CN117036640B - Coronary artery blood vessel model reconstruction method, device, equipment and storage medium - Google Patents
Coronary artery blood vessel model reconstruction method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN117036640B CN117036640B CN202311302080.1A CN202311302080A CN117036640B CN 117036640 B CN117036640 B CN 117036640B CN 202311302080 A CN202311302080 A CN 202311302080A CN 117036640 B CN117036640 B CN 117036640B
- Authority
- CN
- China
- Prior art keywords
- data
- point data
- central
- value
- center point
- 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
- 210000004351 coronary vessel Anatomy 0.000 title claims abstract description 70
- 238000000034 method Methods 0.000 title claims abstract description 51
- 210000004204 blood vessel Anatomy 0.000 claims abstract description 45
- 238000013528 artificial neural network Methods 0.000 claims abstract description 14
- 238000005070 sampling Methods 0.000 claims description 33
- 239000011159 matrix material Substances 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 14
- 238000002586 coronary angiography Methods 0.000 claims description 14
- 238000013075 data extraction Methods 0.000 claims description 4
- 230000001131 transforming effect Effects 0.000 claims description 4
- 230000002792 vascular Effects 0.000 claims description 3
- 238000002601 radiography Methods 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 4
- 208000004434 Calcinosis Diseases 0.000 description 2
- 230000002308 calcification Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 208000031481 Pathologic Constriction Diseases 0.000 description 1
- 238000002583 angiography Methods 0.000 description 1
- 238000004195 computer-aided diagnosis Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000002059 diagnostic imaging Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000007670 refining Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 208000037804 stenosis Diseases 0.000 description 1
- 230000036262 stenosis Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/10—Constructive solid geometry [CSG] using solid primitives, e.g. cylinders, cubes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T19/00—Manipulating 3D models or images for computer graphics
- G06T19/20—Editing of 3D images, e.g. changing shapes or colours, aligning objects or positioning parts
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
- G06T7/344—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods involving models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/68—Analysis of geometric attributes of symmetry
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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; CALCULATING OR 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- Computer Graphics (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Architecture (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The application relates to a coronary artery blood vessel model reconstruction method, a device, equipment and a storage medium, wherein an initial central path of a blood vessel and central point data on the central path are obtained by extracting original three-dimensional image data of coronary artery radiography through a deep neural network, then the central point data are grouped to obtain a plurality of groups of central point data, CT value data and fitting data corresponding to the groups of central point data are constructed in the groups of central point data by calculating CT values of the central points at corresponding positions in the original three-dimensional image data, dislocation sites are positioned in the groups of central point data, corrected central paths and central point data are obtained by correcting all dislocation points, and finally a coronary artery blood vessel model is obtained by reconstructing the corrected central paths and the central point data. The method can automatically correct the central line of the blood vessel so as to save time and labor cost.
Description
Technical Field
The present disclosure relates to the field of medical imaging technologies, and in particular, to a method, an apparatus, a device, and a storage medium for reconstructing a coronary artery blood vessel model.
Background
Computer aided diagnosis techniques based on coronary angiography (CTA) images have played an increasingly important role in medical diagnostics. Centerline-based vessel segmentation is a complex and challenging task. The extraction of the central line of the coronary artery is the basis and premise of realizing coronary artery reconstruction and quantitative stenosis detection. However, the complex geometry of the heart's coronary arteries, the relatively small vessels, and the presence of calcifications make the tracking of the vessel centerline challenging. While the accuracy of the coronary model reconstruction is largely dependent on the accuracy of the centerline determination.
Disclosure of Invention
In view of the above, it is desirable to provide a coronary artery model reconstruction method capable of automatically correcting a blood vessel center line.
A method of coronary vessel model reconstruction, the method comprising:
acquiring original three-dimensional image data of coronary angiography;
extracting an initial central path of a blood vessel and central point data on the central path based on the original three-dimensional image data by using a deep neural network;
grouping the central point data according to three main branches of the coronary artery and the mode of the other side branches to obtain a plurality of groups of central point data;
in each set of center point data, calculating CT values of each center point at corresponding positions in the original three-dimensional image data, constructing CT value data corresponding to each set of center point data according to the CT values corresponding to each center point, and performing curve fitting on the CT value data by adopting a least square method to obtain fitting data corresponding to each set of center point data;
processing according to CT value data and fitting data corresponding to the central point data of each group, positioning dislocation sites in the central point data of each group, and correcting all dislocation sites to obtain corrected central paths and central point data;
and reconstructing according to the corrected central path and the corrected central point data to obtain a coronary artery blood vessel model.
In one embodiment, the processing according to the CT value data and the fitting data corresponding to each set of center point data, and locating the dislocation point in each set of center point data includes:
respectively extracting a CT value and a fitting value corresponding to the same center point from the CT value data and the fitting data;
and calculating a difference value between the CT value and the fitting value corresponding to the same central point, and if the difference value is larger than a first preset threshold value, determining the central point as an error point.
In this embodiment, the processing according to the CT value data and the fitting data corresponding to each set of center point data, and locating the dislocation point in each set of center point data includes:
and calculating the average value of the CT value data corresponding to each group of center point data, and calculating the difference value between each CT value in each group of CT value data and the corresponding average value, wherein if the difference value is larger than a second preset threshold value, the corresponding center point is an error point.
In one embodiment, the center point data includes three-dimensional coordinates of each of the center points and a radius.
In one embodiment, the correcting the corrected center path and the center point data by correcting all the dislocation points includes:
generating a cutting matrix according to the three-dimensional coordinates of the error points, and cutting the original three-dimensional image data by using the cutting matrix to obtain a corresponding two-dimensional section image;
dividing a search area by taking the midpoint of the two-dimensional section image as a circle center and taking a preset initial value as a radius;
sampling is carried out in the search area, so that a plurality of sampling points are obtained;
judging the CT value corresponding to each sampling point, and recording the sampling points corresponding to the CT values meeting the preset conditions of the blood vessels;
taking the average position coordinates of all sampling points meeting the preset conditions of the blood vessels as the coordinates of the central point of the corrected two-dimensional section image corresponding to the error point;
transforming the corrected center point coordinates on the two-dimensional tangent plane image according to the cutting matrix to obtain three-dimensional coordinates corrected by the error points;
and replacing the three-dimensional coordinates of each error point with the corrected three-dimensional coordinates in each group of center point data to obtain corrected center paths and center point data.
In one embodiment, if no sampling points meeting the preset condition of the blood vessel are searched in the searching area divided by taking the preset initial value as the radius, the searching area is divided again by taking the preset initial value which is N times as the radius, and the sampling points meeting the preset condition of the blood vessel are searched in the re-divided searching area, wherein N is more than or equal to 2;
if the sampling points meeting the preset conditions of the blood vessels are not searched in the repartitioned search area, the center point data of the error points are not corrected.
In one embodiment, the reconstructing the coronary artery blood vessel model according to the corrected central path and the central point data includes:
generating a coronary artery pipeline model according to the corrected center point data;
converting the coronary artery pipeline model into binary image data;
and reconstructing by using a level set algorithm based on the binary image data to obtain a coronary artery blood vessel model.
A coronary vessel model reconstruction apparatus, the apparatus comprising:
the image data acquisition module is used for acquiring original three-dimensional image data of coronary angiography;
the central point data extraction module is used for extracting an initial central path of the blood vessel and central point data on the central path based on the original three-dimensional image data by using a deep neural network;
the central point data grouping module is used for grouping the central point data according to three main branches of the coronary artery and the mode of the other side branches to obtain a plurality of groups of central point data;
the intermediate data processing module is used for calculating CT values of the central points at the corresponding positions in the original three-dimensional image data in the central point data of each group, constructing and obtaining CT value data of the central point data of each group according to the CT values corresponding to the central points, and performing curve fitting on the CT value data by adopting a least square method to obtain fitting data of the central point data of each group;
the central point data correction module is used for processing according to CT value data and fitting data corresponding to each group of central point data, locating dislocation sites in each group of central point data, and correcting all dislocation points to obtain corrected central paths and central point data;
and the blood vessel model reconstruction module is used for reconstructing and obtaining a coronary artery blood vessel model according to the corrected central path and the central point data.
A computer device comprising a memory storing a computer program and a processor which when executing the computer program performs the steps of:
acquiring original three-dimensional image data of coronary angiography;
extracting an initial central path of a blood vessel and central point data on the central path based on the original three-dimensional image data by using a deep neural network;
grouping the central point data according to three main branches of the coronary artery and the mode of the other side branches to obtain a plurality of groups of central point data;
in each set of center point data, calculating CT values of each center point at corresponding positions in the original three-dimensional image data, constructing CT value data corresponding to each set of center point data according to the CT values corresponding to each center point, and performing curve fitting on the CT value data by adopting a least square method to obtain fitting data corresponding to each set of center point data;
processing according to CT value data and fitting data corresponding to the central point data of each group, positioning dislocation sites in the central point data of each group, and correcting all dislocation sites to obtain corrected central paths and central point data;
and reconstructing according to the corrected central path and the corrected central point data to obtain a coronary artery blood vessel model.
A computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of:
acquiring original three-dimensional image data of coronary angiography;
extracting an initial central path of a blood vessel and central point data on the central path based on the original three-dimensional image data by using a deep neural network;
grouping the central point data according to three main branches of the coronary artery and the mode of the other side branches to obtain a plurality of groups of central point data;
in each set of center point data, calculating CT values of each center point at corresponding positions in the original three-dimensional image data, constructing CT value data corresponding to each set of center point data according to the CT values corresponding to each center point, and performing curve fitting on the CT value data by adopting a least square method to obtain fitting data corresponding to each set of center point data;
processing according to CT value data and fitting data corresponding to the central point data of each group, positioning dislocation sites in the central point data of each group, and correcting all dislocation sites to obtain corrected central paths and central point data;
and reconstructing according to the corrected central path and the corrected central point data to obtain a coronary artery blood vessel model.
According to the coronary artery blood vessel model reconstruction method, the device, the equipment and the storage medium, the original three-dimensional image data of the obtained coronary artery angiography is extracted through the deep neural network, the initial central path of the blood vessel and the central point data on the central path are obtained, the central point data are grouped according to three main branches of the coronary artery and the mode of the other main branches to obtain multiple groups of central point data, CT values of all central points at corresponding positions in the original three-dimensional image data are calculated in all groups of central point data, CT value data corresponding to all groups of central point data are constructed according to the CT values corresponding to all central points, curve fitting is conducted on the CT value data by a least square method to obtain fitting data corresponding to all groups of central point data, dislocation sites are fixed in all groups of central point data, corrected central paths and central point data are obtained through automatic correction on all dislocation points, and finally a blood vessel model is obtained according to the corrected central path and the central point data. By adopting the method, the center line of the blood vessel can be automatically corrected, thereby avoiding complex manual processing operation, saving time and labor cost and improving the accuracy of the blood vessel model constructed based on the corrected center line.
Drawings
FIG. 1 is a flow chart of a method of reconstructing a coronary vessel model in one embodiment;
FIG. 2 is a schematic diagram of an initial central path extracted using a deep neural network in one embodiment;
FIG. 3 is a schematic diagram of an automatic bit error point determination in one embodiment;
FIG. 4 is a schematic diagram of a two-dimensional section image obtained by cutting original three-dimensional image data with a cutting matrix for an error center point in one embodiment;
FIG. 5 is a schematic diagram of a center point correction performed on a two-dimensional slice image for a false center in one embodiment;
FIG. 6 is a schematic illustration of a corrected vessel center path in one embodiment;
FIG. 7 is a block diagram of a coronary vessel model reconstruction device in one embodiment;
fig. 8 is an internal structural diagram of a computer device in one embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.
Aiming at the problem that the accuracy of coronary artery model reconstruction depends on the accuracy of a central line to a great extent, in the prior art, a correction mode of a blood vessel central line is usually to manually select the central line to be corrected and then carry out interactive modification, but the method needs manual operation and has poor robustness, or the method adopts traversing all central lines, obtains corresponding two-dimensional section data through data of all points on the central line, then respectively carries out tasks of blood vessel segmentation to obtain a central point of segmented data as a new central point, but the method takes longer time, and the method for reconstructing the coronary artery blood vessel model is provided, and comprises the following steps as shown in figure 1:
step S100, original three-dimensional image data of coronary angiography is obtained;
step S110, extracting an initial central path of a blood vessel and central point data on the central path based on the original three-dimensional image data by using a deep neural network;
step S120, grouping the central point data according to three main branches of the coronary artery and the mode of the other side branches to obtain a plurality of groups of central point data;
step S130, calculating CT values of the center points at the corresponding positions in the original three-dimensional image data in the center point data of each group, constructing and obtaining CT value data of the center point data of each group according to the CT values corresponding to the center points, and performing curve fitting on the CT value data by adopting a least square method to obtain fitting data of the center point data of each group;
step S140, processing according to CT value data and fitting data corresponding to each group of center point data, positioning dislocation sites in each group of center point data, and correcting all dislocation sites to obtain corrected center paths and center point data;
and step S150, reconstructing according to the corrected central path and the central point data to obtain a coronary artery blood vessel model.
In this embodiment, by designing a device that can automatically identify the error point on the initially generated vessel centerline and then correct the coordinate of the error point, the purpose of correcting the vessel centerline is achieved, so that the vessel centerline can be obtained quickly and accurately, a more accurate three-dimensional vessel model can be obtained later, and the time and labor cost in the whole process can be greatly reduced.
In step S100, the data processed in the present method is original three-dimensional image data of a coronary artery obtained by using a contrast technique, the original three-dimensional image data including a plurality of coronary angiography images ordered by time.
In step S110, the trained deep neural network is used to process the input original three-dimensional image data of coronary angiography, so as to obtain an initial central path of the coronary vessel, as shown in fig. 2, and point data of each central point on the central path, namely, central point data. Since the coronary vessels are complex, there is a certain error between the obtained initial vessel center path and the actual vessel center line, so that it is necessary to find and correct the center point having the error on the initial vessel center path.
In step S120, in order to facilitate the subsequent identification of the error points, the obtained central data are recombined according to the manner that three vessels (left anterior descending branch, left circumflex branch, right coronary artery) of the coronary artery are taken as main branches and the rest are side branches, so as to obtain multiple sets of central point data, and corresponding multiple vessel central paths can be obtained according to the central point data of each set.
In step S130, after processing each set of center point data, corresponding CT value data and fitting data are obtained, and then processing is performed according to the CT value data and fitting data corresponding to each set of center point data, where locating the offset point in each set of center point data includes: and respectively extracting the CT value and the fitting data corresponding to the same central point from the CT value data and the fitting data, and calculating the difference value between the CT value and the fitting value corresponding to the same central point, wherein if the difference value is larger than a first preset threshold value, the central point is an error point, as shown in fig. 3.
Specifically, for convenience of explanation, the CT value data and the fitting data corresponding to the same set of center point data are replaced by a and B, respectively, and the values of the same center point positions in a and B are taken out and respectively recorded as a and B. If the difference between a and b is greater than the first preset threshold value alpha, the point coordinate is recorded and marked as c, namely the error point.
In one embodiment, the first preset threshold α may be set to 400.
In this embodiment, the processing is performed according to CT value data and fitting data corresponding to each set of center point data, and the positioning of the dislocation point in each set of center point data may further adopt the following manner, which specifically includes: and calculating the average value of the CT value data corresponding to each group of center point data, and calculating the difference value between each CT value in each group of CT value data and the corresponding average value, wherein if the difference value is larger than a second preset threshold value, the corresponding center point is an error point.
Specifically, the CT value data and fitting data corresponding to the same group of center point data are replaced by A and B respectively, the average value of A is calculated and recorded as beta, the difference value of the CT value and the beta of each center point in A is compared, and if the difference value is larger than a set second preset threshold value gamma, the point is recorded as c.
In one embodiment, the second preset threshold γ may be preset to 350.
The method is adopted to traverse the center point data corresponding to all the blood vessel branches, and a set of all error points C is obtained and marked as C.
In step S140, the center point data corresponding to each error point in the error point set C is corrected, where each center point data includes the three-dimensional coordinates and the radius of each center point. In the correction process, the three-dimensional position coordinates of the error point are actually corrected.
In this embodiment, the correction method for all the error points is the same, so that the specific correction step can be described by taking one error point as an example, and the specific steps include generating a cutting matrix according to the three-dimensional coordinates of the error point, cutting the original three-dimensional image data by using the cutting matrix to obtain a corresponding two-dimensional tangent plane image, as shown in fig. 4, dividing a search area by taking a midpoint of the two-dimensional tangent plane image as a center, taking a preset initial value as a radius, sampling in the search area to obtain a plurality of sampling points, judging CT values corresponding to each sampling point, recording the sampling points corresponding to CT values meeting the preset condition of the blood vessel, taking the average position coordinates of all the sampling points meeting the preset condition of the blood vessel as the corrected center point coordinates of the corresponding error point on the two-dimensional tangent plane image, transforming the corrected center point coordinates on the two-dimensional tangent plane image as shown in fig. 5, and obtaining the three-dimensional coordinates after the correction of the error point according to the cutting matrix.
In this embodiment, a corresponding two-dimensional section is obtained from the original three-dimensional image data according to the three-dimensional coordinates of the error point, the center point position is repositioned in the two-dimensional section, and the center point position obtained by repositioning is remapped to the three-dimensional coordinates, so that corrected three-dimensional coordinates of the center point are obtained.
Specifically, a tangent vector, a normal vector and a secondary normal vector of a point c are calculated, a cutting matrix is generated according to the three-dimensional coordinates of the point c, and the original CTA image is cut by using the cutting matrix, so that a two-dimensional tangent plane image is obtained. And taking the midpoint of the two-dimensional tangent plane as the center of a circle, and carrying out annular sampling by using a preset initial radius value to obtain fixed number of sampling point data. And judging the CT value of each sampling point data in turn, recording the position coordinates meeting the blood vessel threshold condition, calculating the average positions of all the points meeting the condition, and recording the average positions as corrected center point coordinates.
In this embodiment, if no sampling points meeting the preset condition of the blood vessel are searched in the search area divided by taking the preset initial value as the radius, the search area is divided again by taking the preset initial value which is N times as the radius, and the sampling points meeting the preset condition of the blood vessel are searched in the divided search area again, wherein N is greater than or equal to 2.
Specifically, if no sampling point meeting the preset condition of the blood vessel is searched in the search area divided by taking the preset value as the radius, the search area is divided again by taking the preset value which is 2 times as the radius, if not found, the search area is divided again by taking the preset value which is 3 times as the radius, and the search is continued until the preset multiple is reached.
If sampling points meeting the preset conditions of the blood vessels are not searched in the search areas divided for a plurality of times, the center point data of the error points are not corrected. This is the case because the two-dimensional cross-sections of the points are calcified, containing severe calcification artefacts. The blood vessel is actually contained within the calcified region and no correction is required.
The method is adopted to correct the center point data corresponding to each error point in the error point combination set C, after corrected three-dimensional coordinate data are obtained, the three-dimensional coordinates of each error point are replaced by the corrected three-dimensional coordinates, and corrected center paths are obtained as shown in figure 6, and the center point data are obtained.
In step S150, reconstructing the coronary artery blood vessel model according to the corrected central path and the central point data includes: firstly, generating a coronary artery pipeline model according to corrected central point data, converting the coronary artery pipeline model into binary image data, and finally, reconstructing the binary image data by using a level set algorithm to obtain a coronary artery vascular model.
Specifically, the specific step of converting the coronary arterial tree conduit model into binary image data comprises: a blank image is generated based on information of original three-dimensional image data including three-dimensional size, pixel pitch, start point coordinate information, and the like of each coronary angiography image. In generating a blank image, pixel values are set to a background color.
And then cutting the coronary artery tree pipeline model data along the Z-axis plane in the three-dimensional coordinate to obtain a plurality of intersecting line data. And according to the connection of each intersecting line data into a polygon, generating a line segment for two point data in the polygon, recording the integral point coordinates of the line segment, and then obtaining the point data meeting the two-dimensional condition. Traversing intersection line data of all Z-axis tangential planes to obtain all point data meeting the conditions on the model.
And assigning the dot data coordinates meeting the conditions on the blank image, and setting the pixel value as the foreground color. In the present embodiment, the background color is set to-1, and the foreground color is set to 1.
Next, reconstructing the vessel model using a level set algorithm includes: taking the binary image data as an initial level set, setting proper parameters, gradually expanding and refining the surface of a blood vessel through evolution of a level set function to obtain a blood vessel model, carrying out a surface reconstruction algorithm, such as a cube algorithm, to obtain a three-dimensional surface of the blood vessel model, and finally reconstructing to obtain the three-dimensional blood vessel model of the coronary artery.
According to the coronary artery blood vessel model reconstruction method, the region with the error tracking of the central line can be automatically positioned, so that the operation of manually selecting the central line to be corrected is avoided, the automatic correction of the central line can be completed in a short time, the accuracy of model reconstruction is improved while higher accuracy and robustness are achieved, the time and labor cost can be effectively reduced when the method is adopted for practical application, and the accuracy and efficiency of follow-up work are improved.
It should be understood that, although the steps in the flowchart of fig. 1 are shown in sequence as indicated by the arrows, the steps are not necessarily performed in sequence as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in fig. 1 may include multiple sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, nor do the order in which the sub-steps or stages are performed necessarily performed in sequence, but may be performed alternately or alternately with at least a portion of other steps or sub-steps of other steps.
In one embodiment, as shown in fig. 7, there is provided a coronary artery blood vessel model reconstruction device including: an image data acquisition module 200, a center point data extraction module 210, a center point data grouping module 220, an intermediate data processing module 230, a center point data correction module 240, and a blood vessel model reconstruction module 250, wherein:
the image data acquisition module 200 is used for acquiring original three-dimensional image data of coronary angiography;
the central point data extraction module 210 is configured to extract an initial central path of a blood vessel and central point data on the central path based on the original three-dimensional image data by using a deep neural network;
a central point data grouping module 220, configured to group the central point data according to three main branches of the coronary artery and the other branch modes, so as to obtain multiple groups of central point data;
the intermediate data processing module 230 is configured to calculate CT values of the center points at corresponding positions in the original three-dimensional image data in each set of center point data, construct CT value data corresponding to each set of center point data according to the CT values corresponding to the center points, and perform curve fitting on the CT value data by using a least square method to obtain fitting data corresponding to each set of center point data;
the central point data correction module 240 is configured to process the CT value data and the fitting data corresponding to each set of central point data, fix dislocation sites in each set of central point data, and obtain corrected central paths and central point data by correcting all dislocation points;
the blood vessel model reconstruction module 250 is configured to reconstruct and obtain a coronary artery blood vessel model according to the corrected central path and the central point data.
For specific limitations of the coronary vessel model reconstruction device, reference may be made to the above limitations of the coronary vessel model reconstruction method, and no further description is given here. The above-described respective modules in the coronary vessel model reconstruction apparatus may be realized in whole or in part by software, hardware, or a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, a computer device is provided, which may be a terminal, and the internal structure thereof may be as shown in fig. 8. The computer device includes a processor, a memory, a network interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a coronary vessel model reconstruction method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, can also be keys, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.
It will be appreciated by those skilled in the art that the structure shown in fig. 8 is merely a block diagram of some of the structures associated with the present application and is not limiting of the computer device to which the present application may be applied, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer device is provided comprising a memory and a processor, the memory having stored therein a computer program, the processor when executing the computer program performing the steps of:
acquiring original three-dimensional image data of coronary angiography;
extracting an initial central path of a blood vessel and central point data on the central path based on the original three-dimensional image data by using a deep neural network;
grouping the central point data according to three main branches of the coronary artery and the mode of the other side branches to obtain a plurality of groups of central point data;
in each set of center point data, calculating CT values of each center point at corresponding positions in the original three-dimensional image data, constructing CT value data corresponding to each set of center point data according to the CT values corresponding to each center point, and performing curve fitting on the CT value data by adopting a least square method to obtain fitting data corresponding to each set of center point data;
processing according to CT value data and fitting data corresponding to the central point data of each group, positioning dislocation sites in the central point data of each group, and correcting all dislocation sites to obtain corrected central paths and central point data;
and reconstructing according to the corrected central path and the corrected central point data to obtain a coronary artery blood vessel model.
In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of:
acquiring original three-dimensional image data of coronary angiography;
extracting an initial central path of a blood vessel and central point data on the central path based on the original three-dimensional image data by using a deep neural network;
grouping the central point data according to three main branches of the coronary artery and the mode of the other side branches to obtain a plurality of groups of central point data;
in each set of center point data, calculating CT values of each center point at corresponding positions in the original three-dimensional image data, constructing CT value data corresponding to each set of center point data according to the CT values corresponding to each center point, and performing curve fitting on the CT value data by adopting a least square method to obtain fitting data corresponding to each set of center point data;
processing according to CT value data and fitting data corresponding to the central point data of each group, positioning dislocation sites in the central point data of each group, and correcting all dislocation sites to obtain corrected central paths and central point data;
and reconstructing according to the corrected central path and the corrected central point data to obtain a coronary artery blood vessel model.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the various embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples merely represent a few embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the invention. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application is to be determined by the claims appended hereto.
Claims (9)
1. A method of reconstructing a coronary vessel model, the method comprising:
acquiring original three-dimensional image data of coronary angiography;
extracting an initial central path of a blood vessel and central point data on the central path based on the original three-dimensional image data by using a deep neural network;
grouping the central point data according to three main branches of the coronary artery and the mode of the other side branches to obtain a plurality of groups of central point data;
in each set of center point data, calculating CT values of each center point at corresponding positions in the original three-dimensional image data, constructing CT value data corresponding to each set of center point data according to the CT values corresponding to each center point, and performing curve fitting on the CT value data by adopting a least square method to obtain fitting data corresponding to each set of center point data;
processing is carried out according to CT value data and fitting data corresponding to each group of center point data, bit error points are determined in each group of center point data, and corrected center paths and center point data are obtained by correcting all error points, wherein the processing comprises the following steps: generating a cutting matrix according to the three-dimensional coordinates of the error points, cutting the original three-dimensional image data by using the cutting matrix to obtain a corresponding two-dimensional section image, and dividing a search area by taking the midpoint of the two-dimensional section image as a circle center and taking a preset initial value as a radius; sampling is carried out in the search area, so that a plurality of sampling points are obtained; judging the CT value corresponding to each sampling point, and recording the sampling points corresponding to the CT values meeting the preset conditions of the blood vessels; taking the average position coordinates of all sampling points meeting the preset conditions of blood vessels as corrected center point coordinates of corresponding error points on a two-dimensional tangent plane image, transforming the corrected center point coordinates on the two-dimensional tangent plane image according to the cutting matrix to obtain corrected three-dimensional coordinates of the error points, and replacing the three-dimensional coordinates of each error point with the corrected three-dimensional coordinates in each set of center point data to obtain corrected center paths and center point data;
and reconstructing according to the corrected central path and the corrected central point data to obtain a coronary artery blood vessel model.
2. The method of reconstructing a coronary vessel model as set forth in claim 1, wherein the processing based on the CT value data and the fitting data corresponding to each set of center point data, locating the error point in each set of center point data includes:
respectively extracting a CT value and a fitting value corresponding to the same center point from the CT value data and the fitting data;
and calculating a difference value between the CT value and the fitting value corresponding to the same central point, and if the difference value is larger than a first preset threshold value, determining the central point as an error point.
3. The method of reconstructing a coronary vessel model as set forth in claim 1, wherein the processing based on the CT value data and the fitting data corresponding to each set of center point data, locating the error point in each set of center point data includes:
and calculating the average value of the CT value data corresponding to each group of center point data, and calculating the difference value between each CT value in each group of CT value data and the corresponding average value, wherein if the difference value is larger than a second preset threshold value, the corresponding center point is an error point.
4. A coronary vessel model reconstruction method according to any one of claims 2 or 3, wherein the center point data includes three-dimensional coordinates of each of the center points and a radius.
5. The method for reconstructing a coronary vessel model as recited in claim 4,
if no sampling points meeting the preset vascular conditions are searched in the searching area divided by taking the preset initial value as the radius, the searching area is divided again by taking the preset initial value which is N times as the radius, and the sampling points meeting the preset vascular conditions are searched in the re-divided searching area, wherein N is more than or equal to 2;
if the sampling points meeting the preset conditions of the blood vessels are not searched in the repartitioned search area, the center point data of the error points are not corrected.
6. The method of claim 5, wherein reconstructing the coronary vessel model from the corrected central path and the central point data comprises:
generating a coronary artery pipeline model according to the corrected center point data;
converting the coronary artery pipeline model into binary image data;
and reconstructing by using a level set algorithm based on the binary image data to obtain a coronary artery blood vessel model.
7. A coronary vessel model reconstruction device, the device comprising:
the image data acquisition module is used for acquiring original three-dimensional image data of coronary angiography;
the central point data extraction module is used for extracting an initial central path of the blood vessel and central point data on the central path based on the original three-dimensional image data by using a deep neural network;
the central point data grouping module is used for grouping the central point data according to three main branches of the coronary artery and the mode of the other side branches to obtain a plurality of groups of central point data;
the intermediate data processing module is used for calculating CT values of the central points at the corresponding positions in the original three-dimensional image data in the central point data of each group, constructing and obtaining CT value data of the central point data of each group according to the CT values corresponding to the central points, and performing curve fitting on the CT value data by adopting a least square method to obtain fitting data of the central point data of each group;
the central point data correction module is used for processing according to CT value data and fitting data corresponding to each group of central point data, determining error points in each group of central point data, and obtaining corrected central paths and central point data by correcting all error points, and comprises the following steps: generating a cutting matrix according to the three-dimensional coordinates of the error points, cutting the original three-dimensional image data by using the cutting matrix to obtain a corresponding two-dimensional section image, and dividing a search area by taking the midpoint of the two-dimensional section image as a circle center and taking a preset initial value as a radius; sampling is carried out in the search area, so that a plurality of sampling points are obtained; judging the CT value corresponding to each sampling point, and recording the sampling points corresponding to the CT values meeting the preset conditions of the blood vessels; taking the average position coordinates of all sampling points meeting the preset conditions of blood vessels as corrected center point coordinates of corresponding error points on a two-dimensional tangent plane image, transforming the corrected center point coordinates on the two-dimensional tangent plane image according to the cutting matrix to obtain corrected three-dimensional coordinates of the error points, and replacing the three-dimensional coordinates of each error point with the corrected three-dimensional coordinates in each set of center point data to obtain corrected center paths and center point data;
and the blood vessel model reconstruction module is used for reconstructing and obtaining a coronary artery blood vessel model according to the corrected central path and the central point data.
8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 6 when the computer program is executed.
9. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311302080.1A CN117036640B (en) | 2023-10-10 | 2023-10-10 | Coronary artery blood vessel model reconstruction method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311302080.1A CN117036640B (en) | 2023-10-10 | 2023-10-10 | Coronary artery blood vessel model reconstruction method, device, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117036640A CN117036640A (en) | 2023-11-10 |
| CN117036640B true CN117036640B (en) | 2023-12-19 |
Family
ID=88639476
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311302080.1A Active CN117036640B (en) | 2023-10-10 | 2023-10-10 | Coronary artery blood vessel model reconstruction method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117036640B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107204041A (en) * | 2017-05-31 | 2017-09-26 | 重庆大学 | Retinal vessel 3 D model construction method and device based on eye fundus image |
| CN111754506A (en) * | 2020-07-01 | 2020-10-09 | 杭州脉流科技有限公司 | Coronary artery stenosis rate calculation method, device and system based on intracavity images and computer storage medium |
| CN112132949A (en) * | 2019-11-19 | 2020-12-25 | 苏州润迈德医疗科技有限公司 | Blood vessel three-dimensional modeling method, device and system with narrow lesion interval |
| CN112419484A (en) * | 2020-11-25 | 2021-02-26 | 苏州润迈德医疗科技有限公司 | Three-dimensional blood vessel synthesis method and system, coronary artery analysis system and storage medium |
| WO2021196536A1 (en) * | 2020-03-31 | 2021-10-07 | 苏州润迈德医疗科技有限公司 | Method and apparatus for accurately extracting vessel centerline, analysis system, and storage medium |
| CN113744272A (en) * | 2021-11-08 | 2021-12-03 | 四川大学 | Automatic cerebral artery delineation method based on deep neural network |
| CN116051738A (en) * | 2022-12-30 | 2023-05-02 | 杭州脉流科技有限公司 | Method for reconstructing coronary artery blood vessel model based on CTA image and readable storage medium |
-
2023
- 2023-10-10 CN CN202311302080.1A patent/CN117036640B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107204041A (en) * | 2017-05-31 | 2017-09-26 | 重庆大学 | Retinal vessel 3 D model construction method and device based on eye fundus image |
| CN112132949A (en) * | 2019-11-19 | 2020-12-25 | 苏州润迈德医疗科技有限公司 | Blood vessel three-dimensional modeling method, device and system with narrow lesion interval |
| WO2021196536A1 (en) * | 2020-03-31 | 2021-10-07 | 苏州润迈德医疗科技有限公司 | Method and apparatus for accurately extracting vessel centerline, analysis system, and storage medium |
| CN111754506A (en) * | 2020-07-01 | 2020-10-09 | 杭州脉流科技有限公司 | Coronary artery stenosis rate calculation method, device and system based on intracavity images and computer storage medium |
| CN112419484A (en) * | 2020-11-25 | 2021-02-26 | 苏州润迈德医疗科技有限公司 | Three-dimensional blood vessel synthesis method and system, coronary artery analysis system and storage medium |
| CN113744272A (en) * | 2021-11-08 | 2021-12-03 | 四川大学 | Automatic cerebral artery delineation method based on deep neural network |
| CN116051738A (en) * | 2022-12-30 | 2023-05-02 | 杭州脉流科技有限公司 | Method for reconstructing coronary artery blood vessel model based on CTA image and readable storage medium |
Non-Patent Citations (1)
| Title |
|---|
| CT冠脉造影中冠状动脉中心线的提取;易建华;何晖光;赵明昌;;计算机科学与探索;第5卷(第11期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117036640A (en) | 2023-11-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108805871B (en) | Blood vessel image processing method and device, computer equipment and storage medium | |
| US11042822B2 (en) | Systems and methods for using geometry sensitivity information for guiding workflow | |
| US11216950B2 (en) | Method and system for automatically segmenting blood vessel in medical image by using machine learning and image processing algorithm | |
| CN110310256B (en) | Coronary stenosis detection method, coronary stenosis detection device, computer equipment and storage medium | |
| US20200320697A1 (en) | Method, system, and device for lung lobe segmentation, model training, model construction and segmentation | |
| CN111369542B (en) | Vessel marking method, image processing system, and storage medium | |
| CN111028248A (en) | Method and device for separating static and dynamic pulses based on CT (computed tomography) image | |
| CN111507965A (en) | Novel coronavirus pneumonia focus detection method, system, device and storage medium | |
| CN110176013B (en) | Blood vessel segmentation method, blood vessel segmentation device, electronic device and storage medium | |
| CN109886973B (en) | Blood vessel extraction method and device and computer readable storage medium | |
| CN112862835A (en) | Coronary vessel segmentation method, device, equipment and computer readable storage medium | |
| CN111312374B (en) | Medical image processing method, medical image processing device, storage medium and computer equipment | |
| CN116051738A (en) | Method for reconstructing coronary artery blood vessel model based on CTA image and readable storage medium | |
| CN117036530B (en) | Cross-modal data-based coronary artery fractional flow reserve prediction method and device | |
| CN116051826A (en) | Coronary vessel segmentation method based on DSA image continuous frame sequence, computer equipment and readable storage medium | |
| CN117036640B (en) | Coronary artery blood vessel model reconstruction method, device, equipment and storage medium | |
| US20150278976A1 (en) | Systems and methods for using geometry sensitivity information for guiding workflow | |
| CN111754522B (en) | Method and device for acquiring coronary artery hemodynamic data | |
| CN116993812A (en) | Coronary vessel centerline extraction method, device, equipment and storage medium | |
| CN113379741B (en) | Retinal blood vessel segmentation method, device and storage medium based on blood vessel characteristics | |
| CN115482261A (en) | Blood vessel registration method, device, electronic equipment and storage medium | |
| CN114170440A (en) | Method and device for determining image feature points, computer equipment and storage medium | |
| CN115908418A (en) | Method, system, equipment and medium for determining central line of aorta CT image | |
| CN112837283B (en) | Pulmonary embolism detection system, medium and electronic equipment | |
| CN115359257A (en) | Deep learning-based spine image segmentation method and operation navigation positioning system |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |