the head to the lower limbs.

[0084] In this embodiment, after obtaining the scanning direction of the segmented mask image, the server can determine the mask layer where the largest connected vessel region is located, and then, based on the determined mask layer where the largest connected vessel region is located and the scanning direction, determine the region of interest layer corresponding to each key point of the vessel to be determined.

[0085] In one specific embodiment, continuing to use the aorta as an example, the mask layer containing the largest vascular connected region is Z.max If the scanning direction is determined to be from the lower limbs to the head, then the server can determine the region of interest layer for key points of the ascending aorta as 2 / 3*Z in the segmentation mask image. shape ~Z shape The mask layer corresponding to the region. Where Z... shape This is the index of the mask layer corresponding to the largest dimension in the Z direction, i.e., the index of the last mask layer scanned from the lower limb to the head. Similarly, the server can determine the region of interest layer for key points of blood vessels in the lower limb as 0 to 1 / 4*Z in the segmentation mask image. shape The mask layer corresponding to the region, and the region of interest layer for key points of branch blood vessels in the head is the Z-axis in the segmentation mask image. max ~Z shape The mask layer corresponding to the region.

[0086] As will be understood by those skilled in the art, 2 / 3*Z shape 1 / 4*Z shape The coefficients 2 / 3 and 1 / 4, etc., can be determined after multiple iterations and updates. For example, the server can set preset coefficients during the first processing, and after multiple processing and updates, the optimal coefficients are determined to be 2 / 3 and 1 / 4.

[0087] In one embodiment, obtaining the scanning direction of the segmentation mask image may include: determining the largest blood vessel connected region in each mask layer of the segmentation mask image; determining a reference surface mask layer from the segmentation mask image based on each largest blood vessel connected region; and determining the scanning direction of the segmentation mask image based on the reference surface mask layer.

[0088] Continuing with the aorta as an example, after the server obtains the segmentation mask image, it can first take a maximum connected component for the aorta segmentation mask to prevent isolated pixels from affecting the result, so as to obtain a clean aorta segmentation mask.

[0089] Specifically, the server can start from layer 0 and perform two-dimensional image morphological connected component analysis on each mask layer of the segmented mask image, calculating the number of connected components in each layer, the label of each connected component, the center coordinates of each connected component, and the size of each connected component, among other attribute information.

[0090] Furthermore, the server can record the largest connected vessel region in each mask layer, and based on the area of ​​the largest connected vessel region in each mask layer, obtain the index of the mask layer containing the largest connected vessel region in the segmented mask image, i.e., Z. max The server can use the mask layer containing the largest connected vessel region in the determined segmentation mask image as the reference mask layer for subsequent processing.

[0091] In this embodiment, the server determines the mask layer Z containing the largest connected vessel region in the segmented mask image. max Subsequently, based on prior knowledge of aortic anatomy, it can be determined that the index with the largest aortic cross-sectional area should be near the aortic arch. The server can then compare the mask layers Z at the midpoint of the segmented mask image. median The mask layer Z containing the largest connected vessel region max The index, i.e., the comparison Z median and Z max Determine the scanning direction of the segmentation mask image, such as Z. median <Z max If the scanning of the segmentation mask image starts from the lower limbs, then it is determined that the scanning starts from the head.

[0092] In this embodiment, once the server determines the scanning direction, it can determine the 0th layer and the last Z layer in the segmentation mask image. shape Then, the layer for the region of interest is determined. See the previous text for details; it will not be repeated here.

[0093] In one embodiment, determining the key point coordinates of vascular key points based on the vascular connected domains in the region of interest (ROI) layer may include: obtaining the morphological parameters of the vascular connected domains of each mask layer in the ROI layer; determining the target mask layer from the ROI layer based on the morphological parameters; and determining the key point coordinates of vascular key points based on the vascular connected domains in the target mask layer.

[0094] The morphological parameters of the vascular connected domains may include at least one of the following: the number of vascular connected domains, their location coordinates, and their area.

[0095] In this embodiment, the number of blood vessel connected components refers to the number of connected components in each mask layer. For example, if the connected component corresponding to pixel value 1 is considered a blood vessel connected component, then it refers to the number of connected components with pixel value 1 in each mask layer. Position coordinates refer to the coordinate position of each blood vessel connected component, which can be the center coordinate position. Region area refers to the area of ​​the blood vessel connected component, which can be determined based on the number of pixels constituting the blood vessel connected component.

[0096] In this embodiment, the server can sequentially perform morphological connectivity processing on each mask layer in each region of interest layer to obtain the number of blood vessel connectivity regions, their location coordinates, and the area of ​​each mask layer.

[0097] In this embodiment, when the server performs morphological connectivity processing on each mask layer in each region of interest layer, it can be a parallel processing process or a serial processing process, and this application does not limit it in this way.

[0098] Furthermore, the server can determine the target mask layer from the region of interest layer based on the morphological parameters of the acquired blood vessel connected regions, and then determine the key point coordinates of the blood vessel key points based on the blood vessel connected regions in the target mask layer.

[0099] Specifically, the target mask layer refers to the mask layer where the key points of the blood vessels are located in the Z direction. After determining the target mask layer, the server can determine the location of the key points of the blood vessels in the Z direction.

[0100] In this embodiment, for each blood vessel key point, the server can determine the corresponding target mask layer based on the corresponding region of interest layer, and then determine the position of each blood vessel key point in the X and Y directions in each target mask layer based on the determined target mask layers, thereby obtaining the key point coordinates of each blood vessel key point.

[0101] In one embodiment, determining a target mask layer from a region of interest layer based on various morphological parameters may include: determining an initial mask layer from the region of interest layer based on the number of vascular connected regions in each mask layer; and determining a target mask layer from the initial mask layer based on the area and / or coordinate position of the vascular connected regions in the initial mask layer.

[0102] The initial mask layer refers to a mask layer whose morphological parameters meet the requirements for the location of key points on the corresponding blood vessels. For example, the number of connected regions of the blood vessels meets the quantity requirements, or the area and coordinate position of the connected regions of the blood vessels meet certain requirements.

[0103] In this embodiment, the server can determine the initial mask layer from the region of interest (ROI) layer based on the number of vascular connected regions in the mask layer within the ROI layer. For example, if the ascending and descending aortas are on the same cross-section, the number of vascular connected regions in the mask layer corresponding to the key points of the ascending aorta should satisfy 2. Similarly, for key points of head branch vessels, the number of vascular connected regions in the corresponding mask layer should be 3.

[0104] In this embodiment, the server can determine the initial mask layer for the corresponding ascending aortic vessel key point based on whether the number of connected vascular regions in each mask layer of the region of interest layer corresponding to the ascending aortic vessel key point is equal to 2. Similarly, the server can determine the initial mask layer for the corresponding head branch vessel key point based on whether the number of connected vascular regions in each mask layer of the region of interest layer corresponding to the head branch vessel key point is equal to 3.

[0105] In this embodiment, the number of connected vessel regions in the mask layer alone is insufficient to accurately determine the specific coordinates of each vessel key point in the Z direction. Therefore, the server can determine the target mask layer from the initial mask layer based on the area and / or coordinate positions of the connected vessel regions in the initial mask layer. For example, when the area and / or coordinate positions of the connected vessel regions in the initial mask layer meet certain preset conditions, the target mask layer for the corresponding vessel key point is determined from the initial mask layer.

[0106] In one embodiment, determining the target mask layer from the initial mask layer based on the area and / or coordinate position of the vascular connected domain in the initial mask layer may include: determining the candidate mask layer from the initial mask image based on the area of ​​the vascular connected domain in the initial mask layer; determining the positional distance between the vascular connected domain and the reference surface based on the coordinate position of the vascular connected domain in the candidate mask layer; and determining the candidate mask layer corresponding to the vascular connected domain with the closest positional distance as the target mask layer for the corresponding ascending aortic vascular key point.

[0107] In this embodiment, when the server determines the target mask layer for the key points of the ascending aorta, due to the influence of the head branch vessels, the initial mask layer determined based on the number of connected regions of the vessels includes the mask layer corresponding to the key points of the head branch vessels. Since there is a significant difference between the vessel area of ​​the ascending aorta and the descending aorta and the vessel area of ​​the head branch vessels, the server can exclude the influence of the head branch vessels based on the area of ​​the connected regions of the vessels in each initial mask layer.

[0108] Specifically, the server can determine whether the area of ​​each blood vessel connected region is greater than a preset area threshold, such as 1000, and whether the ratio of the areas of two blood vessel connected regions is within a preset area ratio range, such as [0.7, 2], based on the area of ​​the blood vessel connected regions in each initial mask layer. In this way, the server can determine whether the initial mask layer is a candidate mask layer for the corresponding ascending aortic blood vessel key point.

[0109] In this embodiment, reference Figure 4 , Figure 4 In the image, (a) represents two connected blood vessels, and (b) also represents two connected blood vessels. After the server performs area determination processing, it can be determined that (a) is the mask layer corresponding to the ascending aorta, while (b) is the mask layer corresponding to the head branch blood vessels.

[0110] In this embodiment, refer to Figure 3The ascending aortic vessel key point of the tested object is closest to the human body surface. After determining the candidate mask layer for the corresponding ascending aortic vessel key point, the server can determine the vessel connected to the human body surface with the closest positional distance based on the coordinate position of the vessel connected region in each candidate mask layer. The corresponding candidate mask layer is the target mask layer for the corresponding ascending aortic vessel key point. (Continue to refer to...) Figure 4 In (a), since the ascending aorta is closer to the body surface, the server can determine the positional distance between the vascular connected domain 401 and the body surface (A surface) to obtain the positional distance between the vascular connected domain and the human body surface in the candidate mask layer. Then, by comparing the positional distances corresponding to multiple candidate mask layers, the server finds the cross-section of the ascending aorta vascular key point, that is, determines the candidate mask layer with the closest (or shortest) positional distance as the candidate mask layer for the corresponding ascending aorta vascular key point.

[0111] In one embodiment, determining the key point coordinates of a vascular key point based on the vascular connected domain in the target mask layer may include: determining the key point coordinates of the ascending aortic vascular key point based on the coordinate position of the vascular connected domain closest to the body surface in the target mask image.

[0112] In this embodiment, after determining the target mask layer corresponding to the key points of the ascending aorta, the server determines the vascular connected domain closest to the body surface as the vascular connected domain corresponding to the ascending aorta based on the coordinate position of the vascular connected domain in the target mask layer.

[0113] Then, based on the coordinates of the center point of the vascular connected region corresponding to the ascending aorta, the server obtains the position coordinates (X and Y) of the key points of the ascending aorta in the target mask image, such as... Figure 6 The central cross intersection position.

[0114] Furthermore, the server obtains the key point coordinates (X, Y, Z) of the ascending aortic vessel key points based on the Z-direction coordinates of the target mask image and the determined position coordinates X and Y.

[0115] In one embodiment, determining the target mask layer from the initial mask layer based on the area and / or coordinate position of the vascular connected domains in the initial mask layer may include: determining the largest vascular connected domain in each initial mask layer based on the area of ​​the vascular connected domains in the initial mask layer; and determining the initial mask layer that satisfies the area ratio condition and / or position distance condition as the target mask layer for the corresponding head branch vascular key point based on the area and / or coordinate position of the largest vascular connected domain in the initial mask layer, and the area and / or coordinate position of each non-largest vascular connected domain.

[0116] In this embodiment, based on the region of interest layer corresponding to the key points of the head branch blood vessels, the server can perform region of interest layer analysis with a fixed step size, for example, by analyzing layer by layer according to a unit step size.

[0117] Specifically, after the server determines an initial mask layer with three connected blood vessels from the region of interest layer of key points of the head branch blood vessels, it can determine the largest connected blood vessel in each initial mask layer based on the area of ​​the connected blood vessels in each initial mask layer.

[0118] Furthermore, after the server determines the largest connected vessel in each initial mask layer, it can determine each non-largest connected vessel in each initial mask layer.

[0119] Furthermore, the server can determine whether the area of ​​the non-maximum connected vessel region in each initial mask layer is greater than 20% of the area of ​​the maximum connected vessel region, and whether the distance between any two connected vessel regions is greater than 70 pixels, based on the area and / or coordinate position of the maximum connected vessel region in the initial mask layer, and the area and / or coordinate position of each non-maximum connected vessel region. In other words, it can determine whether the area ratio condition and the position distance condition are satisfied.

[0120] In other embodiments, the server can determine whether the area of ​​the non-maximum connected vessel region in each initial mask layer is greater than 20% of the area of ​​the maximum connected vessel region based on the area of ​​the maximum connected vessel region and the area of ​​each non-maximum connected vessel region in the initial mask layer, i.e., whether the area ratio condition is met. Alternatively, it can determine whether the distance between any two connected vessel regions in each initial mask layer is greater than 70 pixels based on the coordinate positions of the maximum connected vessel region and the coordinate positions of each non-maximum connected vessel region in the initial mask layer, i.e., whether the positional distance condition is met.

[0121] In this embodiment, when the server determines that the blood vessel connectivity domain in the initial mask layer satisfies the area ratio condition and / or the positional distance condition, the initial mask layer is determined as the target mask layer for the key points of the corresponding head branch blood vessels.

[0122] In one embodiment, determining the key point coordinates of key blood vessels based on the target mask layer may include: determining the key point coordinates of the first head branch blood vessel based on the coordinate position of the largest connected blood vessel in the target mask layer; and determining the key point coordinates of the second and third head branch blood vessels based on the coordinate positions of the non-largest connected blood vessels in the target mask layer that satisfy the area ratio condition and / or position distance condition.

[0123] In this embodiment, after the server obtains the target mask layer for the key points of the head branch vessels, it can determine the key point coordinates of the corresponding key points of the head branch vessels based on the coordinates of the center points of the three connected vessel regions that satisfy the area ratio condition and / or positional distance condition described above, and the Z-direction coordinates of the corresponding target mask layer. The three key points of the head branch vessels determined by the server are as follows: Figure 7 The positions of the cross in (a), (b), and (c) are shown in the diagram.

[0124] In one embodiment, determining the target mask layer from the region of interest layer based on various morphological parameters may include: determining the shortest distance between the vascular connected domain and the reference plane in each mask layer according to the coordinate position of the vascular connected domain in each mask layer in the region of interest layer; determining the nearest distance from the shortest distances; and determining the mask layer corresponding to the nearest distance as the target mask layer for the corresponding lower limb arterial key point.

[0125] Among them, morphological parameters may include, but are not limited to, the number of vascular connected domains, their location coordinates, and the area of ​​the region, as mentioned above.

[0126] A reference surface can refer to a location in the mask layer that indicates the outer surface of the human body, such as the skin surface of the chest or the skin surface of the back.

[0127] In this embodiment, based on the region of interest layer corresponding to the key points of the lower limb arteries, the server can analyze the region of interest layer with a fixed step size, as described above, by analyzing layer by layer according to the unit step size.

[0128] Specifically, the server can determine the shortest distance between the blood vessel connected region in each mask layer and the reference surface in the corresponding mask layer based on the coordinate position of the blood vessel connected region in each mask layer of the region of interest layer.

[0129] Furthermore, the server identifies the cross-section containing the nearest connected component to the body surface, thus determining the mask layer corresponding to the closest distance as the target mask layer for the key points of the corresponding lower limb arteries.

[0130] In one embodiment, determining the key point coordinates of a vascular key point based on the vascular connected regions of the target mask layer may include: determining a first target connected region closest to the body surface based on the coordinate position of the vascular connected regions in the target mask layer; determining the key point position of a first lower limb artery key point based on the coordinate position of the first target connected region; determining candidate connected regions that meet a preset distance based on the positional distance between non-target connected regions and the target connected region in the target mask layer; determining a second target connected region closest to the body surface from the candidate connected regions; and determining the key point position of a second lower limb artery key point based on the coordinate position of the second target connected region.

[0131] In this embodiment, after the server determines the target mask layer of the corresponding lower limb artery key point, it can use the center coordinates of the blood vessel connected domain closest to the body surface in the target mask layer as the location of one key point on one side of the lower limb artery, that is, determine the key point location of the first lower limb artery key point.

[0132] Furthermore, the server can first perform morphological connected component processing on the layer where the key point of the lower limb artery is located, i.e., the target mask layer, and calculate the distance between the center of each blood vessel connected component and the first key point location. If the distance is greater than 100 pixels and is closest to the body surface, it is considered to be a key point location on the other side of the lower limb artery, i.e., the key point location of the second lower limb artery blood vessel key point is determined.

[0133] In this embodiment, the key points of the lower limb arteries determined by the server are as follows: Figure 8 The positions of the cross in (a) and (b) are shown.

[0134] It should be understood that, although Figure 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 2 At least some of the steps in the process 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 executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0135] In one embodiment, such as Figure 9 As shown, a device for determining the location of key points in a blood vessel is provided, comprising: a segmentation mask image acquisition module 100, a region of interest layer determination module 200, and a key point coordinate determination module 300, wherein:

[0136] The segmentation mask image acquisition module 100 is used to acquire the segmentation mask image of blood vessels.

[0137] The Region of Interest (ROI) layer determination module 200 is used to determine the ROI layer of the key points of the corresponding blood vessel from the segmented mask image.

[0138] The key point coordinate determination module 300 is used to determine the key point coordinates of blood vessel key points based on the blood vessel connected domain in the region of interest layer.

[0139] In one embodiment, the region of interest layer determination module 200 may include:

[0140] The scanning direction determination submodule is used to obtain the scanning direction of the segmentation mask image.

[0141] The Region of Interest (ROI) layer determination submodule is used to determine the ROI layer of the corresponding blood vessel key points from the segmented mask image based on the scan direction.

[0142] In one embodiment, the scanning direction determination submodule may include:

[0143] The connected component determination unit is used to determine the largest blood vessel connected component in each mask layer of the segmented mask image.

[0144] The reference mask layer determination unit is used to determine the reference surface mask layer from the segmented mask image based on each largest blood vessel connected region.

[0145] The scanning direction determination unit is used to determine the scanning direction of the segmented mask image based on the reference surface mask layer.

[0146] In one embodiment, the key point coordinate determination module 300 may include:

[0147] The morphological parameter acquisition submodule is used to obtain the morphological parameters of the blood vessel connected regions in each mask layer of the region of interest layer.

[0148] The target mask layer determination submodule is used to determine the target mask layer from the region of interest layer based on various morphological parameters.

[0149] The key point coordinate determination submodule is used to determine the key point coordinates of blood vessel key points based on the blood vessel connected regions in the target mask layer.

[0150] In one embodiment, the morphological parameters of the vascular connected domains may include at least one of the following: the number of vascular connected domains, their location coordinates, and the area of ​​the region.

[0151] In one embodiment, the target mask layer determining submodule may include:

[0152] The initial mask layer determination unit is used to determine the initial mask layer from the region of interest layer based on the number of blood vessel connected regions in each mask layer.

[0153] The first target mask layer determination unit is used to determine the target mask layer from the initial mask layer based on the area and / or coordinate position of the blood vessel connected domain in the initial mask layer.

[0154] In one embodiment, the first target mask layer determining unit may include:

[0155] The candidate mask layer determination sub-unit is used to determine the candidate mask layer from the initial mask image based on the area of ​​the blood vessel connected region in the initial mask layer.

[0156] The position distance determination sub-unit is used to determine the position distance between the blood vessel connected region and the reference surface based on the coordinate position of the blood vessel connected region in the candidate mask layer.

[0157] The first target mask layer determination sub-unit is used to determine the candidate mask layer corresponding to the nearest connected vascular region, which is the target mask layer for the corresponding ascending aortic vascular key point.

[0158] In one embodiment, the key point coordinate determination submodule is used to determine the key point coordinates of the ascending aortic key points based on the coordinate position of the vascular connected domain closest to the body surface in the target mask image.

[0159] In one embodiment, the first target mask layer determining unit may include:

[0160] The maximum connected vessel region determination sub-unit is used to determine the maximum connected vessel region in each initial mask layer based on the area of ​​the connected vessel regions in the initial mask layer.

[0161] The second target mask layer determination subunit is used to determine the target mask layer of the corresponding head branch blood vessel key point based on the area and / or coordinate position of the largest blood vessel connected domain in the initial mask layer, and the area and / or coordinate position of each non-largest blood vessel connected domain.

[0162] In one embodiment, the key point coordinate determination submodule may include:

[0163] The first key point coordinate determination unit is used to determine the key point coordinates of the first head branch blood vessel key point based on the coordinate position of the largest blood vessel connected region in the target mask layer.

[0164] The second key point coordinate determination unit is used to determine the key point coordinates of the second head branch vessel key point and the third head branch vessel key point based on the coordinate position of the non-maximum vessel connectivity region in the target mask layer that satisfies the area ratio condition and / or position distance condition.

[0165] In one embodiment, the target mask layer determining submodule may include:

[0166] The shortest distance determination unit is used to determine the shortest distance between the blood vessel connected region and the reference surface in each mask layer based on the coordinate position of the blood vessel connected region in each mask layer in the region of interest layer.

[0167] The second target mask layer unit is used to determine the nearest distance from the shortest distances, and to determine the mask layer corresponding to the nearest distance as the target mask layer of the corresponding lower limb arterial blood vessel key point.

[0168] In one embodiment, the key point coordinate determination submodule may include:

[0169] The first target connected component determination unit is used to determine the first target connected component closest to the body surface based on the coordinate position of the blood vessel connected component in the target mask layer.

[0170] The first key point location determination unit is used to determine the key point location of the first lower limb artery key point based on the coordinate location of the first target connected domain.

[0171] The alternative connected component determination unit is used to determine alternative connected components that meet a preset distance based on the positional distance between non-target connected components and target connected components in the target mask layer.

[0172] The second target connected component determination unit is used to determine the second target connected component that is closest to the body surface from the candidate connected components.

[0173] The second key point location determination unit is used to determine the key point location of the second lower limb artery key point based on the coordinate location of the second target connected domain.

[0174] Specific limitations regarding the device for determining the location of key blood vessels can be found in the limitations of the method for determining the location of key blood vessels described above, and will not be repeated here. Each module in the aforementioned device for determining the location of key blood vessels 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 in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0175] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 10 As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and the 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 database stores data such as segmented mask images, regions of interest layers, and keypoint coordinates. The network interface is used for communication with external terminals via a network connection. When executed by the processor, the computer program implements a method for determining the location of key points in blood vessels.

[0176] Those skilled in the art will understand that Figure 10 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.

[0177] In one embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to perform the following steps: acquiring a segmented mask image of a blood vessel; determining a region of interest (ROI) layer for key points of the corresponding blood vessel from the segmented mask image; and determining the key point coordinates of the key points of the blood vessel based on the connected components of the blood vessel in the ROI layer.

[0178] In one embodiment, when the processor executes a computer program, it implements a region of interest layer for determining key points of the corresponding blood vessel from a segmented mask image. This may include: obtaining the scanning direction of the segmented mask image; and determining the region of interest layer for the key points of the corresponding blood vessel from the segmented mask image based on the scanning direction.

[0179] In one embodiment, when the processor executes a computer program to obtain the scanning direction of the segmentation mask image, it may include: determining the largest blood vessel connectivity region of each mask layer in the segmentation mask image; determining a reference surface mask layer from the segmentation mask image based on each largest blood vessel connectivity region; and determining the scanning direction of the segmentation mask image based on the reference surface mask layer.

[0180] In one embodiment, when the processor executes a computer program, it determines the key point coordinates of vascular key points based on the vascular connected regions in the region of interest layer. This may include: obtaining the morphological parameters of the vascular connected regions of each mask layer in the region of interest layer; determining a target mask layer from the region of interest layer based on the morphological parameters; and determining the key point coordinates of vascular key points based on the vascular connected regions in the target mask layer.

[0181] In one embodiment, the morphological parameters of the vascular connected domains may include at least one of the following: the number of vascular connected domains, their location coordinates, and the area of ​​the region.

[0182] In one embodiment, when the processor executes a computer program, it determines a target mask layer from a region of interest layer based on various morphological parameters. This may include: determining an initial mask layer from the region of interest layer based on the number of vascular connected regions in each mask layer; and determining a target mask layer from the initial mask layer based on the area and / or coordinate position of the vascular connected regions in the initial mask layer.

[0183] In one embodiment, when the processor executes a computer program, determining a target mask layer from the initial mask layer based on the area and / or coordinate position of the vascular connected domain in the initial mask layer may include: determining a candidate mask layer from the initial mask image based on the area of ​​the vascular connected domain in the initial mask layer; determining the positional distance between the vascular connected domain and the reference surface based on the coordinate position of the vascular connected domain in the candidate mask layer; and determining the candidate mask layer corresponding to the vascular connected domain with the closest positional distance as the target mask layer for the corresponding ascending aortic vessel key point.

[0184] In one embodiment, when the processor executes a computer program, determining the key point coordinates of a vascular key point based on the vascular connected regions in the target mask layer may include: determining the key point coordinates of the ascending aortic vascular key point based on the coordinate position of the vascular connected region closest to the body surface in the target mask image.

[0185] In one embodiment, when the processor executes a computer program, it determines a target mask layer from the initial mask layer based on the area and / or coordinate position of the vascular connected domains in the initial mask layer. This may include: determining the largest vascular connected domain in each initial mask layer based on the area of ​​the vascular connected domains in the initial mask layer; and determining the initial mask layer that satisfies the area ratio condition and / or position distance condition as the target mask layer for the corresponding head branch vascular key point based on the area and / or coordinate position of the largest vascular connected domain in the initial mask layer, and the area and / or coordinate position of each non-largest vascular connected domain.

[0186] In one embodiment, when the processor executes a computer program, it determines the key point coordinates of key points of blood vessels based on a target mask layer. This may include: determining the key point coordinates of a first head branch blood vessel based on the coordinate position of the largest connected blood vessel in the target mask layer; and determining the key point coordinates of a second head branch blood vessel and a third head branch blood vessel based on the coordinate position of a non-largest connected blood vessel that satisfies the area ratio condition and / or position distance condition in the target mask layer.

[0187] In one embodiment, when the processor executes a computer program, it determines a target mask layer from a region of interest layer based on various morphological parameters. This may include: determining the shortest distance between the vascular connected domain and the reference plane in each mask layer according to the coordinate position of the vascular connected domain in each mask layer in the region of interest layer; determining the nearest distance from the shortest distances; and determining the mask layer corresponding to the nearest distance as the target mask layer for the corresponding lower limb arterial vascular key point.

[0188] In one embodiment, when the processor executes the computer program, it determines the key point coordinates of a vascular key point based on the vascular connected regions of the target mask layer. This may include: determining a first target connected region closest to the body surface based on the coordinate position of the vascular connected regions in the target mask layer; determining the key point position of a first lower limb artery key point based on the coordinate position of the first target connected region; determining candidate connected regions that meet a preset distance based on the positional distance between non-target connected regions and the target connected region in the target mask layer; determining a second target connected region closest to the body surface from the candidate connected regions; and determining the key point position of a second lower limb artery key point based on the coordinate position of the second target connected region.

[0189] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, performs the following steps: acquiring a segmented mask image of a blood vessel; determining a region of interest (ROI) layer for key points of the corresponding blood vessel from the segmented mask image; and determining the key point coordinates of the key points of the blood vessel based on the connected components of the blood vessel in the ROI layer.

[0190] In one embodiment, when the computer program is executed by the processor, it implements the determination of the region of interest layer of the key points of the corresponding blood vessel from the segmented mask image, which may include: obtaining the scanning direction of the segmented mask image; and determining the region of interest layer of the key points of the corresponding blood vessel from the segmented mask image based on the scanning direction.

[0191] In one embodiment, when the computer program is executed by the processor, acquiring the scanning direction of the segmentation mask image may include: determining the largest blood vessel connectivity region of each mask layer in the segmentation mask image; determining a reference surface mask layer from the segmentation mask image based on each largest blood vessel connectivity region; and determining the scanning direction of the segmentation mask image based on the reference surface mask layer.

[0192] In one embodiment, when the computer program is executed by the processor, it determines the key point coordinates of blood vessel key points based on the blood vessel connected regions in the region of interest layer. This may include: obtaining the morphological parameters of the blood vessel connected regions of each mask layer in the region of interest layer; determining the target mask layer from the region of interest layer based on the morphological parameters; and determining the key point coordinates of blood vessel key points based on the blood vessel connected regions in the target mask layer.

[0193] In one embodiment, the morphological parameters of the vascular connected domains may include at least one of the following: the number of vascular connected domains, their location coordinates, and the area of ​​the region.

[0194] In one embodiment, when the computer program is executed by the processor, it determines a target mask layer from the region of interest layer based on various morphological parameters. This may include: determining an initial mask layer from the region of interest layer based on the number of vascular connected regions in each mask layer; and determining a target mask layer from the initial mask layer based on the area and / or coordinate position of the vascular connected regions in the initial mask layer.

[0195] In one embodiment, when the computer program is executed by the processor, determining a target mask layer from the initial mask layer based on the area and / or coordinate position of the vascular connected domain in the initial mask layer may include: determining a candidate mask layer from the initial mask image based on the area of ​​the vascular connected domain in the initial mask layer; determining the positional distance between the vascular connected domain and the reference surface based on the coordinate position of the vascular connected domain in the candidate mask layer; and determining the candidate mask layer corresponding to the vascular connected domain with the closest positional distance as the target mask layer for the corresponding ascending aortic vascular key point.

[0196] In one embodiment, when the computer program is executed by the processor, determining the key point coordinates of the vascular key points based on the vascular connected regions in the target mask layer may include: determining the key point coordinates of the ascending aortic vascular key points based on the coordinate position of the vascular connected region closest to the body surface in the target mask image.

[0197] In one embodiment, when the computer program is executed by the processor, it determines a target mask layer from the initial mask layer based on the area and / or coordinate position of the vascular connected domains in the initial mask layer. This may include: determining the largest vascular connected domain in each initial mask layer based on the area of ​​the vascular connected domains in the initial mask layer; and determining the initial mask layer that satisfies the area ratio condition and / or position distance condition as the target mask layer for the corresponding head branch vascular key point based on the area and / or coordinate position of the largest vascular connected domain in the initial mask layer, and the area and / or coordinate position of each non-largest vascular connected domain.

[0198] In one embodiment, when the computer program is executed by the processor, it determines the key point coordinates of key points of blood vessels based on the target mask layer. This may include: determining the key point coordinates of the first head branch blood vessel key point based on the coordinate position of the largest blood vessel connected region in the target mask layer; and determining the key point coordinates of the second head branch blood vessel key point and the third head branch blood vessel key point based on the coordinate position of the non-largest blood vessel connected region in the target mask layer that satisfies the area ratio condition and / or position distance condition.

[0199] In one embodiment, when the computer program is executed by the processor, it determines the target mask layer from the region of interest layer based on various morphological parameters. This may include: determining the shortest distance between the vascular connected region and the reference plane in each mask layer according to the coordinate position of the vascular connected region in each mask layer in the region of interest layer; determining the nearest distance from the shortest distances; and determining the mask layer corresponding to the nearest distance as the target mask layer for the corresponding lower limb arterial key point.

[0200] In one embodiment, when the computer program is executed by the processor, it determines the key point coordinates of a vascular key point based on the vascular connected regions of the target mask layer. This may include: determining a first target connected region closest to the body surface based on the coordinate position of the vascular connected regions in the target mask layer; determining the key point position of a first lower limb artery key point based on the coordinate position of the first target connected region; determining candidate connected regions that meet a preset distance based on the positional distance between non-target connected regions and the target connected region in the target mask layer; determining a second target connected region closest to the body surface from the candidate connected regions; and determining the key point position of a second lower limb artery key point based on the coordinate position of the second target connected region.

[0201] 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, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0202] 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.

[0203] 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 blood vessel key point position determination method, characterized by, The method includes: Obtain a segmentation mask image of a blood vessel, wherein the segmentation mask image is three-dimensional volume data of a cross-section of the blood vessel, including multiple mask layers; From the segmented mask image, determine the region of interest layer corresponding to the key points of the blood vessel to be determined; Based on the blood vessel connected domain in the region of interest layer, determine the key point coordinates of the blood vessel key points; The step of determining the key point coordinates of blood vessel key points based on the blood vessel connected regions in the region of interest layer includes: Obtain the morphological parameters of the blood vessel connected regions in each mask layer of the region of interest layer; Based on the morphological parameters, a target mask layer is determined from the region of interest layer; Based on the connected components of the blood vessels in the target mask layer, determine the key point coordinates of the blood vessel key points; The morphological parameters of the vascular connected regions include the number of vascular connected regions, their location coordinates, and their area; the step of determining the target mask layer from the region of interest layer based on each of the morphological parameters includes: Based on the number of blood vessel connected regions in each of the mask layers, an initial mask layer is determined from the region of interest layer; the number of blood vessel connected regions in each mask layer of the region of interest layer corresponding to different types of blood vessel key points meets different preset number requirements; The target mask layer is determined from the initial mask layer based on the area and / or coordinate position of the blood vessel connected regions in the initial mask layer.

2. The method of claim 1, wherein, The step of determining the region of interest layer corresponding to the key points of the blood vessel to be determined from the segmented mask image includes: Obtain the scanning direction of the segmentation mask image; Based on the scanning direction, a region of interest layer corresponding to the key points of the blood vessel to be determined is determined from the segmentation mask image.

3. The method of claim 2, wherein, The step of obtaining the scanning direction of the segmentation mask image includes: Determine the maximum connected vessel region in each mask layer of the segmented mask image; A reference surface mask layer is determined from the segmented mask image based on each of the largest connected vessel regions. The scanning direction of the segmentation mask image is determined based on the reference surface mask layer.

4. The method of claim 1, wherein, The step of determining the target mask layer from the initial mask layer based on the area and / or coordinate position of the blood vessel connected regions in the initial mask layer includes: Based on the area of ​​the blood vessel connected regions in the initial mask layer, a candidate mask layer is determined from the initial mask layer; Based on the coordinate position of the blood vessel connected region in the candidate mask layer, the positional distance between the blood vessel connected region and the reference surface is determined, where the reference surface refers to the position in the candidate mask layer that indicates the outer surface of the human body. The candidate mask layer corresponding to the blood vessel connected region with the closest location is determined, and it is the target mask layer for the corresponding ascending aortic blood vessel key point.

5. The method of claim 4, wherein, Determining the key point coordinates of blood vessel key points based on the blood vessel connected components in the target mask layer includes: Based on the coordinates of the vascular connected domain closest to the body surface in the target mask layer, determine the key point coordinates of the ascending aorta.

6. The method of claim 1, wherein, The step of determining the target mask layer from the initial mask layer based on the area and / or coordinate position of the blood vessel connected regions in the initial mask layer includes: The largest connected vessel region in each initial mask layer is determined based on the area of ​​the connected vessel region in the initial mask layer. Based on the area and / or coordinates of the largest connected vessel region in the initial mask layer, and the area and / or coordinates of each non-largest connected vessel region, the initial mask layer that satisfies the area ratio condition and / or position distance condition is determined as the target mask layer for the corresponding head branch vessel key point.

7. The method of claim 6, wherein, Determining the key point coordinates of blood vessel key points based on the target mask layer includes: Based on the coordinates of the largest connected vessel region in the target mask layer, determine the key point coordinates of the first head branch vessel key point; Based on the coordinate positions of the non-maximum connected vessel regions in the target mask layer that satisfy the area ratio condition and / or position distance condition, determine the key point coordinates of the second head branch vessel key point and the third head branch vessel key point.

8. The method according to claim 1, characterized in that, The step of determining the target mask layer from the region of interest layer based on each of the morphological parameters includes: Based on the coordinate positions of the blood vessel connected regions in each mask layer in the region of interest layer, determine the shortest distances between the blood vessel connected regions in each mask layer and the reference surface, where the reference surface refers to the position in the mask layer that indicates the outer surface of the human body. The shortest distance is determined from the shortest distances, and the mask layer corresponding to the shortest distance is determined as the target mask layer for the key points of the lower limb arteries.

9. The method according to claim 8, characterized in that, Determining the key point coordinates of blood vessel key points based on the blood vessel connected components of the target mask layer includes: Based on the coordinate position of the blood vessel connected region in the target mask layer, determine the first target connected region closest to the body surface; Based on the coordinate position of the first target connected region, determine the key point position of the first lower limb artery key point; Based on the positional distance between the non-target connected components and the target connected components in the target mask layer, candidate connected components that satisfy the preset distance are determined. Determine the second target connected component that is closest to the body surface from the candidate connected components; Based on the coordinates of the second target connected region, determine the key point locations of the second lower limb arterial blood vessel key points.

10. A device for determining the location of key points in a blood vessel, characterized in that, The device includes: The segmentation mask image acquisition module is used to acquire the segmentation mask image of the blood vessel. The segmentation mask image is three-dimensional volume data of the cross-section of the blood vessel, including multiple mask layers. The region of interest layer determination module is used to determine the region of interest layer corresponding to the key points of the blood vessel to be determined from the segmented mask image; The key point coordinate determination module is used to determine the key point coordinates of key points of blood vessels based on the blood vessel connectivity domain in the region of interest layer. The key point coordinate determination module includes: The morphological parameter acquisition submodule is used to acquire the morphological parameters of the blood vessel connected regions in each mask layer of the region of interest layer; The target mask layer determination submodule is used to determine the target mask layer from the region of interest layer based on the morphological parameters. The key point coordinate determination submodule is used to determine the key point coordinates of blood vessel key points based on the blood vessel connected regions in the target mask layer. The morphological parameters of the blood vessel connected regions include the number of blood vessel connected regions, their location coordinates, and the area of ​​the region; the target mask layer determination submodule includes: The initial mask layer determination unit is used to determine an initial mask layer from the region of interest layer based on the number of blood vessel connected regions in each mask layer; the number of blood vessel connected regions in each mask layer of the region of interest layer corresponding to different types of blood vessel key points meets different preset number requirements; The first target mask layer determination unit is used to determine the target mask layer from the initial mask layer based on the area and / or coordinate position of the blood vessel connected domain in the initial mask layer.

11. 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 9.

12. 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 9.

Images