CN115984239A - Method, system, device and storage medium for extracting central line of cerebral artery blood vessel - Google Patents
Method, system, device and storage medium for extracting central line of cerebral artery blood vessel Download PDFInfo
- Publication number
- CN115984239A CN115984239A CN202310064732.6A CN202310064732A CN115984239A CN 115984239 A CN115984239 A CN 115984239A CN 202310064732 A CN202310064732 A CN 202310064732A CN 115984239 A CN115984239 A CN 115984239A
- Authority
- CN
- China
- Prior art keywords
- image
- cerebral artery
- extracting
- vessel
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Abstract
The invention discloses a method for extracting a central line of a cerebral artery blood vessel, which comprises the following steps: acquiring a target mask image of cerebral artery blood vessels; thinning the target mask image, and preliminarily extracting the central line of the cerebral artery blood vessel to obtain a thinned image; determining a starting point of the internal carotid artery according to the thinned image, and constructing a connected graph according to the thinned image; calculating all minimum cost paths as target center lines according to the initial points of the internal carotid arteries and the connectivity graph; all target centerlines are output. The method can automatically and quickly extract the accurate center line close to the center, and simultaneously increases the robustness of the algorithm for extracting the center line. The invention also discloses a system, equipment and a storage medium for extracting the central line of the cerebral artery blood vessel.
Description
Technical Field
The invention relates to the field of medical image processing, in particular to a method, a system, equipment and a storage medium for extracting a central line of a cerebral artery blood vessel.
Background
With the development and progress of medical Imaging technology, the types of medical images are increasing, and common medical images include X-ray (i.e. X-ray, which is an electromagnetic wave with extremely high frequency, extremely short wavelength and large energy), DSA (short for Digital subtraction angiography), CT (short for Computed Tomography), MRI (short for Magnetic Resonance Imaging), and the like.
Generally, it is difficult for doctors to give objective diagnosis and treatment schemes to the disease condition by judging medical images only through clinical experience and vision; with the development of computer technology, it is becoming more and more the mainstream trend to perform efficient and accurate analysis on medical images by using a computer. For the diagnosis of cerebral artery vascular lesions and the formulation of treatment schemes, quantitative analysis results need to be given clinically, and treatment schemes are planned according to the results, such as whether cerebral artery blood vessels have tumor bodies, the specific positions of the tumor bodies, the lesion degree, the rupture risk and the like, and the premise of quantitative analysis is to extract accurate blood vessel center lines. The extracted cerebral artery vessel center line can be used for diagnosing vessel diseases, calculating lesion positions, lesion degrees and the like, and can also provide tracks, poses and the like of intervention equipment for surgical navigation planning so as to plan a surgical path.
Common blood vessel centerline extraction schemes include a manual calibration method, a topology refinement method and a distance transformation method. The manual calibration method requires an operator to specify a center point according to experience and connect the points into a center line, so that the center line of the two-dimensional image can be accurately extracted, but the operation difficulty for the three-dimensional image is very high, so that the manual calibration method has the problems of low efficiency and dependence on the experience of doctors. The topology refinement method is a method for deleting boundary points from outside to inside to obtain a center line, and is usually realized through template iteration, and the topology refinement method can keep the topological structure of a blood vessel, but has a large complex calculation amount. The distance transformation method is to convert a binary image into a gray image, the gray value is the distance between the pixel point and the boundary point of the image, the method is high in accuracy, but the center point is not continuous.
The existing method for extracting the centerline of the cerebral artery vessel has the problems of dependence on artificial experience, long calculation time consumption, low operation efficiency, discontinuous extracted centerline and insufficient approach to the center.
The invention aims to solve the problems that the existing method for extracting the centerline of the cerebral artery vessel depends on artificial experience, has long calculation time consumption, low operation efficiency, discontinuous extracted centerline and is not close to the center.
On the first hand, the method for extracting the centerline of the cerebral artery vessel automatically and quickly extracts the centerline which is accurate and close to the center, and meanwhile, the robustness of the centerline extraction algorithm is improved.
In order to solve the technical problem, the embodiment of the invention discloses a method for extracting a centerline of a cerebral artery vessel, which comprises the following steps: acquiring a target mask image of cerebral artery blood vessels; thinning the target mask image, and preliminarily extracting the central line of the cerebral artery blood vessel to obtain a thinned image; determining a starting point of the internal carotid artery according to the thinned image, and constructing a connected graph according to the thinned image; calculating all minimum cost paths as target center lines according to the initial points of the internal carotid arteries and the connectivity graph; all target centerlines are output.
By adopting the technical scheme, the central line of the cerebral artery blood vessel is extracted fully automatically, the calculated amount is reduced, the calculation speed is improved, the initial point of the internal carotid artery is determined according to the thinned image, and the connected graph is constructed according to the thinned image, so that the subsequent extraction of the central line path is facilitated, the calculation speed is improved, and the central line of the extracted cerebral artery blood vessel is ensured to be closer to the center.
According to another embodiment of the present invention, determining a starting point of an internal carotid artery from a refined image and constructing a connectivity map from the refined image comprises: determining all end points through 26 neighborhood analysis according to the refined image; and determining the starting point of the internal carotid artery according to all the terminal points, wherein the starting point of the internal carotid artery is the terminal point farthest from the centers of all the terminal points.
According to another embodiment of the present invention, the starting point of the internal carotid artery is determined from all the end points by the formula:
wherein Max { } represents taking the maximum value, | | represents the Euclidean distance, ei represents the ith terminal point, i is a positive integer,
represents the average of all end points.
According to another embodiment of the present invention, determining a starting point of an internal carotid artery from a refined image and constructing a connectivity map from the refined image comprises: constructing a weighted connected graph by adopting a distance transformation algorithm according to the refined image; the farther the distance from the edge of the refined image is, the larger the weight of the connected graph is, and the smaller the weight is; the closer the distance to the edge of the refined image, the smaller the weight of the connected graph and the larger the weight.
According to another embodiment of the present invention, the weight calculation formula of the connection graph is as follows:
W i ＝-W i
wherein W represents weight, the actual value is distance transformation gray scale map value DF (i), i is the position of all mask pixel values in the refined image, W i And representing the weight value of the ith pixel point.
According to another embodiment of the present invention, calculating all minimum cost paths as target centerlines from the starting points of internal carotid arteries and the connectivity map comprises: calculating a minimum cost path between the starting point of the internal carotid artery and each end point by using a Dijkstra algorithm as a target central line; and obtaining all target central lines until the calculation of the minimum cost path between the initial point and all the end points of the internal carotid artery is completed.
According to another embodiment of the present invention, before acquiring the target mask image of the cerebral artery vessel, the method further comprises: acquiring an initial mask image of a cerebral artery blood vessel three-dimensional image; extracting a region of interest in the initial mask image; hole filling is performed on the cerebrovascular image in the region of interest to form a target mask image of the cerebral artery vessels.
According to another embodiment of the present invention, an initial mask image of a three-dimensional image of a cerebral artery vessel is obtained, comprising: inputting a three-dimensional image of a cerebral artery blood vessel; acquiring initial mask images of cerebral aneurysm blood vessels and peripheral blood vessels through a segmentation algorithm according to the three-dimensional images of the cerebral artery blood vessels; the initial position of the initial mask image is the starting point of the internal carotid artery at the proximal end of the parent artery, the ending position is a preset multiple of the diameter of the parent artery at the distal end of the parent artery, which is not shorter than the parent artery at the position of the parent body, the diameter of the cerebral artery blood vessel branch in the initial mask image is larger than a first preset value, and the length of the cerebral artery blood vessel branch is not smaller than a second preset value.
According to another embodiment of the invention, the location of the parent artery is at the beginning of the internal carotid artery.
According to another embodiment of the present invention, extracting a region of interest in an initial mask image comprises: extracting a starting point and an end point of the region of interest; the starting point of the interested region is a point which is a mask pixel value in the first three-dimensional direction in the initial mask image, and the ending point is a point which is a non-mask pixel value in the first three-dimensional direction.
According to another embodiment of the present invention, the thinning processing is performed on the target mask image, and the preliminary extraction of the centerline of the cerebral artery blood vessel to obtain a thinned image includes: thinning the target mask image by adopting a distance transformation algorithm to obtain a distance transformation gray level image; performing a point deletion operation by 26 neighborhood analysis according to the distance transformation gray-scale map, comprising: judging whether each pixel point is communicated with 26 neighborhoods: if yes, the pixel point is reserved; if not, deleting the pixel point.
In a second aspect, an embodiment of the present invention discloses a system for extracting a centerline of a cerebral artery vessel, including: the data receiving module is used for receiving a three-dimensional image of a cerebral artery vessel input by a user; the data processing module is connected with the data receiving module and used for acquiring a target mask image of the cerebral artery blood vessel; thinning the target mask image, and preliminarily extracting the central line of the cerebral artery blood vessel to obtain a thinned image; determining a starting point of the internal carotid artery according to the thinned image, and constructing a connected graph according to the thinned image; calculating all minimum cost paths as target center lines according to the initial points of the internal carotid arteries and the connectivity graph; and the data output module is connected with the data processing module and is used for outputting all the target center lines.
By adopting the technical scheme, under the synergistic action of the data receiving module, the data processing module and the data output module, the center line of the cerebral artery blood vessel is fully automatically extracted through the center line extracting system of the cerebral artery blood vessel, so that the calculated amount is reduced, the calculating speed is improved, and the extracted center line of the cerebral artery blood vessel is closer to the center.
In a third aspect, an embodiment of the present invention discloses an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor, when executing the computer program, implements the method for extracting a centerline of a cerebral artery vessel in any embodiment of the first aspect.
By adopting the technical scheme, the electronic equipment automatically and quickly extracts the central line of the cerebral artery blood vessel which is accurate and close to the center, the efficiency is improved, and the time is saved.
In a fourth aspect, an embodiment of the present invention discloses a computer-readable storage medium, where a computer program is stored, and when the computer program is executed by a processor, the method for extracting a centerline of a cerebral artery vessel in any embodiment of the first aspect is implemented.
Drawings
Fig. 1 shows a first flowchart of a method for extracting a centerline of a cerebral artery vessel in an embodiment of the invention;
FIG. 2 is a flow chart of a method for extracting a centerline of a cerebral artery vessel according to an embodiment of the present invention;
FIG. 3 is a flow chart III of a method for extracting the centerline of a cerebral artery vessel according to an embodiment of the present invention;
FIG. 4 is a fourth flowchart of a centerline extraction method for cerebral artery vessels according to an embodiment of the present invention;
FIG. 5 is a flow chart of a fifth method for extracting a centerline of a cerebral artery vessel according to an embodiment of the present invention;
fig. 6 shows a sixth flowchart of a centerline extraction method for cerebral artery vessels in an embodiment of the present invention;
fig. 7 shows a seventh flowchart of a centerline extraction method of a cerebral artery vessel in an embodiment of the present invention;
fig. 8 shows a schematic diagram of a region of interest of a cerebral artery vessel extracted in an embodiment of the present invention;
FIG. 9 is a schematic diagram illustrating holes in a mask image of cerebral artery vessels in an embodiment of the present invention;
FIG. 10 is a schematic diagram showing the structure of a centerline extraction system for cerebral artery vessels in an embodiment of the present invention;
fig. 11 is a schematic structural diagram of an electronic device for extracting the centerline of a cerebral artery vessel in the embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention is provided for illustrative purposes, and other advantages and effects of the present invention will become apparent to those skilled in the art from the present disclosure. While the invention will be described in conjunction with the preferred embodiments, it is not intended that the features of the invention be limited to that embodiment. On the contrary, the invention has been described in connection with the embodiments for the purpose of covering alternatives or modifications as may be extended based on the claims of the invention. In the following description, numerous specific details are included to provide a thorough understanding of the invention. The invention may be practiced without these particulars. Moreover, some of the specific details have been left out of the description in order to avoid obscuring or obscuring the focus of the present invention. It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict.
It should be noted that in this specification, like reference numerals and letters refer to like items in the following drawings, and thus, once an item is defined in one drawing, it need not be further defined and explained in subsequent drawings.
The terms "first," "second," and the like are used solely to distinguish one from another and are not to be construed as indicating or implying relative importance.
In the description of the present embodiment, it should be further noted that, unless explicitly stated or limited otherwise, the terms "disposed," "connected" and "connected" should be interpreted broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. Specific meanings of the above terms in the present embodiment can be understood as specific cases by those of ordinary skill in the art.
To make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
In a first aspect, referring to fig. 1, an embodiment of the present invention provides a method for extracting a centerline of a cerebral artery blood vessel, where the method includes:
s1: and acquiring a target mask image of the cerebral artery blood vessel.
Because the target mask image of the cerebral artery blood vessel only occupies a part of the initial mask image of the cerebral artery blood vessel, namely 1/4 to 1/3, the calculated amount in the subsequent processing flow is greatly reduced, and the efficiency of extracting the central line of the cerebral artery blood vessel is improved; the noise of the target mask image is removed, so that the extracted central line of the cerebral artery blood vessel is more accurate.
S2: and thinning the target mask image, and preliminarily extracting the central line of the cerebral artery blood vessel to obtain a thinned image.
The thinning processing ensures that the center line of the cerebral artery blood vessel which is extracted preliminarily is a continuous center line and ensures the connectivity of the thinned image.
S3: and determining a starting point of the internal carotid artery according to the refined image, and constructing a connected graph according to the refined image.
The starting point of the internal carotid artery can be determined automatically according to the refined image without any interaction. On the other hand, compared with the prior art, the method needs human-computer interaction and determines the starting point of the internal carotid artery after traversing the terminal point of the thinned image, and the method determines the starting point of the internal carotid artery fully automatically according to the thinned image, so that the calculation amount is reduced, and the calculation time is saved.
And constructing a connected graph according to the refined image, so that the subsequent extraction of the central line path of the cerebral artery blood vessel is facilitated, the centrality of the central line path of the cerebral artery blood vessel can be ensured, and the extracted central line of the cerebral artery blood vessel is more centered.
S4: and calculating all minimum cost paths as target central lines according to the initial points of the internal carotid arteries and the connectivity graph.
And calculating a refined image based on the connectivity and a weighted connectivity graph based on a distance transformation algorithm by using the minimum cost path, and ensuring the centrality of the central line of the extracted cerebral artery blood vessel on the basis of maintaining the connectivity.
S5: all target centerlines are output.
By adopting the technical scheme, in the step S2, the target mask image of the cerebral artery blood vessel is directly refined, the calculated amount is reduced, the calculation speed is improved, the robustness of the algorithm is increased, and the central line of the cerebral artery blood vessel is extracted as a continuous central line after the target mask image is refined. And step S3: the starting point of the internal carotid artery is determined according to the thinned image, and the connected graph is constructed according to the thinned image, so that the subsequent extraction of the central line path is facilitated, the calculation speed is improved, and the central line of the extracted cerebral artery blood vessel is ensured to be closer to the center. In the step S4, a target central line is extracted by adopting a minimum cost path calculation method in combination with the weighted refined graph, and the accuracy of the central line is improved while the calculation speed is improved.
The method for extracting the central line of the cerebral artery blood vessel is used for optimizing the extraction of the blood vessel central line of cerebral aneurysm planning software, can determine the specific position of a tumor body according to the blood vessel central line aiming at the cerebral aneurysm planning software, calculates the lesion degree of the tumor body, estimates the rupture risk and can plan an interventional operation path.
In some possible embodiments provided by the present invention, referring to fig. 2 in combination with fig. 1, S1: before acquiring the target mask image of the cerebral artery blood vessel, the method further comprises the following steps:
s01: and acquiring an initial mask image of the cerebral artery blood vessel three-dimensional image.
The initial mask image is an image of the cerebral artery blood vessel from which redundant bifurcation blood vessels are removed, so that the subsequent calculation amount is reduced conveniently, and the calculation efficiency is improved.
S02: extracting a region of interest in the initial mask image.
Usually, the region of interest of the cerebral aneurysm only occupies 1/4 to 1/3 of the initial mask image, and after the region of interest is extracted, the calculation amount of a subsequent processing flow can be greatly reduced, and the calculation efficiency is improved.
S03: hole filling is performed on the cerebrovascular image in the region of interest to form a target mask image of the cerebral artery vessels.
In the region of interest of the extracted initial mask image, a hole appears at the partial filling inside the cerebral artery blood vessel, that is, in the region of interest, a hole (such as the hole at B shown in fig. 9) appears in the cerebral artery blood vessel, and the existence of the hole results in that the center line of the cerebral artery blood vessel cannot be accurately extracted. Specifically, according to the definition of the center line, the extracted center line usually keeps the original structure of the image, so in the region of interest, in order to keep the structure of the hole, the center line near the extracted hole is a circle surrounding the hole, and the circle is not the original topology of the center line of the cerebral artery blood vessel, thereby causing the error of the extracted center line. In step S03, the hole filling is carried out on the cerebrovascular image in the region of interest, so that preparation is made for extracting accurate central lines of cerebral artery blood vessels subsequently.
Illustratively, the hole filling algorithm is a flood filling method. In other possible embodiments, the invention does not limit the hole filling algorithm.
In some possible embodiments provided by the present invention, referring to fig. 3 in conjunction with fig. 2, S01: acquiring an initial mask image of a cerebral artery blood vessel three-dimensional image, wherein the initial mask image comprises:
s011: inputting a three-dimensional image of a cerebral artery vessel.
S012: and acquiring initial mask images of the cerebral aneurysm blood vessel and peripheral blood vessels by a segmentation algorithm according to the three-dimensional image of the cerebral artery blood vessel.
In step S012, redundant bifurcated vessels are removed from the three-dimensional image of cerebral artery vessels by a segmentation algorithm, and an initial mask image is obtained, so that the subsequent calculation amount is reduced, and the calculation efficiency is improved.
In this embodiment, the initial mask image has a starting position of the proximal internal carotid artery of the parent artery, a stopping position of the proximal internal carotid artery of the parent artery, a diameter of the distal end of the parent artery, which is not less than a preset multiple of the diameter of the parent artery at the position of the parent body, a diameter of the cerebral artery blood vessel branch in the initial mask image is greater than a first preset value, and a length of the cerebral artery blood vessel branch is not less than a second preset value. Illustratively, the preset multiple is 3 times, or the preset multiple is 5 times. Illustratively, the first preset value may range from 1mm to 3mm, for example, the first preset value is 1mm, i.e. the diameter of the cerebral artery vessel branch in the initial mask image is greater than 1mm. Illustratively, the second preset value may range from 3mm to 5mm, for example, the second preset value is 3mm, that is, the length of the cerebral artery vessel branch in the initial mask image is not less than 3mm.
Illustratively, for example, the segmentation algorithm employs AI techniques; for another example, the segmentation algorithm adopts the conventional image processing technology, and the adopted segmentation algorithm is not limited by the invention.
In some possible embodiments provided by the present invention, the location of the parent artery is at the beginning of the internal carotid artery.
In some possible embodiments provided by the present invention, referring to fig. 4 in combination with fig. 2, S02: extracting a region of interest in an initial mask image, comprising:
s022: and extracting a starting point and an end point of the region of interest.
The starting point of the region of interest is a point with the first mask pixel value in the three-dimensional direction of the initial mask image, namely the X, Y, Z direction; the end points of the region of interest are the points of the respective first non-mask pixel values in the three-dimensional direction in the initial mask image, i.e., in the X, Y, Z direction. The range of the region of interest is accurately determined by extracting the starting point and the end point of the region of interest, so that the subsequent calculation amount is reduced, and the calculation efficiency is improved.
In some possible embodiments provided by the invention, referring to fig. 4 in combination with fig. 2, S022: before extracting the start point and the end point of the region of interest, the method further comprises:
s021: and carrying out maximum connected domain processing on the initial mask image.
Step S021 eliminates noise caused by the initial mask image processing algorithm, and solves the problem that the extracted region of interest is inaccurate.
In some possible embodiments provided by the invention, referring to fig. 5 in combination with fig. 1, S2: thinning the target mask image, and preliminarily extracting the central line of the cerebral artery blood vessel to obtain a thinned image, wherein the thinning image comprises the following steps:
s21: and thinning the target mask image by adopting a distance transformation algorithm to obtain a distance transformation gray level image.
The distance transformation algorithm is adopted to refine the target mask image, the operation speed is high, but the extracted central line of the cerebral artery blood vessel is discontinuous.
S22: performing a point deletion operation by 26 neighborhood analysis according to the distance transformation gray-scale map, comprising: judging whether each pixel point is communicated with 26 neighborhoods: if yes, the pixel point is reserved; if not, deleting the pixel point.
The multi-neighborhood of the three-dimensional image includes a 6-neighborhood, an 18-neighborhood, a 26-neighborhood, etc., which are not limited by the present invention. The 6 fields of the three-dimensional image represent 6 faces that can be connected. The 18 neighborhood indicates that the directions are connected except for the four corners of the cube. The 26 neighborhood, i.e. 26 voxel points in the three-dimensional space adjacent to the central point, are all considered as its neighborhood points, including the voxels connected directly and diagonally to it, for example, if the coordinates of the central voxel point are (i, j, k), then (i ± 1,j ± 1,k ± 1) have 27 points, except the original central point (i, j, k), and the total of the surroundings is 26 points. Of the 26 points, 6 points with a distance of 1 from the central point, namely 6 points directly connected with the central point (exactly corresponding to 6 neighborhoods, which indicate that 6 surfaces of the three-dimensional image can be connected) are provided; a distance from the center point of
12 points (four corners of three planes passing through the center point) in diagonal relation with the center point; in the distanceThe distance of the heart point is->
I.e. 8 points forming a body diagonal of a small cube with the center point (two levels above and below, 4 cubes per level).
Compared with the 6 neighborhood and the 18 neighborhood, the 26 neighborhood ensures that any direction of the pixel points can be communicated, the 360-degree directivity can be achieved, analysis is carried out through the 26 neighborhood according to any directivity of the central line of the cerebral artery blood vessel, and the 26 neighborhood property deletion pixel points are ensured to accord with the definition of extracting the central line of the cerebral artery blood vessel.
Through the step S22, the deleting operation is carried out through 26 neighborhood analysis according to the distance transformation gray-scale map, the problem that the central line of the cerebral artery blood vessel extracted in the step S21 is discontinuous is solved, and the extracted central line of the cerebral artery blood vessel is a continuous central line.
In some possible embodiments provided by the present invention, referring to fig. 6 in combination with fig. 1, S3: determining a starting point of the internal carotid artery according to the refined image, and constructing a connected graph according to the refined image, wherein the method comprises the following steps:
s31: all end points are determined by 26 neighborhood analysis from the refined image.
Through the step S2, the connectivity of the refined image obtained after the target mask image is refined is maintained. Since the end point of the refined image usually has only one neighbor node in the 26-neighborhood, all the end points of the refined image can be found by the 26-neighborhood analysis.
S32: the starting point of the internal carotid artery is determined from all the end points, and the starting point of the internal carotid artery is the end point farthest from the center of all the end points.
The starting point of the internal carotid artery is defined as the end point farthest from the centers of all the end points, the starting point of the internal carotid artery can be automatically determined through the step S32, the starting point of the internal carotid artery does not need to be determined again according to all the end points through manual experience, and complexity increase caused by calculation of the end points is reduced.
In some possible embodiments provided by the present invention, in step S32, a starting point of the internal carotid artery is determined according to all the terminal points, and the formula for determining the starting point is as follows:
wherein Max { } represents taking the maximum value, | | represents the Euclidean distance, ei represents the ith terminal point, i is a positive integer,
represents the average of all end points. The formula for determining the starting point shows that the starting point S of the internal carotid artery is at the center from all the end points->
The most distal end point.
In some possible embodiments provided by the present invention, referring to fig. 6 in combination with fig. 1, S3: determining a starting point of the internal carotid artery according to the refined image, and constructing a connected graph according to the refined image, wherein the method comprises the following steps:
s33: and constructing a weighted connected graph by adopting a distance transformation algorithm according to the refined image.
And constructing a connected graph, so that the central line of the cerebral artery vessel can be conveniently extracted subsequently. And calculating the weight by adopting a distance transformation algorithm to construct a weighted connected graph, so that the central line of the extracted cerebral artery blood vessel is closer to the center. The calculation of the weight adopts a distance transformation graph as a basis, and the distance transformation graph has the principle that the farther the distance from the edge of the refined image is, the heavier the weight of the connected graph is, the smaller the weight is; the closer the distance to the edge of the refined image, the smaller the weight of the connected graph and the larger the weight.
In some possible embodiments provided by the present invention, in step S33, the weight calculation formula of the connectivity graph is as follows:
W i ＝-W i
wherein W represents weight, the actual value is distance transformation gray scale map value DF (i), i is the position of all mask pixel values in the refined image, W i And representing the weight value of the ith pixel point.
In some possible embodiments provided by the present invention, referring to fig. 7 in conjunction with fig. 1, S4: calculating all minimum cost paths as target center lines according to the starting points of the internal carotid artery and the connectivity graph, wherein the minimum cost paths comprise:
s41: the Dijkstra algorithm is used to calculate the minimum cost path between the starting point and each end point of the internal carotid artery as the target centerline.
The method specifically defines and accurately determines the initial point and each terminal point of the internal carotid artery, and accurately extracts the minimum cost path as the target center line, so that the extracted target center line is more accurate and closer to the center.
S42: and obtaining all target central lines until the calculation of the minimum cost path between the initial point and all the end points of the internal carotid artery is completed.
And iteratively extracting the minimum cost path between the starting point and all the end points of the internal carotid artery so as to obtain all accurate target central lines.
In other possible embodiments provided by the present invention, other algorithms such as depth-first search are adopted to extract the path between the start point and each end point of the internal carotid artery as the target centerline.
In some possible embodiments provided by the invention, with reference to fig. 7, S4: after calculating all minimum cost paths as target center lines according to the starting points of the internal carotid artery and the connectivity map, the method further comprises the following steps:
s43: all the target center lines are subjected to a filtering process (not shown in the figure). All the extracted target center lines are smoothed, and further, the 6D poses of subsequent navigation planning can be guided according to all the target center lines after filtering processing, so that the robustness of the algorithm is improved.
Illustratively, the filtering algorithm is mean filtering; in another example, the filtering algorithm is based on the mask image edge to perform further filtering operations. The filtering algorithm adopted by the invention for smoothing the central line is not limited.
Referring to fig. 1 to 9, an exemplary embodiment of a centerline extraction method of a cerebral artery blood vessel is described below.
Referring to fig. 3, S011: inputting a three-dimensional image of a cerebral artery vessel.
S012: and acquiring initial mask images of the cerebral aneurysm blood vessel and peripheral blood vessels by a segmentation algorithm according to the three-dimensional image of the cerebral artery blood vessel.
In this embodiment, the initial mask image has a starting position of the proximal internal carotid artery of the parent artery, a stopping position of the proximal internal carotid artery of the parent artery, a diameter of the distal end of the parent artery, which is not less than a preset multiple of the diameter of the parent artery at the position of the parent body, a diameter of the cerebral artery blood vessel branch in the initial mask image is greater than a first preset value, and a length of the cerebral artery blood vessel branch is not less than a second preset value. The initial mask image obtained in step S012 removes redundant bifurcated vessels from the three-dimensional image of cerebral artery vessels to reduce the amount of subsequent calculation and improve the calculation efficiency.
Referring to fig. 2, step S01 is performed: and acquiring an initial mask image of the cerebral artery blood vessel three-dimensional image. Preparation is made for extracting the region of interest.
Step S02 is executed: extracting a region of interest in an initial mask image, comprising:
referring to fig. 4, step S021 is performed: and carrying out maximum connected domain processing on the initial mask image. So as to eliminate noise caused by the initial mask image processing algorithm and ensure that the extracted region of interest is accurate.
S022: and extracting a starting point and an end point of the region of interest. The initial point of the interested region is a point in which the first of the X, Y, Z directions in the initial mask image is a mask pixel value; the end points of the region of interest are the points of the respective first non-mask pixel values in the X, Y, Z direction in the initial mask image. The range of the region of interest is accurately determined by extracting the starting point and the end point of the region of interest, so that the subsequent calculation amount is reduced, and the calculation efficiency is improved.
With continued reference to fig. 2, step S03 is performed: hole filling is performed on the cerebrovascular image in the region of interest to form a target mask image of the cerebral artery vessels. Referring to fig. 9, the holes are shown at B in fig. 9. As described above, hole filling is performed in step S03, so as to prepare for subsequent extraction of an accurate centerline of a cerebral artery.
Referring to fig. 1, step S1 is performed: and acquiring a target mask image of the cerebral artery blood vessel. The central line of the cerebral artery blood vessel is extracted from the target mask image, so that the calculation amount for extracting the central line is reduced, and the time for extracting the central line is saved.
Executing the step S2: thinning the target mask image, and preliminarily extracting the central line of the cerebral artery blood vessel to obtain a thinned image, wherein the thinning process comprises the following steps: as shown in steps S21 and S22 of fig. 5.
Specifically, referring to fig. 5, S21: and thinning the target mask image by adopting a distance transformation algorithm to obtain a distance transformation gray level image. Although the distance transformation algorithm is fast, the extracted center line is not continuous, and thus the step S22 is continuously performed to overcome the problem of discontinuous extracted center line.
S22: according to the distance variable gray level graph, point deletion operation is carried out through 26 neighborhood analysis, and the method comprises the following steps: judging whether each pixel point is connected with 26 neighborhoods: if yes, the pixel point is reserved; if not, deleting the pixel point. To extract the centerline of the continuous cerebral arterial vessel.
By adopting the technical scheme, the thinning processing ensures that the central line of the cerebral artery blood vessel is initially extracted as a continuous central line, and the connectivity of the thinned image is ensured.
Referring to fig. 1, step S3 is performed: determining a starting point of the internal carotid artery from the refined image, and constructing a connectivity graph from the refined image, including steps S31, S32, and S33 as shown in fig. 6.
Specifically, referring to fig. 6, step S31 is performed: all end points are determined by 26 neighborhood analysis from the refined image.
The refined image obtained by step S2 maintains connectivity. Since the end point of the refined image usually has only one neighbor node in the 26-neighborhood, all the end points of the refined image can be found by the 26-neighborhood analysis.
Step S32 is executed: the starting point of the internal carotid artery is determined from all the end points, and the starting point of the internal carotid artery is the end point farthest from the center of all the end points. Determining the starting point of the internal carotid artery provides for subsequent extraction of the target centerline through the least cost path.
In the present embodiment, the formula is used
A starting point S is determined. Wherein Max { } represents taking the maximum value, | | represents the Euclidean distance, ei represents the ith terminal point, i is a positive integer, and/or>
Represents the average of all end points. The formula for determining the starting point shows that the starting point S of the internal carotid artery is at the center from all the end points->
The most distal end point. Referring to fig. 8, the starting point of the internal carotid artery is shown as point a in fig. 8.
Referring to fig. 6, execution continues with step S33: and constructing a weighted connected graph by adopting a distance transformation algorithm according to the refined image. So that the center line of the extracted cerebral artery blood vessel is closer to the center.
In this embodiment, the weight calculation formula of the connected graph is: w is a group of i ＝-W i . Wherein W represents weight, the actual value is distance transformation gray scale map value DF (i), i is the position of all mask pixel values in the refined image, W i And representing the weight value of the ith pixel point.
With continuing reference to fig. 1, step S4 is performed: and calculating all minimum cost paths as target central lines according to the initial points of the internal carotid arteries and the connectivity graph. Step S4 includes step S41, step S42 as shown in fig. 7, and step S43 not shown in the figure.
S41: the Dijkstra algorithm is used to calculate the minimum cost path between the starting point and each end point of the internal carotid artery as the target centerline. The method comprises the steps of taking a starting point of an internal carotid artery as a starting point and each end point as an end point, and calculating a minimum cost path of the starting point and each end point through a Dijkstra algorithm to be used as a target central line, so that the extracted target central line is more accurate and closer to the center.
S42: and obtaining all target central lines until the calculation of the minimum cost path between the initial point and all the end points of the internal carotid artery is completed. That is, after each minimum cost path between one end point (i.e. end point) and the start point (i.e. start point) of the internal carotid artery is obtained, it is determined whether there is an end point (i.e. end point) where the minimum cost path is not obtained:
if yes, go to step S41: calculating a minimum cost path between the starting point of the internal carotid artery and each end point by using a Dijkstra algorithm as a target central line;
if not, step S42 is executed to obtain all target center lines.
Next, step S43 is executed: all target centerlines are averaged (not shown). All the extracted target center lines are smoothed, and further, the 6D poses of subsequent navigation planning can be guided according to all the target center lines after filtering processing, so that the robustness of the algorithm is improved.
Referring to fig. 1, step S5 is performed: all target centerlines are output.
By adopting the technical scheme, the preprocessing step is added in the process of extracting the central line of the cerebral artery blood vessel, for example, noise is removed in the step S01 to obtain the initial mask image, and holes are filled in the step S03, so that the problem of inaccuracy of data in the process of obtaining the data source is solved, the calculation speed is improved, and the robustness of the algorithm is also improved. And a post-processing step is added in the process of extracting the central line of the cerebral artery blood vessel, for example, all target central lines are filtered in the step S43, so that the robustness of the algorithm is improved.
The method for extracting the central line of the cerebral artery blood vessel realizes the automatic and rapid extraction of the accurate central line close to the center, simultaneously increases the robustness of the algorithm for extracting the central line, optimizes the extraction of the blood vessel central line of cerebral aneurysm planning software, can accurately determine the specific position of a tumor body, calculates the lesion degree of the tumor body, estimates the rupture risk and plans the interventional operation path.
In a second aspect, referring to fig. 10, the present invention provides a centerline extraction system 1 for cerebral artery blood vessels, comprising: a data receiving module 11, a data processing module 12 and a data output module 13, wherein,
the data receiving module 11 is used for receiving a three-dimensional image of a cerebral artery vessel input by a user;
a data processing module 12 connected with the data receiving module 11 for
Acquiring a target mask image of a cerebral artery blood vessel;
thinning the target mask image, and preliminarily extracting the central line of the cerebral artery blood vessel to obtain a thinned image;
determining a starting point of the internal carotid artery according to the thinned image, and constructing a connected graph according to the thinned image;
calculating all minimum cost paths as target center lines according to the initial points of the internal carotid arteries and the connectivity graph;
and the data output module 13 is connected with the data processing module 12 and is used for outputting all the target center lines.
By adopting the technical scheme, under the synergistic action of the data receiving module 11, the data processing module 12 and the data output module 13, the central line of the cerebral artery blood vessel is fully automatically extracted by the central line extracting system 1 of the cerebral artery blood vessel, so that the calculated amount is reduced, the calculating speed is improved, and the extracted central line of the cerebral artery blood vessel is closer to the center.
In a third aspect, referring to fig. 11, the present invention provides an electronic device 2, including a memory 21, a processor 22, and a computer program stored in the memory 21 and executable on the processor 22, wherein the processor 22, when executing the computer program, implements the centerline extraction method for the cerebral artery blood vessel in any of the foregoing embodiments. The memory 21 may include, for example, a system memory, a fixed nonvolatile storage medium, and the like. The system memory stores, for example, an operating system, an application program, a Boot Loader (Boot Loader), and other programs.
In the embodiment, the electronic device 2 automatically and quickly extracts the central line of the cerebral artery blood vessel close to the center, so that the efficiency is improved, and the time is saved.
In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program is executed by a processor to implement any one of the foregoing methods for extracting a centerline of a cerebral artery vessel.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable non-transitory storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention has been described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While the invention has been shown and described with reference to certain preferred embodiments thereof, it will be understood by those skilled in the art that the foregoing is a more particular description of the invention than is possible with reference to the specific embodiments, and the specific embodiments of the invention are not to be considered as limited to those descriptions. Various changes in form and detail, including simple deductions or substitutions, may be made by those skilled in the art without departing from the spirit and scope of the invention.
Claims (14)
1. A method for extracting centerline of cerebral artery blood vessel, comprising:
acquiring a target mask image of cerebral artery blood vessels;
thinning the target mask image, and preliminarily extracting the central line of the cerebral artery blood vessel to obtain a thinned image;
determining a starting point of an internal carotid artery according to the refined image, and constructing a connected graph according to the refined image;
calculating all minimum cost paths as target center lines according to the starting points of the internal carotid artery and the connectivity graph;
outputting all of the target centerlines.
2. The method for extracting the centerline of the cerebral artery vessel as claimed in claim 1, wherein the determining the starting point of the internal carotid artery according to the refined image and constructing the connectivity graph according to the refined image comprises:
determining all end points through 26 neighborhood analysis according to the refined image;
determining the starting point of the internal carotid artery according to all the end points, wherein the starting point of the internal carotid artery is the end point farthest from the centers of all the end points.
3. The method for extracting centerline of cerebral artery vessel as claimed in claim 2, wherein the starting point of the internal carotid artery is determined according to all the end points by the formula:
wherein Max { } represents taking the maximum value, | | represents the Euclidean distance, ei represents the ith terminal point, i is a positive integer,
represents the average of all end points.
4. The method for extracting the centerline of the cerebral artery vessel as claimed in claim 3, wherein the determining the starting point of the internal carotid artery according to the refined image and constructing the connectivity map according to the refined image comprises:
constructing a weighted connected graph by adopting a distance transformation algorithm according to the refined image;
the farther the distance from the edge of the refined image is, the larger the weight of the connected graph is, and the smaller the weight is; the closer the distance to the edge of the refined image is, the smaller the weight of the connected graph is, and the larger the weight is.
5. The method for extracting centerline of cerebral artery vessel according to claim 4, wherein the weight calculation formula of the connectivity graph is as follows:
W i ＝-W i
wherein W represents weight, the actual value is distance transformation gray scale map value DF (i), i is the position of all mask pixel values in the refined image, W i And representing the weight value of the ith pixel point.
6. The method for extracting centerline of cerebral artery vessel according to claim 5, wherein the calculating all minimum cost paths as target centerline according to the starting point of the internal carotid artery and the connectivity map comprises:
calculating a minimum cost path between a starting point of the internal carotid artery and each end point as a target centerline by using Dijkstra algorithm;
and obtaining all target central lines until the calculation of the minimum cost paths between the starting point of the internal carotid artery and all the end points is completed.
7. The method for extracting centerline of cerebral artery vessel as claimed in claim 1, wherein before the obtaining of the target mask image of cerebral artery vessel, the method further comprises:
acquiring an initial mask image of a cerebral artery blood vessel three-dimensional image;
extracting a region of interest in the initial mask image;
and filling holes in the cerebrovascular image in the region of interest to form a target mask image of the cerebral artery blood vessels.
8. The method for extracting the centerline of the cerebral artery vessel as claimed in claim 7, wherein the step of obtaining the initial mask image of the three-dimensional image of the cerebral artery vessel comprises:
inputting a three-dimensional image of a cerebral artery blood vessel;
acquiring initial mask images of cerebral aneurysm blood vessels and peripheral blood vessels by a segmentation algorithm according to the three-dimensional images of the cerebral artery blood vessels;
the initial mask image is used for determining the initial mask image, the initial mask image is used for determining the starting point of the internal carotid artery at the proximal end of the parent artery, the stopping point is used for determining the preset multiple of the diameter of the parent artery at the position of the parent artery, the diameter of the cerebral artery blood vessel branch in the initial mask image is larger than a first preset value, and the length of the cerebral artery blood vessel branch is not smaller than a second preset value.
9. The method of claim 8, wherein the location of the parent artery is at the beginning of the internal carotid artery.
10. The method for extracting centerline of cerebral artery vessel as claimed in claim 7, wherein the extracting the region of interest in the initial mask image comprises:
extracting a starting point and an end point of the region of interest;
the starting point of the interested region is a point which is a mask pixel value in the first three-dimensional direction in the initial mask image, and the ending point is a point which is a non-mask pixel value in the first three-dimensional direction.
11. The method for extracting centerline of cerebral artery vessel as claimed in claim 1, wherein the refining the target mask image to preliminarily extract the centerline of cerebral artery vessel to obtain a refined image comprises:
thinning the target mask image by adopting a distance transformation algorithm to obtain a distance transformation gray level image;
performing a point deletion operation by 26 neighborhood analysis according to the distance transformation gray-scale map, comprising: judging whether each pixel point is communicated with 26 neighborhoods: if yes, the pixel point is reserved; if not, deleting the pixel point.
12. A system for extracting a centerline of a cerebral artery vessel, comprising:
the data receiving module is used for receiving a three-dimensional image of a cerebral artery vessel input by a user;
a data processing module connected with the data receiving module and used for
Acquiring a target mask image of cerebral artery blood vessels;
thinning the target mask image, and preliminarily extracting the central line of the cerebral artery blood vessel to obtain a thinned image;
determining a starting point of an internal carotid artery according to the refined image, and constructing a connected graph according to the refined image;
calculating all minimum cost paths as target center lines according to the initial points of the internal carotid artery and the connectivity graph;
and the data output module is connected with the data processing module and is used for outputting all the target center lines.
13. An electronic device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor when executing the computer program implements the method for extracting a centerline of a cerebral artery vessel according to any one of claims 1 to 11.
14. A computer-readable storage medium, in which a computer program is stored, which, when being executed by a processor, implements the method for extracting a centerline of a cerebral artery vessel according to any one of claims 1 to 11.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310064732.6A CN115984239A (en) | 2023-01-12 | 2023-01-12 | Method, system, device and storage medium for extracting central line of cerebral artery blood vessel |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310064732.6A CN115984239A (en) | 2023-01-12 | 2023-01-12 | Method, system, device and storage medium for extracting central line of cerebral artery blood vessel |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115984239A true CN115984239A (en) | 2023-04-18 |
Family
ID=85970278
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310064732.6A Pending CN115984239A (en) | 2023-01-12 | 2023-01-12 | Method, system, device and storage medium for extracting central line of cerebral artery blood vessel |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115984239A (en) |
-
2023
- 2023-01-12 CN CN202310064732.6A patent/CN115984239A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Tobon-Gomez et al. | Benchmark for algorithms segmenting the left atrium from 3D CT and MRI datasets | |
| Li et al. | Optimal surface segmentation in volumetric images-a graph-theoretic approach | |
| EP2677505B1 (en) | Modelling of tubular structures | |
| JP5129480B2 (en) | System for performing three-dimensional reconstruction of tubular organ and method for operating blood vessel imaging device | |
| JP2019193808A (en) | Diagnostically useful results in real time | |
| KR102050649B1 (en) | Method for extracting vascular structure in 2d x-ray angiogram, computer readable medium and apparatus for performing the method | |
| CN112652052B (en) | Coronary artery three-dimensional reconstruction method and system based on blood vessel branch registration | |
| CN109478327B (en) | Method for automatic detection of systemic arteries in Computed Tomography Angiography (CTA) of arbitrary field of view | |
| CN111798451A (en) | 3D guide wire tracking method and device based on blood vessel 3D/2D matching | |
| CN113409456B (en) | Modeling method, system, device and medium for three-dimensional model before craniocerebral puncture operation | |
| CN111260704B (en) | Vascular structure 3D/2D rigid registration method and device based on heuristic tree search | |
| CN111784751B (en) | 3D/2D registration-based guide wire 3D simulation tracking method and device | |
| Lidayova et al. | Skeleton-based fast, fully automated generation of vessel tree structure for clinical evaluation of blood vessel systems | |
| CN112651984A (en) | Blood vessel lumen intimal contour extraction method and device, ultrasonic equipment and storage medium | |
| CN108010056B (en) | Vascular motion tracking method based on four-dimensional medical image | |
| CN115984239A (en) | Method, system, device and storage medium for extracting central line of cerebral artery blood vessel | |
| CN113096166B (en) | Medical image registration method and device | |
| CN114931435A (en) | Three-dimensional model processing method and device and electronic equipment | |
| CN112669370B (en) | Coronary artery radius calculation method, terminal and storage medium | |
| Freiman et al. | Nearly automatic vessels segmentation using graph-based energy minimization | |
| CN114049282B (en) | Coronary artery construction method, device, terminal and storage medium | |
| Araujo et al. | Computer aided detection of deep inferior epigastric perforators in computed tomography angiography scans | |
| Perera-Bel et al. | Segmentation of the placenta and its vascular tree in Doppler ultrasound for fetal surgery planning | |
| Li et al. | Development of preoperative liver and vascular system segmentation and modeling tool for image-guided surgery and surgical planning | |
| EP4315237A1 (en) | Systems and methods for automatic blood vessel extraction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |