CN115908821A

CN115908821A - Blood vessel image segmentation method, device, electronic device and storage medium

Info

Publication number: CN115908821A
Application number: CN202211729808.4A
Authority: CN
Inventors: 黄星胜; 马骏; 郑凌霄; 兰宏志
Original assignee: Shenzhen Raysight Intelligent Medical Technology Co Ltd
Current assignee: Shenzhen Raysight Intelligent Medical Technology Co Ltd
Priority date: 2022-12-30
Filing date: 2022-12-30
Publication date: 2023-04-04

Abstract

The invention provides a blood vessel image segmentation method, a blood vessel image segmentation device, electronic equipment and a storage medium. The method comprises the following steps: acquiring an initial blood vessel image, and respectively determining a coronary artery rough segmentation image and a coronary artery skeleton image of the initial blood vessel image; determining each skeleton point on a skeleton line in the coronary skeleton image, and determining a skeleton branch point in each skeleton point based on the degree of each skeleton point; identifying a vein branch image in an initial blood vessel image based on branch point positions of the skeleton branch points in the coronary artery rough segmentation image/the coronary artery skeleton image; determining a coronary segmentation image based on the coarse coronary segmentation image and the vein branch image. By the technical scheme disclosed by the invention, the problem of low accuracy of coronary artery segmentation on the blood vessel in the prior art is solved, and the accuracy of coronary artery segmentation is improved.

Description

Blood vessel image segmentation method, device, electronic device and storage medium

Technical Field

The present invention relates to the field of medical image processing technologies, and in particular, to a method and an apparatus for segmenting a blood vessel image, an electronic device, and a storage medium.

Background

With the increase in imaging speed and scanning accuracy of CT (Computed Tomography) apparatuses, CT medical images have been widely used for cardiac examination and plaque diagnosis. The coronary artery segmentation based on the CT medical image is widely used, can extract the contours of coronary arteries and plaques in cavities, is convenient for doctors to observe the situations of stenosis, calcification, plaques and the like, and provides basis for early prevention and diagnosis of cardiovascular diseases for the doctors.

At present, in the heart coronary artery segmentation process, because the image characteristics (close distance, even fit, close CT value and the like) of veins are similar to the coronary artery, the veins are easy to be identified by mistake, and the accuracy of heart coronary artery segmentation is low.

Disclosure of Invention

The invention provides a blood vessel image segmentation method, a blood vessel image segmentation device, electronic equipment and a storage medium, which solve the problem of low accuracy of coronary artery segmentation on blood vessels in the prior art by removing vein branches in a coronary artery rough segmentation result and realize improvement of the accuracy of coronary artery segmentation.

In a first aspect, an embodiment of the present invention provides a blood vessel image segmentation method, including:

acquiring an initial blood vessel image, and respectively determining a coronary artery rough segmentation image and a coronary artery skeleton image of the initial blood vessel image;

determining each skeleton point on a skeleton line in the coronary skeleton image, and determining a skeleton branch point in each skeleton point based on the degree of each skeleton point;

identifying a vein branch image in an initial blood vessel image based on branch point positions of the skeleton branch points in the coronary artery rough segmentation image/the coronary artery skeleton image;

determining a coronary segmentation image based on the coarse coronary segmentation image and the vein branch image.

Optionally, the determining the coronary artery rough segmentation image and the coronary artery skeleton image of the initial blood vessel image respectively includes:

inputting the initial blood vessel image to a coronary artery rough segmentation model trained in advance to obtain a coronary artery rough segmentation image output by the coronary artery rough segmentation model;

and inputting the initial blood vessel image into a pre-trained framework segmentation model to obtain a coronary artery framework image output by the framework segmentation model.

Optionally, the determining a skeleton branch point in each skeleton point based on the degree of each skeleton point includes:

for any skeleton point, determining each neighborhood point of the current skeleton point, and determining the degree of the current skeleton point based on the skeleton points contained in the neighborhood points;

and acquiring a preset degree threshold value, and determining a skeleton branch point in each skeleton point based on the degree of the skeleton point and the preset degree threshold value.

Optionally, determining a skeleton branch point in each skeleton point based on the degree of each skeleton point, further includes:

determining a skeleton endpoint and a plurality of candidate branch points in the skeleton points based on the degree of each skeleton point, wherein the skeleton endpoint comprises a root endpoint and a plurality of end branch endpoints;

and determining the end point serial number of each tail end point, respectively determining a skeleton communication path from the root end point to any two tail end points with adjacent serial numbers, and screening candidate branch points based on the skeleton communication paths to obtain skeleton branch points in the candidate branch point set.

Optionally, the screening at least one candidate branch point based on the skeleton communication path includes:

determining candidate branch points which are traversed repeatedly in each framework communication path, and respectively determining a framework distance between each candidate branch point and a root end point;

and determining the skeleton branch point in the candidate branch points based on the comparison result of the skeleton distances.

Optionally, the identifying, based on a branch point position of the skeleton branch point in the coronary coarse segmentation image, a vein branch image in an initial blood vessel image includes:

determining a branch point segmentation position of the skeleton branch point in the coronary artery rough segmentation image, and determining a coronary artery branch region in the coronary artery rough segmentation image based on the branch point segmentation position;

obtaining a pre-trained vein branch recognition model, and carrying out vein branch recognition on the coronary artery branch region based on the vein branch recognition model to obtain a vein branch image.

Optionally, the identifying, based on a branch point position of the skeleton branch point in the coronary skeleton image, a vein branch image in an initial blood vessel image includes:

determining a branch point skeleton position of the skeleton branch point in the coronary skeleton image, and determining a skeleton branch region in the coronary skeleton image based on the branch point skeleton position;

acquiring a pre-trained vein branch skeleton recognition model, and performing vein branch skeleton recognition on the skeleton branch region based on the vein branch skeleton recognition model to obtain a vein branch skeleton;

and acquiring branch expansion parameters of the vein branch skeleton, and performing expansion processing on the vein branch skeleton based on the branch expansion parameters to obtain a vein branch image corresponding to the vein branch skeleton.

In a second aspect, an embodiment of the present invention further provides a blood vessel image segmentation apparatus, including:

the image determining module is used for acquiring an initial blood vessel image and respectively determining a coronary artery coarse segmentation image and a coronary artery skeleton image of the initial blood vessel image;

a skeleton branch point determination module, configured to determine skeleton points on a skeleton line in the coronary artery skeleton image, determine skeleton end points and a candidate branch point set in each skeleton point based on a degree of each skeleton point, and determine skeleton branch points in the candidate branch point set based on each skeleton end point;

a vein branch image determination module, configured to identify a vein branch image in an initial blood vessel image based on a branch point position of the skeleton branch point in the coronary coarse segmentation image/the coronary skeleton image;

a coronary artery segmentation image determination module for determining a coronary artery segmentation image of the initial blood vessel image based on the coarse coronary artery segmentation image and the vein branch image.

In a third aspect, an embodiment of the present invention further provides an electronic device, including:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein, the first and the second end of the pipe are connected with each other,

the memory stores a computer program executable by the at least one processor, the computer program being executable by the at least one processor to enable the at least one processor to perform the vessel image segmentation method according to any one of the embodiments of the present invention.

In a fourth aspect, the embodiment of the present invention further provides a computer-readable storage medium, where computer instructions are stored, and the computer instructions are configured to, when executed, cause a processor to implement the blood vessel image segmentation method according to any embodiment of the present invention.

According to the technical scheme provided by the embodiment of the invention, an initial blood vessel image is obtained, and a coronary artery rough segmentation image and a coronary artery skeleton image of the initial blood vessel image are respectively determined; determining each skeleton point on a skeleton line in a coronary skeleton image, and determining a skeleton branch point in each skeleton point based on the degree of each skeleton point; identifying a vein branch image in the initial blood vessel image based on branch point positions of the skeleton branch points in the coronary artery rough segmentation image/coronary artery skeleton image; determining a coronary artery segmentation image based on the coronary artery rough segmentation image and the vein branch image. According to the technical scheme, the blood vessel image to be segmented is processed to obtain each end point in the blood vessel, candidate branch points are determined based on the end points, and each branch point is further determined; and respectively identifying blood vessels between the branch points to obtain vein branches, and then optimizing the coronary artery segmentation result based on the identified vein branches to obtain a final coronary artery segmentation result, so that the accuracy of the coronary artery segmentation is improved.

It should be understood that the statements in this section are not intended to identify key or critical features of the embodiments of the present invention, nor are they intended to limit the scope of the invention. Other features of the present invention will become apparent from the following description. .

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the description below are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.

Fig. 1 is a flowchart of a blood vessel image segmentation method according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of a coronary artery rough segmentation image provided according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of a coronary skeleton image provided in accordance with an embodiment of the present invention;

FIG. 4 is a schematic diagram of a skeletal point provided in accordance with an embodiment of the present invention;

FIG. 5 is a schematic illustration of another skeletal point provided in accordance with an embodiment of the present invention;

FIG. 6 is a schematic diagram of another image of a coarse coronary segmentation according to an embodiment of the present invention;

FIG. 7 is a schematic diagram of a branch point of a skeleton according to an embodiment of the present invention;

FIG. 8 is a schematic view of a coronary branch region provided in accordance with an embodiment of the present invention;

FIG. 9 is a schematic diagram of a vein branch image provided in accordance with an embodiment of the present invention;

FIG. 10 is a schematic view of a branch region of a framework provided in accordance with an embodiment of the present invention;

FIG. 11 is a schematic view of a venous branch skeleton according to an embodiment of the present invention;

FIG. 12 is a schematic diagram of a coronary segmentation image provided in accordance with an embodiment of the present invention;

fig. 13 is a schematic structural diagram of a blood vessel image segmentation apparatus according to an embodiment of the present invention;

fig. 14 is a schematic structural diagram of an electronic device implementing the blood vessel image segmentation method according to the embodiment of the present invention.

Detailed Description

In order to make those skilled in the art better understand the technical solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, shall fall within the protection scope of the present invention.

It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein.

The names of messages or information exchanged between devices in the embodiments of the present disclosure are for illustrative purposes only, and are not intended to limit the scope of the messages or information.

It is understood that, before the technical solutions disclosed in the embodiments of the present disclosure are used, the user should be informed of the type, the use range, the use scene, etc. of the personal information related to the present disclosure in a proper manner according to the relevant laws and regulations and obtain the authorization of the user.

For example, in response to receiving an active request from a user, a prompt message is sent to the user to explicitly prompt the user that the requested operation to be performed would require the acquisition and use of personal information to the user. Thus, the user can autonomously select whether to provide personal information to software or hardware such as an electronic device, an application program, a server, or a storage medium that performs the operations of the technical solution of the present disclosure, according to the prompt information.

As an alternative but non-limiting implementation manner, in response to receiving an active request from the user, the manner of sending the prompt information to the user may be, for example, a pop-up window manner, and the prompt information may be presented in a text manner in the pop-up window. In addition, a selection control for providing personal information to the electronic device by the user's selection of "agreeing" or "disagreeing" can be carried in the popup.

It is understood that the above notification and user authorization process is only illustrative and is not intended to limit the implementation of the present disclosure, and other ways of satisfying the relevant laws and regulations may be applied to the implementation of the present disclosure.

It will be appreciated that the data referred to in this disclosure, including but not limited to the data itself, the acquisition or use of the data, should comply with the requirements of the applicable laws and regulations and related regulations.

Fig. 1 is a flowchart of a blood vessel image segmentation method according to an embodiment of the present invention, which is applicable to a case of accurately segmenting a coronary artery in a blood vessel.

In the prior art, a blood vessel image is segmented, and obtained coronary artery segmentation results are often mixed with vein branches interlaced with coronary arteries, so that the segmentation results are inaccurate. In view of the above technical problem, the present embodiment provides an image segmentation method, which obtains end points in a blood vessel by processing an image of the blood vessel to be segmented, determines candidate branch points based on the end points, and further determines branch points; and respectively identifying blood vessels between the branch points to obtain vein branches, and then optimizing the coronary artery segmentation result based on the identified vein branches to obtain a final coronary artery segmentation result, so that the accuracy of the coronary artery segmentation is improved.

The method can be executed by a blood vessel image segmentation device, the blood vessel image segmentation device can be realized in a hardware and/or software mode, and the blood vessel image segmentation device can be configured in an intelligent terminal or a cloud server. As shown in fig. 1, the method includes:

and S110, acquiring an initial blood vessel image, and respectively determining a coronary artery coarse segmentation image and a coronary artery skeleton image of the initial blood vessel image.

In the embodiment of the present invention, the initial blood vessel image may be understood as a blood vessel image obtained by scanning blood vessels around the heart. The blood vessel initial image comprises a vein image for collecting blood flowing back into the heart and a coronary image for transmitting the blood provided by the heart to the whole body. The coronary artery rough segmentation image is a coronary artery rough segmentation result of the initial blood vessel image, and the coronary artery rough segmentation result may contain a vein branch which is identified by mistake, so that the vein branch identification needs to be carried out on the coronary artery rough segmentation result to obtain an accurate coronary artery segmentation result. The coronary skeleton image may be understood as an image including a coronary skeleton line obtained by performing result thinning processing on the result of the coarse coronary segmentation, and the coronary skeleton line includes information of each branch in the result of the coarse coronary segmentation.

Specifically, the method for obtaining the initial blood vessel image may be to read image data in a local database or a database of a cloud server to obtain an original initial blood vessel image, or may also be to scan a heart portion of a scanned object based on a medical scanning device to obtain the initial blood vessel image, which is not limited in this embodiment. Alternatively, the medical scanning apparatus may be, but is not limited to, a CT (Positron Emission Tomography) -CT (Computed Tomography) apparatus, a bt (CT apparatus), and an MRI (Magnetic Resonance Imaging) apparatus. The corresponding obtained initial vessel image may include, but is not limited to, at least one of CT image data, PET image data, and MRI image data.

On the basis of obtaining the initial blood vessel image, the technical scheme of the embodiment can obtain a pre-trained coronary artery rough segmentation model, and input the initial blood vessel image into the coronary artery rough segmentation model to obtain a coronary artery rough segmentation image output by the coronary artery rough segmentation model. Optionally, in this embodiment, the coronary artery rough segmentation processing may be performed on the initial blood vessel image by using other existing segmentation technologies to obtain a coronary artery rough segmentation image, and the method for obtaining the coronary artery rough segmentation image is not limited in this embodiment. Referring to fig. 2, fig. 2 is an exemplary coarse coronary segmentation image after coarse coronary segmentation is performed on the initial blood vessel image.

Specifically, a pre-trained framework segmentation model is obtained, the initial blood vessel image is input into the framework segmentation model, and a coronary framework image which is output after the framework segmentation model carries out segmentation and thinning processing on the initial blood vessel image is obtained. Optionally, on the basis of obtaining the coronary artery rough segmentation image, a thinning processing method, such as an optimal path algorithm, may be further adopted to perform thinning processing on the coronary artery rough segmentation image, so as to obtain a coronary artery skeleton image. Referring to fig. 3, fig. 3 is a coronary skeleton image corresponding to the initial blood vessel image in the present embodiment after segmentation and refinement.

And S120, determining each skeleton point on a skeleton line in the coronary skeleton image, and determining a skeleton branch point in each skeleton point based on the degree of each skeleton point.

In the present embodiment, the coronary skeleton image includes skeleton lines representing coronary blood vessel skeleton information. Referring to fig. 3, for example, a skeleton line can be viewed as being composed of an infinite number of skeleton points. Specifically, the degree of each skeleton point in the skeleton line is determined, and the type of each skeleton point is determined according to the degree of each skeleton point. In this embodiment, the skeleton point types include, but are not limited to, skeleton branch points, skeleton center points, skeleton end points, and the like.

It should be explained that, for any node in the undirected graph, the number of other nodes associated with the node is referred to as the degree of the node. Based on this, the degree of the skeleton point in the present embodiment may be understood as the number of points having an association relationship between the current skeleton point and other skeleton points of the skeleton line where the current skeleton point is located.

Optionally, in this embodiment, the method for determining the skeleton branch point based on the degree of the skeleton point may include: for any skeleton point, determining each neighborhood point of the current skeleton point, and determining the degree of the current skeleton point based on the skeleton points contained in the neighborhood points; and acquiring a preset degree threshold value, and determining a skeleton branch point in each skeleton point based on the degree of the skeleton point and the preset degree threshold value.

The neighborhood points can be understood as neighboring points of the current skeleton point in each spatial direction. Specifically, whether the neighborhood point of the current skeleton point coincides with other skeleton points in the current skeleton line is judged, and the degree of the current skeleton point is determined based on the judgment result. Illustratively, for example, if 26 neighborhood points exist in a current skeleton point, if none of the 26 neighborhood points is overlapped with other skeleton points, the degree of the current skeleton point is 0; on the contrary, if the neighborhood points which are overlapped with other skeleton points exist in the 26 neighborhood points, the degree of the current skeleton point is determined based on the number of the overlapped neighborhood points.

In this embodiment, the degree threshold may be understood as a threshold used for dividing the type of the skeleton point. Different skeletal point types for the skeletal points may be determined based on different degree thresholds. Specifically, a degree threshold corresponding to the skeleton branch point is obtained, and whether the current skeleton point is the skeleton branch point is determined based on the degree threshold and the degree of the current skeleton point. Optionally, the degree of the current skeleton point is compared with a degree threshold, and if the degree of the current skeleton point is within the range of the degree threshold, the current skeleton point is determined to be a skeleton branch point on the skeleton line; and otherwise, determining the current skeleton point as other types of skeleton points in the skeleton line.

Illustratively, a degree threshold 3 corresponding to the skeleton branch point is obtained, and if the degree of the current skeleton point is determined to be 3, the current skeleton point can be determined to be the skeleton branch point based on the degree threshold; otherwise, if it is determined that the degree of the current skeleton point is 2, 4 or other numerical value, it may be determined that the current skeleton point is other types of skeleton points based on the degree threshold. Referring to fig. 4, the circle points represent skeleton branch points, and the rectangle points and the square points represent other types of skeleton points, such as skeleton end points and skeleton center points. The reason why the degree threshold corresponding to the skeleton branch point is set to 3 is that: if the degree of the skeleton point is 3, it indicates that 3 neighborhood points in the current skeleton point coincide with other skeleton points, that is, three neighboring points exist in the current skeleton point in the skeleton line, and based on this, it can be determined that the current skeleton point is located at a branch in the skeleton line, that is, it can be determined that the current skeleton point is a skeleton branch point.

It should be noted that, in practical applications, partial adhesion of the blood vessel segmentation result at the branch may inevitably occur, so that a plurality of branch points with neighborhood points greater than 3 may occur at the branch, and such branch points are defined as candidate branch points in the present embodiment. If a plurality of candidate branch points appear at one branch, the candidate branch points need to be screened to obtain a final skeleton branch point, and other candidate branch points are optimized to be a skeleton center point, so as to avoid the situation that a segmentation result appears a bulge.

Specifically, candidate branch points may be filtered based on skeleton endpoints among the skeleton points and skeleton communication paths between the endpoints to obtain skeleton branch points.

Optionally, the method for obtaining the framework branch point by screening the candidate branch points may be: determining a skeleton endpoint and a plurality of candidate branch points in the skeleton points based on the degree of each skeleton point, wherein the skeleton endpoint comprises a root endpoint and a plurality of tail branch endpoints; and determining the end point serial numbers of the end branch end points, respectively determining a skeleton communication path from the root end point to the end branch end points with any two adjacent serial numbers, and screening the candidate branch points based on the skeleton communication paths to obtain the skeleton branch points in the candidate branch point set.

Specifically, a degree threshold corresponding to a skeleton endpoint is preset to be 1, and a degree threshold corresponding to a candidate branch point is larger than 3; further, the type of the current skeleton point is determined based on the degree of the current skeleton point and the degree threshold corresponding to each type of skeleton point, and then skeleton endpoints and candidate branch points in each skeleton point can be determined. Referring to fig. 5, a plurality of circular points at the bifurcation in fig. 5 represent a plurality of candidate bifurcation points, and a square point represents a skeleton end point; of course, other types of skeleton points are also included in the graph, such as skeleton center points, etc. The reason why the degree threshold corresponding to the skeleton endpoint is set to 1 is that: if the degree of the skeleton point is 1, it is described that 1 neighborhood point in the current skeleton point coincides with other skeleton points, that is, 1 neighborhood point exists in the current skeleton point in the skeleton line, and based on this, the end point of the current skeleton point in the skeleton line can be determined to be out, that is, the current skeleton point can be determined as the skeleton end point.

In this embodiment, the root end point may be understood as a branch end point of a coronary branch branching from the aorta. The root endpoints comprise a root left endpoint and a root right endpoint. In practical application, two coronary branches are branched at the left side and the right side of the aorta respectively, and the branched regions are a left sinus region and a right sinus region respectively, so that a point of the left sinus region connected with the left side coronary branch is a root left end point, and a point of the right sinus region connected with the right side coronary branch is a root right end point. An end-branch point is understood to be the end-point of the branch end of each coronary branch.

Referring to fig. 6 exemplarily, the thickest blood vessel in the middle of fig. 6 is a divided aorta blood vessel, and two branches divided from the aorta are a left coronary branch and a right coronary branch respectively, wherein a point where the left branch is connected with the aorta is a root left end point, and a point where the right branch is connected with the aorta is a root right end point. With continued reference to fig. 6, the branch-to-end points of the branches of the coronary artery are the end-branch points.

Optionally, the method for determining a root endpoint and at least one end branch endpoint in the skeleton endpoints in this embodiment may include: determining the aorta position in the coronary artery coarse segmentation image, and determining a root endpoint and at least one tail branch endpoint in each skeleton endpoint based on the aorta position and the position of each skeleton endpoint.

Specifically, a pre-trained endpoint classification model may be obtained, and a root endpoint and a terminal branch endpoint in the skeleton endpoints are determined based on the endpoint positions of the skeleton endpoints and the artery position of the aorta and the endpoint classification model, optionally, the technical solution of this embodiment may also be that for each skeleton endpoint, whether each endpoint is connected with the left sinus or the right sinus in the aorta is determined, and if yes, the endpoint is determined as a root left endpoint or a root right endpoint; otherwise, if not, the terminal is determined as the terminal of the last branch. The embodiment may also adopt other methods to identify the endpoint, and the comparison is not limited.

Exemplary continuing reference to fig. 5, fig. 5 is a schematic diagram of the left coronary branch, where the upper-right-most square point in the diagram is the root endpoint and the other square points of the coronary branch in the diagram are the end-branch endpoints.

Specifically, the sequence number of each last branch endpoint is determined, for example, referring to fig. 5, fig. 5 includes 4 last branch endpoints, and each last branch endpoint is sequentially encoded to obtain the sequence number of each last branch code. Further, regarding the end point 3 and the end point 4, which are two adjacent end points, a skeleton communicating path between the root left end point and the end point 3 and a skeleton communicating path between the root left end point and the end point 4 are respectively determined, and a skeleton branch point in candidate branch points at branches in the paths is determined based on the determined skeleton communicating paths.

Alternatively, the method for determining the skeleton branch point in the candidate branch points at the branches in the path may be: determining candidate branch points which are traversed repeatedly in each framework communication path, and respectively determining the framework distance between each candidate branch point and the root end point; and determining the skeleton branch points in the candidate branch points based on the comparison result of the skeleton distances.

Specifically, for the skeleton communicating paths corresponding to the two end branch end points respectively, each candidate branch point at the branch position in the path is determined, further, the candidate branch point traversed repeatedly by the two skeleton communicating paths is determined, and the skeleton distance from the candidate branch point to the root end point is determined. Optionally, if the traversal is repeated for a plurality of candidate branch points, the candidate branch point corresponding to the minimum skeleton distance in each skeleton distance is determined as the skeleton branch point. Referring to FIG. 7, the candidate branch points pointed by arrows in FIG. 7 are the screened skeleton branch points. Optionally, based on the foregoing manner, candidate branch points included in each skeleton point are screened to obtain a skeleton branch point, and other candidate branch points are optimized to be a skeleton center point.

And S130, identifying a vein branch image in the initial blood vessel image based on the branch point position of the skeleton branch point in the coronary artery coarse segmentation image/coronary artery skeleton image.

In an embodiment of the present invention, upon determining each of the branch points in the skeleton point, the vein branch is identified based on the location of the skeleton branch point. Optionally, the identification may be performed based on a branch point position of the skeleton branch point in the coronary artery rough segmentation image, or may be performed based on a branch point position of the skeleton branch point in the coronary artery skeleton image, so as to obtain the vein branch image.

Optionally, identifying a vein branch image in the initial blood vessel image based on a branch point position of the skeleton branch point in the coronary artery rough segmentation image includes: determining a branch point segmentation position of a skeleton branch point in the coronary artery rough segmentation image, and determining a coronary artery branch region in the coronary artery rough segmentation image based on the branch point segmentation position; obtaining a pre-trained vein branch recognition model, and carrying out vein branch recognition on a coronary artery branch region based on the vein branch recognition model to obtain a vein branch image.

The branch point segmentation position can be understood as a segmentation center point position of the skeleton branch point in the coronary artery rough segmentation image. The coronary artery branch region can be understood as a region of interest which takes a skeleton branch point as the center and contains preset pixel points in a coronary artery rough segmentation image.

Specifically, the position of a branch point of the skeleton branch point in the coronary skeleton line is determined, and the branch point division position of the skeleton branch point in the coronary roughly-divided image is determined based on the correspondence between the coronary skeleton line and the coronary roughly-divided image and the determined position of the branch point. And then taking the branch point dividing position as a central point, acquiring a region of interest with a preset size, and taking the region of interest as a coronary artery branch region. Furthermore, a pre-trained vein branch recognition model is obtained, the coronary artery branch region is input into the vein branch recognition model, vein branch recognition is carried out on the coronary artery branch region based on the vein branch recognition model, and a vein branch image output by the model is obtained.

Referring to fig. 8, fig. 8 illustrates a coronary branch region determined based on a branch point. As can be seen from the figure, the coronary branch region includes a plurality of blood vessel branches bounded by branch points, and each blood vessel branch may be a vein branch, may be a coronary branch, or includes both a coronary branch and a vein branch. Therefore, vein recognition needs to be performed on each blood vessel branch in the coronary branch region, so as to further process the coronary coarse segmentation image according to the vein recognition result, thereby obtaining an accurate coronary segmentation result.

Specifically, the coronary artery branch region in fig. 8 is input into a pre-trained vein branch recognition model, and then the model recognizes each blood vessel branch to obtain a vein branch recognition result of the current coronary artery branch region. Furthermore, a vein branch image corresponding to the coronary artery rough segmentation image is obtained based on the recognition result of each coronary artery branch region. Illustratively, a vein branch image is shown in fig. 9.

Optionally, in this embodiment, the method for identifying a vein branch image in an initial blood vessel image based on a branch point position of a skeleton branch point in a coronary artery skeleton image may include: determining a branch point skeleton position of a skeleton branch point in the coronary skeleton image, and determining a skeleton branch region in the coronary skeleton image based on the branch point skeleton position; acquiring a pre-trained vein branch skeleton recognition model, and performing vein branch skeleton recognition on a skeleton branch region based on the vein branch skeleton recognition model to obtain a vein branch skeleton; and acquiring branch expansion parameters of the vein branch skeleton, and performing expansion processing on the vein branch skeleton based on the branch expansion parameters to obtain a vein branch image corresponding to the vein branch skeleton.

Here, the branch point skeleton position may be understood as a position of the skeleton branch point in the coronary skeleton line. The skeleton branch region can be understood as a region of interest containing preset pixel points and centered at a skeleton branch point in the coronary artery skeleton image.

Specifically, the position of a branch point of a skeleton branch point in a coronary artery skeleton line is determined, and then a region of interest with a preset size is obtained by taking the division position of the branch point as a central point, and the region of interest is taken as a skeleton branch region. Exemplary skeletal branching regions may be shown in fig. 10. Referring to fig. 10, the skeleton branch region includes a plurality of skeleton branches bounded by branch points, and each skeleton branch may be a vein branch, a coronary branch, or both a coronary branch and a vein branch. Therefore, vein recognition needs to be performed on each skeleton branch in the skeleton branch region, so as to further process the coronary artery rough segmentation image according to the vein recognition result, thereby obtaining an accurate coronary artery segmentation result.

Furthermore, a pre-trained vein branch recognition model is obtained, the skeleton branch region is input into the vein branch recognition model, vein branch recognition is carried out on the skeleton branch region based on the vein branch recognition model, and a vein branch recognition result of the current skeleton branch region is obtained. Further, a vein branch skeleton in the coronary skeleton line is obtained based on the recognition result of each skeleton branch region. Illustratively, the identified vein branch skeleton is shown in fig. 11.

Furthermore, branch expansion is carried out on the vein branch skeleton based on the branch expansion parameters, and a vein branch image is obtained.

In this embodiment, the branch expansion parameter may be understood as an expansion diameter parameter of the vein branch skeleton that needs to be expanded. The branch extension parameters may include an endpoint extension parameter and a center extension parameter of the venous branch skeleton.

In this embodiment, the vein branch image obtained by performing the expansion processing on the expansion parameter based on the expansion parameter by the vein branch skeleton needs to completely cover the vein branch in the coronary artery rough segmentation image. In this case, vein removal processing can be performed on the vein branch in the coronary artery rough segmentation image based on the vein branch image. Therefore, branch expansion parameters for expanding the vein branch skeleton need to be determined according to the rough segmentation result in the coronary artery rough segmentation image.

Specifically, the initial blood vessel image may be subjected to coronary artery rough segmentation by using the existing segmentation technology to obtain a coronary artery rough segmentation image, or the initial blood vessel image may be subjected to segmentation processing by using a conventional image processing algorithm to obtain a coronary artery rough segmentation image, which is not limited in the present embodiment.

Further, on the basis of determining each vein branch skeleton, the vein branch skeleton is subjected to expansion processing to obtain a vein branch image. Optionally, the method for only expanding processing in this embodiment may include: for any end point of any vein branch skeleton, the end point expansion parameter is determined based on the line segment end point parameter of the vein branch skeleton and the vessel parameters of other vessels connected with the vein branch in the coronary rough segmentation image. Furthermore, on the basis of determining endpoint extension parameters of two segment endpoints of the vein branch skeleton, fitting is performed on the extension parameters of a middle point of the vein branch skeleton based on the two endpoint extension parameters to determine at least one central point extension parameter of the vein branch skeleton, and then extension processing is performed on the vein branch skeleton by taking each endpoint extension parameter and each central extension parameter as extension targets to obtain a vein branch image of the vein branch skeleton. Alternatively, all vein branch skeletons are traversed based on the above embodiment, and a vein branch image as shown in fig. 9 is obtained.

And S140, determining a coronary artery segmentation image based on the coronary artery rough segmentation image and the vein branch image.

In the embodiment of the invention, on the basis of determining the vein branch image and the coronary artery rough segmentation image, vein removal processing is carried out on the vein branch in the coronary artery rough segmentation image based on the vein branch image, so as to obtain the processed coronary artery segmentation image.

Specifically, a branch region overlapping with the vein branch image in the coronary artery rough segmentation image may be determined, and the branch region may be removed from the coronary artery segmentation image, and the extracted image may be the coronary artery segmentation image from which the vein branch is removed.

For example, after obtaining the vein branch image shown in fig. 9, the coronary artery rough segmentation image shown in fig. 2 is subjected to vein branch removal processing based on the vein branch image, and the coronary artery segmentation image shown in fig. 12 is obtained.

According to the technical scheme provided by the embodiment of the invention, an initial blood vessel image is obtained, and a coronary artery rough segmentation image and a coronary artery skeleton image of the initial blood vessel image are respectively determined; determining each skeleton point on a skeleton line in the coronary artery skeleton image, and determining a skeleton branch point in each skeleton point based on the degree of each skeleton point; identifying a vein branch image in the initial blood vessel image based on branch point positions of the skeleton branch points in the coronary artery rough segmentation image/coronary artery skeleton image; determining a coronary artery segmentation image based on the coronary artery rough segmentation image and the vein branch image. According to the technical scheme, the blood vessel image to be segmented is processed to obtain each end point in the blood vessel, candidate branch points are determined based on the end points, and each branch point is further determined; and respectively identifying blood vessels between the branch points to obtain vein branches, and then optimizing the coronary artery segmentation result based on the identified vein branches to obtain a final coronary artery segmentation result, so that the accuracy of the coronary artery segmentation is improved.

Fig. 13 is a schematic structural diagram of a blood vessel image segmentation apparatus according to an embodiment of the present invention. As shown in fig. 13, the apparatus includes: an image determination module 210, a skeleton branch point determination module 220, a vein branch image determination module 230, and a coronary segmentation image determination module 240; wherein the content of the first and second substances,

the image determining module 210 is configured to obtain an initial blood vessel image, and respectively determine a coronary artery coarse segmentation image and a coronary artery skeleton image of the initial blood vessel image;

a skeleton branch point determining module 220, configured to determine skeleton points on a skeleton line in the coronary artery skeleton image, and determine a skeleton branch point in each skeleton point based on a degree of each skeleton point;

a vein branch image determining module 230, configured to identify a vein branch image in an initial blood vessel image based on a branch point position of the skeleton branch point in the coronary coarse segmentation image/the coronary skeleton image;

a coronary segmentation image determination module 240 for determining a coronary segmentation image of the initial blood vessel image based on the coarse coronary segmentation image and the vein branch image.

On the basis of the foregoing embodiment, optionally, the image determining module 210 includes:

a coronary artery rough segmentation image obtaining unit, configured to input the initial blood vessel image to a coronary artery rough segmentation model trained in advance, and obtain a coronary artery rough segmentation image output by the coronary artery rough segmentation model;

and the coronary artery skeleton image obtaining unit is used for inputting the initial blood vessel image into a skeleton segmentation model trained in advance to obtain a coronary artery skeleton image output by the skeleton segmentation model.

On the basis of the foregoing embodiment, optionally, the skeleton branch point determining module 220 includes:

the framework point degree determining unit is used for determining each neighborhood point of the current framework point for any framework point and determining the degree of the current framework point based on the framework points contained in the neighborhood points;

and the skeleton point type determining unit is used for acquiring a preset degree threshold value and determining a skeleton branch point in each skeleton point based on the degree of the skeleton point and the preset degree threshold value.

an end point and candidate branch point determining unit, configured to determine a skeleton end point and a plurality of candidate branch points in the skeleton points based on a degree of each skeleton point, where the skeleton end point includes a root end point and a plurality of end branch end points;

and the skeleton branch point determining unit is used for determining the end point number of each tail end point, respectively determining a skeleton communicating path between the root end point and any two tail end points with adjacent numbers, and screening candidate branch points based on the skeleton communicating paths to obtain skeleton branch points in the candidate branch point set.

On the basis of the foregoing embodiment, optionally, the skeleton branch point determining unit includes:

the skeleton distance determining subunit is configured to determine candidate branch points that are repeatedly traversed in each skeleton communication path, and determine skeleton distances between the candidate branch points and root end points respectively;

and the skeleton branch point determining unit is used for determining a skeleton branch point in the candidate branch points based on the comparison result of each skeleton distance.

On the basis of the foregoing embodiment, optionally, the vein branch image determining module 230 includes:

a coronary branch region determination unit configured to determine a branch point division position of the skeleton branch point in the coronary artery rough-divided image, and determine a coronary branch region in the coronary artery rough-divided image based on the branch point division position;

and the first vein branch image obtaining unit is used for obtaining a pre-trained vein branch recognition model, and performing vein branch recognition on the coronary artery branch region based on the vein branch recognition model to obtain a vein branch image.

a skeleton branch region determination unit configured to determine a branch point skeleton position of the skeleton branch point in the coronary skeleton image, and determine a skeleton branch region in the coronary skeleton image based on the branch point skeleton position;

the vein branch framework determining unit is used for acquiring a pre-trained vein branch framework recognition model and carrying out vein branch framework recognition on the framework branch region based on the vein branch framework recognition model to obtain a vein branch framework;

and the second vein branch image obtaining unit is used for obtaining branch expansion parameters of the vein branch skeleton, and performing expansion processing on the vein branch skeleton based on the branch expansion parameters to obtain a vein branch image corresponding to the vein branch skeleton.

The blood vessel image segmentation device provided by the embodiment of the invention can execute the blood vessel image segmentation method provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method.

FIG. 14 illustrates a schematic structural diagram of an electronic device 10 that may be used to implement an embodiment of the present invention. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the inventions described and/or claimed herein.

As shown in fig. 14, the electronic device 10 includes at least one processor 11, and a memory communicatively connected to the at least one processor 11, such as a Read Only Memory (ROM) 12, a Random Access Memory (RAM) 13, and the like, wherein the memory stores a computer program executable by the at least one processor, and the processor 11 may perform various suitable actions and processes according to the computer program stored in the Read Only Memory (ROM) 12 or the computer program loaded from a storage unit 18 into the Random Access Memory (RAM) 13. In the RAM 13, various programs and data necessary for the operation of the electronic apparatus 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other via a bus 14. An input/output (I/O) interface 15 is also connected to bus 14.

A number of components in the electronic device 10 are connected to the I/O interface 15, including: an input unit 16 such as a keyboard, a mouse, or the like; an output unit 17 such as various types of displays, speakers, and the like; a storage unit 18 such as a magnetic disk, optical disk, or the like; and a communication unit 19 such as a network card, modem, wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information/data with other devices via a computer network such as the internet and/or various telecommunication networks.

The processor 11 may be a variety of general and/or special purpose processing components having 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 dedicated Artificial Intelligence (AI) computing chips, various processors running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, and so forth. The processor 11 performs the various methods and processes described above, such as a blood vessel image segmentation method.

In some embodiments, the blood vessel image segmentation method may be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and/or installed onto the electronic device 10 via the ROM 12 and/or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the above described vessel image segmentation method may be performed. Alternatively, in other embodiments, the processor 11 may be configured to perform the blood vessel image segmentation method by any other suitable means (e.g. by means of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuitry, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), system on a chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, receiving data and instructions from, and transmitting data and instructions to, a storage system, at least one input device, and at least one output device.

A computer program for implementing 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, special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions/acts specified in the flowchart and/or block diagram block or blocks to be performed. A computer program can execute entirely on a machine, partly on a machine, as a stand-alone software package partly on a machine and partly on a remote machine or entirely on a remote machine or server.

In the context of the present invention, a computer-readable storage medium may be a tangible medium that can contain, or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. A computer readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, the computer readable storage medium may be a machine readable signal medium. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here 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 a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which a user can provide input to the electronic device. Other kinds of devices may also be used to provide for interaction with a 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 may be received in any form, including acoustic, speech, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), blockchain networks, and the internet.

The computing system may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical host and VPS service are overcome.

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

The above-described embodiments should not be construed as limiting the scope of the invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made, depending on design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A blood vessel image segmentation method is characterized by comprising the following steps:

2. The method according to claim 1, wherein the determining the coronary coarse segmentation image and the coronary skeleton image of the initial blood vessel image respectively comprises:

3. The method of claim 1, wherein determining a skeleton branch point in each of the skeleton points based on the degree of each of the skeleton points comprises:

4. The method of claim 1 or 3, wherein determining a skeleton branch point in each of the skeleton points based on the degree of each of the skeleton points further comprises:

5. The method of claim 4, wherein screening at least one candidate branch point based on the skeleton communication path comprises:

determining candidate branch points which are traversed repeatedly in each framework communication path, and respectively determining the framework distance between each candidate branch point and a root endpoint;

and determining a skeleton branch point in the candidate branch points based on the comparison result of the skeleton distances.

6. The method of claim 1, wherein identifying a vein branch image in an initial vessel image based on branch point locations of the skeletal branch points in the coronary coarse segmentation image comprises:

7. The method of claim 1, wherein identifying a vein branch image in an initial blood vessel image based on branch point locations of the skeletal branch points in the coronary skeleton image comprises:

8. A blood vessel image segmentation apparatus, comprising:

the image determining module is used for acquiring an initial blood vessel image and respectively determining a coronary artery rough segmentation image and a coronary artery skeleton image of the initial blood vessel image;

a skeleton branch point determining module, configured to determine skeleton points on a skeleton line in the coronary skeleton image, determine skeleton end points and a candidate branch point set in the skeleton points based on degrees of the skeleton points, and determine skeleton branch points in the candidate branch point set based on the skeleton end points;

a coronary artery segmentation image determination module for determining a coronary artery segmentation image of the initial blood vessel image based on the coronary artery rough segmentation image and the vein branch image.

9. An electronic device, characterized in that the electronic device comprises:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,

the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the vessel image segmentation method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing computer instructions for causing a processor to perform the vessel image segmentation method according to any one of claims 1 to 7 when executed.