A simple and reliable method for determining the left-right coronary artery shunt fraction and calculating FFR

Abstract

A method for simply and reliably determining the left and right coronary artery shunt fraction and a method for calculating FFR, belonging to the field of numerical simulation, comprises the following steps: constructing a personalized three-dimensional model based on a real coronary artery tomographic image; performing model gridding preprocessing based on the personalized model; extracting target parameters based on the preprocessing model and establishing a personalized shunt model, i.e., a personalized shunt model based on cross-sectional area; (based on cross-sectional area) a personalized left and right coronary artery flow method is: #imgabs0# The cross-sectional area refers to the cross-sectional area at the interface with the aorta. Q-flow value, S-cross-sectional area, LCA-left coronary artery, RCA-right coronary artery. The method of the present invention can more accurately calculate the personalized FFR.

Description

Technical Field

[0001] The invention belongs to the field of numerical simulation of coronary hemodynamics and is a simple and reliable personalized method for determining the left and right coronary artery shunt fraction. Background Art

[0002] FFR is a reliable indicator that directly describes the degree of functional myocardial ischemia. It can be obtained through numerical simulation in related studies. The determination of the left and right coronary artery shunt fraction is particularly important in the field of numerical simulation of coronary hemodynamics. A simple and reliable left and right coronary artery shunt fraction allows more accurate simulation results. This left and right coronary flow determination method is intended to provide a personalized shunt basis to improve the accuracy of non-invasive FFR (defined as the ratio of the mean pressure at the distal end of the stenosis (Pd) to the mean pressure at the aortic root (Pa)), and provide a simpler and more reliable estimation of the left and right coronary flow, thereby accurately determining the afterload parameters of the personalized coronary geometric multi-scale model system. Numerical simulation methods for hemodynamics are emerging in an endless stream, but the existing simulation studies are all numerical simulations that lack a reliable basis for determining the left and right coronary artery shunt fraction, and have the disadvantages of non-personalization and poor accuracy. This method proposes a method for allocating left and right coronary artery flow based on cross-sectional area parameters to achieve personalized left and right coronary shunts and improve the accuracy of numerical simulation results. It has certain theoretical value for the diagnosis and treatment strategy research of personalized coronary artery physiological diseases. Summary of the invention

[0003] The present invention provides a simpler and more reliable method for personalized left and right coronary artery shunt fractions. Compared with the current situation in which the personalized left and right coronary artery shunt fractions are not achieved in many coronary artery simulations, the flow estimation method can be applied to all studies including but not limited to left and right coronary artery simulations, and is generally applicable to determining and personalizing left and right coronary artery flows or any calculation of afterload based on a geometric multi-scale calculation model. The simple and reliable method for personalized left and right coronary artery flows includes the following steps: including the construction of a personalized three-dimensional model, model gridding preprocessing, extraction of target parameters, establishment of a shunt model, determination of a zero-dimensional afterload model, establishment of a zero-dimensional-three-dimensional coupled simulation system, calculation and extraction of FFR values, comparative analysis of results, and objective evaluation of the relative advantages of a personalized coronary artery flow estimation method based on cross-sectional area.

[0004] To achieve the above object, the present invention is implemented through the following technical solutions:

[0005] Step A1: constructing a personalized three-dimensional model based on real coronary artery tomography images;

[0006] Step A2: Perform model grid preprocessing based on the personalized model;

[0007] Step A3: extracting target parameters based on the preprocessing model and establishing a personalized shunt model, that is, a personalized shunt model based on cross-sectional area;

[0008] The FFR is further calculated using the personalized left and right coronary flow fractions obtained from the above-mentioned personalized flow shunt model based on cross-sectional area:

[0009] Step A4: determining a zero-dimensional afterload model based on the structural features and extracted parameters of the personalized three-dimensional model in step A1;

[0010] Step A5: establishing a zero-dimensional-three-dimensional coupling simulation system based on the personalized three-dimensional model of step A1 and the zero-dimensional model of step A4, coupling and connecting the zero-dimensional afterload model at the end of each left and right coronary artery branch of the personalized three-dimensional model;

[0011] Step A6: The left and right coronary flow shunt fraction is personalized based on the cross-sectional area, and the FFR value is extracted based on non-invasive calculation according to the existing technology;

[0012] As a further technical solution of the present invention, the features described in step A1 first construct a personalized three-dimensional model based on the individual real coronary artery tomographic image, and the requirements for constructing the model are as follows: retain the complete aortic trunk, the left coronary artery and the left coronary artery branch connected to the aortic trunk, the right coronary artery and the right coronary artery branch connected to the aortic trunk, and use a cross-sectional area of ​​1mm 2 As a criterion for determining whether to keep or remove a branch, it should be greater than or equal to 1mm 2 Leave;

[0013] As a further technical solution of the present invention, the feature described in step A2 is to perform model meshing preprocessing based on the personalized model. The computing software supports the input form of the mesh file, so the following preprocessing steps are performed before the model is meshed:

[0014] Step B1: In Mimics software, three-dimensional entity reconstruction is performed based on the tomographic image data of the coronary artery. First, in the three-view editable operation interface, based on the difference in grayscale thresholds of CTA images of different tissues and combined with the threshold selection technology (Thresholding), the target tissue is manually selected, namely the aorta and the left coronary artery and its branches, the right coronary artery and its branches that penetrate it, and the original target mask is calculated by the software; second, the three-dimensional region growing technology (3DRegion Growing) is used to obtain a new mask. Third, the newly generated mask is imported and calculated in the three-dimensional entity (3D Object) menu bar to obtain the three-dimensional reconstruction model of the selected entity structure area. Finally, in order to repair the image, the contour interpolation method that is more suitable for medical image processing is selected in the TheCalculate 3D interface. The smooth model is achieved by reducing the matrix, the number of triangles, and the contour elements to improve the quality of the target model.

[0015] Step B2: The model is smoothed by Freeform software to obtain a high-quality three-dimensional solid remodel suitable for calculation;

[0016] Step B3: Import the processed model in step B2 into Geomagic and SolidWorks software to perform port flattening and preliminary meshing on the target model;

[0017] Step B4: Use the meshing sub-function module of ANSYS Workbench 15.0 to perform pre-calculation meshing on the processed target model. This step requires a sensitivity analysis of the mesh to eliminate the influence of the mesh factor on the calculation results.

[0018] As a further technical solution of the present invention, the features described in step A3 extract the target parameters based on the preprocessing model and establish the personalized diversion model proposed by the present method. The personalized diversion model proposed by the present method is implemented based on the cross-sectional area parameters of the target model. The diversion models of the cross-sectional areas proposed by the present method are established respectively according to the above parameters.

[0019] This method proposes a personalized left and right coronary flow method (based on cross-sectional area):

[0020] Q LCA :Q RCA =S LCA 3 / 4:S RCA 3 / 4

[0021] The cross-sectional area in the above formula refers to the cross-sectional area at the interface with the aorta.

[0022] The relevant letters in the above-mentioned formula are given the following definitions: Q-flow value, S-cross-sectional area value, LCA-Left coronary artery, i.e. left coronary artery, RCA-Right coronary artery, i.e. right coronary artery.

[0023] As a further technical solution of the present invention, the features described in step A4 are based on the target three-dimensional model structure features established in step A2, and combined with step A3 to extract the target parameters to determine the corresponding zero-dimensional afterload model. The zero-dimensional coronary afterload model is composed of basic electronic components such as inductors and resistors. The equivalent relationship between hemodynamic parameters and physical electrical parameters is shown in Table 1.

[0024] Table 1 Equivalent relationship between hemodynamic parameters and physical and electrical parameters

[0025]

[0026] It can be assumed that the length of a blood vessel is l, the cross-sectional area of ​​the blood vessel is A, and the corresponding parameters of a blood vessel in zero dimension and three dimension are shown in Figure 5 , then according to the elastic cavity theory established by Frank, the quantitative relationship between the flow, pressure and related resistance in this section of blood vessels can be expressed as:

[0027] ΔP=Q*R'

[0028]

[0029] In the formula, ΔP represents the pressure drop in this section of blood vessel, Q represents the blood flow of this branch of blood vessel, R' represents the viscous resistance generated by this section of blood vessel; μ represents the blood viscosity value under normal physiological conditions, and its value is 3.5e-3Pa·s; l represents the length of this section of blood vessel, and A represents the cross-sectional area of ​​this section of blood vessel.

[0030] The self-inductance of the inductor is equivalent to the flow inertia of the blood flow, and the resistance effect of the resistor is equivalent to the microcirculatory resistance of each branch of the coronary artery. Then the values ​​of the resistor R and the inductor L in the zero-dimensional model are:

[0031]

[0032] In the formula, P 末 represents the end pressure of the blood vessel, and ρ represents the blood flow density, which is defined as 1060 kg / m 3 .

[0033] As a further technical solution of the present invention, the feature described in step A5 is to build a zero-dimensional-three-dimensional coupled simulation system model based on the three-dimensional model grid file obtained in step A2 and the zero-dimensional model built in step A4. The specific implementation process includes the following steps:

[0034] Step C1: in the fluid setting window after the model is imported into the calculation software, assign the fluid calculation parameters blood flow density and blood flow dynamic viscosity values;

[0035] Step C2: The boundary conditions are determined based on the calculation results of the pressure-flow-afterload relationship of the three-dimensional model, the mean arterial pressure is given at the fluid inlet, the virtual flow boundary is defined at each branch outlet, and the zero-dimensional afterload model is coupled;

[0036] Step C3: In the numerical simulation calculation, 0.8 s is taken as a complete cardiac cycle, the time step is set to 0.0025 s, and the simulation calculation process is executed.

[0037] As a further technical solution of the present invention, the features described in step A6 are based on clinical invasive FFR measurement values, and are compared and analyzed with the FFR values ​​extracted by non-invasive calculation.

[0038] Based on certain indicators, the calculation results of this personalized flow method and other two existing flow allocation methods were evaluated and analyzed to verify the superiority of this personalized coronary flow estimation method. Based on certain indicators, the results of three types of left and right coronary artery shunts were evaluated and analyzed to verify the superiority of the predicted personalized coronary flow estimation scheme.

[0039] The method of the present invention can calculate the personalized FFR more accurately. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 : Flow chart of the method of the present invention.

[0041] Figure 2 : Target three-dimensional model diagram.

[0042] Figure 3 : Data chart of three types of diversion methods.

[0043] Figure 4 : Grid division model diagram.

[0044] Figure 5 : Diagram of the zero-dimensional afterload model.

[0045] Figure 6 : Diagram of the complete zero-dimensional-three-dimensional coupled simulation system.

[0046] Figure 7 : FFR calculation cloud diagram based on cross-sectional area.

[0047] Figure 8 : MSE comparative analysis bar chart of three diversion methods. DETAILED DESCRIPTION

[0048] The present invention will be further described below in conjunction with examples, but the present invention is not limited to the following examples.

[0049] Example 1

[0050] The method flow is shown in Figure 1 .

[0051] Step A1: construct a personalized 3D model based on the real coronary artery tomography images of the case;

[0052] Step A2: Perform model grid preprocessing based on the personalized model;

[0053] Step A3: extracting target parameters based on the preprocessing model and establishing the personalized shunt model proposed by the method for determining left and right coronary artery flow based on cross-sectional area parameters and two other control group shunt models;

[0054] Step A4: determining a zero-dimensional afterload model based on the personalized three-dimensional model structural features and extracted parameters;

[0055] Step A5: constructing a zero-dimensional-three-dimensional coupling simulation system based on the three-dimensional model and the zero-dimensional model; coupling and connecting the zero-dimensional afterload model at the end of each left and right coronary artery branch of the personalized three-dimensional model;

[0056] Step A6: Comparative analysis based on the invasive FFR value and the FFR value extracted by non-invasive calculation;

[0057] Step A7: Based on certain indicators, the calculation results of this personalized flow method for determining the left and right coronary flow based on the cross-sectional area and the other two existing flow distribution methods are evaluated and analyzed to verify the superiority of this personalized coronary flow estimation method.

[0058] As a further technical solution of the present invention, the features described in step A1 are firstly used to construct a personalized three-dimensional model based on the real coronary artery tomographic image of the case. The requirements for constructing the model are as follows: retain the complete aortic trunk and the left coronary artery and its branches, the right coronary artery and its branches (with a cross-sectional area of ​​1 mm 2 as a criterion for deciding whether to keep or remove a branch);

[0059] As a further technical solution of the present invention, the feature described in step A2 is to perform model meshing preprocessing based on the personalized model. The computing software only supports the input form of mesh files. Therefore, the following preprocessing steps are performed before the model is meshed:

[0060] Step B1: In Mimics software, three-dimensional entity reconstruction is performed based on the tomographic image data of the coronary artery. First, in the three-view editable operation interface, based on the difference in grayscale thresholds of CTA images of different tissues and combined with the threshold selection technology (Thresholding), the target tissue is manually selected, namely the aorta and the left coronary artery and its branches, the right coronary artery and its branches that penetrate it, and the original target mask is calculated by the software. Second, the three-dimensional region growing technology (3DRegion Growing) is used to obtain a new mask. Third, the newly generated mask is imported and calculated in the three-dimensional entity (3D Object) menu bar to obtain the three-dimensional reconstruction model of the selected entity structure area. Finally, in order to repair the image, the contour interpolation method that is more suitable for medical image processing is selected in the TheCalculate 3D interface. The smooth model is achieved by reducing the matrix, the number of triangles, and the contour elements to improve the quality of the target model.

[0061] Step B2: The model is smoothed by Freeform software to obtain a high-quality three-dimensional solid remodel suitable for calculation;

[0062] Step B3: Import the processed model in step B2 into Geomagic and SolidWorks software to perform port flattening and preliminary meshing on the target model;

[0063] Step B4: Use the meshing sub-function module of ANSYS Workbench 15.0 to perform pre-calculation meshing on the processed target model (see Figure 4 ), this step requires a sensitivity analysis of the grid to eliminate the influence of the grid factor on the calculation results.

[0064] As a further technical solution of the present invention, the features described in step A3 extract the target parameters based on the preprocessing model and establish the personalized shunt model based on the cross-sectional area parameters of the target model proposed by the method and two other control group shunt models. Other necessary parameters also include the left and right crown volumes, cross-sectional areas and cross-sectional areas and lengths of each branch of the target model. According to the above parameters, the cross-sectional area shunt model B proposed by the method and the control group shunt models A and C based on volume and fixed ratio are respectively established.

[0065] This method proposes a personalized left and right coronary flow method (based on cross-sectional area):

[0066] Q LCA :Q RCA =S LCA 3 / 4:S RCA 3 / 4

[0067] The cross-sectional area in the above formula refers to the cross-sectional area at the interface with the aorta.

[0068] The existing method for determining left and right coronary artery flow is:

[0069] The volume-based determination method is:

[0070] Q LCA :Q RCA =V LCA 3 / 4:V RCA 3 / 4

[0071] The method for determining the fixed diversion ratio is:

[0072] Q LCA :Q RCA =6:4 / 5:5 (left dominant / right dominant)

[0073] The letters in the above formula are defined as follows: Q-flow value, S-cross-sectional area value, V-volume value, LCA-Left coronary artery, i.e., left coronary artery, RCA-Right coronary artery, i.e., right coronary artery. As a further technical solution of the present invention, the features described in step A4 are based on the target three-dimensional model established in step A2 (see Figure 2) structural features, and combined with step A3 to extract parameters to determine the corresponding zero-dimensional afterload model. The zero-dimensional coronary afterload model is composed of basic electronic components inductance and resistance. The equivalent relationship between hemodynamic parameters and physical and electrical parameters is shown in Table 1.

[0074] It can be assumed that the length of a blood vessel is l and the diameter is A. According to the elastic cavity theory established by Frank, the quantitative relationship between the flow, pressure and related resistance in this blood vessel can be expressed as:

[0075] ΔP=Q*R'

[0076]

[0077] In the above formula, ΔP represents the pressure drop in the blood vessel, Q represents the blood flow in the blood vessel, R' represents the viscous resistance generated by the blood vessel, and μ represents the blood viscosity value under normal physiological conditions, which is 3.5e -3 Pa·s; l represents the length of the blood vessel segment, and A represents the cross-sectional area of ​​the blood vessel segment.

[0078] The self-inductance of the inductor is equivalent to the flow inertia of the blood flow, and the resistance effect of the resistor is equivalent to the microcirculatory resistance of each branch of the coronary artery. Then the values ​​of the resistor R and the inductor L in the zero-dimensional model are:

[0079]

[0080] In the above formula, P 末 represents the end pressure of the blood vessel, and ρ represents the blood flow density, which is defined as 1060 kg / m 3 .

[0081] As a further technical solution of the present invention, the feature described in step A5 is to construct a zero-dimensional-three-dimensional coupled simulation system based on the three-dimensional model grid file obtained in step A2 and the zero-dimensional model built in step A4 (see Figure 6 ), the specific implementation process includes the following steps:

[0082] Step C1: in the fluid setting window after the model is imported into the calculation software, assign blood flow density and blood flow dynamic viscosity values ​​in the fluid calculation parameters;

[0083] Step C2: The boundary conditions are determined based on the calculation results of the pressure-flow-afterload relationship of the three-dimensional model, the mean arterial pressure is given at the fluid inlet, the virtual flow boundary is defined at each branch outlet, and the zero-dimensional afterload model is coupled;

[0084] Step C3: In the numerical simulation calculation, 0.8 s is taken as a complete cardiac cycle, the time step is set to 0.0025 s, and the simulation calculation process is executed.

[0085] As a further technical solution of the present invention, the feature described in step A6 is based on the clinical invasive FFR measurement value, and is compared with the FFR value extracted by non-invasive calculation (see Figure 7 ) for comparative analysis: the average pressure of the cross-section about 30 mm downstream of the stenotic lesion was extracted, and the FFR value was calculated according to the formula FFR=Pd / Pa, and a two-way comparative analysis was performed based on the clinical invasive FFR value.

[0086] As a further technical solution of the present invention, the characteristics described in step A7 are used to perform corresponding evaluation and analysis on the three types of left and right coronary artery shunt results based on certain indicators (see Figure 3 Corresponding table), verifying the superiority of the predicted personalized coronary flow estimation scheme.

[0087] The goal of the present invention is to achieve personalization of the left and right coronary artery shunt fraction in coronary artery simulation calculations, use real case CTA data to reconstruct the model, extract parameters such as the left and right coronary volume, cross-sectional area, and cross-sectional area and length of each branch of the target model, and based on the above parameters, establish three types of left and right coronary artery shunt fraction models based on volume, cross-sectional area and fixed shunt ratio ABC, then determine the zero-dimensional afterload model based on the extracted parameters and the structural characteristics of the target model, finally, establish a zero-dimensional-three-dimensional coupled simulation system to simulate the coronary hemodynamic process, realize coronary hemodynamic simulation calculations, obtain FFR values ​​non-invasively, and compare and analyze the calculation results, and objectively evaluate the relative advantages of personalized coronary flow estimation methods based on cross-sectional area.

[0088] This left and right coronary flow estimation method improves the calculation accuracy of non-invasive FFR values ​​and accurately determines the afterload parameters of the personalized coronary geometry multi-scale model system. Before this method was verified, there were endless methods for estimating left and right coronary flow in hemodynamic numerical simulations, and scholars in various fields often disagreed. Simulation studies were all numerical simulations without a definite and reliable basis for left and right coronary shunt fractions, which had the disadvantages of non-personalization and poor accuracy. This method used invasive FFR as the evaluation criterion for MSE (mean square error) analysis (see Figure 8 ) verified the effectiveness of the personalized method for estimating left and right coronary artery flow based on cross-sectional area, and provided a reference for determining the boundaries of simulation research in the coronary artery field, which has high application value and guiding significance.

Claims

1. A personalized method for determining left-right coronary artery shunt fraction, characterized in that: The following steps are involved: Step A1: constructing a personalized three-dimensional model based on real coronary artery tomography images; Step A2: Perform model mesh preprocessing based on the personalized three-dimensional model; Step A3: extracting target parameters based on the preprocessed model and establishing a personalized diversion model, that is, a personalized diversion model based on cross-sectional area; The formula for the left and right coronary artery flow shunt fraction determined based on the cross-sectional area personalized shunt model is: Q LCA :Q RCA =S LCA 3 / 4 :S RCA 3 / 4 The cross-sectional area in the above formula refers to the cross-sectional area at the interface with the aorta; The relevant letters in the above formula are given the following definitions: Q-flow value, S-cross-sectional area value, subscript LCA-Left coronary artery, i.e. left coronary artery, subscript RCA-Right coronary artery, i.e. right coronary artery; In step A1, a personalized 3D model is constructed based on the real coronary artery tomographic image, with the following requirements: retaining the complete aortic trunk, the left coronary artery and the left coronary artery branch connected to the aortic trunk, the right coronary artery and the right coronary artery branch connected to the aortic trunk, with a cross-sectional area of ​​1 mm 2 As a criterion for determining whether to keep or remove a branch, it should be greater than or equal to 1mm 2 Leave; In step A2, model meshing preprocessing is performed based on the personalized three-dimensional model. The computing software supports the input form of mesh files, so the following preprocessing steps are performed before the model is meshed: Step B1: In Mimics software, 3D entity reconstruction is performed based on the tomographic image data of the coronary artery. First, in the three-view editable operation interface, based on the difference in grayscale thresholds of CTA images of different tissues and combined with the threshold selection technology, the target tissues, namely the aorta trunk and the left coronary artery and its branches, the right coronary artery and its branches connected thereto, are manually selected, and the original target mask is calculated by the software; second, the 3D region growing technology, namely 3D Region Growing, is used to obtain a new mask; third, the newly generated mask is imported and calculated in the 3D entity, namely 3D Object menu bar, to obtain a 3D reconstructed model of the selected entity structure area; fourth, in order to repair the image, the contour interpolation method that is more suitable for medical image processing is selected in the Calculate 3D interface, and a smooth model is achieved by reducing the matrix, the number of triangles, and the contour element operation, thereby improving the quality of the target model; Step B2: The model obtained in step B1 is smoothed by Freeform software to obtain a three-dimensional solid reconstruction model suitable for calculation; Step B3: import the three-dimensional solid reconstruction model processed in step B2 into Geomagic and SolidWorks software to perform port flattening and preliminary meshing on the target model; Step B4: Use the meshing sub-function module of ANSYS Workbench 15.0 to perform pre-calculation meshing on the target model. This step requires a sensitivity analysis of the mesh to eliminate the influence of the mesh factor on the calculation results.

2. A method for calculating FFR using the left and right coronary artery shunt fraction obtained by the method of claim 1, comprising the following steps: Step A4: determining a zero-dimensional afterload model based on the structural features and extracted parameters of the personalized three-dimensional model in step A1; Step A5: establishing a zero-dimensional-three-dimensional coupling simulation system based on the personalized three-dimensional model of step A1 and the zero-dimensional afterload model of step A4, coupling and connecting the zero-dimensional afterload model at the end of each left and right coronary artery branch of the personalized three-dimensional model; Step A6: Individualize the left and right coronary flow shunt fraction based on cross-sectional area and extract the FFR value based on noninvasive calculation.

3. The method according to claim 2, characterized in that In step A4, based on the target three-dimensional model structure features established in step A2 and in combination with step A3 to extract target parameters, a corresponding zero-dimensional afterload model is determined. The zero-dimensional afterload model is composed of basic electronic components such as inductors and resistors.

4. The method according to claim 2, characterized in that: Step A4, assuming that the length of a blood vessel is l and the cross-sectional area of ​​the blood vessel is A, then according to the elastic cavity theory established by Frank, the quantitative relationship between the flow, pressure and related resistance on the blood vessel is expressed as: ΔP=Q*R' In the formula, ΔP represents the pressure drop in the blood vessel, Q represents the blood flow in the blood vessel, R' represents the viscous resistance in the blood vessel; μ represents the blood viscosity under normal physiological conditions, which is 3.5e-3Pa·s; l represents the length of the blood vessel, and A represents the cross-sectional area of ​​the blood vessel. The self-inductance of the inductor is equivalent to the flow inertia of the blood flow, and the resistance effect of the resistor is equivalent to the microcirculatory resistance of each branch of the coronary artery; then the values ​​of the resistor R and the inductor L in the zero-dimensional afterload model are: In the formula, P 末 represents the end pressure of the blood vessel, and ρ represents the blood flow density, which is defined as 1060 kg / m 3 .

5. The method according to claim 2, characterized in that: In step A5, a zero-dimensional-three-dimensional coupled simulation system model is constructed based on the personalized three-dimensional model in step A1 and the zero-dimensional model constructed in step A4. The specific implementation process includes the following steps: Step C1: in the fluid setting window after the model is imported into the calculation software, assign the fluid calculation parameters blood flow density and blood flow dynamic viscosity values; Step C2: The boundary conditions are determined based on the calculation results of the pressure-flow-afterload relationship of the three-dimensional model, the mean arterial pressure is given at the fluid inlet, the virtual flow boundary is defined at each branch outlet, and the zero-dimensional afterload model is coupled; Step C3: In the numerical simulation calculation, 0.8 s is taken as a complete cardiac cycle, the time step is set to 0.0025 s, and the simulation calculation process is executed.

Images