CN108022237B - Blood vessel extraction method, system and storage medium - Google Patents
Blood vessel extraction method, system and storage medium Download PDFInfo
- Publication number
- CN108022237B CN108022237B CN201711245630.5A CN201711245630A CN108022237B CN 108022237 B CN108022237 B CN 108022237B CN 201711245630 A CN201711245630 A CN 201711245630A CN 108022237 B CN108022237 B CN 108022237B
- Authority
- CN
- China
- Prior art keywords
- blood vessel
- model
- radius
- preset
- target blood
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/005—Tree description, e.g. octree, quadtree
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30101—Blood vessel; Artery; Vein; Vascular
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- Computer Graphics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The embodiment of the invention discloses a blood vessel extraction method, a blood vessel extraction system and a storage medium. The blood vessel extraction method comprises the following steps: in a medical image, acquiring a plurality of predicted direction sampling points in a preset direction range of a current target blood vessel; obtaining modeling parameters of a blood vessel model corresponding to the predicted direction sampling points, and establishing the blood vessel model according to the modeling parameters, wherein each blood vessel model corresponds to one predicted direction sampling point; fitting the blood vessel model to determine an effective blood vessel model; and updating the extracted target blood vessel data according to the effective blood vessel model. The technical problem that the traditional blood vessel extraction method is poor in adaptability is solved, and the technical effect of accurately and completely extracting the blood vessel from the image is achieved.
Description
Technical Field
The embodiment of the invention relates to an image processing technology, in particular to a blood vessel extraction method, a blood vessel extraction system and a storage medium.
Background
The extraction and reconstruction of blood vessels is of great importance in CT angiography and medical diagnostics. Although the CT angiography technique can significantly enhance the CT value and contrast of the blood vessel, it is limited by image noise, uneven brightness of the blood vessel, and factors of more branches and uneven thickness of a specific scene, and it needs to be extracted by a certain image processing technique.
Existing methods for extracting blood vessels include region growing methods, level set-based methods, graph cutting methods, and the like. The region growing method is sensitive to CT value, and the threshold value for determining the growth termination is difficult and has poor adaptability to data. Because the image may have noise, the blood vessel may be locally darkened, and the level set method and the image segmentation method may have the problem of under-segmentation.
Therefore, the conventional blood vessel extraction method has a problem of poor adaptability.
Disclosure of Invention
The embodiment of the invention provides a blood vessel extraction method, a blood vessel extraction system and a storage medium, which aim to solve the technical problem that the existing blood vessel extraction method is poor in adaptability.
In a first aspect, an embodiment of the present invention provides a blood vessel extraction method, where the method includes: in a medical image, acquiring a plurality of predicted direction sampling points in a preset direction range of a current target blood vessel;
obtaining modeling parameters of a blood vessel model corresponding to the predicted direction sampling points, and establishing the blood vessel model according to the modeling parameters, wherein each blood vessel model corresponds to one predicted direction sampling point;
fitting the blood vessel model to determine an effective blood vessel model;
and updating the extracted target blood vessel data according to the effective blood vessel model.
In a second aspect, an embodiment of the present invention further provides a medical imaging system, where the system includes:
projection data acquisition means for acquiring projection data of a medical image containing a target blood vessel;
a computer device connected to the projection data acquisition means, the computer device comprising a memory and one or more processors and a computer program stored on the memory and executable on the processors, characterized in that the processors, when executing the program, generate a medical image from the projection data, perform the vessel extraction method according to the first aspect on the medical image.
In a third aspect, the present invention further provides a computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the blood vessel extraction method according to the first aspect.
According to the technical scheme of the blood vessel extraction method provided by the embodiment of the invention, a plurality of prediction direction sampling points and modeling parameters thereof in the preset direction range of the current target blood vessel in the medical image are obtained, a blood vessel model is established based on the modeling parameters, a blood vessel model which is possibly inosculated with the actual blood vessel is established, the blood vessel model is fitted to determine an effective blood vessel model, and the extracted target blood vessel data is updated according to the effective blood vessel model, so that the blood vessel can be completely extracted from the medical image, and the method has high adaptability, sensitivity and robustness.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a flowchart of a blood vessel extraction method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of forward prediction according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of modeling multiple preset radii according to an embodiment of the present invention;
FIG. 4 is a flowchart of a modeling parameter calculation method according to a second embodiment of the present invention;
fig. 5 is a flowchart of a method for obtaining a center point and a radius of a start end according to a second embodiment of the present invention;
FIG. 6 is a schematic diagram of modeling multiple preset step sizes according to a second embodiment of the present invention;
FIG. 7 is a schematic diagram of a polar coordinate prediction sphere provided in the second embodiment of the present invention;
FIG. 8 is a schematic diagram of a grid predictor sphere provided in accordance with a second embodiment of the present invention;
FIG. 9 is a flowchart of a tree structure generation method according to a third embodiment of the present invention;
fig. 10 is a flowchart of a reverse retroactive blood vessel naming identification method according to the fourth embodiment of the present invention;
fig. 11 is a block diagram of a blood vessel extraction device according to a fifth embodiment of the present invention;
fig. 12 is a block diagram illustrating a medical imaging system according to a sixth embodiment of the present invention;
fig. 13 is a block diagram of a computer device according to a sixth embodiment of the present invention.
Icon:
30-existing blood vessels; 301-target vessel extracted at previous moment; 31-target blood vessel; 32-a vascular model; 33-preset direction range; 41-polar coordinate predictor sphere; 42-mesh prediction sphere.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described through embodiments with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example one
Fig. 1 is a flowchart of a blood vessel extraction method provided in this embodiment, which is applicable to a case of extracting a blood vessel from a medical image, particularly extracting a hepatic vein with complex branches and radius changes, and which can be executed by software/hardware configured in a processor of an intelligent device, such as a medical server, a general computer, a doctor workstation or a medical imaging device with computer function, and includes:
s11, acquiring a plurality of prediction direction sampling points in the preset direction range of the current target blood vessel in the medical image.
The medical image may be a CT image, an MR image, a PET image, or other medical image containing blood vessel information, and the present embodiment is described by taking a CT angiography image as an example.
In order to accurately extract a target blood vessel from a CT angiography image, a preset range of directions in which the target blood vessel may exist, i.e., a set of directions in which the target blood vessel may appear, is generally determined. Optionally, the present embodiment adopts a forward prediction method, as shown in fig. 2, that is, the preset direction range is located in front of the existing blood vessel 30, and the existing blood vessel 30 may be a target blood vessel 301 extracted at a previous time, or may be a blood vessel segment obtained by another method. The starting end in this embodiment refers to a blood vessel cross section at which the blood vessel extraction starts, the end refers to a blood vessel cross section at which the blood vessel extraction ends, and the preset direction range may be set according to an empirical value or a statistical rule of the direction of the blood vessel and the direction of the branch.
The predicted direction samples of the present embodiment represent possible scatter positions of the blood vessel model, and in view of the fact that the target blood vessel is located in the preset direction range, it is only necessary to obtain the possible scatter positions of the blood vessel model and the modeling parameters required by each scatter position in the preset direction range in order to increase the blood vessel extraction speed.
And S12, obtaining modeling parameters of the blood vessel model corresponding to the predicted direction sampling points, and establishing the blood vessel model according to the modeling parameters, wherein each blood vessel model corresponds to one predicted direction sampling point.
The modeling parameters may include a target vessel radius r, a target vessel direction
Coordinates of center point of target blood vessel
The contrast k of the pixel values inside and outside the blood vessel segment, the average value m of the pixels outside the blood vessel segment, and the modeling parameter may be information manually input by a user, may also be partially input by the user (usually, a starting point coordinate, or a combination of the starting point coordinate and the blood vessel direction), may be partially obtained by computer calculation, or may be completely obtained by automatic calculation.
And establishing a blood vessel model based on each acquired prediction direction sampling point and the corresponding modeling parameter, so that each blood vessel model corresponds to one prediction direction sampling point.
Optionally, as shown in fig. 3, the existing blood vessel 30 in this embodiment includes a target blood vessel 301 extracted at a previous time, and corresponding to the extracted target blood vessel in this embodiment, the target blood vessel 31 is located in front of the end of the existing blood vessel 30, or the target blood vessel 31 is located in front of the end of the target blood vessel 301 extracted at the previous time, a preset direction range 33 is determined in front of the end of the target blood vessel 301 extracted at the previous time, a blood vessel model 32 is established within the preset direction range 33, when the estimated radius in the obtained modeling parameters is multiple, each predicted direction sampling point may correspond to multiple blood vessel models 32, and the number of the corresponding blood vessel models 32 is equal to the number of the blood vessel radii corresponding to the predicted direction sampling points.
And S13, fitting the blood vessel model to determine an effective blood vessel model.
The method adopts a linear least square method to fit the outside-tube mean value and the inside-tube and outside-tube contrast of the blood vessel model; and fitting the radius, the central point and the vessel direction of the target vessel in each tomographic image by adopting a nonlinear least square method according to the outside-of-tube mean value and the inside-of-tube contrast to construct a fitting vessel model.
In order to improve the accuracy of blood vessel extraction, a fitting blood vessel model with a high matching degree with a target blood vessel needs to be found from all fitting blood vessel models, namely, which one or more blood vessel models are effective to be fitted is judged, therefore, the embodiment performs t test on the blood vessel contrast, tests the significance of the blood vessel contrast, judges whether the model score of the fitting blood vessel model is greater than a preset score value, and if so, takes the fitting blood vessel model as an effective blood vessel model; and if not, deleting the fitted blood vessel model, wherein the model score is the ratio of the blood vessel contrast to the blood vessel contrast t-test square root difference.
The preset score can be set according to an empirical value, and the selection of the preset score is related to the number of the predicted direction sampling points, for example, a certain radius of a blood vessel is 1.5mm, the number of the predicted direction sampling points is more than 4000, and the preset score can be set to 15.
And S14, updating the extracted target blood vessel data according to the effective blood vessel model.
It can be understood that, in the present embodiment, the target blood vessels are extracted segment by segment, and the target blood vessels are connected in sequence to form the blood vessels to be extracted, so that after the current target blood vessel is extracted, the target blood vessel data is added to the extracted target blood vessel data to form new extracted target blood vessel data.
According to the technical scheme of the blood vessel extraction method provided by the embodiment of the invention, a plurality of prediction direction sampling points and modeling parameters thereof in the preset direction range of the current target blood vessel in the medical image are obtained, a blood vessel model is established based on the modeling parameters, a blood vessel model possibly inosculated with the actual blood vessel is scattered, the blood vessel model is fitted, an effective blood vessel model is determined, and the extracted target blood vessel data is updated according to the effective blood vessel model, so that the blood vessel can be completely extracted from the medical image, and the method has high adaptability, sensitivity and robustness.
Example two
Fig. 4 is a flowchart of a modeling parameter obtaining method according to the second embodiment, and this embodiment optimizes the obtaining of the modeling parameter of the blood vessel model corresponding to the predicted direction sampling point on the basis of the above embodiments. As shown in fig. 4, the modeling parameter solving method includes:
s121, obtaining a starting end central point and a starting end radius of the current target blood vessel, and generating at least one preset radius according to the starting end radius.
To build a vessel model, modeling parameters of a target vessel, such as a starting end center point and a starting end radius of the target vessel, are acquired, and then at least one preset radius is generated according to the starting end vessel radius.
For the case that the variation of the blood vessel is small, the preset radius may be set as the radius of the blood vessel starting end multiplied by a large scale factor, for example, when the radius of the blood vessel starting end is r, the selectable range of the preset radius is 0.6r-r, and the preset radius may be set as: 0.6r, 0.8r and 1.0 r. For the case that the radius of the blood vessel changes in a large range, such as the hepatic portal vein, the radius of the blood vessel changes in a range of 0.5-8mm, the preset radius may be set as the radius of the blood vessel starting end multiplied by a small scale factor, for example, when the radius of the blood vessel starting end is r, the radius of the blood vessel starting end may be set to 0.4r-r, and the preset radius may be set as: 0.4r, 0.6r, 0.8r and 1.0 r. By setting a plurality of preset radiuses, the hepatic portal vein can be accurately and completely extracted, and the method has high applicability and robustness.
Optionally, the method for obtaining the starting end center point and the starting end radius in this embodiment (as shown in fig. 5) includes:
s1211, determining whether the extracted target blood vessel data exists, if so, performing step S1212, otherwise, performing step S1213.
And S1212, taking the extracted tail end central point and tail end radius of the target blood vessel as the starting end central point and starting end radius of the current target blood vessel.
In this embodiment, the target blood vessels are extracted segment by segment, and the blood vessels to be extracted are formed by sequentially connecting the target blood vessels, so that the currently extracted end center point and end radius of the target blood vessel can be used as the start end center point and start end radius of the current target blood vessel.
S1213, determining the starting end center point and the starting end radius of the current target blood vessel through a blood vessel filter.
And when the extracted target blood vessel data does not exist, namely the starting end of the current target blood vessel is the starting end of the blood vessel in the medical image, taking the starting end central point of the whole blood vessel as the starting end central point of the target blood vessel.
Due to the difference of the radius of the blood vessels and the concentration of the contrast agent in the blood vessels, in a CT angiography image, the gray value of the thicker blood vessels is higher, while the gray value of the thinner blood vessels is lower, although the larger blood vessels can be highlighted through gray conversion, the thinner blood vessels are difficult to be distinguished from the surrounding tissues, and the accuracy of blood vessel extraction is easily reduced. The embodiment further enhances the blood vessels by a Hessian matrix eigenvalue algorithm, and extracts the blood vessels by using the tubular structures of the blood vessels.
Since the liver blood vessels are composed of large blood vessels and numerous thin blood vessels of varying thickness and varying diameters, the blood vessel parameters cannot be accurately obtained using single-scale enhancement. In this embodiment, the vessel parameters are obtained through n layers of scale spaces, that is, a multi-scale space, and the multi-scale space theory may be used to represent signal features of different sizes in the same space.
Based on the method, for the tomograms with the preset number of blood vessel starting ends, the Hessian matrix and the characteristic value of each pixel are calculated under each scale, whether the characteristic value represents the blood vessel structure or not is judged, namely, the characteristic value lambda of the Hessian matrix of each pixel is judged1、λ2And λ3Whether or not 0 ≈ λ is satisfied1|＜＜|λ2|≈|λ3If yes, the pixel belongs to the blood vessel, and if not, the pixel does not belong to the blood vessel. Based on Hessian matrix constructs a vessel filter, calculates the filter response value of each pixel belonging to the vessel, and takes the scale value corresponding to the maximum response value as the radius of the starting end of the target vessel, and the characteristic value lambda corresponding to the maximum response value1The corresponding characteristic vector is taken as the direction of the blood vessel, and the pixel coordinate corresponding to the maximum response value is taken as the center point of the blood vessel.
And S122, establishing a predicted sphere by taking the center point of the starting end as the center of the sphere and at least one preset step length as the radius.
In this embodiment, the preset step is a radius of a prediction sphere and may also be used as a length of a blood vessel model, when the contrast of a certain segment of an image of a blood vessel is low, the prediction sphere is established by a single preset step, and it may be impossible to determine whether some prediction direction sampling points are located at a blood vessel center point, and then a situation of ending blood vessel extraction in advance occurs.
For example, as shown in fig. 8, assume that point a is a first coordinate point adjacent to the first origin V in the spherical grid coordinate system. Connecting the first origin V with the point A, wherein the length of the line segment is a vector corresponding to the point A
The direction of point a is the direction in which the first origin points to point a. If the vector corresponding to point A
Direction vector of known vessel segment
Is less than a specified angle threshold, point a is in a known vessel segmentIn the forward tracking direction.
The method comprises the steps of analyzing prediction direction sample points generated by prediction balls based on different preset step lengths respectively, judging whether blood vessel center points detected by the different preset step lengths are located on the same radius, if so, establishing a blood vessel model based on a smaller preset step length, if not, adopting two preset step lengths to establish the blood vessel model respectively, and judging whether the established blood vessel model is effective, so that the situation that the blood vessel extraction cannot be completely extracted due to the fact that whether the prediction direction sample points are located on the blood vessel center points cannot be judged is avoided.
Optionally, the preset step length in this embodiment is 1.0 to 1.7 times of the radius of the extracted end of the target blood vessel, and optionally, when a prediction sphere is established by using multiple preset step lengths, the preset step lengths may be set to be 1.0 time, 1.5 times, and 1.7 times of the radius of the end of the target blood vessel, and of course, other preset step length setting manners may also be used according to the change situation of the radius of the blood vessel.
And S123, generating a plurality of predicted direction sampling points on the surface of the predicted ball in the preset direction range.
The distribution rule of the blood vessels is generally related to the radius of the blood vessels, and the embodiment adopts different methods for generating the sample points in the prediction direction for the target blood vessels with different radii, specifically:
when the target blood vessel is a thin blood vessel, generating a plurality of prediction direction sampling points within a preset direction range of the thin blood vessel by adopting a polar coordinate condition, as shown by a polar coordinate prediction sphere 41 in fig. 7; when the target blood vessel is a thick blood vessel, a plurality of predicted direction sampling points are generated within the range of the preset direction of the non-thin blood vessel by adopting the spherical grid condition, as shown by a grid predicted ball 42 in fig. 8.
The division of the thin blood vessels and the non-thin blood vessels can be performed according to the statistical values or the empirical values of the radius and the distribution rule of the blood vessels at different positions, for example, the thin blood vessels of the hepatic veins in this embodiment are blood vessels with the radius smaller than 1.1-1.5 mm.
For the thin blood vessel, polar coordinate conditions are adopted, a plurality of predicted direction sampling points are generated only in the preset direction range of the thin blood vessel, or the predicted direction sampling points distributed on the whole surface of the predicted sphere are generated, but only the predicted direction sampling points in the preset direction range are analyzed.
Similarly, for non-fine blood vessels, a spherical grid condition is adopted, a plurality of predicted direction sampling points are generated only in a preset direction range of the blood vessel, or the predicted direction sampling points distributed on the whole surface of the predicted sphere are generated, but only the predicted direction sampling points in the preset direction range are analyzed.
In this embodiment, the preset direction range of the non-thin blood vessel is larger than the preset direction range of the thin blood vessel, and in addition, whether the target blood vessel belongs to the thin blood vessel is determined, and whether the target blood vessel belongs to the thin blood vessel can be indirectly obtained through the radius of the tail end of the target blood vessel extracted from the previous section.
It is understood that the present embodiment may also divide the blood vessel into a plurality of blood vessel radius ranges according to the blood vessel radius, and then set different preset direction ranges based on different blood vessel radius ranges.
And S124, constructing a tubular structure based on the central point of the starting end, the sample point of the prediction direction and the preset radius.
In this embodiment, a sample point pointing to the predicted direction from the center point of the start end is used as the direction of the center line, and a tubular structure is constructed by using a preset radius as the radius, so as to limit the pixel distribution range analyzed by the vascular filter.
And S125, constructing a blood vessel filter based on the Hessian matrix, and analyzing pixels in the tubular structure by using the blood vessel filter to obtain blood vessel model parameters.
Constructing n-layer scale space, calculating the Hessian matrix and the characteristic value of each pixel of each tomographic image in the tubular structure under each scale, and judging whether the characteristic value represents the vascular structure, namely judging the characteristic value lambda of the Hessian matrix of each pixel1、λ2And λ3Whether or not 0 ≈ λ is satisfied1|＜＜|λ2|≈|λ3If yes, the pixel belongs to the blood vessel, and if not, the pixel does not belong to the blood vessel. Constructing a vessel filter, such as a Frangi's filter, based on the Hessian matrix, calculating filter response values of pixels belonging to vessels in each tomographic image, and calculating a value of the filter response for each pixel belonging to a vessel in each tomographic imageThe scale value corresponding to the maximum response value is used as the estimated radius of the target blood vessel in the current tomographic image, and the characteristic value lambda corresponding to the maximum response value1The corresponding characteristic vector is the estimated direction, and the pixel coordinate corresponding to the maximum response value is the center point of the blood vessel. And taking the central point, the estimated radius and the estimation of the target blood vessel corresponding to each tomography image as blood vessel model parameters.
In order to improve the accuracy of the estimated radius, the preset radius of the tubular structure is usually larger than the radius of the blood vessel model to prevent the tubular structure from directly excluding some pixels belonging to the blood vessel, but when the preset radius of the tubular structure is too large relative to the radius of the blood vessel model, the data calculation amount is easily increased, therefore, when the same prediction direction sampling point corresponds to the tubular structure range determined by a plurality of preset radii and the prediction direction sampling point, the tubular structure with a smaller preset radius can be obtained, the estimated radius and the estimated direction of the blood vessel in each defined tomographic image are determined to judge whether the preset radius of the tubular structure is smaller than or equal to the estimated radii which are obtained, if so, the tubular structure determined by the larger preset radius and the prediction direction sampling point is determined, and the size relation between the corresponding estimated radius and the preset radius of the tubular structure is determined, and taking the estimated radius, the estimated direction and the blood vessel center point of the blood vessel corresponding to the tubular structure as blood vessel model parameters until the preset radius of the first tubular structure is larger than all the estimated radii corresponding to the tubular structure.
According to the method, the modeling parameters of the blood vessel model are obtained through the prediction direction sampling points distributed on the surface of the prediction sphere, the data calculation amount is reduced through multiple preset radiuses, the modeling accuracy of the blood vessel model is improved, the accuracy of blood vessel extraction is improved, and the technical effect of completely and accurately extracting the blood vessel is achieved.
EXAMPLE III
Fig. 9 is a flowchart of a tree structure generation method provided in the third embodiment, and the tree structure generation method is added after the extracted target blood vessel data is updated according to the effective blood vessel model in the above embodiments. As shown in fig. 9, the tree structure generation method includes:
s151, determining whether at least one of the radius, angle, or model score of each branch vessel meets a predetermined branch condition, if yes, performing step S152, otherwise, performing step S153.
The present embodiment determines whether the branch vessel meets the condition at least from one or more of the factors such as the radius of the vessel, the angle, or the score of the vessel model, i.e. whether the extracted branch vessel exists in the medical image, for example, the radius of the branch vessel is usually smaller than the radius of the vessel before the branch; for the same blood vessel, the included angle between the branch blood vessels has a certain range, and the range can be obtained by a statistical method; or obtained through empirical values; the model score in the model verification process reflects the matching degree between the blood vessel model and the target blood vessel, and may also be used as a reference factor.
S152, reserving and identifying, and putting the branch blood vessels into a tree structure.
If a branch vessel meets the branch vessel condition, the branch vessel is retained, and the branch vessel is identified and classified into a tree structure.
S153, removing the branch blood vessel.
If a branch vessel is judged not to meet the branch vessel condition, the branch vessel is deleted.
In this embodiment, the analysis and determination of the branch blood vessel may be started from the starting end, or may be started from the end of the extracted target blood vessel.
Optionally, in this embodiment, each branch vessel is analyzed from the starting end of the extracted target vessel, in order to increase the speed of vessel extraction and tree structure generation, vessel extraction and tree structure generation are performed synchronously, at this time, when it is detected that a certain branch vessel does not meet the preset branch condition, the branch vessel may be stopped and deleted, thereby reducing the data computation amount of vessel extraction.
In the embodiment, the reserved branch blood vessels meeting the preset branch conditions are represented by the tree structure, and the shape of the extracted target blood vessel can be visually and vividly represented.
Example four
Fig. 10 is a flowchart of a blood vessel extraction method provided by the fourth embodiment, in the present embodiment, a reverse retrospective blood vessel naming identification method is added after the extracted target blood vessel data is updated according to the effective blood vessel model in the foregoing embodiment, as shown in fig. 10, the blood vessel extraction method includes:
s11, acquiring a plurality of prediction direction sampling points in the preset direction range of the current target blood vessel in the medical image.
And S12, obtaining modeling parameters of the blood vessel model corresponding to the predicted direction sampling points, and establishing the blood vessel model according to the modeling parameters, wherein each blood vessel model corresponds to one predicted direction sampling point.
And S13, fitting the blood vessel model to determine an effective blood vessel model.
And S14, updating the extracted target blood vessel data according to the effective blood vessel model.
The embodiment carries out blood vessel naming based on blood vessel tracking, and a blood vessel section generated by the blood vessel tracking forms a directed acyclic graph and presents a tree structure, thereby facilitating topology analysis. Taking the positioning of hepatic vein branches as an example, starting from an initial point near the first hepatic portal, searching a first bifurcation point along the directed graph, positioning as a portal artery main trunk, then searching the next bifurcation point of the sub-branches, and distributing and naming the hepatic left portal vein and the hepatic right portal vein.
Optionally, as shown in fig. 10, the reverse retroactive blood vessel naming and identifying method in this embodiment includes:
s161, determining whether there is a target blood vessel to be extracted at the end of the extracted target blood vessel, if not, executing step S162, and if so, executing step S11.
And determining whether the branch vessel where the extracted target vessel is located is extracted completely by judging whether the target vessel to be extracted exists at the tail end of the extracted target vessel.
And S162, tracking the end of the extracted target blood vessel to the starting end of the extracted target blood vessel and performing tracking identification.
When the extraction of a branch blood vessel is finished, the tail end of the branch blood vessel can be tracked to the starting end, and the tracking path is tracked and marked, so that naming check can be conveniently carried out at the later stage.
And S163, naming and identifying the extracted target blood vessel which is matched with the preset naming identification condition and is tracked and identified.
And naming the extracted target blood vessel which meets the preset naming condition according to a preset naming identification condition, wherein the preset naming identification condition is the length of the blood vessel or the preset relative length optionally.
Illustratively, for hepatic vein branch location, since the regional angiography of the hepatic vein into the inferior vena cava is uneven, when tracking in the forward direction, three branches of the hepatic vein are not in a connected graph, the naming accuracy is reduced, while the embodiment tracks from the end to the beginning of the extracted target blood vessel, the degree of incidence of all vertexes of the tracked directed graph is less than 2, so that a unique path exists for each branch blood vessel, no bifurcation exists, and when the preset naming identification condition is set to be a relative length, two longest blood vessels in all backtracking paths are named as the hepatic right vein and the hepatic middle vein according to positions.
The reverse tracing blood vessel naming identification method provided by this embodiment tracks and identifies the start end of the extracted target blood vessel along the end of the extracted target blood vessel, and the introductions of all vertexes of the tracked directed graph are less than 2, so that each branch blood vessel has a unique path, no bifurcation exists, and high accuracy, and thus blood vessels can be accurately named according to preset naming identification conditions, and accuracy of blood vessel naming is improved.
EXAMPLE five
Fig. 11 is a block diagram of a blood vessel extraction device according to a fifth embodiment, which is used to execute the blood vessel extraction method according to any of the embodiments described above, and can be implemented by software/hardware. As shown in fig. 11, the apparatus includes:
the prediction sampling point acquisition module 11 is configured to acquire a plurality of prediction direction sampling points within a preset direction range of a current target blood vessel in a medical image;
the modeling module 12 is configured to obtain modeling parameters of a blood vessel model corresponding to the predicted direction sampling points, and establish a blood vessel model according to the modeling parameters, where each blood vessel model corresponds to one predicted direction sampling point;
an effective blood vessel model determining module 13, configured to fit the blood vessel model to determine an effective blood vessel model;
and an updating module 14, configured to update the extracted target blood vessel data according to the effective blood vessel model.
Further, this embodiment further includes:
a tree structure generating module 15, configured to determine whether at least one of a radius, an angle, or a model score of each branch blood vessel meets a preset branch condition; if so, reserving and marking, and putting the branch blood vessel into a tree structure; if not, the branch vessel is removed.
A naming module 16, configured to determine whether a current target blood vessel exists at the end of the extracted target blood vessel; if not, tracking the end of the extracted target blood vessel to the starting end of the extracted target blood vessel and carrying out tracking identification; and naming and identifying the extracted target blood vessel which is consistent with the preset naming and identifying conditions and is tracked and identified.
The blood vessel extraction method and device provided by the embodiment of the invention can execute the blood vessel extraction method provided by any embodiment of the invention, and have corresponding functional modules and beneficial effects of the execution method.
According to the technical scheme of the blood vessel extraction device provided by the embodiment of the invention, a plurality of prediction direction sampling points and modeling parameters thereof in the preset direction range of the current target blood vessel in the medical image are obtained, a blood vessel model is established based on the modeling parameters, a blood vessel model which is possibly inosculated with the actual blood vessel is established, the blood vessel model is fitted to determine an effective blood vessel model, and the extracted target blood vessel data is updated according to the effective blood vessel model, so that the blood vessel can be completely extracted from the medical image, and the blood vessel extraction device has high adaptability, sensitivity and robustness.
EXAMPLE six
Fig. 12 is a block diagram of a medical imaging system according to a sixth embodiment of the present invention. As shown in fig. 12, the system comprises an image acquisition apparatus 2 and a computer device 3, wherein the image acquisition apparatus 2 is used for acquiring a medical image containing blood vessel data; the computer device 3 is connected to the image capturing apparatus 2, and as shown in fig. 13, the computer device 3 includes a processor 51, a memory 52, an input device 53, and an output device 54; the number of the processors 51 in the computer device 3 may be one or more, and one processor 51 is taken as an example in fig. 13; the processor 51, the memory 52, the input device 53, and the output device 54 in the computer apparatus 3 may be connected by a bus or other means, and the bus connection is exemplified in fig. 13.
The memory 52 is a computer-readable storage medium, and can be used for storing software programs, computer-executable programs, and modules, such as program instructions/modules (e.g., the predicted sampling point obtaining module 11, the modeling module 12, the effective blood vessel model determining module 13, and the updating module 14) corresponding to the blood vessel extracting method in the embodiment of the present invention. The processor 51 executes various functional applications of the device and data processing by running software programs, instructions and modules stored in the memory 52, that is, implements the blood vessel extraction method described above.
The memory 52 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the terminal, and the like. Further, the memory 52 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some examples, the memory 52 may further include memory located remotely from the processor 301, which may be connected to the device over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The input device 53 may be used to receive input numeric or character information and generate key signal inputs relating to user settings and function control of the apparatus.
The output device 54 may include a display device such as a display screen, for example, of a user terminal.
Example eight
An eighth embodiment of the present invention further provides a storage medium containing computer-executable instructions, which when executed by a computer processor, are configured to perform a blood vessel extraction method, including:
in a medical image, acquiring a plurality of predicted direction sampling points in a preset direction range of a current target blood vessel;
obtaining modeling parameters of a blood vessel model corresponding to the predicted direction sampling points, and establishing the blood vessel model according to the modeling parameters, wherein each blood vessel model corresponds to one predicted direction sampling point;
fitting the blood vessel model to determine an effective blood vessel model;
and updating the extracted target blood vessel data according to the effective blood vessel model.
Of course, the storage medium provided by the embodiment of the present invention contains computer-executable instructions, and the computer-executable instructions are not limited to the method operations described above, and may also perform related operations in the blood vessel extraction method provided by any embodiment of the present invention.
From the above description of the embodiments, it is obvious for those skilled in the art that the present invention can be implemented by software and necessary general hardware, and certainly, can also be implemented by hardware, but the former is a better embodiment in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a FLASH Memory (FLASH), a hard disk or an optical disk of a computer, and includes several instructions to enable a computer device (which may be a personal computer, a server, or a network device) to execute the blood vessel extraction method according to the embodiments of the present invention.
It should be noted that, in the embodiment of the blood vessel extraction device, the included units and modules are only divided according to the functional logic, but are not limited to the above division as long as the corresponding functions can be realized; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present invention.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (8)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711245630.5A CN108022237B (en) | 2017-11-30 | 2017-11-30 | Blood vessel extraction method, system and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711245630.5A CN108022237B (en) | 2017-11-30 | 2017-11-30 | Blood vessel extraction method, system and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108022237A CN108022237A (en) | 2018-05-11 |
| CN108022237B true CN108022237B (en) | 2021-07-13 |
Family
ID=62077869
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711245630.5A Active CN108022237B (en) | 2017-11-30 | 2017-11-30 | Blood vessel extraction method, system and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108022237B (en) |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109544543B (en) * | 2018-11-29 | 2021-08-27 | 上海联影医疗科技股份有限公司 | Blood vessel identification method, terminal and readable medium |
| CN110223271B (en) * | 2019-04-30 | 2022-11-15 | 深圳市阅影科技有限公司 | Method and device for automatic level set segmentation of blood vessel images |
| CN112307804B (en) * | 2019-07-25 | 2024-09-17 | 宏碁股份有限公司 | Vascular state evaluation method and vascular state evaluation device |
| CN111640124B (en) * | 2020-05-25 | 2023-06-02 | 浙江同花顺智能科技有限公司 | Blood vessel extraction method, device, equipment and storage medium |
| CN111738986B (en) * | 2020-06-01 | 2021-02-09 | 数坤（北京）网络科技有限公司 | Fat attenuation index generation method and device and computer readable medium |
| CN112381758B (en) * | 2020-10-09 | 2024-01-30 | 北京师范大学 | Method for calculating similarity of blood vessel tree |
| CN112184690B (en) * | 2020-10-12 | 2021-11-02 | 推想医疗科技股份有限公司 | Coronary vessel trend prediction method, prediction model training method and device |
| CN112150465B (en) * | 2020-11-24 | 2022-05-13 | 数坤(北京)网络科技股份有限公司 | Blood vessel naming method and device |
| CN113421218B (en) * | 2021-04-16 | 2024-02-23 | 深圳大学 | Extraction method of vascular network branch point |
| CN114723684B (en) * | 2022-03-22 | 2023-03-24 | 推想医疗科技股份有限公司 | Model training method and device, and vascular structure generation method and device |
| CN115953495B (en) * | 2023-03-14 | 2023-08-01 | 北京唯迈医疗设备有限公司 | Intelligent path planning device, system and storage medium based on two-dimensional contrast image |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1954340A (en) * | 2004-05-18 | 2007-04-25 | 皇家飞利浦电子股份有限公司 | Image processing system for automatic segmentation of a 3-d tree-like tubular surface of an object, using 3-d deformable mesh models |
| CN101393644A (en) * | 2008-08-15 | 2009-03-25 | 华中科技大学 | A method and system for modeling hepatic portal vein vascular tree |
| CN104182956A (en) * | 2013-05-21 | 2014-12-03 | 上海联影医疗科技有限公司 | Blood vessel extraction method |
| CN104361626A (en) * | 2014-09-29 | 2015-02-18 | 北京理工大学 | Subcutaneous vein three-dimensional reconstruction method based on hybrid matching strategy |
| CN105286779A (en) * | 2015-10-30 | 2016-02-03 | 温州医科大学 | Method of imaging of in vivo retina haemodynamics and measuring of absolute flow velocity |
| CN107067409A (en) * | 2017-05-09 | 2017-08-18 | 上海联影医疗科技有限公司 | A blood vessel separation method and system |
| CN107203741A (en) * | 2017-05-03 | 2017-09-26 | 上海联影医疗科技有限公司 | Vessel extraction method, device and its system |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7602970B2 (en) * | 2005-03-21 | 2009-10-13 | Siemens Medical Solutions Usa, Inc. | System and method for Kalman filtering in vascular segmentation |
| US8315812B2 (en) * | 2010-08-12 | 2012-11-20 | Heartflow, Inc. | Method and system for patient-specific modeling of blood flow |
-
2017
- 2017-11-30 CN CN201711245630.5A patent/CN108022237B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1954340A (en) * | 2004-05-18 | 2007-04-25 | 皇家飞利浦电子股份有限公司 | Image processing system for automatic segmentation of a 3-d tree-like tubular surface of an object, using 3-d deformable mesh models |
| CN101393644A (en) * | 2008-08-15 | 2009-03-25 | 华中科技大学 | A method and system for modeling hepatic portal vein vascular tree |
| CN104182956A (en) * | 2013-05-21 | 2014-12-03 | 上海联影医疗科技有限公司 | Blood vessel extraction method |
| CN104361626A (en) * | 2014-09-29 | 2015-02-18 | 北京理工大学 | Subcutaneous vein three-dimensional reconstruction method based on hybrid matching strategy |
| CN105286779A (en) * | 2015-10-30 | 2016-02-03 | 温州医科大学 | Method of imaging of in vivo retina haemodynamics and measuring of absolute flow velocity |
| CN107203741A (en) * | 2017-05-03 | 2017-09-26 | 上海联影医疗科技有限公司 | Vessel extraction method, device and its system |
| CN107067409A (en) * | 2017-05-09 | 2017-08-18 | 上海联影医疗科技有限公司 | A blood vessel separation method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108022237A (en) | 2018-05-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108022237B (en) | Blood vessel extraction method, system and storage medium | |
| US20210272030A1 (en) | Systems and methods for using geometry sensitivity information for guiding workflow | |
| CN109741344B (en) | Blood vessel segmentation method, blood vessel segmentation device, medical imaging equipment and storage medium | |
| US10115039B2 (en) | Method and system for machine learning based classification of vascular branches | |
| CN111696089B (en) | Arteriovenous determination method, device, equipment and storage medium | |
| CN107945168B (en) | Medical image processing method and medical image processing system | |
| US12165309B2 (en) | Deformity edge detection | |
| CN109712131B (en) | Method and device for quantifying lung nodule characteristics, electronic equipment and storage medium | |
| US20140355858A1 (en) | Vascular anatomy modeling derived from 3-dimensional medical image processing | |
| CN111145173A (en) | Plaque identification method, device, equipment and medium for coronary angiography image | |
| CN110929728B (en) | Image region-of-interest dividing method, image segmentation method and device | |
| CN116051580A (en) | A blood vessel separation method and system | |
| CN107274402A (en) | A kind of Lung neoplasm automatic testing method and system based on chest CT image | |
| CN111768411A (en) | Coronary artery central line extraction method and device, computer equipment and storage medium | |
| CN111178420B (en) | Coronary artery segment marking method and system on two-dimensional contrast image | |
| CN110211200B (en) | Dental arch wire generating method and system based on neural network technology | |
| CN110738702B (en) | Three-dimensional ultrasonic image processing method, device, equipment and storage medium | |
| US10354349B2 (en) | Systems and methods for using geometry sensitivity information for guiding workflow | |
| JPWO2016159379A1 (en) | Blood vessel shape construction apparatus, method and computer software program | |
| KR101126224B1 (en) | Prostate segmentation method using dynamic mr images and system thereof | |
| CN117830531A (en) | Simulation model construction method and system based on coronary artery ablation | |
| CN114972257A (en) | Coronary artery segmentation method, device, storage medium and electronic device | |
| JPWO2020054019A1 (en) | Image processing equipment, image processing methods, and programs | |
| CN114880960A (en) | Method for evaluating radioactive embolism dose and injection position based on fluid dynamics | |
| CN120107514B (en) | Multistage vascular network reconstruction and segmentation method based on drainage basin analysis |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 201807 Shanghai City, north of the city of Jiading District Road No. 2258 Applicant after: Shanghai Lianying Medical Technology Co., Ltd Address before: 201807 Shanghai City, north of the city of Jiading District Road No. 2258 Applicant before: SHANGHAI UNITED IMAGING HEALTHCARE Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |