CN114049282B - Coronary artery construction method, device, terminal and storage medium - Google Patents

Abstract

The embodiment of the application provides a coronary artery construction method, a terminal and a storage medium, wherein the coronary artery construction method comprises the following steps: calculating a dynamic threshold value based on first point cloud data of the coronary artery reconstructed by the collected multiple thoracic cavity tomography images and based on the first point cloud data; determining a predicted location of calcified plaque on the first point cloud data based on a dynamic threshold; determining a coronary artery segment where the predicted position is located on the first point cloud data, and generating a cross section along the position of a central point in the coronary artery segment; reconstructing a coronary vessel flow-path region in a cross-section; and constructing second point cloud data of the coronary artery based on the coronary artery blood vessel flow passage area.

Description

Coronary artery construction method, device, terminal and storage medium

technical field

The present application relates to, but is not limited to, the technical field of medical image processing, and in particular, relates to a method, device, terminal and storage medium for constructing a coronary artery.

Background technique

In the recognition and processing technology of medical images, the post-processing analysis of digital images such as Computed Tomography (CT) images and Magnetic Resonance Imaging (MRI) images has been widely used in clinical disease diagnosis. Especially for coronary artery disease, doctors currently agree to calculate the Fractional Flow Reserve (FFR) to measure the degree of disease, and then give patients appropriate advice, such as whether to cure coronary artery disease through surgery, The calculation of FFR is to allocate the blood supply of the branches through the coronary artery model. Therefore, how to accurately and efficiently extract coronary arteries in medical imaging is of great significance for FFR calculation.

Coronary reconstruction methods in the related art are to use deep learning methods or machine learning methods such as region growth, filtering processing or morphological processing, etc., starting from CT images and using centerline guidance to reconstruct coronary arteries, or directly using spatial correlation. information to reconstruct coronary arteries, or use measurements to create simulated model matches, etc. However, in the above method, there is at least an artifact caused by calcified plaque, which leads to the inaccuracy of the finally constructed coronary artery.

SUMMARY OF THE INVENTION

Embodiments of the present application provide a method, device, terminal, and storage medium for constructing a coronary artery, so as to solve the problem of inaccuracy in the final constructed coronary artery caused by artifacts caused by calcified plaque in the related art.

The technical solutions of the embodiments of the present application are implemented as follows:

In a first aspect, an embodiment of the present application provides a method for constructing a coronary artery, the method comprising:

Calculate the dynamic threshold based on the first point cloud data of the coronary artery reconstructed from the collected multiple thoracic tomographic images, and based on the first point cloud data;

determining the predicted position of the calcified plaque on the first point cloud data based on the dynamic threshold;

On the first point cloud data, determine the coronary artery segment where the predicted position is located, and generate a cross-section along the position of the center point in the coronary artery segment;

reconstructing a coronary vascular flow channel region in said cross-section;

Based on the coronary vessel flow channel region, second point cloud data of the coronary arteries are constructed.

In a second aspect, an embodiment of the present application provides a device for constructing a coronary artery, the device comprising:

a processing module, configured to calculate the dynamic threshold based on the first point cloud data of the coronary artery reconstructed based on the collected multiple thoracic tomographic images, and based on the first point cloud data;

a determination module, configured to determine the predicted position of the calcified plaque on the first point cloud data based on the dynamic threshold;

The determining module is further configured to determine, on the first point cloud data, the coronary artery segment where the predicted position is located;

a generation module for generating a cross-section along the position of a center point within the coronary artery segment;

a reconstruction module for reconstructing a coronary vessel flow channel region in the cross-section;

a building module for building second point cloud data of the coronary artery based on the coronary vessel flow channel region.

In a third aspect, an embodiment of the present application provides a terminal, where the terminal includes: a processor, a memory, and a communication bus;

The communication bus is used to realize the communication connection between the processor and the memory;

The processor is configured to execute the coronary artery construction program stored in the memory, so as to realize the steps of the above-mentioned coronary artery construction method.

In a third aspect, embodiments of the present application provide a storage medium, where one or more programs are stored in the storage medium, and the one or more programs can be executed by one or more processors to implement the above-mentioned coronary arteries Steps to build a method.

The following beneficial effects are achieved by applying the embodiments of the present application: based on the medical coronary model, the dynamic threshold is increased to identify the calcified plaque in the arterial model, and the vascular flow channel region of the coronary artery segment with the calcified plaque is repaired, It provides a more accurate basis for establishing an accurate three-dimensional model of coronary arteries, and then provides accurate blood flow distribution for calculating FFR.

The coronary artery construction method, device, terminal, and storage medium provided by the embodiments of the present application use the first point cloud data of the coronary artery reconstructed based on the collected multiple thoracic tomographic images, and based on the first point cloud data, calculate Dynamic threshold; determine the predicted position of the calcified plaque on the first point cloud data based on the dynamic threshold; on the first point cloud data, determine the coronary artery segment where the predicted position is located, and generate along the position of the center point in the coronary artery segment cross section; reconstruct the coronary vascular flow channel area in the cross section; based on the coronary vascular flow channel area, construct the second point cloud data of the coronary artery; that is, the present application is based on the first point cloud data of the coronary artery, The dynamic threshold corresponding to the thoracic tomographic image is adaptively calculated, and then based on the dynamic threshold, the red calcified plaque of the three-dimensional coronary model can be accurately identified. An accurate three-dimensional model of the coronary artery is established based on the repaired vascular flow channel area; in this way, the artifact caused by the calcified plaque in the related art is solved, which leads to the inaccuracy of the finally constructed coronary artery, which improves the understanding of the coronary artery. Accurate modeling is performed to provide accurate blood flow distribution for calculating FFR. At the same time, the scheme has good robustness and scalability.

Description of drawings

1 is a schematic flowchart of a method for constructing a coronary artery according to an embodiment of the present application;

FIG. 2 is a schematic diagram of a reconstructed first point cloud data of a coronary artery according to an embodiment of the present application;

3 is a schematic flowchart of another method for constructing a coronary artery according to an embodiment of the present application;

4 is a schematic flowchart of another method for constructing a coronary artery according to an embodiment of the present application;

5 is a schematic diagram of a predicted position of a calcified plaque determined based on a dynamic threshold according to an embodiment of the present application;

6 is a schematic diagram of a manually selected position of a calcified plaque in a coronary artery according to an embodiment of the present application;

7 is a schematic flowchart of a method for constructing a coronary artery according to another embodiment of the present application;

FIG. 8 is a schematic diagram of a point on the central axis of a coronary artery according to an embodiment of the present application;

FIG. 9 is a schematic diagram of a cross section generated in a coronary artery segment according to an embodiment of the present application;

10 is a schematic diagram of a cross-sectional level set segmentation result provided by an embodiment of the present application;

11 is a schematic flowchart of another method for constructing a coronary artery according to another embodiment of the present application;

12 is a schematic diagram of a level set segmentation result displayed in a three-dimensional space according to an embodiment of the present application;

13 is a schematic diagram of a result of traversing all target points in a coronary vascular flow channel region in order of distance from an entry point according to an embodiment of the present application;

14 is a schematic diagram of a coronary artery model obtained based on a repaired vascular flow channel region provided by an embodiment of the present application;

15 is a schematic structural diagram of a coronary artery construction device provided in an embodiment of the application;

FIG. 16 is a schematic structural diagram of a terminal according to an embodiment of the present application.

Detailed ways

In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be described in further detail below with reference to the accompanying drawings. All other embodiments obtained under the premise of creative work fall within the scope of protection of the present application.

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the technical field to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application, and are not intended to limit the present application.

In related technologies, with the explosion of image big data, image processing technology shows the advantages of high processing accuracy, good reproducibility, high flexibility, and strong versatility. Shape analysis and recognition of objects play an increasingly important role. The main idea is to determine the object through key point positioning, and the drawing of an object's shape must be inseparable from the extraction of its bones.

Generally speaking, the process of obtaining the bones of an image is the process of 'refining' the image, which can effectively reflect the connectivity and topology of the original object shape. The current bone extraction algorithm is to iteratively calculate from the boundary, and evenly peel off the boundary of the graph layer by layer until the innermost one-dimensional bone remains.

Image skeleton extraction technology is a very important transformation in image analysis and shape description, and it is an important topological description in image geometry. The premise of the application of bone extraction technology in image target shape analysis, feature extraction, pattern recognition, etc. In the fields of virtual navigation, shape matching, fingerprint recognition, medical image processing and other fields, the curve bone extraction algorithm has already become a research hotspot.

In the field of medical imaging, color Doppler cardiovascular imaging, magnetic resonance imaging MRI, digital subtraction angiography (DSA), morphological technology, and machine learning technology have emerged one after another. The degree of digitization of medical images is getting higher and higher, and the types are more diverse. Medical imaging technology has not only provided reconstructed models of various human organs, but also reflected the changes in hemodynamics through time-series blood flow velocity fields. The development of these technologies has greatly improved the efficiency and accuracy of doctors' diagnosis, while also reducing unnecessary surgical risks for patients.

In the recognition and processing technology of medical images, the post-processing analysis of digital images such as Computed Tomography (CT) images and Magnetic Resonance Imaging (MRI) images has been widely used in clinical disease diagnosis. Especially for coronary artery disease, doctors currently agree to calculate the Fractional Flow Reserve (FFR) to measure the degree of disease, and then give patients appropriate advice, such as whether to cure coronary artery disease through surgery, The calculation of FFR is to allocate the blood supply of the branches through the coronary artery model. Therefore, how to accurately and efficiently extract coronary arteries in medical imaging is of great significance for FFR calculation.

Coronary reconstruction methods in the related art are to use deep learning methods or machine learning methods such as region growth, filtering processing or morphological processing, etc., starting from CT images and using centerline guidance to reconstruct coronary arteries, or directly using spatial correlation. information to reconstruct coronary arteries, or use measurements to create simulated model matches, etc. However, in the above methods, at least there are artifacts caused by calcified plaque and/or insufficient contrast agent caused by vascular stenosis, which leads to inaccuracy of the final coronary model constructed. And the resulting vascular stenosis, at least have a huge impact on the calculation of FFR.

An embodiment of the present application provides a method for constructing a coronary artery. The method is applied to a terminal. Referring to FIG. 1 , the method includes:

Step 101: Calculate the dynamic threshold based on the first point cloud data of the coronary artery reconstructed based on the collected multiple thoracic tomographic images, and based on the first point cloud data.

In the embodiment of the present application, the multiple thoracic tomographic images may be multiple CT images obtained based on CT technology. Here, the CT images may be the coronary arteries scanned by a multi-slice spiral CT machine after intravenous injection of an appropriate contrast agent, thereby Obtained coronary CT images. The multiple thoracic tomographic images can also be angiography images obtained based on coronary angiography; the multiple thoracic tomographic images can also be angiography images obtained based on X-ray technology. The tomographic images can be determined based on any medical imaging information that can reflect the coronary arteries. The data sources of the multiple thoracic tomographic images collected are not specifically limited in the embodiments of the present application, so as to realize the coronary arteries provided by the present application. The build method prevails.

In the embodiment of the present application, the first point cloud data of the coronary arteries may be data obtained by the terminal performing three-dimensional reconstruction based on multiple collected thoracic tomographic images to obtain an initial coronary artery model, and based on the initial coronary artery model. Here, the first point cloud data includes a tissue density value corresponding to each target point in the region where the coronary arteries are located.

Here, the point cloud data can be a two-dimensional array with H rows and 3 columns in the form of spatial volume coordinates (x, y, z) (H is the number of points contained in the point cloud data). In this embodiment of the present application, first, after acquiring the point cloud data, the terminal can convert the point cloud array into a binarized three-dimensional matrix with a certain size.

In this embodiment of the present application, the dynamic threshold is used to predict the position of the calcified plaque on the first point cloud data.

In the embodiment of the present application, the terminal acquires multiple thoracic tomographic images, and performs three-dimensional reconstruction based on the multiple thoracic tomographic images to obtain an initial coronary artery model; further, the terminal generates a reconstructed coronary artery based on the initial coronary artery model and based on the first point cloud data, a dynamic threshold for predicting the position of the calcified plaque on the first point cloud data is calculated.

In an achievable application scenario, referring to FIG. 2 , FIG. 2 shows a schematic diagram of the reconstructed first point cloud data of the coronary artery. First, the terminal acquires multiple thoracic tomographic images collected, and arranges the multiple thoracic tomographic images into a three-dimensional form in a spatial order to obtain a three-dimensional image Image3d; secondly, the terminal processes the three-dimensional image Image3d to obtain a reconstructed image. The initial coronary model; further, the terminal obtains the first point cloud data of the reconstructed coronary artery, that is, the reconstruction result of the coronary artery, based on the initial coronary model. Finally, the terminal calculates, based on the first point cloud data Coronary, a dynamic threshold in the region where the coronary artery is located for predicting the position of the calcified plaque on the first point cloud data.

Step 102: Determine the predicted position of the calcified plaque on the first point cloud data based on the dynamic threshold.

Step 103: On the first point cloud data, determine the coronary artery segment where the predicted position is located, and generate a cross section along the position of the center point in the coronary artery segment.

In the embodiment of the present application, the coronary artery segment is the coronary artery segment corresponding to the location of the calcified plaque. It should be noted that the coronary artery includes the proximal right coronary artery, the middle right coronary artery, the distal right coronary artery, and the right coronary artery. Posterior-descending-artery (PDA), left main (LM), proximal left anterior descending artery, middle left anterior descending artery, distal left anterior descending artery, first diagonal branch, second diagonal branch , proximal left circumflex branch, blunt marginal branch, distal left circumflex branch.

In the embodiment of the present application, the terminal determines the first point based on the dynamic threshold when the terminal calculates the dynamic threshold based on the first point cloud data of the coronary artery reconstructed based on the collected multiple thoracic tomographic images, and based on the first point cloud data The predicted position of the calcified plaque on the cloud data; further, on the first point cloud data, the terminal determines the segment of the coronary artery where the predicted location is located, that is, determines the segment of the coronary artery where the calcified plaque is located from multiple coronary segments, and Cross-sections are generated along the location of the center point within the coronary segment.

Step 104 , reconstruct the coronary vessel flow channel region in the cross section.

In the embodiments of the present application, reconstructing the coronary vascular flow channel region in the cross-section can be understood as segmenting the calcified plaque region in the coronary artery and the vascular flow channel region in the coronary artery in the cross-section, based on the segmented coronary vascular flow channel region. A vascular flow channel area in an artery, reconstructing a coronary vascular flow channel area in cross-section.

Step 105 , construct second point cloud data of the coronary artery based on the coronary vascular flow channel region.

In the embodiment of the present application, the second point cloud data may be understood as point cloud data after repair of the blood vessel flow channel in the coronary artery.

In the embodiment of the present application, when the terminal reconstructs the coronary blood vessel flow channel region in the cross section, the terminal constructs the second point cloud data of the coronary artery based on the reconstructed coronary blood vessel flow channel region, that is, according to the second point cloud data to construct a target coronary model after vascular flow channel repair.

The coronary artery construction method provided by the embodiment of the present application uses the first point cloud data of the coronary artery reconstructed based on the collected multiple thoracic tomographic images, and calculates the dynamic threshold based on the first point cloud data; The predicted location of the calcified plaque on the first point cloud data; on the first point cloud data, determine the coronary segment where the predicted location is located, and generate a cross-section along the position of the center point within the coronary segment; in the cross-section Reconstruct the coronary vascular flow channel area; based on the coronary vascular flow channel area, construct the second point cloud data of the coronary artery; that is to say, this application is based on the first point cloud data of the coronary artery, adaptive calculation and thoracic cavity The dynamic threshold corresponding to the tomographic image, and then based on the dynamic threshold to accurately identify the red calcified plaque of the three-dimensional coronary model, and at the same time, repair the vascular flow channel area of the coronary artery segment with calcified plaque, based on the repaired vascular flow channel. In this way, the artifact caused by calcified plaque in the related art is solved, which leads to the inaccuracy of the final coronary artery, and the accurate modeling of the coronary artery is improved. Computational FFR provides accurate blood flow distribution, and at the same time, the scheme has good robustness and scalability.

An embodiment of the present application provides a method for constructing a coronary artery. The method is applied to a terminal. Referring to FIG. 3 , the method includes:

Step 201 , reconstructing the first point cloud data of the coronary artery based on the acquired multiple thoracic tomographic images.

Step 202: Acquire a tissue density value corresponding to each target point in the first point cloud data.

In the embodiment of the present application, the tissue density value can be understood as the absorption rate of X-rays in the tissue or organ in the region where the coronary artery is located. The tissue density value can also become the CT value, and the unit of the CT value is the Hounsfield unit (HU). It should be noted that plaques in coronary arteries can be classified according to the CT values of tissues or organs, so that plaques can be divided into soft plaques, fibrous plaques and calcified plaques. Here, soft plaques and fibrous plaques The plaques are collectively referred to as non-calcified plaques.

Step 203: Determine the mean value of all tissue density values, which is the mean value of tissue density in the region where the coronary arteries are located.

Step 204 , select a threshold calculation formula corresponding to the mean tissue density, and calculate the dynamic threshold.

In the embodiment of the present application, after acquiring the tissue density value corresponding to each target point in the first point cloud data, the terminal calculates the mean value of all tissue density values, and uses the mean value as the tissue density mean value of the region where the coronary arteries are located; further, The terminal selects a threshold calculation formula corresponding to the mean tissue density, and calculates a dynamic threshold for predicting the position of the calcified plaque on the first point cloud data.

In the embodiment of the present application, as shown in FIG. 4 , step 204 selects a threshold calculation formula corresponding to the mean tissue density, and calculates the dynamic threshold, which can be achieved by the following steps:

Step A1: Based on the tissue density value range to which the mean tissue density value belongs, select a threshold value calculation formula corresponding to the tissue density value range.

In the embodiment of the present application, step A1 selects the threshold value calculation formula corresponding to the tissue density value range based on the tissue density value range to which the tissue density mean value belongs, which can be implemented in the following manner:

If the mean tissue density is less than or equal to the first parameter, select the first threshold calculation formula.

If the mean tissue density is greater than the first parameter and less than the second parameter, the second threshold calculation formula is selected.

If the mean tissue density is greater than or equal to the second parameter, select the third threshold calculation formula.

；第二阈值计算公式为，

；第三阈值计算公式为，

。Among them, the first threshold calculation formula is,

; The second threshold calculation formula is,

; The third threshold calculation formula is,

.

为动态阈值，

为组织密度均值，

为第一参数，

为第二参数，

为第三参数，

为第四参数，

为第五参数，

均为正数。in,

is the dynamic threshold,

is the mean tissue density,

is the first parameter,

is the second parameter,

is the third parameter,

is the fourth parameter,

is the fifth parameter,

All are positive numbers.

In the embodiment of the present application, the first parameter, the second parameter, the third parameter, the fourth parameter, and the fifth parameter are the empirical threshold ranges obtained by analyzing and processing a large amount of case data.

Step A2: Calculate the dynamic threshold based on the tissue density mean and the threshold calculation formula.

In this embodiment of the present application, the threshold calculation formula includes a first threshold calculation formula, a second threshold calculation formula, and a third threshold calculation formula.

、第二参数

、第三参数

，第四参数

，第五参数

为例进行说明。终端基于采集到的多张胸腔断层图像重构的冠状动脉的第一点云数据

之后，获取第一点云数据中每一目标点对应的组织密度值，计算所有组织密度值的均值，并确定该均值为冠状动脉所在区域的组织密度均值。这里，组织密度均值可以通过如下公式计算得到：In an achievable application scenario, the first parameter

, the second parameter

, the third parameter

, the fourth parameter

, the fifth parameter

Take an example to illustrate. The terminal reconstructs the first point cloud data of the coronary artery based on the collected multiple thoracic tomographic images

After that, the tissue density value corresponding to each target point in the first point cloud data is obtained, the mean value of all tissue density values is calculated, and the mean value is determined as the mean tissue density value of the region where the coronary arteries are located. Here, the mean tissue density can be calculated by the following formula:

为组织密度均值，

表示计算矩阵中所有组织密度值的平均值的函数，

为第一点云数据。in,

is the mean tissue density,

represents a function that computes the mean of all tissue density values in the matrix,

is the first point cloud data.

所属的组织密度值范围，即判断组织密度均值是否小于等于第一参数

，或判断组织密度均值是否大于第一参数

小于第二参数

，或判断组织密度均值是否大于等于第二参数

；若终端确定若组织密度均值小于等于第一参数，选择第一阈值计算公式

；若终端确定组织密度均值大于第一参数小于第二参数，选择第二阈值计算公式

；若终端确定组织密度均值大于等于第二参数，选择第三阈值计算公式

。如此，根据冠状动脉中的组织密度均值动态的选择动态阈值，以便准确快速地预测出第一点云数据上钙化斑块所在的位置；同时，选择动态阈值的方法是在完全的自动化的流程下实现的，全程无需人工干预或其他操作，且避免了使用复杂的计算方法如深度学习，导致结果不可控、不稳定的方法。同时，该方法计算速度快，准确率高、可靠性和可扩展性，易于应用到各个领域。Further, the terminal determines the mean tissue density

The range of tissue density values to which it belongs, that is, to determine whether the mean value of tissue density is less than or equal to the first parameter

, or determine whether the mean tissue density is greater than the first parameter

less than the second parameter

, or determine whether the mean tissue density is greater than or equal to the second parameter

; If the terminal determines that if the mean value of tissue density is less than or equal to the first parameter, select the first threshold calculation formula

; If the terminal determines that the mean value of tissue density is greater than the first parameter and less than the second parameter, select the second threshold calculation formula

; If the terminal determines that the mean value of tissue density is greater than or equal to the second parameter, select the third threshold calculation formula

. In this way, the dynamic threshold is dynamically selected according to the mean tissue density in the coronary arteries, so as to accurately and quickly predict the location of the calcified plaque on the first point cloud data; at the same time, the method of selecting the dynamic threshold is under a completely automated process. It is realized without manual intervention or other operations in the whole process, and avoids the use of complex computing methods such as deep learning, resulting in uncontrollable and unstable results. At the same time, the method has fast calculation speed, high accuracy, reliability and scalability, and is easy to apply to various fields.

Step 205: From all the tissue density values, determine the position of the target point whose tissue density value is greater than the dynamic threshold, which is the predicted position of the calcified plaque on the first point cloud data.

Here, referring to FIG. 5 , FIG. 5 shows a schematic diagram of a predicted position of a calcified plaque determined based on a dynamic threshold.

In other embodiments of the present application, the terminal may also predict the patch position with a pre-specified or locally selected threshold. In one case, according to the reconstructed first point cloud data of the coronary artery, the terminal manually selects the corresponding position in the three-dimensional space according to the shape and contour of the coronary artery, and the position with high probability that calcified plaque may appear, and obtains The three-dimensional coordinates of this location. Referring to FIG. 6 , FIG. 6 shows the manually selected three-dimensional coordinates of the position of a clearly protruding calcified plaque in the right coronary artery in the coronary artery, such as (206, 358, 157).

Step 206 , on the first point cloud data, determine the coronary artery segment where the predicted position is located, and generate a cross-section along the position of the center point in the coronary artery segment.

In the embodiment of the present application, referring to FIG. 7 , in step 206, on the first point cloud data, determining the coronary artery segment where the predicted position is located can be achieved by the following steps:

Step B1: Determine the centerline point set and the segmentation information of the coronary arteries based on the first point cloud data.

Wherein, the centerline point set includes points on the centerline of the first point cloud data.

In the embodiment of the present application, the centerline point set can be understood as the point on the central axis of the coronary artery that is directly extracted by the coronary bone extraction algorithm, and the centerline point set can also be understood as a tree structure representing all the point cloud data of the coronary artery The points on the central axis of the coronary arteries, exemplarily, use a script language such as the three-dimensional centerline extraction algorithm in the Python language to directly extract the points on the central axis of the coronary arteries, as shown in FIG. Schematic diagram of a point on the central axis. Here, the point on the central axis of the coronary artery can also be called the keel point of the bone.

In the embodiment of the present application, the segmentation information of the coronary arteries is to name the coronary arteries through medical nomenclature to achieve the purpose of segmentation.

Step B2: Based on the segmentation information, from the centerline point set, determine the position of the center point closest to the predicted position as the coronary artery segment where the predicted position is located.

In the embodiment of the present application, after the terminal determines the centerline point set and the segment information of the coronary arteries based on the first point cloud data, based on the segment information, from the centerline point set, it is determined that the location of the center point closest to the predicted position is: The coronary artery segment where the predicted position is located, that is, the coronary artery segment where the calcified plaque is currently located is located. Referring to Fig. 5, Fig. 5 shows a schematic diagram of determining the predicted positions of calcified plaques through dynamic thresholds. Here, the predicted positions of calcified plaques on the first point cloud data have three P1, P2 and P3, and three The predicted locations are in the left anterior descending and right coronary arteries, respectively.

In this embodiment of the present application, the terminal determines the centerline point set and the segment information of the coronary arteries based on the first point cloud data, and based on the segment information, from the centerline point set, determines the location of the center point closest to the predicted position as the prediction After the coronary artery segment where the location is located, a plurality of cross-sections are generated along the location of the center point within the coronary artery segment, and the extent of the cross-section includes the cross-sectional extent of the coronary artery at the predicted location. Referring to FIG. 9, A in FIG. 9 shows a schematic diagram of a cross section generated in a coronary artery segment, and B in FIG. Schematic representation of the cross section, C in FIG. 9 shows a schematic representation of a longitudinal section generated in a coronary artery segment.

Step 207: Remove the target points corresponding to the calcified plaques in the cross-section to obtain the reconstructed coronary vessel flow channel area.

In the embodiment of the present application, the terminal determines the coronary artery segment where the predicted position is located on the first point cloud data, and generates a cross-section along the position of the center point in the coronary artery segment, and uses the distance regularization level set method (DRLSE). , Distance Regularized Level Set Evolution) image segmentation of each cross section to obtain the calcified plaque area and the coronary vessel flow channel area; further, the terminal removes the target point in the area where the calcified plaque is located in the cross section to obtain a reconstruction The coronary vascular flow channel region, that is, the coronary vascular flow channel region reconstructed in cross-section. Referring to FIG. 10 , A in FIG. 10 shows the level set segmentation result of the cross-section, and B in FIG. 10 shows the level set segmentation result of the limited coronary artery region.

Step 208 , construct second point cloud data of the coronary artery based on the coronary vascular flow channel region.

In the embodiment of the present application, as shown in FIG. 11 , step 208 constructs the second point cloud data of the coronary artery based on the coronary vascular flow channel region, which can be implemented by the following steps:

Step C1: Take the position of the entry point in the coronary vascular flow channel region as the starting point, and traverse all the target points in the coronary vascular flow channel region in the order of distance from the entry point from near to far to obtain the closed region of the coronary artery segment target point cloud data.

Step C2: Replace the point cloud data corresponding to the coronary artery segment of the first point cloud data with the target point cloud data to obtain the second point cloud data.

In the embodiment of the present application, the coronary vascular flow channel region is the above-mentioned reconstructed coronary vascular flow channel region.

In the embodiment of the present application, the terminal determines the entry point in the coronary vascular flow channel region, and takes the position of the entry point as the starting point, and traverses all the targets in the coronary vascular flow channel region in order of distance from the entry point from near to far point to obtain the target point cloud data of the closed area of the coronary artery segment; further, the terminal acquires the point cloud data corresponding to the coronary artery segment of the first point cloud data, and replaces the point cloud data corresponding to the coronary artery segment with the target point cloud data to obtain the second point cloud data, and then based on the second point cloud data, the target coronary model after the repair of the vascular flow channel is constructed, as shown in Figure 12, Figure 13 and Figure 14, Figure 12 shows the three-dimensional space A schematic diagram of the level set segmentation results shown in Figure 13 is a schematic diagram of the results of traversing all target points in the coronary vessel flow channel region in the order of distance from the entry point, A in Figure 14 shows the A schematic diagram of the second point cloud data of the coronary artery closure region, B in FIG. 14 shows a schematic diagram of the target coronary artery model obtained based on the repaired vascular flow channel region.

It can be seen from the above that in the embodiment of the present application, the terminal calculates the average tissue density of the coronary artery based on the first point cloud data of the coronary artery, and based on the range to which the average tissue density belongs, adaptively calculates the dynamic threshold corresponding to the thoracic tomographic image, and further. Accurately identify the red calcified plaques in the three-dimensional coronary model based on dynamic thresholds. Further, for the coronary segments with calcified plaques, multiple cross-sections are generated along the position of the center point, and the calcified plaque areas in the cross-sections and The vascular flow channel area is segmented, and the target points corresponding to the calcified plaque area in the cross section are removed to repair the vascular flow channel region of the coronary artery, and then an accurate three-dimensional model of the coronary artery is established based on the repaired vascular flow channel region; , which solves the inaccurate problem of the final constructed coronary artery caused by the artifact caused by calcified plaque in the related art, and improves the accurate modeling of the coronary artery, thereby providing accurate blood flow distribution for the calculation of FFR. At the same time, The scheme has good robustness and scalability.

It should be noted that, for the description of the same steps and the same content in this embodiment as in other embodiments, reference may be made to the descriptions in other embodiments, and details are not repeated here.

Based on the foregoing embodiments, the present application provides a coronary artery construction device, which can be used to implement a coronary artery construction method corresponding to FIG. 1 , FIG. 3 to FIG. 4 , FIG. 7 and FIG. 11 . , as shown in FIG. 15 , the coronary artery construction device 15 includes:

The processing module 1501 is configured to calculate the dynamic threshold based on the first point cloud data of the coronary artery reconstructed based on the collected multiple thoracic tomographic images, and based on the first point cloud data;

The determining module 1502 is further configured to determine the predicted position of the calcified plaque on the first point cloud data based on the dynamic threshold;

The determining module 1502 is further configured to determine the coronary artery segment where the predicted position is located on the first point cloud data;

a generating module 1503 for generating a cross-section along the position of the center point within the coronary artery segment;

a reconstruction module 1504 for reconstructing the coronary vessel flow channel region in cross-section;

The construction module 1505 is configured to construct the second point cloud data of the coronary artery based on the coronary vessel flow channel region.

In other embodiments of the present application, the processing module 1501 is further configured to obtain the tissue density value corresponding to each target point in the first point cloud data; determine the mean value of all tissue density values, which is the mean tissue density value of the region where the coronary arteries are located; select The threshold calculation formula corresponding to the mean value of tissue density is used to calculate the dynamic threshold.

In other embodiments of the present application, the processing module 1501 is further configured to select a threshold value calculation formula corresponding to the tissue density value range based on the tissue density value range to which the tissue density mean value belongs; and calculate the dynamic threshold value based on the tissue density mean value and the threshold value calculation formula.

；第二阈值计算公式为，

；第三阈值计算公式为，

；其中，

为动态阈值，

为组织密度均值，

为第一参数，

为第二参数，

为第三参数，

为第四参数，

为第五参数，

均为正数。In other embodiments of the present application, the processing module 1501 is further configured to select the first threshold calculation formula if the mean tissue density is less than or equal to the first parameter; if the mean tissue density is greater than the first parameter and less than the second parameter, select the second threshold calculation formula ; If the mean value of tissue density is greater than or equal to the second parameter, select the third threshold calculation formula; wherein, the first threshold calculation formula is,

; The second threshold calculation formula is,

; The third threshold calculation formula is,

;in,

is the dynamic threshold,

is the mean tissue density,

is the first parameter,

is the second parameter,

is the third parameter,

is the fourth parameter,

is the fifth parameter,

All are positive numbers.

In other embodiments of the present application, the determining module 1502 is further configured to determine, from all tissue density values, the position of the target point whose tissue density value is greater than the dynamic threshold as the predicted position of the calcified plaque on the first point cloud data.

In other embodiments of the present application, the determining module 1502 is further configured to determine the centerline point set and the segmentation information of the coronary arteries based on the first point cloud data; wherein the centerline point set includes the points on the center axis of the first point cloud data. point; based on the segmentation information, from the centerline point set, determine the position of the center point closest to the predicted position as the coronary artery segment where the predicted position is located.

In other embodiments of the present application, the reconstruction module 1504 is further configured to remove the target points corresponding to the calcified plaques in the cross section to obtain the reconstructed coronary vascular flow channel area; the construction module 1505 is further configured to use the coronary vascular flow The position of the entry point in the channel area is the starting point, and in the order of distance from the entry point from near to far, all the target points in the coronary vascular flow channel area are traversed to obtain the target point cloud data of the closed area of the coronary artery segment; The point cloud data corresponding to the coronary artery segment of the one point cloud data is replaced with the target point cloud data to obtain the second point cloud data.

An embodiment of the present application provides a terminal, which can be used to implement a method for constructing a coronary artery provided by the embodiments corresponding to FIG. 1 , FIG. 3 to FIG. 4 , FIG. 7 , and FIG. 11 . Referring to FIG. 16 , The terminal 16 (the terminal 16 in FIG. 16 corresponds to the coronary artery construction device 15 in FIG. 15 ) includes: a processor 1601 , a memory 1602 and a communication bus 1603 , wherein:

The communication bus 1603 is used to realize the communication connection between the processor 1601 and the memory 1602 .

The processor 1601 is configured to execute the coronary artery construction program stored in the memory 1602 to realize the following steps:

Based on the first point cloud data of the coronary artery reconstructed from the collected multiple thoracic tomographic images, and based on the first point cloud data, the dynamic threshold is calculated;

Determine the predicted position of the calcified plaque on the first point cloud data based on the dynamic threshold;

On the first point cloud data, determine the coronary artery segment where the predicted position is located, and generate a cross-section along the position of the center point in the coronary artery segment;

Reconstructing the coronary vascular flow channel region in cross-section;

Based on the coronary vessel flow channel area, the second point cloud data of the coronary arteries are constructed.

In other embodiments of the present application, the processor 1601 is configured to execute the program stored in the memory 1602 to implement the following steps:

Obtain the tissue density value corresponding to each target point in the first point cloud data; determine the mean value of all tissue density values, which is the tissue density mean value of the area where the coronary arteries are located; select the threshold value calculation formula corresponding to the tissue density mean value, and calculate the dynamic threshold value.

In other embodiments of the present application, the processor 1601 is configured to execute the program stored in the memory 1602 to implement the following steps:

Based on the tissue density value range to which the tissue density mean value belongs, the threshold value calculation formula corresponding to the tissue density value range is selected; the dynamic threshold value is calculated based on the tissue density mean value and the threshold value calculation formula.

In other embodiments of the present application, the processor 1601 is configured to execute the program stored in the memory 1602 to implement the following steps:

；第二阈值计算公式为，

；第三阈值计算公式为，

；其中，

为动态阈值，

为组织密度均值，

为第一参数，

为第二参数，

为第三参数，

为第四参数，

为第五参数，

均为正数。If the mean tissue density is less than or equal to the first parameter, select the first threshold calculation formula; if the mean tissue density is greater than the first parameter and less than the second parameter, select the second threshold calculation formula; if the mean tissue density is greater than or equal to the second parameter, select the third threshold Threshold calculation formula; wherein, the first threshold calculation formula is,

; The second threshold calculation formula is,

; The third threshold calculation formula is,

;in,

is the dynamic threshold,

is the mean tissue density,

is the first parameter,

is the second parameter,

is the third parameter,

is the fourth parameter,

is the fifth parameter,

All are positive numbers.

In other embodiments of the present application, the processor 1601 is configured to execute the program stored in the memory 1602 to implement the following steps:

From all the tissue density values, determine the position of the target point whose tissue density value is greater than the dynamic threshold as the predicted position of the calcified plaque on the first point cloud data.

In other embodiments of the present application, the processor 1601 is configured to execute the program stored in the memory 1602 to implement the following steps:

The centerline point set and the segmentation information of the coronary arteries are determined based on the first point cloud data; wherein, the centerline point set includes points on the central axis of the first point cloud data; based on the segmentation information, from the centerline point set, determine The position of the center point closest to the predicted position is the coronary artery segment where the predicted position is located.

In other embodiments of the present application, the processor 1601 is configured to execute the program stored in the memory 1602 to implement the following steps:

Remove the target points corresponding to the calcified plaques in the cross-section to obtain the reconstructed coronary vascular flow channel area; take the position of the entry point in the coronary vascular flow channel area as the starting point, and in the order of distance from the entry point from near to far, Traverse all the target points in the coronary vascular flow channel area to obtain the target point cloud data of the closed area of the coronary artery segment; replace the point cloud data corresponding to the coronary artery segment of the first point cloud data with the target point cloud data, and obtain the first point cloud data. Two point cloud data.

Embodiments of the present application provide a storage medium, where one or more programs are stored in the storage medium, and the one or more programs can be executed by one or more processors to implement the following steps:

Based on the first point cloud data of the coronary artery reconstructed from the collected multiple thoracic tomographic images, and based on the first point cloud data, the dynamic threshold is calculated;

Determine the predicted position of the calcified plaque on the first point cloud data based on the dynamic threshold;

On the first point cloud data, determine the coronary artery segment where the predicted position is located, and generate a cross-section along the position of the center point in the coronary artery segment;

Reconstructing the coronary vascular flow channel region in cross-section;

Based on the coronary vessel flow channel area, the second point cloud data of the coronary arteries are constructed.

In other embodiments of the present application, the one or more programs may be executed by one or more processors to implement the following steps:

Obtain the tissue density value corresponding to each target point in the first point cloud data; determine the mean value of all tissue density values, which is the tissue density mean value of the area where the coronary arteries are located; select the threshold value calculation formula corresponding to the tissue density mean value, and calculate the dynamic threshold value.

In other embodiments of the present application, the one or more programs may be executed by one or more processors to implement the following steps:

Based on the tissue density value range to which the tissue density mean value belongs, the threshold value calculation formula corresponding to the tissue density value range is selected; the dynamic threshold value is calculated based on the tissue density mean value and the threshold value calculation formula.

In other embodiments of the present application, the one or more programs may be executed by one or more processors to implement the following steps:

；第二阈值计算公式为，

；第三阈值计算公式为，

；其中，

为动态阈值，

为组织密度均值，

为第一参数，

为第二参数，

为第三参数，

为第四参数，

为第五参数，

均为正数。If the mean tissue density is less than or equal to the first parameter, select the first threshold calculation formula; if the mean tissue density is greater than the first parameter and less than the second parameter, select the second threshold calculation formula; if the mean tissue density is greater than or equal to the second parameter, select the third threshold Threshold calculation formula; wherein, the first threshold calculation formula is,

; The second threshold calculation formula is,

; The third threshold calculation formula is,

;in,

is the dynamic threshold,

is the mean tissue density,

is the first parameter,

is the second parameter,

is the third parameter,

is the fourth parameter,

is the fifth parameter,

All are positive numbers.

In other embodiments of the present application, the one or more programs may be executed by one or more processors to implement the following steps:

From all the tissue density values, determine the position of the target point whose tissue density value is greater than the dynamic threshold as the predicted position of the calcified plaque on the first point cloud data.

In other embodiments of the present application, the one or more programs may be executed by one or more processors to implement the following steps:

The centerline point set and the segmentation information of the coronary arteries are determined based on the first point cloud data; wherein, the centerline point set includes points on the central axis of the first point cloud data; based on the segmentation information, from the centerline point set, determine The position of the center point closest to the predicted position is the coronary artery segment where the predicted position is located.

In other embodiments of the present application, the one or more programs may be executed by one or more processors to implement the following steps:

Remove the target points corresponding to the calcified plaques in the cross-section to obtain the reconstructed coronary vascular flow channel area; take the position of the entry point in the coronary vascular flow channel area as the starting point, and in the order of distance from the entry point from near to far, Traverse all the target points in the coronary vascular flow channel area to obtain the target point cloud data of the closed area of the coronary artery segment; replace the point cloud data corresponding to the coronary artery segment of the first point cloud data with the target point cloud data, and obtain the first point cloud data. Two point cloud data.

It should be noted that the above-mentioned computer storage medium/memory may be a read-only memory (Read OnlyMemory, ROM), a programmable read-only memory (Programmable Read-Only Memory, PROM), an erasable programmable read-only memory (Erasable Programmable Read Only Memory, ROM) -Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Magnetic Random Access Memory (FRAM), Flash Memory (Flash Memory), Magnetic surface memory, optical disk, or memory such as Compact Disc Read-Only Memory (CD-ROM); it can also be various terminals including one or any combination of the above memories, such as mobile phones, computers, tablet devices, Personal digital assistants, etc.

In the several embodiments provided in this application, it should be understood that the disclosed apparatus and method may be implemented in other manners. The device embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components may be combined, or Can be integrated into another system, or some features can be ignored, or not implemented. In addition, the coupling, or direct coupling, or communication connection between the components shown or discussed may be through some interfaces, and the indirect coupling or communication connection of devices or units may be electrical, mechanical or other forms. of.

In addition, each functional unit in each embodiment of the present application may all be integrated into one processing module, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above integration The unit can be implemented either in the form of hardware or in the form of hardware plus software functional units. Those of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments may be completed by program instructions related to hardware, the aforementioned program may be stored in a computer-readable storage medium, and when the program is executed, execute It includes the steps of the above method embodiments; and the aforementioned storage medium includes: a mobile storage device, a read-only memory (Read-Only Memory, ROM), a random access memory (Random Access Memory, RAM), a magnetic disk or an optical disk and other various A medium on which program code can be stored.

The features disclosed in several method or device embodiments provided in this application can be combined arbitrarily without conflict to obtain new method embodiments or device embodiments.

The above are only specific embodiments of the present application, but the protection scope of the present application is not limited to this. should be covered within the scope of protection of this application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims (10)

1. a construction method of coronary artery, is characterized in that, described method comprises: Calculate the dynamic threshold based on the first point cloud data of the coronary artery reconstructed from the collected multiple thoracic tomographic images, and based on the first point cloud data; determining the predicted position of the calcified plaque on the first point cloud data based on the dynamic threshold; On the first point cloud data, determine the coronary artery segment where the predicted position is located, and generate a cross-section along the position of the center point in the coronary artery segment; removing the target point corresponding to the calcified plaque in the cross section to obtain a reconstructed coronary vascular flow channel area; Based on the coronary vessel flow channel region, second point cloud data of the coronary arteries are constructed. 2. The method according to claim 1, wherein the calculating a dynamic threshold based on the first point cloud data comprises: Obtain the tissue density value corresponding to each target point in the first point cloud data; Determine the mean value of all tissue density values, which is the mean value of tissue density in the region where the coronary arteries are located; A threshold calculation formula corresponding to the mean tissue density is selected, and the dynamic threshold is calculated. 3. The method according to claim 2, wherein the selecting a threshold value calculation formula corresponding to the tissue density mean value, and calculating the dynamic threshold value, comprises: Selecting a threshold value calculation formula corresponding to the tissue density value range based on the tissue density value range to which the tissue density mean value belongs; The dynamic threshold is calculated based on the mean tissue density and the threshold calculation formula. 4. The method according to claim 2, wherein the selecting a threshold value calculation formula corresponding to the tissue density value range based on the tissue density value range to which the tissue density mean value belongs, comprising: If the mean tissue density is less than or equal to the first parameter, select the first threshold calculation formula; If the mean tissue density is greater than the first parameter and less than the second parameter, select a second threshold calculation formula; If the mean tissue density is greater than or equal to the second parameter, select a third threshold calculation formula; Wherein, the first threshold calculation formula is,

; The second threshold calculation formula is,

; The third threshold calculation formula is,

; Among them, the

is the dynamic threshold, the meanCoronary is the mean value of the tissue density, the

for the first parameter, the

for the second parameter, the

is the third parameter, the

is the fourth parameter, the

is the fifth parameter, the

All are positive numbers. The method according to any one of claims 2 to 4, wherein the determining the predicted position of the calcified plaque on the first point cloud data based on the dynamic threshold comprises: From all the tissue density values, the position of the target point whose tissue density value is greater than the dynamic threshold is determined as the predicted position of the calcified plaque on the first point cloud data. 6. The method according to claim 1, wherein, on the first point cloud data, determining the coronary artery segment where the predicted position is located, comprising: Determine a centerline point set and segment information of coronary arteries based on the first point cloud data; wherein, the centerline point set includes points on the central axis of the first point cloud data; Based on the segment information, from the centerline point set, it is determined that the position of the center point closest to the predicted position is the coronary artery segment where the predicted position is located. 7. The method according to claim 1, wherein the constructing the second point cloud data of the coronary artery based on the coronary vascular flow channel region comprises: Taking the position of the entry point in the coronary vascular flow channel area as a starting point, traverse all the target points in the coronary vascular flow channel area in order of distance from the entry point from near to far, and obtain the coronary vascular flow channel area. The target point cloud data of the closed area of the arterial segment; The point cloud data corresponding to the coronary artery segment of the first point cloud data is replaced with the target point cloud data to obtain the second point cloud data. 8. A device for constructing coronary arteries, wherein the device comprises: a processing module, used for the first point cloud data of the coronary arteries reconstructed from the collected multiple thoracic tomographic images, and calculating a dynamic threshold based on the first point cloud data; a determination module, configured to determine the predicted position of the calcified plaque on the first point cloud data based on the dynamic threshold; The determining module is further configured to determine, on the first point cloud data, the coronary artery segment where the predicted position is located; a generation module for generating a cross-section along the position of a center point within the coronary artery segment; a reconstruction module, configured to remove the target point corresponding to the calcified plaque in the cross section to obtain a reconstructed coronary vascular flow channel area; a building module for building second point cloud data of the coronary artery based on the coronary vessel flow channel region. 9. A terminal, wherein the terminal comprises: memory for storing executable instructions; A processor, configured to execute the executable instructions stored in the memory, to implement the method for constructing a coronary artery according to any one of claims 1 to 7. 10. A storage medium, characterized in that it stores executable instructions, when the executable instructions are executed, for causing a processor to execute the method for constructing a coronary artery according to any one of claims 1 to 7 .

Images