Vessel centerline determination system, device, and storage medium

Abstract

This invention discloses a system, apparatus, and storage medium for determining the vascular centerline. The system includes a processor configured to execute a vascular centerline determination method comprising: determining an initial vascular centerline image corresponding to a medical image of a target object, and an initial vascular centerline in the initial vascular centerline image; performing connected component processing on the initial vascular centerline to obtain a connected component result, and determining a target connected component in the connected component result that meets set connected component conditions; determining two seed points based on the target connected component, and growing the vascular centerline along the connected component away from the seed points, starting from the two seed points, and adding the grown pixels to a corresponding growth list; and determining a target vascular centerline corresponding to the initial vascular centerline based on the target connected component and the corresponding growth list. This achieves the technical effect of improving the accuracy of vascular centerline determination results in medical images.

Description

Technical Field

[0001] This invention relates to the field of medical image processing technology, and in particular to a system, device, and storage medium for determining the centerline of blood vessels. Background Technology

[0002] Cardiovascular and cerebrovascular diseases are associated with high morbidity, disability, and mortality rates. Early diagnosis and risk assessment of these diseases can significantly improve life expectancy and quality of life. Coronary angiography is widely used in cardiac examinations and disease diagnosis because it provides a very clear view of the diameter, course, distribution, morphology, and extent of lesions in the coronary arteries. However, the images presented to doctors by coronary angiography often show numerous tortuous vessels intertwined and obstructing each other, making it difficult for doctors to directly observe lesions such as stenosis and calcification.

[0003] The vascular centerline is the skeleton reflecting the spatial topology of blood vessels, and its accurate extraction is a crucial step in the quantitative analysis of angiographic images. Among commonly used deep learning-based segmentation methods, point cloud-based segmentation methods utilize the topological structure of blood vessels and can effectively segment various vessel categories. However, for complex coronary arteries with diverse vessel types, point cloud segmentation methods still suffer from undersegmentation and oversegmentation.

[0004] In summary, the accuracy of existing methods or systems for determining the vascular centerline is relatively low. Summary of the Invention

[0005] This invention provides a system, device, and storage medium for determining the centerline of a blood vessel, in order to solve the problem of low accuracy in determining the centerline of a blood vessel by existing systems.

[0006] According to one aspect of the present invention, a system for determining the centerline of a blood vessel is provided, the system comprising a processor configured to perform a method for determining the centerline of a blood vessel, comprising:

[0007] Determine the initial vascular centerline image corresponding to the medical image of the target object, and the initial vascular centerline in the initial vascular centerline image;

[0008] The initial blood vessel centerline is processed to obtain a connected component result, and the target connected component in the connected component result that meets the set connected component condition is determined.

[0009] Two seed points are determined based on the target connected region. Starting from the two seed points, the blood vessel centerline is grown along the direction away from the connected region where the seed points are located, and the grown pixels are added to the corresponding growth list.

[0010] The target vessel centerline corresponding to the initial vessel centerline is determined based on the target connected region and the growth list corresponding to the target connected region.

[0011] According to another aspect of the present invention, a device for determining the centerline of a blood vessel is provided, comprising:

[0012] An initial vascular centerline module is used to determine the initial vascular centerline image corresponding to the medical image of the target object, and the initial vascular centerline in the initial vascular centerline image;

[0013] The connected component module is used to perform connected component processing on the initial blood vessel centerline to obtain connected component results, and to determine the target connected component in the connected component results that meets the set connected component conditions.

[0014] The growth module is used to determine two seed points based on the target connected region, and grow the blood vessel centerline along the direction away from the connected region where the seed points are located, starting from the two seed points, and add the grown pixels to the corresponding growth list.

[0015] The result module is used to determine the target vessel centerline corresponding to the initial vessel centerline based on the target connected region and the growth list corresponding to the target connected region.

[0016] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the method for determining the vascular centerline according to any embodiment of the present invention.

[0017] The technical solution of the vascular centerline determination system provided in this invention is used to determine the connected component result corresponding to the initial vascular centerline determined by the prior art, determine a seed point based on the target connected component in the connected component result, and grow the vascular centerline based on the seed point along the direction away from the target connected component. This achieves accurate growth results by growing the vascular centerline based on the target connected component, and the accurate target vascular centerline can be determined based on the growth result corresponding to the seed point of the target connected component.

[0018] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a structural block diagram of a blood vessel centerline determination system provided according to an embodiment of the present invention;

[0021] Figure 2A This is a flowchart of a method for determining the centerline of a blood vessel according to an embodiment of the present invention;

[0022] Figure 2B This is an initial blood vessel centerline image including the largest connected region provided according to an embodiment of the present invention;

[0023] Figure 2C This is a partial initial blood vessel centerline image, excluding the largest connected region, provided according to an embodiment of the present invention.

[0024] Figure 2D This is yet another initial blood vessel centerline image including the largest connected region provided according to an embodiment of the present invention;

[0025] Figure 2E This is a schematic diagram of the growth path corresponding to an initial branch vessel according to an embodiment of the present invention;

[0026] Figure 3 This is a flowchart of another method for determining the centerline of a blood vessel provided according to an embodiment of the present invention;

[0027] Figure 4 This is a flowchart of another method for determining the centerline of a blood vessel provided according to an embodiment of the present invention;

[0028] Figure 5A This is a flowchart of the circle deletion method provided in the embodiments of the present invention;

[0029] Figure 5B This is a schematic diagram of the circle determination result provided by an embodiment of the present invention;

[0030] Figure 5C This is a schematic diagram of the portion to be deleted according to an embodiment of the present invention;

[0031] Figure 6A This is a schematic diagram of the structure of the blood vessel centerline determination device provided in an embodiment of the present invention;

[0032] Figure 6B This is a schematic diagram of the structure of another blood vessel centerline determination device provided according to an embodiment of the present invention. Detailed Implementation

[0033] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0034] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0035] Figure 1 The vascular centerline determination system provided for embodiments of the present invention can be a digital computer of various forms, such as a laptop computer, desktop computer, workbench, personal digital assistant, server, blade server, mainframe computer, and other suitable computer. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0036] like Figure 1 As shown, the vascular centerline determination system 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the vascular centerline determination system 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0037] Multiple components in the vascular centerline determination system 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, optical disk, etc.; and a communication unit 19, such as a network card, modem, wireless transceiver, etc. The communication unit 19 allows the vascular centerline determination system 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0038] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described below, such as the method for determining the centerline of a blood vessel.

[0039] Figure 2A This is a flowchart of a method for determining the centerline of a blood vessel according to an embodiment of the present invention. This embodiment is applicable to determining the centerline of a target blood vessel in a medical image, and is particularly suitable for determining the centerline of the coronary arteries in a medical image containing coronary vessels. This method can be executed by a blood vessel centerline determining device, which can be implemented in hardware and / or software, and can be configured in the processor of a blood vessel centerline determining system. Figure 2A As shown, the method includes:

[0040] S110. Determine the initial vascular centerline image corresponding to the medical image of the target object, and the initial vascular centerline in the initial vascular centerline image.

[0041] Among them, medical images include medical images of blood vessels, such as coronary angiography images or other angiography images.

[0042] In one embodiment, the initial vascular centerline image is a vascular centerline image corresponding to a medical image determined by existing technology.

[0043] In one embodiment, the initial vascular centerline image includes at least one initial main vascular centerline and at least one initial branch vascular centerline, see [link to documentation]. Figure 2B As shown, the initial vascular centerline image includes the initial main vascular centerline located in the middle and the centerlines of two initial branch vessels distributed on both sides of the initial main vascular centerline.

[0044] In one embodiment, the initial vascular centerline image includes at least one initial main vessel and at least one initial branch vessel, each labeled with a corresponding main branch vessel label and branch vessel label. For example, initial vascular centerline image A includes an initial main vessel centerline labeled as the left anterior descending artery (LAD); initial vascular centerline image B includes an initial main vessel centerline labeled as the left anterior descending artery, an initial main vessel centerline labeled as the left circumflex artery (LCX), and an initial main vessel centerline labeled as the right coronary artery (RCA); initial vascular centerline image C includes an initial main vessel centerline of the right coronary artery, an initial branch vessel centerline labeled as the diagonal branch (D), an initial branch vessel centerline labeled as the obtuse marginal branch (OM), and an initial branch vessel centerline labeled as the right posterior descending artery (RPDA).

[0045] S120. Perform connected component processing on the initial blood vessel centerline to obtain connected component results, and determine the target connected component in the connected component results that meets the set connected component conditions.

[0046] In one embodiment, the connectivity condition is set to the maximum connectivity (see [link]). Figure 2B The largest thick solid line in the diagram). Here, the largest connected component refers to the connected component with the largest area, or the connected component with the most pixels. In this embodiment, connected component processing is performed on the initial blood vessel centerline to obtain a connected component result. This connected component result includes at least two connected components, and the largest connected component in the connected component result is taken as the target connected component that meets the set conditions.

[0047] In one embodiment, the connectivity condition is set to the middle position of the corresponding connectivity region. In this embodiment, the connectivity result includes at least two connectivity regions. The connectivity region located at the middle position of the initial blood vessel centerline is taken as the target connectivity region that meets the set connectivity condition.

[0048] In one embodiment, the initial vascular centerline image includes at least one initial main vascular centerline and at least one initial branch vascular centerline. The connected regions of all initial vascular centerlines in the initial vascular centerline image can be determined in parallel; alternatively, each initial vascular centerline can be determined sequentially according to a set order. Specifically, the current initial vascular centerline in the initial vascular centerline image can be determined based on the order of initializing the main vascular centerline first and then the branch vascular centerline.

[0049] By adopting the order of initializing the main vessel first and then the branch vessels, the centerline of the branch vessels can be determined by referring to the centerline of the main vessel, which can improve the accuracy of the subsequent determination of the centerlines of the branch vessels.

[0050] S130. Determine two seed points based on the target connected region. Starting from these two seed points, grow the blood vessel centerline along the direction away from the connected region where the seed points are located, and add the grown pixels to the corresponding growth list.

[0051] This step aims to determine the corresponding vessel centerline using an automatic growth method, starting from the seed point.

[0052] In one embodiment, the two endpoints of the target connected region are directly used as two seed points. Starting from each of these seed points, growth is performed along the corresponding vessel centerline in directions away from the target connected region (see...). Figure 2B , Figure 2C , Figure 2D and Figure 2E (The black arrow in the diagram) and during the growth process, the grown pixels are added to the corresponding growth list until the stopping condition is met. The stopping condition is that the currently growing pixel is an endpoint with only one neighbor, or the current seed point is an endpoint with only one neighbor. It can be understood that if an endpoint of the target connected component is the starting or ending endpoint of the corresponding initial blood vessel centerline, the number of grown pixels corresponding to that starting or ending endpoint, which serves as the seed point, is zero, and the number of pixels in the corresponding growth list is also zero.

[0053] In one embodiment, any two pixels in the target connected region are used as two seed points. Then, starting from each of the seed points, growth is performed along the corresponding blood vessel centerline in a direction away from the other seed point. During the growth process, the grown pixels are added to the corresponding growth list until the stopping condition is met.

[0054] In one embodiment, the number of seed points is two, and the growth pixels corresponding to these two seed points are added to different pixel groups in a growth list. A growth list includes two pixel groups, and each pixel group is configured to be written into the growth list corresponding to one seed point.

[0055] See Figure 2B As shown, the central line of the blood vessel is the initial main blood vessel centerline, and an initial branch blood vessel centerline is connected to each side of the initial main blood vessel centerline. The maximum connected region of this initial main blood vessel centerline is ( Figure 2B The two endpoints of the thick solid line (in the diagram) are set as seed points. The two black arrows in the diagram indicate the growth direction of the vessel centerline. Therefore, as... Figure 2C As shown, the growth method can be used to obtain... Figure 2BThe central lines of the main blood vessels above and below the largest connected region are defined. Therefore, this largest connected region corresponds to two growth lists: all pixels in one growth list form the central lines of the main blood vessels above the largest connected region; all pixels in the other growth list form the central lines of the main blood vessels below the largest connected region.

[0056] See Figure 2D As shown, the central vessel centerline is the initial main vessel centerline, and each side of the initial main vessel centerline is connected to an initial branch vessel centerline. For the initial branch vessel centerlines located to the right of the initial main vessel centerline, their maximum connected region (… Figure 2D The two endpoints of the thick solid line (in the diagram) are set as seed points. The two black arrows in the diagram indicate the growth direction of the vessel centerline. Therefore, as... Figure 2E As shown, the growth method can be used to obtain... Figure 2D The partial branch vessel centerlines above and below the maximum connected region are defined. Therefore, this maximum connected region corresponds to two growth lists: all pixels in one growth list form the partial branch vessel centerlines above the maximum connected region; all pixels in the other growth list form the partial branch vessel centerlines below the maximum connected region.

[0057] S140. Determine the target vessel centerline corresponding to the initial vessel centerline based on the target connected region and the growth list corresponding to the target connected region.

[0058] Among them, the target vessel centerline is the desired vessel centerline, which can be used for quantitative analysis or clinical diagnosis.

[0059] The paths corresponding to all pixels in the growth list are the growth paths. The union of the target connected region and the growth path is taken as the target blood vessel centerline corresponding to the initial blood vessel centerline.

[0060] Specifically, when the seed point is the endpoint of the target connected region, the target connected region is spliced ​​with the growth path corresponding to the seed point of the target connected region, and the splicing result is used as the target blood vessel center.

[0061] When a seed point is not an endpoint of the target connected region, the connected region between two seed points is determined. The growth paths corresponding to various seed points and the connected regions between the two types of seed points are spliced ​​together, and the splicing result is used as the centerline of the target blood vessel.

[0062] The technical solution of the vascular centerline determination system provided in this invention is used to determine the connected component result corresponding to the initial vascular centerline determined by the prior art, determine a seed point based on the target connected component in the connected component result, and grow the vascular centerline based on the seed point along the direction away from the target connected component. This achieves accurate growth results by growing the vascular centerline based on the target connected component, and the accurate target vascular centerline can be determined based on the growth result corresponding to the seed point of the target connected component.

[0063] Figure 3 This is a flowchart of a method for determining the centerline of blood vessels provided in an embodiment of the present invention. This embodiment refines the initial method for determining the centerline of blood vessels in the aforementioned embodiments. For example... Figure 3 As shown, the method includes:

[0064] S2101. Obtain angiographic images of the target scanning area.

[0065] The target scanning area refers to the area that includes the blood vessel to be examined. For example, if the blood vessel to be examined is the coronary artery, then the target scanning area is the chest; if the blood vessel to be examined is the cerebral blood vessel, then the target scanning area is the head.

[0066] Angiography is a medical imaging technique used to visualize the interior or lumen of blood vessels and organs in the human body, especially arteries, veins, and coronary arteries. It has been applied to radionuclide angiography and newer vascular imaging techniques such as CT angiography and MR angiography.

[0067] S2102. Determine the binary blood vessel segmentation image corresponding to the angiography image.

[0068] In one embodiment, angiography images are input into a trained segmentation model to obtain binary vessel segmentation images. The segmentation model can employ a CNN (Convolutional Neural Network) structure, such as U-net. Training samples are input into the segmentation model to obtain segmentation results. The segmentation model iterates continuously based on the segmentation results and the binary vessel labels carried by the training samples, gradually reducing the loss (such as cross-entropy loss, Dice loss) until the model converges.

[0069] S2103. Determine all vessel centerlines in the binary vessel segmentation image, and the feature set including the point cloud features of all vessel centerlines.

[0070] After the binary segmentation image is determined, the skeletonization algorithm is used to determine the center lines of all blood vessels in the binary segmentation image; then the point cloud features of each blood vessel center line are determined, and the union of the point cloud features of all blood vessel center lines is taken as the feature set.

[0071] S2104. Input the feature set into the trained point cloud segmentation model to obtain an initial blood vessel centerline image. The initial blood vessel centerline image includes at least one initial main blood vessel centerline and at least one initial branch blood vessel centerline, and each initial blood vessel centerline is labeled with a corresponding label, which includes at least one of blood vessel identifier and blood vessel category.

[0072] In one embodiment, this step aims to determine the label of each initial vessel centerline in the binary vessel segmentation image based on the target point cloud features. In this embodiment, each initial vessel centerline in the initial vessel centerline image is identified with a corresponding vessel name or vessel category, such as "left anterior descending artery", "branch vessel", or "diagonal branch (branch vessel)".

[0073] In one embodiment, the point cloud segmentation model may employ a point cloud segmentation network, such as PointNet or Point Transformer.

[0074] S220. Perform connected component processing on the initial blood vessel centerline to obtain connected component results, and determine the target connected component in the connected component results that meets the set connected component conditions.

[0075] S230. Determine two seed points based on the target connected region. Starting from these two seed points, grow the blood vessel centerline along the direction away from the connected region where the seed points are located, and add the grown pixels to the corresponding growth list.

[0076] S240. Determine the target vessel centerline corresponding to the initial vessel centerline based on the target connected region and the growth list corresponding to the target connected region.

[0077] In one embodiment, a method can be adopted to first perform blood vessel segmentation, and then determine the blood vessel centerline and the label of each blood vessel centerline in the segmentation result. Specifically, after obtaining the binary blood vessel segmentation image, the point cloud features in the binary blood vessel segmentation image can be determined first, and then the point cloud features can be input into the trained point cloud segmentation model to obtain the initial blood vessel image. The initial blood vessel image includes blood vessel labels, and then the blood vessel centerline of the blood vessel carrying each blood vessel label can be determined to obtain the initial blood vessel centerline image.

[0078] This invention first determines a binary vessel segmentation image corresponding to an angiography image, and then inputs the point cloud features of each vessel centerline in the binary vessel segmentation image into a trained point cloud segmentation model to obtain an initial vessel centerline image including initial vessel centerline labels. This improves the accuracy of the initial vessel centerline image and prepares the system for managing vessels through vessel centerline labels, or for selecting different processing methods for different initial vessel centerlines based on the vessel centerline labels.

[0079] Figure 4 This is a flowchart illustrating a method for determining the vascular centerline according to an embodiment of the present invention. This embodiment further refines the vascular centerline growth described in the foregoing embodiments. Figure 4 As shown, the method includes:

[0080] S310. Determine the initial vascular centerline image corresponding to the medical image of the target object, and the initial vascular centerline in the initial vascular centerline image.

[0081] S320. Perform connected component processing on the initial blood vessel centerline to obtain connected component results, and determine the target connected component in the connected component results that meets the set connected component conditions.

[0082] S3301. Determine two seed points based on the target connected component. Starting from these two seed points, grow the blood vessel centerline along the direction away from the connected component where the seed points are located. Add the grown pixels to the corresponding growth list and determine whether a bifurcation point is encountered or whether the stopping condition is met.

[0083] Blood vessels are not straight lines, but rather resemble a tree with a trunk and branches. Each branch has a fork in the tree; both the trunk and branches have start and end points. Therefore, during the growth of the blood vessel centerline based on seed points determined by the target connected region, at least one of the following will be encountered: a fork in the tree, a start point, or an end point.

[0084] In this embodiment, during the growth of the vascular centerline, it is detected in real time whether a bifurcation point is encountered or whether the stopping condition is met. If a bifurcation point is encountered, S3302 is executed; if the stopping condition is met, S3303 is executed; if neither a bifurcation point is encountered nor the stopping condition is met, the current growth of the vascular centerline continues.

[0085] If the current pixel on the blood vessel centerline determined by the blood vessel centerline growth method is detected as a single neighbor pixel, then the pixel is considered to be the start or end point of the initial blood vessel centerline, and therefore the current blood vessel centerline growth is determined to meet the stopping condition.

[0086] S3302. When encountering a bifurcation point, determine the growth direction corresponding to the bifurcation point based on the set bifurcation growth strategy, continue to grow along the growth direction corresponding to the bifurcation point, and return to add the grown pixels to the corresponding growth list, and determine whether a bifurcation point has been encountered or the stopping condition is met.

[0087] The branching growth strategy is set as follows: grow along the branching path with the smallest angle to the current growth direction.

[0088] like Figure 2CThe bifurcation point in the equation is a bifurcation point within a 3-neighborhood, denoted as m. b Determine the coordinates of several pixels that have completed blood vessel centerline growth before the bifurcation point, denoted as m. b-5 The direction vector of the bifurcation point is... Similarly, the direction vectors of the two bifurcated paths can be obtained. and Then calculate and The included angle, and and The angle between the two growth directions is determined, and the branch path corresponding to the smallest angle is taken as the branch path with the smallest angle to the current growth direction, that is, the branch path corresponding to the smoother direction.

[0089] S3303. If the stopping conditions are met, stop the growth of the vascular centerline.

[0090] Once the vascular centerline stops growing, the corresponding growth list also stops updating.

[0091] In one embodiment, a stop character is added to the corresponding growth list while the growth of the vessel centerline is stopped. This stop character indicates that the growth list stops updating data because the vessel centerline has stopped growing.

[0092] S340. Determine the target vessel centerline corresponding to the initial vessel centerline based on the target connected region and the growth list corresponding to the target connected region.

[0093] In this embodiment, when a bifurcation point is encountered during the growth of the vascular centerline, growth is carried out along the growth direction corresponding to the bifurcation point determined based on the set bifurcation growth strategy, which can improve the accuracy of vascular centerline growth.

[0094] Figure 5A This is a flowchart illustrating a method for determining the centerline of a blood vessel according to an embodiment of the present invention. This embodiment adds the step of deleting the circle formed by the centerlines of branch vessels and the main vessel from the target vessel centerline, based on the previous embodiment. Figure 5A As shown, the method includes:

[0095] S450. For any target branch vessel centerline, determine whether both endpoints of the target branch vessel centerline are on the target main vessel centerline; if so, execute S460; wherein, the target branch vessel centerline is the target vessel centerline corresponding to the initial branch vessel centerline, and the target main vessel centerline is the target vessel centerline corresponding to the initial main vessel centerline.

[0096] Because blood flows in a unidirectional direction, the starting point of a branch vessel usually originates from the main vessel, while the ending point of a branch vessel does not end at the main vessel; that is, the starting point of the branch vessel's centerline usually originates from the main vessel's centerline, while the ending point of the branch vessel's centerline usually does not end at the main vessel's centerline.

[0097] Therefore, for any target branch vessel centerline, it is determined whether both endpoints of the target branch vessel centerline lie on the target main vessel centerline. If so, it is determined that the target branch vessel centerline and the target main vessel centerline form a circle (see...). Figure 5B (The thick solid line part in the text), therefore S460 is executed.

[0098] This embodiment determines whether the corresponding endpoint of the target branch vessel centerline is on the target main vessel centerline by determining whether the last grown pixel in the two growth lists corresponding to the branch vessel centerline is on the target branch vessel centerline.

[0099] S460. Determine the bifurcation points in the two growth lists corresponding to the center line of the target branch blood vessel that meet the set bifurcation conditions, and delete the pixels whose growth time is later than that of the bifurcation point from the growth list where the bifurcation point is located.

[0100] In one embodiment, the following steps are used to determine the bifurcation points in two growth lists that meet the set bifurcation conditions corresponding to the centerline of the target branch vessel:

[0101] Step c1: Determine all bifurcation points included in the two growth lists corresponding to the centerline of the target branch vessel.

[0102] Step c2: Determine the maximum absolute value among the target direction coordinates of all bifurcation points, where the target direction is the direction of the initial main blood vessel centerline and the coordinate axis with the smallest angle in the current coordinate system.

[0103] Step c3: Take the bifurcation point corresponding to the maximum absolute value as the bifurcation point in the two growth lists that meet the set bifurcation conditions corresponding to the center line of the target branch blood vessel.

[0104] In one embodiment, the target direction is the Y-axis direction, i.e., the vertical direction. For example... Figure 5C As shown, the maximum absolute value in the Y-axis direction is the maximum coordinate value in the Y-axis direction, which is... Figure 5C The coordinates of the lowest bifurcation point of the central line of the included annular blood vessel in the vertical direction are used as the bifurcation point that meets the set bifurcation conditions.

[0105] Understandably, the choice of target direction is related to the specific coordinate system of the angiography image, as long as it is aligned with the coordinate axis of the largest component of the main vessel centerline.

[0106] After identifying the branching point that meets the set branching conditions in the corresponding growth list, determine the pixels that grow after that branching point and delete the pixels that grow after that branching point from the corresponding growth list. See [link to relevant documentation]. Figure 5C The thick solid line portion in the text.

[0107] This embodiment improves the accuracy of the target branch vessel centerline by deleting part of the vessel centerline within the circle formed between the centerline of each branch vessel and the centerline of the main vessel, thereby improving the accuracy of the target vessel centerline.

[0108] Figure 6A This is a schematic diagram of the structure of the blood vessel centerline determination device provided in an embodiment of the present invention. Figure 6A As shown, the device includes:

[0109] The initial vascular centerline module 610 is used to determine the initial vascular centerline image corresponding to the medical image of the target object, and the initial vascular centerline in the initial vascular centerline image.

[0110] The connected component module 620 is used to perform connected component processing on the initial blood vessel centerline to obtain a connected component result, and to determine the target connected component in the connected component result that meets the set connected component conditions.

[0111] The growth module 630 is used to determine two seed points based on the target connected region, grow the blood vessel centerline from the two seed points along the direction away from the connected region where the seed points are located, and add the grown pixels to the corresponding growth list.

[0112] Result module 640 is used to determine the target vessel centerline corresponding to the initial vessel centerline based on the target connected region and the growth list corresponding to the target connected region.

[0113] In one embodiment, the initial centerline module 610 is specifically used for:

[0114] Acquire angiographic images of the target scanning area;

[0115] Determine the binary blood vessel segmentation image corresponding to the angiography image;

[0116] Determine all vessel centerlines in the binary vessel segmentation image, and a feature set including point cloud features of all vessel centerlines;

[0117] The feature set is input into the trained point cloud segmentation model to obtain an initial blood vessel centerline image. The initial blood vessel centerline image includes at least one initial main blood vessel centerline and at least one initial branch blood vessel centerline, and each initial blood vessel centerline is marked with a corresponding label, the label including at least one of blood vessel identifier and blood vessel category.

[0118] In one embodiment, the initial centerline module 610 is specifically used for:

[0119] Based on the order of initializing the main vessel first and then the branch vessels, the current initial vessel centerline in the initial vessel centerline image is determined.

[0120] In one embodiment, the growth module 630 is specifically used for:

[0121] The blood vessel centerline is grown along a direction other than the location of the largest connected region. The grown pixels are added to the corresponding growth list, and it is determined whether a bifurcation point is encountered or whether the stopping condition is met.

[0122] When a bifurcation point is encountered, the growth direction corresponding to the bifurcation point is determined based on the set bifurcation growth strategy, and growth continues along the growth direction corresponding to the bifurcation point. Then, the process of adding the grown pixels to the corresponding growth list and determining whether a bifurcation point has been encountered or whether the stopping condition is met is returned.

[0123] If the stopping conditions are met, the growth of the vascular centerline will be stopped.

[0124] In one embodiment, the set bifurcation growth strategy is to grow along the bifurcation path with the smallest angle to the current growth direction.

[0125] In one embodiment, such as Figure 6B As shown, the device also includes a de-circling module 650, which includes:

[0126] The determination unit is used to determine, for any target branch vessel centerline, whether both ends of the target branch vessel centerline are on the target main vessel centerline, wherein the target branch vessel centerline is the target vessel centerline corresponding to the initial branch vessel centerline, and the target main vessel centerline is the target vessel centerline corresponding to the initial main vessel centerline.

[0127] The deletion unit is used to determine, if so, the two growth lists corresponding to the center line of the target branch blood vessel that meet the set bifurcation conditions, and to delete pixels whose growth time is later than that of the bifurcation point from the growth list where the bifurcation point is located.

[0128] In one embodiment, the deletion unit is specifically used for:

[0129] Identify all bifurcation points included in the two growth lists corresponding to the centerline of the target branch vessel;

[0130] Determine the maximum absolute value among the target direction coordinates of all bifurcation points, where the target direction is the direction of the initial main blood vessel centerline and the coordinate axis with the smallest angle in the current coordinate system;

[0131] The bifurcation point corresponding to the maximum absolute value is taken as the bifurcation point in the two growth lists that meet the set bifurcation conditions corresponding to the center line of the target branch blood vessel.

[0132] In one embodiment, the target connected component that meets the set connected component conditions is the largest connected component in the connected component result.

[0133] In one embodiment, the binary vessel segmentation image is a binary coronary artery segmentation image.

[0134] The technical solution of the vessel centerline determination device provided in this invention is used to determine the connected component result corresponding to the initial vessel centerline determined by the prior art, determine a seed point based on the target connected component in the connected component result, and grow the vessel centerline based on the seed point along the direction away from the target connected component. This achieves accurate growth results by growing the vessel centerline based on the target connected component, and the accurate target vessel centerline can be determined based on the growth result corresponding to the seed point of the target connected component.

[0135] The vascular centerline determination device provided in the embodiments of the present invention can execute the vascular centerline determination method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the method.

[0136] In some embodiments, the vascular centerline determination method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as... Figure 1 The storage unit 18 is used in the system. In some embodiments, part or all of the computer program may be loaded and / or installed onto the vascular centerline determination system 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the vascular centerline determination method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the vascular centerline determination method by any other suitable means (e.g., by means of firmware).

[0137] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0138] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0139] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0140] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0141] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0142] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0143] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0144] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A system for determining the centerline of a blood vessel, characterized in that, The system includes a processor configured to perform the following method for determining the centerline of a blood vessel: Determine the initial vascular centerline image corresponding to the medical image of the target object, and the initial vascular centerline in the initial vascular centerline image; The initial blood vessel centerline is processed to obtain a connected component result, and the target connected component in the connected component result that meets the set connected component condition is determined. Two seed points are determined based on the target connected component. Starting from the two seed points, the growth of the blood vessel centerline is carried out along the direction away from the connected component where the seed points are located. The grown pixels are added to the corresponding growth list until the stopping condition is met. The stopping condition means that the currently grown pixel is the endpoint of only one neighborhood, or the current seed point is the endpoint of only one neighborhood. The target vessel centerline corresponding to the initial vessel centerline is determined based on the target connected region and the growth list corresponding to the target connected region.

2. The system according to claim 1, characterized in that, The initial vascular centerline image corresponding to the medical image of the target object, and the initial vascular centerline in the initial vascular centerline image, include: Obtain angiographic images of the target scanning area of ​​the target object; Determine the binary blood vessel segmentation image corresponding to the angiography image; Determine all vessel centerlines in the binary vessel segmentation image, and a feature set including point cloud features of all vessel centerlines; The feature set is input into the trained point cloud segmentation model to obtain an initial blood vessel centerline image. The initial blood vessel centerline image includes at least one initial main blood vessel centerline and at least one initial branch blood vessel centerline, and each initial blood vessel centerline is marked with a corresponding label, the label including at least one of blood vessel identifier and blood vessel category.

3. The system according to claim 2, characterized in that, The initial vascular centerline in the initial vascular centerline image is determined by the following steps: Based on the order of initializing the main vessel first and then the branch vessels, the current initial vessel centerline in the initial vessel centerline image is determined.

4. The system according to claim 1, characterized in that, The step of growing the blood vessel centerline along a direction away from the connected region where the seed point is located, and adding the grown pixels to the corresponding growth list, includes: The blood vessel centerline is grown along the direction away from the connected region where the seed point is located, and the grown pixels are added to the corresponding growth list. It is then determined whether a bifurcation point is encountered or whether the stopping condition is met. When a bifurcation point is encountered, the growth direction corresponding to the bifurcation point is determined based on the set bifurcation growth strategy, and growth continues along the growth direction corresponding to the bifurcation point. Then, the process of adding the grown pixels to the corresponding growth list and determining whether a bifurcation point has been encountered or whether the stopping condition is met is returned. If the stopping conditions are met, the growth of the vascular centerline will be stopped.

5. The system according to claim 4, characterized in that, The set bifurcation growth strategy is to grow along the bifurcation path with the smallest angle to the current growth direction.

6. The system according to claim 2, characterized in that, Also includes: For any target branch vessel centerline, determine whether both endpoints of the target branch vessel centerline are on the target main vessel centerline, wherein the target branch vessel centerline is the target vessel centerline corresponding to the initial branch vessel centerline, and the target main vessel centerline is the target vessel centerline corresponding to the initial main vessel centerline; If so, then identify the two bifurcation points in the growth lists corresponding to the center line of the target branch blood vessel that meet the set bifurcation conditions, and delete pixels whose growth time is later than that of the bifurcation point from the growth list where the bifurcation point is located.

7. The system according to claim 6, characterized in that, The process of determining the bifurcation points in the two growth lists corresponding to the centerline of the target branch vessel that meet the set bifurcation conditions includes: Identify all bifurcation points included in the two growth lists corresponding to the centerline of the target branch vessel; Determine the maximum absolute value among the target direction coordinates of all bifurcation points, where the target direction is the direction of the initial main blood vessel centerline and the coordinate axis with the smallest angle in the current coordinate system; The bifurcation point corresponding to the maximum absolute value is taken as the bifurcation point in the two growth lists that meet the set bifurcation conditions corresponding to the center line of the target branch blood vessel.

8. The system according to claim 1, characterized in that, The target connected component that meets the set connected component conditions is the largest connected component in the connected component results.

9. The system according to claim 2, characterized in that, The binary vessel segmentation image is a binary coronary artery segmentation image.

10. A device for determining the centerline of a blood vessel, characterized in that, include: An initial vascular centerline module is used to determine the initial vascular centerline image corresponding to the medical image of the target object, and the initial vascular centerline in the initial vascular centerline image; The connected component module is used to perform connected component processing on the initial blood vessel centerline to obtain connected component results, and to determine the target connected component in the connected component results that meets the set connected component conditions. The growth module is used to determine two seed points based on the target connected component, and grow the blood vessel centerline from the two seed points along the connected component away from the seed points, and add the grown pixels to the corresponding growth list until the stopping condition is met. The stopping condition means that the currently grown pixel is the endpoint of only one neighborhood, or the current seed point is the endpoint of only one neighborhood. The result module is used to determine the target vessel centerline corresponding to the initial vessel centerline based on the target connected region and the growth list corresponding to the target connected region.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the following method for determining the centerline of a blood vessel, including: Determine the initial vascular centerline image corresponding to the medical image of the target object, and the initial vascular centerline in the initial vascular centerline image; The initial blood vessel centerline is processed to obtain a connected component result, and the target connected component in the connected component result that meets the set connected component condition is determined. Two seed points are determined based on the target connected component. Starting from the two seed points, the growth of the blood vessel centerline is carried out along the direction away from the connected component where the seed points are located. The grown pixels are added to the corresponding growth list until the stopping condition is met. The stopping condition means that the currently grown pixel is the endpoint of only one neighborhood, or the current seed point is the endpoint of only one neighborhood. The target vessel centerline corresponding to the initial vessel centerline is determined based on the target connected region and the growth list corresponding to the target connected region.

Images