A method and device for overall assessment of coronary plaque risk based on medical images

Abstract

The application provides a method and device for comprehensively evaluating coronary plaque risk based on medical images. The method for comprehensively evaluating coronary plaque risk based on medical images comprises the following steps: acquiring a coronary strain parameter according to a coronary tree segmented image; acquiring plaque morphology parameter information according to a plaque segmented image; acquiring hemodynamic parameter information according to a resting state time sequence image; and acquiring a plaque risk index according to the coronary strain parameter, the plaque morphology parameter information and the hemodynamic parameter information. The application firstly segments a heart structure and calculates a coronary strain parameter according to the morphology of the coronary artery and the strain difference between different time phases. The plaque morphology parameter is calculated according to the position, morphology and texture structure of the plaque. Then, the computational fluid dynamics method is used to calculate the hemodynamic parameter of the coronary artery. Finally, a comprehensive index is designed to jointly determine the coronary plaque risk.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image processing, in particular to a method for comprehensively evaluating the risk of coronary plaque based on medical images and a device for comprehensively evaluating the risk of coronary plaque based on medical images. BACKGROUND

[0002] Coronary plaque shedding leading to cardiac ischemia is an important cause of myocardial infarction. Therefore, judging the vulnerability of coronary plaque and predicting the probability and risk of coronary plaque shedding has great significance for the prevention of myocardial infarction. The present application proposes a method for comprehensively judging the risk of coronary plaque according to the strain parameters of coronary artery in different phases of load state and resting state, the morphological parameters of plaque, and the hemodynamic parameters.

[0003] There are a large number of literatures in the prior art for predicting the risk of plaque, but they mainly focus on a single type of index, such as based on hemodynamics or based on plaque morphology. However, since the human body is a complex system, the risk of plaque is affected by many complex factors, and a single index cannot represent the risk of plaque, which is far from the actual situation. Some literatures also use multiple methods to jointly evaluate the risk of plaque, and the research results are as follows:

[0004] Some prior art discloses a method for predicting the vulnerability of plaque based on plaque morphology parameters and hemodynamic parameters. This method is mainly based on CT and OCT images and uses convolutional neural network to predict. This method directly predicts the result using neural network, and the black box effect makes it difficult for doctors to judge the specific influence degree of various indexes and make targeted judgments. Moreover, this method requires the use of OCT, which reduces the scope of adaptation. This method also does not use the strain parameters of coronary artery between different phases.

[0005] Some prior art discloses a research based on the radial wall strain (RWS) of coronary artery in angiography, which mainly focuses on using RWS to evaluate the stability of plaque in patients with coronary heart disease and predict the risk of future acute myocardial infarction (AMI) events. However, this method only evaluates the strain of coronary artery and does not consider the influence of plaque and hemodynamic factors.

[0006] Or combine plaque morphological parameters with hemodynamic parameters to analyze and evaluate the cardiovascular function and plaque characteristics of patients with acute coronary syndrome, including plaque and hemodynamic quantitative analysis of each lesion and each blood vessel, which can identify criminal lesions and help identify patients with increased risk of adverse cardiac events. This method has been verified in multi-country and multi-center studies, but this method does not take into account the differences in the shape of human coronary artery itself, and the same plaque may not perform the same in different positions of different coronary arteries.

[0007] In summary, plaque risk and coronary situation, plaque situation, and blood flow situation are all very relevant, but the prior art often only focuses on one side and cannot comprehensively evaluate plaque risk. SUMMARY

[0008] The present application provides a method for comprehensively evaluating coronary plaque risk based on medical images, which comprises:

[0009] obtaining coronary strain parameters according to the coronary tree segmented image;

[0010] obtaining plaque morphology parameter information according to the plaque segmented image;

[0011] obtaining hemodynamic parameter information according to the resting state time series image;

[0012] obtaining a plaque risk index according to the coronary strain parameters, plaque morphology parameter information, and hemodynamic parameter information.

[0013] Optionally, the obtaining of the coronary strain parameters according to the coronary tree segmented image comprises:

[0014] obtaining a load state time series image and a resting state time series image;

[0015] obtaining segmented coronary tree images at different time phases according to the load state time series image and the resting state time series image, respectively;

[0016] performing the following operations on the segmented coronary tree images at different time phases:

[0017] extracting the center line of the coronary artery in the coronary tree image and straightening and unfolding the coronary artery to obtain a straightened coronary artery image;

[0018] detecting calcified plaque and non-calcified plaque on the coronary artery according to the obtained coronary artery image, when calcified plaque or non-calcified plaque is detected, performing image segmentation on the detected calcified plaque or the detected non-calcified plaque, thereby obtaining a calcified plaque image or a non-calcified plaque image, the calcified plaque image or the non-calcified plaque image constitutes the plaque segmented image; wherein one calcified plaque has a plurality of calcified plaque images at different time phases, and one non-calcified plaque has a plurality of non-calcified plaque images at different time phases;

[0019] obtaining the corresponding coronary strain parameters of each calcified plaque according to the calcified plaque images at different time phases of each calcified plaque;

[0020] obtaining the corresponding coronary strain parameters of each non-calcified plaque according to the non-calcified plaque images at different time phases of each non-calcified plaque.

[0021] Optionally, the acquiring of the corresponding coronary artery strain parameter of each calcified plaque according to the calcified plaque images of different phases of each calcified plaque comprises:

[0022] For each of the calcified plaque images of different phases belonging to the same calcified plaque, the following operations are performed:

[0023] The minimum diameter of the coronary artery lumen of each calcified plaque image is calculated respectively to obtain a sequence of the minimum diameter of the coronary artery, and the maximum value D max and the minimum value D min of the sequence of the minimum diameter of the coronary artery are taken, and the coronary artery strain parameter of each calcified plaque is acquired by the following formula:

[0024] e = (D max -D min ) / D max ;

[0025] Wherein, e is the coronary artery strain parameter.

[0026] Optionally, the acquiring of the corresponding coronary artery strain parameter of each non-calcified plaque according to the non-calcified plaque images of different phases of each non-calcified plaque comprises:

[0027] For each of the non-calcified plaque images of different phases belonging to the same non-calcified plaque, the following operations are performed:

[0028] The minimum diameter of the coronary artery lumen of each non-calcified plaque image is calculated respectively to obtain a sequence of the minimum diameter of the coronary artery, and the maximum value D max and the minimum value D min of the sequence of the minimum diameter of the coronary artery are taken, and the coronary artery strain parameter of each non-calcified plaque is acquired by the following formula:

[0029] e = (D max -D min ) / D max ;

[0030] Wherein, e is the coronary artery strain parameter.

[0031] Optionally, the plaque morphology parameter information comprises plaque component information, plaque morphology information, plaque position information and plaque density information.

[0032] Optionally, the acquiring of the hemodynamic parameter information according to the resting state time sequence images comprises:

[0033] The segmented coronary tree images of each phase are compared with each other to obtain a resting state time sequence image of the phase with the smallest diameter of the coronary artery lumen, and the hemodynamics of the coronary artery in the group of resting state time sequence images is simulated to obtain the hemodynamic parameter.

[0034] Optionally, the plaque risk index is obtained according to the coronary strain parameter, plaque morphology parameter information and hemodynamic parameter information by the following formula:

[0035] F(X) = a1Q + a2(ΔP) + a3R + a4τ + a5(ΔR) + a6(ΔQ) + a7L + … + b1x1 + … + b n x n ;

[0036] Wherein, a, b are to be fitted parameters; x is other coronary strain parameters, plaque morphology parameters and hemodynamic parameters; Q is flow rate; R is the coronary lumen radius at the most narrow place; ΔP is the pressure difference before and after the stenosis; L is the length of the stenosis section; μ is the blood viscosity; τ is the wall shear stress; ΔR is the difference of the lumen in systole and diastole, that is, strain; ΔQ is the difference between the flow rate assuming that the lumen R remains unchanged in systole and the flow rate assuming that the lumen R remains unchanged in diastole.

[0037] Optionally, the difference of the lumen in systole and diastole ΔR is obtained by the following formula:

[0038]

[0039] Wherein, Q is flow rate, R is the coronary lumen radius at the most narrow place, ΔP is the pressure difference before and after the stenosis, L is the length of the stenosis section, μ is the blood viscosity, and τ is the wall shear stress.

[0040] Optionally, the flow rate Q is obtained by the following formula:

[0041]

[0042] Wherein, Q is flow rate, R is the coronary lumen radius at the most narrow place, ΔP is the pressure difference before and after the stenosis, L is the length of the stenosis section, μ is the blood viscosity, and τ is the wall shear stress.

[0043] The wall shear stress τ is obtained by the following formula:

[0044]

[0045] Wherein, Q is flow rate, R is the coronary lumen radius at the most narrow place, ΔP is the pressure difference before and after the stenosis, L is the length of the stenosis section, μ is the blood viscosity, and τ is the wall shear stress.

[0046] The application also provides a device for comprehensively evaluating coronary plaque risk based on medical images, characterized in that the device for comprehensively evaluating coronary plaque risk based on medical images comprises:

[0047] a coronary strain parameter acquisition module, configured to acquire a coronary strain parameter according to the coronary tree segmentation image;

[0048] a plaque morphology parameter information acquisition module, configured to acquire plaque morphology parameter information according to the plaque segmentation image;

[0049] a hemodynamic parameter information acquisition module, configured to acquire hemodynamic parameter information according to the resting state time sequence image;

[0050] a plaque risk index acquisition module, configured to acquire a plaque risk index according to the coronary strain parameter, the plaque morphology parameter information and the hemodynamic parameter information.

[0051] The method for comprehensively evaluating coronary plaque risk based on medical images according to the present application first segments the heart structure to segment the coronary artery, the left ventricle and the plaque on the coronary artery. According to the morphology of the coronary artery and the strain difference between different time phases, the coronary strain parameter is calculated. According to the position, morphology and texture structure of the plaque, the plaque morphology parameter is calculated. Then, the computational fluid dynamics (CFD) method is used to calculate the hemodynamic parameters of the coronary artery, including the pressure, flow, flow velocity, wall shear stress, FFR, etc. Finally, a comprehensive index is designed to combine the coronary strain parameter, the plaque morphology parameter and the hemodynamic parameter to jointly judge the coronary plaque risk. BRIEF DESCRIPTION OF DRAWINGS

[0052] The accompanying drawings are included to provide a further understanding of the present application, and constitute a part of the specification, illustrate the present application and are used to explain the present application together with the embodiments of the present application, and do not constitute a limitation to the present application. In the drawings:

[0053] Figure 1 a flowchart of the method for comprehensively evaluating coronary plaque risk based on medical images according to an embodiment of the present application;

[0054] Figure 2 a schematic diagram of the coronary tree segmentation image according to an embodiment of the present application;

[0055] Figure 3 a schematic diagram of the straightened coronary artery according to an embodiment of the present application;

[0056] Figure 4 a schematic diagram of the coronary artery diameter according to an embodiment of the present application;

[0057] Figure 5 a CFD meshing schematic diagram according to an embodiment of the present application. DETAILED DESCRIPTION

[0058] The preferred embodiments of the present application are described below in conjunction with the accompanying drawings, it should be understood that the preferred embodiments described herein are only used to explain and illustrate the present application, and are not used to limit the present application.

[0059] As shown in the method for comprehensively evaluating coronary plaque risk based on medical images, Figure 1

[0060] Obtaining coronary strain parameters according to the coronary tree segmented images;

[0061] Obtaining plaque morphology parameter information according to the plaque segmented images;

[0062] Obtaining hemodynamic parameter information according to the resting state time series images;

[0063] Obtaining plaque risk index according to the coronary strain parameters, plaque morphology parameter information and hemodynamic parameter information.

[0064] In the present embodiment, the obtaining of the coronary strain parameters according to the coronary tree segmented images comprises:

[0065] Obtaining the stress state time series images and the resting state time series images;

[0066] Obtaining the segmented coronary tree images at different time phases respectively according to the stress state time series images and the resting state time series images;

[0067] Performing the following operations on the segmented coronary tree images at different time phases:

[0068] Extracting the center line of the coronary artery in the coronary tree image and straightening and unfolding the coronary artery to obtain a straightened coronary artery image;

[0069] Detecting calcified plaque and non-calcified plaque on the coronary artery according to the obtained coronary artery image, when detecting the calcified plaque or the non-calcified plaque, performing image segmentation on the detected calcified plaque or the detected non-calcified plaque, thereby obtaining a calcified plaque image or a non-calcified plaque image, the calcified plaque image or the non-calcified plaque image constitutes the plaque segmented images; wherein one calcified plaque has a plurality of calcified plaque images at different time phases, and one non-calcified plaque has a plurality of non-calcified plaque images at different time phases;

[0070] Obtaining the corresponding coronary strain parameters of each calcified plaque according to the calcified plaque images at different time phases of each calcified plaque;

[0071] Obtaining the corresponding coronary strain parameters of each non-calcified plaque according to the non-calcified plaque images at different time phases of each non-calcified plaque.

[0072] ​In the embodiment, the method further comprises the following steps of:

[0073] For each of the calcified plaque images of different time phases belonging to the same calcified plaque, the following operations are performed:

[0074] The minimum diameter of the coronary lumen of each calcified plaque image is calculated respectively to obtain a sequence of the minimum diameters of the coronary lumen, and the maximum value D max and the minimum value D min of the sequence of the minimum diameters of the coronary lumen are taken, and the coronary strain parameter of each calcified plaque is obtained by the following formula:

[0075] e=(D max -D min ) / D max ;

[0076] wherein, e is the coronary strain parameter.

[0077] In the embodiment, the method further comprises the following steps of:

[0078] For each of the non-calcified plaque images of different time phases belonging to the same non-calcified plaque, the following operations are performed:

[0079] The minimum diameter of the coronary lumen of each non-calcified plaque image is calculated respectively to obtain a sequence of the minimum diameters of the coronary lumen, and the maximum value D max and the minimum value D min of the sequence of the minimum diameters of the coronary lumen are taken, and the coronary strain parameter of each non-calcified plaque is obtained by the following formula:

[0080] e=(D max -D min ) / D max ;

[0081] wherein, e is the coronary strain parameter.

[0082] In the embodiment, the plaque morphology parameter information comprises plaque component information, plaque morphology information, plaque position information and plaque density information.

[0083] In the embodiment, the method further comprises the following steps of:

[0084] The segmented coronary tree images of different time phases are compared with each other to obtain a resting state time sequence image of a time phase with the minimum diameter of the coronary lumen, and the coronary blood flow dynamics simulation is performed on the coronary artery in the group of resting state time sequence images to obtain the blood flow dynamics parameters.

[0085] In the embodiment, the plaque risk index is obtained according to the coronary strain parameters, plaque morphology parameter information and hemodynamic parameter information by the following formula:

[0086] F(X) = a1Q + a2(ΔP) + a3R + a4τ + a5(ΔR) + a6(ΔQ) + a7L + … + b1x1 + … + b n x n ;

[0087] Wherein, a, b are to be fitted parameters; x is other coronary strain parameters, plaque morphology parameters and hemodynamic parameters, such as plaque volume, plaque maximum diameter, etc.; Q is flow rate; R is the coronary lumen radius at the most stenosis; ΔP is the pressure difference before and after the stenosis; L is the length of the stenosis segment, μ is the blood viscosity; τ is the wall shear stress; ΔR is the difference of the lumen in systole and diastole, that is, strain, ΔQ is the difference between the flow rate assuming that the lumen R remains unchanged in systole and the flow rate assuming that the lumen R remains unchanged in diastole.

[0088] In the embodiment, the difference of the lumen in systole and diastole ΔR is obtained by the following formula:

[0089]

[0090] Wherein, Q is flow rate, R is the coronary lumen radius at the most stenosis, ΔP is the pressure difference before and after the stenosis, L is the length of the stenosis segment, μ is the blood viscosity, τ is the wall shear stress.

[0091] In the embodiment, the flow rate Q is obtained by the following formula:

[0092]

[0093] Wherein, Q is flow rate, R is the coronary lumen radius at the most stenosis, ΔP is the pressure difference before and after the stenosis, L is the length of the stenosis segment, μ is the blood viscosity, τ is the wall shear stress.

[0094] The wall shear stress τ is obtained by the following formula:

[0095]

[0096] Wherein, Q is flow rate, R is the coronary lumen radius at the most stenosis, ΔP is the pressure difference before and after the stenosis, L is the length of the stenosis segment, μ is the blood viscosity, τ is the wall shear stress.

[0097] The application will be further described in detail below with reference to the accompanying drawings. It can be understood that the examples do not constitute any limitation on the application.

[0098] Reference is made to Figure 1, obtain the segmented coronary tree. The segmentation of the coronary tree can adopt a deep learning method based on nn-Unet and topological analysis, first segment the coronary artery by nn-Unet, and then perform topological analysis on the coronary artery to ensure the continuity of the coronary artery and remove impurities that are not coronary arteries.

[0099] Referring to Figure 2 Then, the center line of the coronary artery is extracted, and the coronary artery is straightened and unfolded (specifically, based on the skeletonization method, the coronary artery segmentation result is skeletonized into a center line, and then according to the normal direction of the center line, the cross section at each point on the center line is intercepted, and the two-dimensional cross section is stacked into a three-dimensional body data, which is a three-dimensional body after the coronary artery is straightened, and the coronary artery is located at the center of the three-dimensional body). Obtain the straightened coronary artery.

[0100] Referring to Figure 3 , adopt CNN neural network to detect calcified and non-calcified plaques on the center line. At the detected plaque, adopt nn-Unet to segment the plaque. Then, the plaque is removed from the coronary tree to obtain the coronary tree after the plaque is removed, and the coronary of the coronary tree represents the coronary lumen through which the blood flows. On the cross section of the coronary lumen at the plaque, the maximum diameter and the minimum diameter of the lumen can be calculated, wherein the minimum diameter represents the diameter of the coronary at the stenosis.

[0101] On different time phases, the minimum diameter of the coronary lumen at the same plaque is calculated respectively to obtain a sequence of coronary minimum diameters, Dt1, Dt2,..., Dtn, and the maximum value D max and the minimum value D min . The coronary strain can be defined as e=(D max -D min ) / D max ; wherein e is the coronary strain parameter.

[0102] In this embodiment, the plaque morphology parameters are calculated by the following method:

[0103] In the time series images, the calcified and non-calcified plaques on the coronary artery are detected according to the obtained coronary images, the sequence images showing the clearest are selected from the images corresponding to each plaque, and the morphology parameters of the plaques are calculated. Including but not limited to the following parameters:

[0104] Plaque composition:

[0105] Lipid core size: larger lipid core is related to plaque vulnerability.

[0106] Fiber cap thickness: thin fiber cap (usually <65 microns) is a feature of vulnerable plaques.

[0107] Calcification degree: calcified plaques are relatively stable, but mixed calcification may increase vulnerability.

[0108] Plaque morphology:

[0109] Plaque volume: Larger plaques are more likely to rupture.

[0110] Plaque area: Area of plaque on cross-section of vessel.

[0111] Positive remodeling: Plaque expands outward, potentially increasing vulnerability.

[0112] Negative remodeling: Plaque contracts inward, potentially decreasing vulnerability.

[0113] Plaque location:

[0114] Proximal or bifurcation plaques: These are more likely to rupture.

[0115] Plaque eccentricity: Eccentric plaques are more vulnerable than concentric plaques.

[0116] Plaque density:

[0117] CT values (HU units): Low-density plaques (<30 HU) are associated with lipid cores, and high-density plaques (>130 HU) are associated with calcification.

[0118] In this embodiment, the calculation of hemodynamic parameters can adopt the following technical solutions:

[0119] In the time series images, according to the coronary lumen calculated in step one, the sequence with the minimum minimum diameter value of the lumen at the plaque is selected, and on the basis of the segmented coronary tree, the coronary blood flow is simulated, and the steps are as follows:

[0120] 1. Mesh division.

[0121] Mesh division can adopt the delauney mesh division algorithm, which can ensure that the triangular mesh divided is an acute triangle. When mesh division, attention should also be paid to according to the radian to distinguish the mesh size, in the aorta and other larger volume, can adopt larger mesh, the number of mesh is sparse. While in the coronary, especially in the fine branches of the coronary, smaller mesh should be used, and the number of mesh is dense, so as to well represent the characteristics of the fine part of the blood vessel, and the mesh divided coronary blood vessel is as shown in Figure 4 .

[0122] 2. Simulate by CFD method

[0123] CFD is the abbreviation of computational fluid dynamics. After meshing, the present application uses the method of computational fluid dynamics to calculate the blood flow parameters in the coronary artery. First, the blood flow related boundary condition parameters need to be calculated. These parameters can be calculated from the image or measured separately from the image. The blood flow related parameters that need to be extracted by the present application include but are not limited to the patient's blood pressure, cardiac output, coronary artery inlet and outlet area, etc. After the blood flow related parameters are extracted, they are input into the CFD calculation solver as boundary conditions and meshes, and the blood flow results at each grid point, such as blood pressure, flow rate, wall shear stress, flow velocity and blood flow direction, etc. can be calculated.

[0124] 3. Calculate the hemodynamic parameters

[0125] After the CFD simulation, the hemodynamic parameters can be directly obtained. In another embodiment, the results of CFD can be used as a gold standard to train an AI neural network, and then the AI neural network is used to directly calculate the hemodynamic parameters. The advantage of the AI method is faster than using CFD simulation.

[0126] In this embodiment, the hemodynamic equation is introduced into the regression fitting process, and the correlation between various parameters is ensured at the same time of linear regression. In this way, the risk can be judged according to the risk index to judge the prognosis of the patient, and on the other hand, due to the sufficient correlation between various parameters, the user can understand the influence of various parameters on the prognosis of the patient according to the weight of each parameter, and make a comprehensive judgment.

[0127] First, a hemodynamic model of various parameters reflecting the risk of plaque is established. The difference between this model and the conventional hemodynamic model is that the strain of the coronary artery is considered.

[0128] According to the dynamic equation of Poiseuille flow in a circular tube, the relationship equation between the blood flow parameters at the stenosis and the coronary morphology parameters can be obtained:

[0129]

[0130] Where Q is the flow rate, R is the coronary lumen radius at the most stenotic site, ΔP is the pressure difference before and after the stenosis, L is the length of the stenosis, μ is the blood viscosity, and τ is the wall shear stress.

[0131] We introduce the coronary strain parameter into the model. The coronary strain can be defined as the difference between the lumen at the most stenotic site during systole and diastole, i.e. ΔR. The relationship between blood flow and strain is:

[0132] Where Q is the flow rate, R is the coronary lumen radius at the most stenotic site, ΔP is the pressure difference before and after the stenosis, L is the length of the stenosis, μ is the blood viscosity, and τ is the wall shear stress.

[0133] Then, we establish a constraint-based linear regression model, based on the above constraints, to solve the linear regression model:

[0134] F(X)=a1Q+a2(ΔP)+a3R+a4τ+a5(ΔR)+a6(ΔQ)+a7L+…+b1x1+…+b n x n ;

[0135] Where a, b are to be fitted parameters; x is other coronary strain parameters, plaque morphology parameters and hemodynamic parameters; Q is flow rate; R is the radius of the most narrow coronary lumen; ΔP is the pressure difference before and after the stenosis; L is the length of the stenosis; μ is the blood viscosity; τ is the wall shear stress; ΔR is the difference between the systolic and diastolic lumen, that is, the strain; ΔQ is the difference between the flow rate assuming that the lumen R remains unchanged in the systolic period and the flow rate assuming that the lumen R remains unchanged in the diastolic period.

[0136] Solving the constraint optimization equation, the constraint least squares method can be used to solve it. Since the above constraint equation is a nonlinear equation, it is more complex to solve, and can be calculated by introducing the iterative numerical calculation method of the Lagrange function.

[0137] The present application proposes a method for comprehensively evaluating the risk of coronary plaque based on medical images, which has the following advantages:

[0138] 1. Combining the load state time series image and the resting state time series image, extracting the coronary artery, and calculating the strain of the coronary artery at different phases, more accurate strain values can be obtained.

[0139] 2. Considering the coronary strain, plaque morphology parameters, and hemodynamic parameters at the same time for evaluating the risk of plaque.

[0140] The present application also provides a device for comprehensively evaluating the risk of coronary plaque based on medical images, which comprises a coronary strain parameter acquisition module, a plaque morphology parameter information acquisition module, a hemodynamic parameter information acquisition module, and a plaque risk index acquisition module, wherein,

[0141] The coronary strain parameter acquisition module is used to acquire the coronary strain parameter according to the coronary tree segmentation image;

[0142] The plaque morphology parameter information acquisition module is used to acquire the plaque morphology parameter information according to the plaque segmentation image;

[0143] The hemodynamic parameter information acquisition module is used to acquire the hemodynamic parameter information according to the resting state time series image;

[0144] The plaque risk index acquisition module is configured to acquire a plaque risk index according to the coronary strain parameter, plaque morphology parameter information, and hemodynamic parameter information.

[0145] Obviously, various modifications and changes can be made to the present application without departing from the spirit and scope thereof. Accordingly, it is intended that all such modifications and changes be included within the scope of the application as long as the modified and changed embodiments fall within the scope of the claims and their equivalents.

Claims

1. A method for overall assessment of coronary plaque risk based on medical images, characterized by, The method for comprehensively evaluating coronary plaque risk based on medical images comprises: acquiring coronary strain parameters according to coronary tree segmented images; acquiring plaque morphology parameter information according to plaque segmented images; acquiring hemodynamic parameter information according to resting state time series images; acquiring plaque risk index according to the coronary strain parameters, plaque morphology parameter information and hemodynamic parameter information; the acquiring coronary strain parameters according to coronary tree segmented images comprises: acquiring load state time series images at different time phases and resting state time series images at different time phases; obtaining segmented coronary tree images at different time phases respectively according to the load state time series images and resting state time series images; performing the following operations on the segmented coronary tree images at different time phases: extracting the center line of the coronary artery in the coronary tree image and straightening and unfolding the coronary artery to obtain a straightened coronary artery image; detecting calcified plaque and non-calcified plaque on the coronary artery according to the obtained coronary artery image, when detecting the calcified plaque or non-calcified plaque, performing image segmentation on the detected calcified plaque or non-calcified plaque, thereby obtaining a calcified plaque image or non-calcified plaque image, the calcified plaque image or non-calcified plaque image constitutes the plaque segmented image; wherein one calcified plaque has a plurality of calcified plaque images at different time phases, and one non-calcified plaque has a plurality of non-calcified plaque images at different time phases; acquiring the corresponding coronary strain parameters of each calcified plaque according to the calcified plaque images at different time phases of each calcified plaque; acquiring the corresponding coronary strain parameters of each non-calcified plaque according to the non-calcified plaque images at different time phases of each non-calcified plaque; the plaque morphology parameter information comprises plaque component information, plaque morphology information, plaque position information and plaque density information; the acquiring plaque risk index according to the coronary strain parameters, plaque morphology parameter information and hemodynamic parameter information is acquired by the following formula: F(X) = a1Q + a2(ΔP) + a3R + a4τ + a5(ΔR) + a6(ΔQ) + a7L +... + b1x1 +... + b n x n ; wherein a and b are to-be-fitted parameters; x is other coronary strain parameters, plaque morphology parameter information and hemodynamic parameters; Q is flow rate; R is the coronary lumen radius at the most narrow place; ΔP is the pressure difference before and after the stenosis; L is the length of the stenosis section, μ is the blood viscosity; τ is the wall shear stress; ΔR is the difference of the lumen in the systole and diastole, that is, the strain, and ΔQ is the difference between the flow rate assuming that the lumen R remains unchanged in the systole and the flow rate assuming that the lumen R remains unchanged in the diastole.

2. The method for assessing the risk of coronary plaque based on medical images according to claim 1, wherein, the acquiring the corresponding coronary strain parameters of each calcified plaque according to the calcified plaque images at different time phases of each calcified plaque comprises: performing the following operations on each of the calcified plaque images at different time phases belonging to the same calcified plaque: The minimum diameter of the coronary vessel lumen of each calcified plaque image is calculated respectively to obtain a sequence of the minimum diameter of the coronary vessel, and the maximum value D1 in the sequence of the minimum diameter of the coronary vessel is taken max and the minimum value D1 min The coronary strain parameter of each calcified plaque is obtained through the following formula: e1 = (D1 max -D1 min ) / D1 max ; wherein e1 is the coronary strain parameter.

3. The method for assessing the risk of coronary plaque based on medical images according to claim 2, wherein, the acquiring the corresponding coronary strain parameters of each non-calcified plaque according to the non-calcified plaque images at different time phases of each non-calcified plaque comprises: performing the following operations on each of the non-calcified plaque images at different time phases belonging to the same non-calcified plaque: The minimum diameter of the coronary vessel lumen of each non-calcified plaque image is calculated respectively to obtain a sequence of the minimum diameters of the coronary vessels, and the maximum value D2 in the sequence of the minimum diameters of the coronary vessels is taken max and the minimum value D2 min The coronary strain parameter of each non-calcified plaque is obtained through the following formula: e2= (D2 max -D2 min ) / D2 max ; wherein e2 is the coronary strain parameter.

4. The method for assessing the risk of coronary plaque based on medical images according to claim 3, wherein, the acquiring hemodynamic parameter information according to resting state time series images comprises: The rest state time sequence images of the coronary artery lumen in each phase are obtained by comparing the segmented coronary tree images in each phase with each other, and the hemodynamic parameters are obtained by performing hemodynamic simulation on the coronary arteries in the group of rest state time sequence images.

5. The method for assessing the risk of coronary plaque based on medical images according to claim 4, wherein, The difference ΔR of the lumen in the systole and diastole is obtained by the following formula: ； Wherein, Q is the flow, R is the radius of the coronary artery lumen at the most narrow place, ΔP is the pressure difference before and after the stenosis, L is the length of the stenosis section, μ is the blood viscosity, and τ is the wall shear stress.

6. The method for assessing the risk of coronary plaque based on medical images according to claim 5, wherein, The flow Q is obtained by the following formula: ； Wherein, Q is the flow, R is the radius of the coronary artery lumen at the most narrow place, ΔP is the pressure difference before and after the stenosis, L is the length of the stenosis section, μ is the blood viscosity, and τ is the wall shear stress. The wall shear stress τ is obtained by the following formula: ； Wherein, Q is the flow, R is the radius of the coronary artery lumen at the most narrow place, ΔP is the pressure difference before and after the stenosis, L is the length of the stenosis section, μ is the blood viscosity, and τ is the wall shear stress.

7. A device for comprehensively assessing the risk of coronary plaques based on medical images, characterized in that, The device for comprehensively evaluating the risk of coronary artery plaque based on medical images comprises: a coronary artery strain parameter acquisition module, which is configured to acquire coronary artery strain parameters according to coronary tree segmentation images; a plaque morphology parameter information acquisition module, which is configured to acquire plaque morphology parameter information according to plaque segmentation images; a hemodynamic parameter information acquisition module, which is configured to acquire hemodynamic parameter information according to rest state time sequence images; a plaque risk index acquisition module, which is configured to acquire a plaque risk index according to the coronary artery strain parameters, plaque morphology parameter information, and hemodynamic parameter information; the coronary artery strain parameters are acquired according to the coronary tree segmentation images, which comprises: obtaining load state time sequence images in different phases and rest state time sequence images in different phases; obtaining segmented coronary tree images in different phases according to the load state time sequence images and rest state time sequence images respectively; the segmented coronary tree images in different phases are subjected to the following operations: extracting the center line of the coronary arteries in the coronary tree images and straightening and unfolding the coronary arteries to obtain straightened coronary artery images; detecting calcified plaques and non-calcified plaques on the coronary arteries according to the obtained coronary artery images, performing image segmentation on the detected calcified plaques or non-calcified plaques when the calcified plaques or non-calcified plaques are detected, thereby obtaining calcified plaque images or non-calcified plaque images, and the calcified plaque images or non-calcified plaque images constitute the plaque segmentation images; wherein one calcified plaque has multiple calcified plaque images in different phases, and one non-calcified plaque has multiple non-calcified plaque images in different phases; acquiring the corresponding coronary artery strain parameters of each calcified plaque according to the calcified plaque images of each calcified plaque in different phases; acquiring the corresponding coronary artery strain parameters of each non-calcified plaque according to the non-calcified plaque images of each non-calcified plaque in different phases; the plaque morphology parameter information comprises plaque component information, plaque morphology information, plaque position information, and plaque density information; The plaque risk index is obtained according to the coronary strain parameter, plaque morphology parameter information and hemodynamic parameter information by the following formula: F(X) = a1Q + a2(ΔP) + a3R + a4τ + a5(ΔR) + a6(ΔQ) + a7L +... + b1x1 +... + b n x n ; Wherein, a, b are to be fitted parameters; x is other coronary strain parameter, plaque morphology parameter information and hemodynamic parameter; Q is flow; R is the coronary lumen radius at the most narrow place; ΔP is the pressure difference before and after the stenosis; L is the length of the stenosis section, μ is the blood viscosity; τ is the wall shear stress; ΔR is the difference of the lumen in the systole and diastole, that is, the strain, and ΔQ is the difference between the flow assuming that the lumen R remains unchanged in the systole and the flow assuming that the lumen R remains unchanged in the diastole.

Images