Method for constructing a heart model based on multi-physics and geometric multiscale

Abstract

The present application relates to the technical field of model construction, and particularly relates to a heart model construction method based on multi-physical field and geometric multi-scale, comprising: carrying out segmentation processing and three-dimensional model reconstruction on a heart image to generate an initial heart geometric model, carrying out anisotropic attribute assignment on the initial heart geometric model based on a regularization algorithm to obtain a myocardial tissue structure; embedding a cell electrophysiology model into a calculation grid of the myocardial tissue structure through parameter mapping to output a heart model with multi-scale geometric characteristics; respectively constructing an electrophysiology field equation, a mechanics field equation and a heart fluid dynamics field equation to realize coupling between the multi-physical fields; determining whether the feature weight of the myocardial tissue structure texture needs to be increased; determining whether the transition function width of the cell electrophysiology model parameter mapping needs to be increased; and determining a pressure feedback relaxation factor in the coupling iteration. The present application improves the stability of heart model construction.

Description

Technical Field

[0001] This invention relates to the field of model building technology, and in particular to a method for building a heart model based on multiphysics and geometric multiscale. Background Technology

[0002] In an era where precision medicine is advancing and the demand for personalized treatment is constantly escalating, research on cardiac physiological and pathological mechanisms, early disease diagnosis, and optimization of treatment plans place higher demands on the multi-dimensional reconstruction capabilities of cardiac models. Traditional single-scale or single-physics-field cardiac models can no longer meet these needs. The core value of cardiac model construction methods based on multi-physics fields and geometric multi-scales lies in their ability to integrate multi-scale geometric features of organs, tissues, and cells, and couple multiple physical fields such as electrophysiology, hemodynamics, and mechanics to achieve accurate replication of cardiac physiological and disease processes. This drives the upgrade of cardiac models from single-dimensional simulation to multi-dimensional collaborative reconstruction, and has significant clinical application and research value. However, existing construction methods still generally suffer from problems such as unnatural integration of multi-scale geometric features and insufficient adaptation of physiological parameters, making it difficult to fully meet the accuracy requirements of models in the diagnosis and treatment of complex cardiovascular diseases.

[0003] Chinese Patent Publication No. CN111243084A discloses a method for constructing a three-dimensional cardiac model in cardiac radiotherapy, comprising the following steps: acquiring a three-dimensional cardiac electrophysiological image and a corresponding set of three-dimensional images of cardiac cross-sections; performing a registration operation on the current three-dimensional cardiac electrophysiological image and the set of three-dimensional images of cardiac cross-sections to obtain a preliminary three-dimensional cardiac model; determining whether there are artifacts in the current preliminary three-dimensional cardiac model; if so, reacquiring the three-dimensional cardiac electrophysiological image and / or the set of three-dimensional images of cardiac cross-sections, and then re-performing the registration operation; if not, obtaining the three-dimensional structural contour of the target body based on the electrophysiological data and graphic data in the current preliminary three-dimensional cardiac model; and outputting the preliminary three-dimensional cardiac model including the three-dimensional structural contour of the target body as the final three-dimensional cardiac model. It is evident that the method for constructing a three-dimensional cardiac model in cardiac radiotherapy suffers from insufficient stability in the constructed cardiac model due to the lack of a quantitative recognition threshold in the artifact detection step, which prevents the adaptive differentiation between variations in cardiac physiological structure and real artifacts. Summary of the Invention

[0004] To address this, the present invention provides a method for constructing a heart model based on multi-physics and geometric multi-scale, which overcomes the problem in the prior art where the lack of a quantitative recognition threshold in the artifact detection process makes it impossible to adaptively distinguish between variations in cardiac physiological structure and real artifacts, resulting in insufficient stability in the construction of the heart model.

[0005] To achieve the above objectives, this invention provides a method for constructing a heart model based on multiphysics and geometric multiscale, comprising:

[0006] Cardiac images are segmented and three-dimensional models are reconstructed to generate an initial cardiac geometric model. Based on a regularization algorithm, anisotropic attributes are assigned to the initial cardiac geometric model to obtain the myocardial tissue structure.

[0007] The cell electrophysiological model is embedded into the computational grid of the myocardial tissue structure through parameter mapping to output a heart model with multi-scale geometric properties.

[0008] Electrophysiological field equations, mechanical field equations, and cardiac fluid dynamics field equations are constructed respectively. The membrane potential activation time output by the electrophysiological field is input into the mechanical field equation to obtain the cardiac wall deformation. The cardiac wall deformation is input into the cardiac fluid dynamics field equation to obtain the blood flow.

[0009] The pressure generated by blood flow is used to apply a load to the myocardial wall, and the load is then input into the mechanical equation in reverse to achieve coupling between multiple physical fields.

[0010] Obtain the structural integrity rate of the initial cardiac geometric model, and determine whether the stability of the constructed cardiac model meets the requirements based on the structural integrity rate of the initial cardiac geometric model;

[0011] If the stability of the constructed heart model does not meet the requirements, it is determined whether it is necessary to increase the feature weight of the myocardial tissue structure texture.

[0012] If it is not necessary to increase the feature weight of the myocardial tissue structure texture, then obtain the maximum gradient of the membrane potential after mapping the electrophysiological field equation to determine whether the stability of the anisotropic properties of the myocardial tissue meets the requirements.

[0013] If the stability of the anisotropic properties of the myocardial tissue does not meet the requirements, then determine whether it is necessary to increase the width of the transition function for the parameter mapping of the cell electrophysiological model.

[0014] If it is not necessary to increase the width of the transition function for the parameter mapping of the cell electrophysiological model, then the pressure feedback relaxation factor in the coupling iteration is determined based on the force-fluid coupling residual.

[0015] Furthermore, based on the structural integrity rate of the initial cardiac geometric model, the stability of the constructed cardiac model is determined to meet the requirements, including:

[0016] The structural integrity rate of the initial cardiac geometric model was compared with the preset second integrity rate;

[0017] If the structural integrity rate of the initial cardiac geometric model is greater than the preset second integrity rate, then the stability of the cardiac model construction is determined to meet the requirements.

[0018] If the structural integrity rate of the initial cardiac geometry model is less than or equal to the preset second integrity rate, it is determined that the stability of the cardiac model construction does not meet the requirements, and it is determined whether the feature weight of the myocardial tissue structure texture needs to be increased based on the structural integrity rate of the initial cardiac geometry model.

[0019] Further, determine whether it is necessary to increase the feature weights of myocardial tissue structural texture, including:

[0020] The structural integrity rate of the initial cardiac geometric model is compared with the preset first integrity rate and the preset second integrity rate, respectively;

[0021] If the structural integrity rate of the initial cardiac geometry model is less than or equal to the preset first integrity rate, then it is determined that the feature weights of the myocardial tissue structure texture need to be increased.

[0022] If the structural integrity rate of the initial cardiac geometry model is greater than the preset first integrity rate and less than or equal to the preset second integrity rate, then it is determined that there is no need to increase the feature weight of the myocardial tissue structure texture.

[0023] Furthermore, the increase in the feature weight of the myocardial tissue structure texture is determined by the difference between the preset first integrity rate and the structural integrity rate of the initial cardiac geometric model.

[0024] Furthermore, the stability of the anisotropic properties of myocardial tissue, determined based on the maximum gradient of the membrane potential mapped from the electrophysiological field equations, is assessed to determine whether it meets the requirements, including:

[0025] The maximum gradient of the membrane potential after mapping the electrophysiological field equation is compared with the preset first gradient;

[0026] If the maximum gradient of the membrane potential after mapping the electrophysiological field equation is less than or equal to the preset first gradient, then it is determined that the stability of the anisotropic properties of myocardial tissue meets the requirements, and it is determined whether the feature weights of the myocardial tissue structure texture meet the requirements.

[0027] If the maximum gradient of the membrane potential after mapping the electrophysiological field equation is greater than the preset first gradient, it is determined that the stability of the anisotropic properties of the myocardial tissue does not meet the requirements, and it is determined whether the width of the transition function for mapping the cell electrophysiological model parameters needs to be increased based on the maximum gradient of the membrane potential after mapping the electrophysiological field equation.

[0028] Further, determine whether it is necessary to increase the width of the transition function for mapping the cell electrophysiological model parameters, including:

[0029] The maximum gradient of the membrane potential after mapping the electrophysiological field equation is compared with the preset first gradient and the preset second gradient, respectively;

[0030] If the maximum gradient of the membrane potential after mapping the electrophysiological field equation is greater than the preset first gradient and less than or equal to the preset second gradient, then it is determined that the width of the transition function for mapping the cell electrophysiological model parameters needs to be increased.

[0031] If the maximum gradient of the membrane potential after mapping the electrophysiological field equation is greater than the preset second gradient, then it is determined that it is not necessary to increase the width of the transition function for mapping the cell electrophysiological model parameters.

[0032] Furthermore, the increase in the width of the transition function mapped by the cell electrophysiological model parameters is determined by the difference between the maximum gradient of the membrane potential after mapping by the electrophysiological field equation and the preset first gradient.

[0033] Furthermore, the pressure feedback relaxation factor in the coupled iteration is determined based on the force-fluid coupling residual, including:

[0034] The force-fluid coupling residual is compared with the preset residual;

[0035] If the force-fluid coupling residual is less than or equal to the preset residual, then it is determined that the stability of the multiphysics coupling iteration meets the requirements, and it is not necessary to reduce the pressure feedback relaxation factor in the coupling iteration. It is also determined whether the transition function width of the cell electrophysiological model parameter mapping meets the requirements.

[0036] If the force-fluid coupling residual is greater than the preset residual, it is determined that the stability of the multiphysics coupling iteration does not meet the requirements, and the pressure feedback relaxation factor in the coupling iteration needs to be reduced.

[0037] Furthermore, the force-fluid coupling residual is the difference between the actual calculation result and the theoretical calculation result of the feedback iteration between the mechanical field and the fluid dynamics field.

[0038] Furthermore, the reduction magnitude of the pressure feedback relaxation factor in the coupling iteration is determined by the difference between the force-flow coupling residual and the preset residual.

[0039] Compared with existing technologies, the beneficial effects of this invention are as follows: The method of this invention adjusts the feature weights of myocardial tissue structure texture based on the structural integrity rate of the initial cardiac geometric model. Because the key structural information of fiber orientation and tissue boundaries carried by the myocardial tissue texture is not fully highlighted during multi-scale geometric feature fusion, its adaptability to macroscopic geometric contour features is insufficient, leading to the loss or distortion of some fine structures during model construction. By increasing the feature weights of myocardial tissue structure texture, its ability to represent structural details can be strengthened, improving the restoration degree of key details during multi-scale feature fusion. Furthermore, the method adjusts the width of the transition function mapping of cell electrophysiological model parameters based on the maximum gradient of membrane potential after mapping the electrophysiological field equation. Because the electrophysiological characteristics of cells in different regions of myocardial tissue have inherent differences, the transition function cannot adapt to this regional heterogeneity in cell model parameter mapping, leading to the electrophysiological field equation... The membrane potential gradient distribution after process mapping shows significant abrupt changes. By increasing the width of the transition function and optimizing it, the parameter mapping can be effectively smoothed, making the changes in membrane potential gradient between different regions smoother. This more realistically reflects the electrophysiological characteristics and physiological differences of different regions of myocardial tissue. The pressure feedback relaxation factor in the coupling iteration is adjusted according to the force-fluid coupling residual. Due to the high-frequency oscillation phenomenon in the pressure feedback link during the multi-physics coupling iteration process, the mechanical field pressure transmission signal is prone to superposition and conflict with the steady-state response signal of other physical fields when synchronized. By reducing the pressure feedback relaxation factor, the accumulation of force-fluid coupling residuals caused by pressure fluctuations can be reduced, making the pressure feedback response and the feedback iteration of other physical fields closer to the residual control standard. This reduces the coupling imbalance caused by residual exceeding the threshold, promotes the convergence of multi-physics feedback interaction towards a low residual steady state, and improves the stability of the cardiac model construction.

[0040] Furthermore, the method of the present invention adjusts the feature weights of myocardial tissue structure texture by setting a preset first integrity rate and a preset second integrity rate. Since the key structural information of fiber orientation and tissue boundary carried by the myocardial tissue texture is not fully highlighted during the multi-scale geometric feature fusion process, and the adaptability with macroscopic geometric contour features is insufficient, some fine structures are missing or distorted during model construction. By increasing the feature weights of myocardial tissue structure texture, its ability to represent structural details can be strengthened, the restoration degree of key details during multi-scale feature fusion can be improved, and the stability of cardiac model construction can be further improved.

[0041] Furthermore, the method described in this invention adjusts the width of the transition function for mapping the parameters of the cell electrophysiological model by setting a preset first gradient and a preset second gradient. Since there are inherent differences in the electrophysiological characteristics of cells in different regions of myocardial tissue, the transition function cannot adapt to this regional heterogeneity in the cell model parameter mapping, resulting in significant abrupt changes in the membrane potential gradient distribution after the electrophysiological field equation is mapped. Increasing the width of the transition function can effectively smooth the parameter mapping by optimizing the width of the transition function, making the changes in the membrane potential gradient between different regions smoother, thereby more realistically reflecting the electrophysiological characteristics and physiological differences of different regions of myocardial tissue, and further improving the stability of the heart model construction.

[0042] Furthermore, the method of the present invention adjusts the pressure feedback relaxation factor in the coupled iteration by setting a preset residual. Since there is a high-frequency oscillation phenomenon in the pressure feedback link during the multi-physics coupling iteration process, the mechanical field pressure transmission signal is prone to superposition and conflict with the steady-state response signal of other physical fields when synchronized. By reducing the pressure feedback relaxation factor, the accumulation of force-fluid coupling residuals caused by pressure fluctuations can be reduced, so that the pressure feedback response and the feedback iteration of other physical fields are close to the residual control standard, reducing the coupling imbalance caused by residual exceeding the threshold, promoting the convergence of multi-physics feedback interaction towards a low residual steady state, and further improving the stability of the heart model construction. Attached Figure Description

[0043] Figure 1 This is an overall flowchart of the method for constructing a heart model based on multiphysics and geometric multiscale in an embodiment of the present invention;

[0044] Figure 2 This is a flowchart illustrating the process of determining whether to increase the feature weights of myocardial tissue structure texture in the heart model construction method based on multiphysics and geometric multiscale in this embodiment of the invention.

[0045] Figure 3 This is a flowchart illustrating the process of determining whether to increase the width of the transition function for mapping the parameters of the cellular electrophysiological model in the method for constructing a cardiac model based on multiphysics and geometric multiscale in this embodiment of the invention.

[0046] Figure 4 This is a flowchart illustrating the process of determining the pressure feedback relaxation factor in the coupled iteration of the method for constructing a heart model based on multiphysics and geometric multiscale in an embodiment of the present invention. Detailed Implementation

[0047] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0048] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0049] Please see Figure 1 As shown, it is an overall flowchart of the method for constructing a heart model based on multiphysics and geometric multiscale in an embodiment of the present invention.

[0050] This invention provides a method for constructing a heart model based on multiphysics and geometric multiscale, comprising:

[0051] Step S1: The cardiac image is segmented and a three-dimensional model is reconstructed to generate an initial cardiac geometric model. Based on a regularization algorithm, the anisotropic attributes of the initial cardiac geometric model are assigned to obtain the myocardial tissue structure.

[0052] Step S2: The cellular electrophysiological model is embedded into the computational grid of the myocardial tissue structure through parameter mapping to output a heart model with multi-scale geometric properties.

[0053] Step S3: Construct the electrophysiological field equation, the mechanical field equation, and the cardiac fluid dynamics field equation respectively. Input the membrane potential activation time output by the electrophysiological field into the mechanical field equation to obtain the cardiac wall deformation. Input the cardiac wall deformation into the cardiac fluid dynamics field equation to obtain the blood flow.

[0054] Step S4: Apply a load to the myocardial wall based on the pressure generated by the blood flow, and input the load in reverse into the mechanical equation to achieve coupling between multiple physical fields;

[0055] Step S5: Obtain the structural integrity rate of the initial cardiac geometric model, and determine whether the stability of the constructed cardiac model meets the requirements based on the structural integrity rate of the initial cardiac geometric model;

[0056] Step S6: If the stability of the constructed heart model does not meet the requirements, determine whether it is necessary to increase the feature weight of the myocardial tissue structure texture.

[0057] Step S7: If it is not necessary to increase the feature weight of the myocardial tissue structure texture, then obtain the maximum gradient of the membrane potential after mapping the electrophysiological field equation to determine whether the stability of the anisotropic properties of the myocardial tissue meets the requirements.

[0058] Step S8: If the stability of the anisotropic properties of the myocardial tissue does not meet the requirements, determine whether it is necessary to increase the width of the transition function for the parameter mapping of the cell electrophysiological model.

[0059] Step S9: If it is not necessary to increase the width of the transition function for the parameter mapping of the cell electrophysiological model, then determine the pressure feedback relaxation factor in the coupling iteration based on the force-fluid coupling residual.

[0060] Specifically, cardiac imaging includes coronary angiography, cardiac CT, and cardiac diffusion tensor imaging.

[0061] Specifically, the initial cardiac geometric model is an organ-scale cardiac model with detailed anatomical structures.

[0062] Specifically, the process of assigning anisotropic attributes to the initial cardiac geometric model based on a regularization algorithm to obtain the myocardial tissue structure involves constructing a regularized coordinate field in the initial cardiac geometric model, obtaining the continuous gradient direction from the endocardium to the epicardium by solving the Laplace-Dirichlet equation, and then mapping the fiber angle and lamella angle to each grid node based on cardiac diffusion tensor imaging data, thereby establishing a myocardial tissue structure with anisotropy at the real tissue scale.

[0063] Specifically, the regularized coordinate field is a standardized coordinate system obtained in the initial cardiac geometric model by spatially transforming and normalizing the cardiac structure to define the anisotropic properties of the cardiac wall.

[0064] Specifically, the process of obtaining the continuous gradient direction from the endocardium to the epicardium by solving the Laplace-Dirichlet equation involves solving the Laplace equation according to preset boundary conditions: By setting φ(r) = 0 and applying Dirichlet boundary conditions, a continuous scalar field is obtained. The gradient direction from the endocardium to the epicardium was calculated.

[0065] Specifically, For the Laplace operator.

[0066] Specifically, The scalar fields to be determined are the specific values ​​of the fiber angle and the sheet angle. Specifically, the fiber angle is the angle of inclination of the myocardial fibers in the heart wall relative to the plane of the heart wall.

[0067] Specifically, the lamella angle is the angle of the myocardial tissue layer in the heart wall relative to the plane of the heart wall.

[0068] Specifically, true tissue-scale anisotropy means that the heart wall has different physical properties in different directions of its structural features at the tissue level.

[0069] Specifically, the cell electrophysiological model is as follows:

[0070] ;

[0071] Used to describe the changes in electrical potential of heart muscle cell membranes.

[0072] in: Let be the membrane potential, representing the membrane potential value at position r and time t.

[0073] D is electrical conductivity, which reflects the speed at which electrical signals propagate and is usually determined by the electrophysiological properties of tissues.

[0074] The rate of change of membrane potential in space.

[0075] The current density term is a nonlinear function determined by the film potential V.

[0076] Specifically, the process of embedding the cell electrophysiological model into the myocardial tissue structure computational grid through parameter mapping involves mapping the parameters of the cell electrophysiological model to each node based on the grid node positions of the initial cardiac geometry model, and adjusting the cell electrophysiological model parameters in combination with the myocardial tissue structure and its anisotropic properties.

[0077] Specifically, the electrophysiological field equation is:

[0078] ;

[0079] Where I represents blood flow / current,

[0080] R is a parameter reflecting the resistance of blood vessels to blood flow, set as follows: , Let be the viscosity of the blood, l be the length of the blood vessel, and r be the radius of the blood vessel.

[0081] L represents inductance, reflecting the damping effect on blood flow.

[0082] C is a parameter reflecting vascular compliance, set as follows: Where V is the volume of blood in the blood vessel, and P is the pressure on the inner wall of the blood vessel.

[0083] V(t) is the voltage waveform generated by the heart pumping blood.

[0084] Specifically, the mechanical field equations are:

[0085] ;

[0086] Where u represents the displacement field, which signifies the deformation of the myocardium under the action of force.

[0087] The density of the myocardium,

[0088] Let be the acceleration term, representing the second derivative of displacement u with respect to time t.

[0089] The Cauchy stress tensor is used to represent the distribution of internal forces.

[0090] f represents the force exerted by external volume forces on the myocardial tissue. These external volume forces include gravity and blood pressure.

[0091] Specifically, the cardiac hydrodynamic field equations are:

[0092] ;

[0093] Where v is the velocity field vector, representing the blood flow velocity.

[0094] For the density of blood,

[0095] It is a pressure field.

[0096] For the dynamic viscosity of blood,

[0097] External volume force

[0098] This represents the gradient of blood flow velocity, used to describe the viscous diffusion of blood flow.

[0099] Since blood is considered an incompressible fluid, therefore:

[0100] ;

[0101] That is, the fluid density remains constant.

[0102] Specifically, the process of inputting the membrane potential activation time output by the electrophysiological field into the mechanical field equation to obtain the amount of heart wall deformation is as follows:

[0103] Calculate muscle contraction stress:

[0104] ;

[0105] in, For the stress of the myocardium,

[0106] This is a proportionality factor for stress-corresponding potential, used to reflect the relationship between membrane potential activation and muscle contraction.

[0107] Let be the membrane potential activation function, representing the time delay of membrane potential activation.

[0108] This is the derivative of the change in intracellular calcium ion concentration.

[0109] This stress affects the mechanical field of the heart, specifically manifested as deformation of the heart wall.

[0110] Specifically, the equation for calculating blood flow is obtained by inputting the amount of heart wall deformation into the cardiac fluid dynamics field equation:

[0111] The amount of cardiac wall deformation is usually expressed by the pressure distribution generated by cardiac wall contraction and relaxation, and is input as an external force f into the following cardiac fluid dynamics field equations:

[0112] ;

[0113] Where: v is the velocity field of the blood.

[0114] For the density of blood,

[0115] For the pressure of blood,

[0116] The viscosity coefficient of blood.

[0117] The equation yields the blood flow velocity v and the blood volume. , where n is the unit normal vector passing through the surface of the heart chamber. This represents the area of ​​the surface elements of the heart chambers.

[0118] Specifically, the process of applying a load to the myocardial wall based on the pressure generated by blood flow and then inputting this load in reverse into the mechanical equations to achieve coupling between multiple physical fields involves updating the Cauchy stress tensor in the mechanical field equations based on the pressure-stress conversion relationship. The Cauchy stress tensor update equation is as follows:

[0119] ;

[0120] in, For the stress of the myocardium itself,

[0121] Stress generated by blood flow pressure The pressure generated by blood flow (unit: Pa) is denoted by n, which is the normal vector of the heart wall surface.

[0122] The above calculations are performed iteratively until the solutions of each physical field converge, thus completing the coupling between the multiphysics fields.

[0123] Specifically, the feature weights of myocardial tissue texture are dimensionless parameters used to quantify the importance of local texture features in myocardial images.

[0124] Specifically, the transition function width for mapping cell electrophysiological model parameters is the spatial distance of the smooth transition region set to avoid spatial discontinuities and abrupt changes in model parameters when mapping cell-level electrophysiological dynamics model parameters from local to tissue-scale grid.

[0125] Specifically, the pressure feedback relaxation factor in coupled iteration is an important parameter that controls the update magnitude of the pressure load feedback transferred from the fluid dynamics field to the mechanical field in the iterative solution of the coupled fluid dynamics field and mechanical field.

[0126] In implementation, the method of this invention adjusts the feature weights of myocardial tissue texture based on the structural integrity rate of the initial cardiac geometric model. Because the key structural information of fiber orientation and tissue boundaries carried by the myocardial tissue texture is not fully highlighted during multi-scale geometric feature fusion, its adaptability to macroscopic geometric contour features is insufficient, leading to the loss or distortion of some fine structures during model construction. By increasing the feature weights of myocardial tissue texture, its ability to represent structural details can be strengthened, improving the restoration of key details during multi-scale feature fusion. The transition function width of the cell electrophysiological model parameter mapping is adjusted based on the maximum gradient of the membrane potential after mapping the electrophysiological field equation. Because the electrophysiological characteristics of cells in different regions of myocardial tissue have inherent differences, the transition function cannot adapt to this regional heterogeneity in the cell model parameter mapping, resulting in the membrane potential width after mapping the electrophysiological field equation being insufficient. Significant abrupt changes in the potential gradient distribution can be addressed by increasing the width of the transition function. Optimizing the width of the transition function can effectively smooth the parameter mapping, making the changes in membrane potential gradient between different regions smoother. This allows for a more realistic reflection of the electrophysiological characteristics and physiological differences in various regions of myocardial tissue. Adjusting the pressure feedback relaxation factor in the coupled iteration based on the force-fluid coupling residuals is crucial. Due to the high-frequency oscillations in the pressure feedback loop during multi-physics coupling iteration, the pressure transmission signal from the mechanical field is prone to superposition and conflict with the steady-state response signals of other physical fields when synchronized. Reducing the pressure feedback relaxation factor can decrease the accumulation of force-fluid coupling residuals caused by pressure fluctuations, bringing the pressure feedback response closer to the residual control standard in the feedback iteration with other physical fields. This reduces coupling imbalance caused by residual exceeding the threshold, promotes the convergence of multi-physics feedback interaction towards a low-residual steady state, and improves the stability of the cardiac model construction.

[0127] Please continue reading. Figure 2 As shown, it is a flowchart illustrating the process of determining whether to increase the feature weights of myocardial tissue structure texture in the heart model construction method based on multiphysics and geometric multiscale according to an embodiment of the present invention.

[0128] Specifically, determining whether the stability of the constructed heart model meets the requirements based on the structural integrity rate of the initial heart geometric model includes:

[0129] The structural integrity rate of the initial cardiac geometric model was compared with the preset second integrity rate;

[0130] If the structural integrity rate of the initial cardiac geometric model is greater than the preset second integrity rate, then the stability of the cardiac model construction is determined to meet the requirements.

[0131] If the structural integrity rate of the initial cardiac geometry model is less than or equal to the preset second integrity rate, it is determined that the stability of the cardiac model construction does not meet the requirements, and it is determined whether the feature weight of the myocardial tissue structure texture needs to be increased based on the structural integrity rate of the initial cardiac geometry model.

[0132] One reason why the stability of the constructed heart model might not meet requirements could be that the anisotropic properties of the myocardial tissue are not stable enough, or that the feature weights of the myocardial tissue structure texture are not appropriate. The next step is to determine which specific cause it is, which is essentially the process of deciding whether to increase the feature weights of the myocardial tissue structure texture.

[0133] Specifically, determining whether to increase the feature weights of myocardial tissue texture includes:

[0134] The structural integrity rate of the initial cardiac geometric model is compared with the preset first integrity rate and the preset second integrity rate, respectively;

[0135] If the structural integrity rate of the initial cardiac geometry model is less than or equal to the preset first integrity rate, then it is determined that the feature weights of the myocardial tissue structure texture need to be increased.

[0136] If the structural integrity rate of the initial cardiac geometry model is greater than the preset first integrity rate and less than or equal to the preset second integrity rate, then it is determined that there is no need to increase the feature weight of the myocardial tissue structure texture.

[0137] Specifically, when the structural integrity rate of the initial cardiac geometric model is less than or equal to the preset first integrity rate, it is determined that the reason for the inadequate stability of the cardiac model construction is that the feature weights of the myocardial tissue structure texture do not meet the requirements, thus requiring an increase in the feature weights of the myocardial tissue structure texture. When the structural integrity rate of the initial cardiac geometric model is greater than the preset first integrity rate but less than or equal to the preset second integrity rate, it can be preliminarily determined that the stability of the anisotropic properties of the myocardial tissue does not meet the requirements. The next step is to use the maximum gradient of the membrane potential mapped from the electrophysiological field equation to finally determine whether the stability of the myocardial tissue anisotropic properties meets the requirements, i.e., to determine whether the reason for the inadequate stability of the cardiac model construction is the inadequate stability of the myocardial tissue anisotropic properties.

[0138] It is understandable that the preset first completeness rate is lower than the preset second completeness rate. The three intervals divided by the preset first completeness rate and the preset second completeness rate correspond to three different scenarios:

[0139] The first interval is when the structural integrity rate of the initial cardiac geometric model is less than or equal to the preset first integrity rate. The corresponding situation is: due to the multi-scale geometric feature fusion process, the key structural information of the fiber orientation and tissue boundary carried by the myocardial tissue texture is not fully highlighted, and the adaptability with the macro geometric contour features is insufficient, resulting in the loss or distortion of some fine structures when the model is built. At this time, it is necessary to adjust the feature weights of the myocardial tissue structure texture.

[0140] The second interval is when the structural integrity rate of the initial cardiac geometric model is greater than the preset first integrity rate and less than or equal to the preset second integrity rate. The corresponding situation is: due to the inherent differences in the electrophysiological characteristics of cells in different regions of myocardial tissue, the transition function cannot adapt to this regional heterogeneity in the cell model parameter mapping, resulting in a significant abrupt change in the membrane potential gradient distribution after the electrophysiological field equation is mapped. At this time, it is necessary to further determine whether the stability of the anisotropic properties of myocardial tissue meets the requirements.

[0141] The third interval is when the structural integrity rate of the initial cardiac geometric model is greater than the preset second integrity rate. The corresponding situation is that the stability of the cardiac model construction meets the requirements, and no adjustment is needed.

[0142] Understandably, in the construction of a heart model based on multiphysics and geometric multiscales, the stability of the heart model construction is characterized by preset first integrity rate and preset second integrity rate. The core logic is to transform the abstract stability of the heart model construction into a quantifiable integrity rate range by correlating the structural integrity rate of the initial heart geometric model with the stability of the model construction. The first integrity rate serves as a dividing line to distinguish stability, and its core function is to determine whether there are integrity issues at the basic structural level. The second integrity rate serves as a dividing line to distinguish whether the construction stability meets the requirements, and its core function is to clarify the quantitative standard for achieving the model's structural integrity. The preset first integrity rate and preset second integrity rate can be set according to actual working conditions. The setting of the preset first integrity rate and preset second integrity rate aims to ensure the stability and practicality of the heart model construction. Optionally, the preset first integrity rate and preset second integrity rate are determined through a limited number of experiments by evaluating the effects of different integrity rates on the construction of the heart model. The determined preset first integrity rate and preset second integrity rate should satisfy the condition that they are neither too small nor cause excessive interference to the heart model construction process. For example, the preset first completeness rate is generally selected in the range of [89%, 91%], and the preset second completeness rate is generally selected in the range of [94%, 96%].

[0143] Preferably, the first integrity rate is 90% in the preferred embodiment, and the second integrity rate is 95% in the preferred embodiment.

[0144] Specifically, the structural integrity rate of the initial cardiac geometry model is the ratio of the number of effective cardiac structural mesh elements that are fully reconstructed in the cardiac model to the total number of structural mesh elements that should theoretically be included.

[0145] Specifically, the effective heart structure is a model structure with biological, geometric, and numerical significance, which can participate in subsequent multiphysics field solutions.

[0146] Specifically, the increase in the feature weights of the myocardial tissue structure texture is determined by the difference between the preset first integrity rate and the structural integrity rate of the initial cardiac geometric model.

[0147] Specifically, when the difference between the preset first integrity rate and the structural integrity rate of the initial cardiac geometry model is within 3%, the feature weight of the myocardial tissue structure texture is increased to 1.1 times the original value. When the difference between the preset first integrity rate and the structural integrity rate of the initial cardiac geometry model exceeds 3%, the feature weight of the myocardial tissue structure texture increases by 0.1 for every 1% increase beyond the original value of 1.1 times. For example, when the difference between the preset first integrity rate and the structural integrity rate of the initial cardiac geometry model is 5%, the current feature weight of the myocardial tissue structure texture is 1.5, and the increased feature weight of the myocardial tissue structure texture is 1.5×1.1+0.1×2=1.85.

[0148] In practice, the method of the present invention adjusts the feature weights of myocardial tissue structure texture by setting a preset first integrity rate and a preset second integrity rate. Since the key structural information of fiber orientation and tissue boundary carried by myocardial tissue texture is not fully highlighted during the multi-scale geometric feature fusion process, and the adaptability with macroscopic geometric contour features is insufficient, some fine structures are missing or distorted when the model is built. By increasing the feature weights of myocardial tissue structure texture, its ability to represent structural details can be strengthened, the restoration degree of key details during multi-scale feature fusion can be improved, and the stability of cardiac model construction can be further improved.

[0149] Please continue reading. Figure 3 As shown, it is a logical flowchart of the process of determining whether it is necessary to increase the width of the transition function for mapping the parameters of the cell electrophysiological model in the method for constructing a heart model based on multi-physics field and geometric multi-scale according to an embodiment of the present invention.

[0150] Specifically, determining whether the stability of the anisotropic properties of myocardial tissue meets the requirements based on the maximum gradient of the membrane potential after mapping the electrophysiological field equation includes: comparing the maximum gradient of the membrane potential after mapping the electrophysiological field equation with a preset first gradient;

[0151] If the maximum gradient of the membrane potential after mapping the electrophysiological field equation is less than or equal to the preset first gradient, then it is determined that the stability of the anisotropic properties of myocardial tissue meets the requirements, and it is determined whether the feature weights of the myocardial tissue structure texture meet the requirements.

[0152] If the maximum gradient of the membrane potential after mapping the electrophysiological field equation is greater than the preset first gradient, it is determined that the stability of the anisotropic properties of the myocardial tissue does not meet the requirements, and it is determined whether the width of the transition function for mapping the cell electrophysiological model parameters needs to be increased based on the maximum gradient of the membrane potential after mapping the electrophysiological field equation.

[0153] Specifically, when the maximum gradient of the membrane potential after mapping the electrophysiological field equation is less than or equal to the preset first gradient, it is determined that the stability of the anisotropic properties of myocardial tissue meets the requirements. However, if the stability of the previously determined cardiac model does not meet the requirements, it is necessary to further determine whether the feature weights of the myocardial tissue structure texture meet the requirements.

[0154] In practice, the feature weights of the actual myocardial tissue structure texture are compared with the predetermined feature weight threshold to determine whether the feature weights of the myocardial tissue structure texture meet the requirements. If the feature weights of the actual myocardial tissue structure texture are less than the predetermined feature weight threshold, the feature weights of the myocardial tissue structure texture are determined to be unacceptable. The predetermined feature weight threshold is the average value of the feature weights of the myocardial tissue structure texture monitored in the previous three months of the historical period.

[0155] If the feature weights of the myocardial tissue structure texture do not meet the requirements, then increase the feature weights of the myocardial tissue structure texture; if the feature weights of the myocardial tissue structure texture meet the requirements, then re-acquire the gradient norm fluctuation amplitude of the back diffusion process and re-determine whether the stability of the heart model construction meets the requirements.

[0156] If the maximum gradient of the membrane potential after mapping the electrophysiological field equations is greater than the preset first gradient and less than or equal to the preset second gradient, then the reason for the instability of the heart model construction can be determined to be the instability of the anisotropic properties of myocardial tissue. The reasons for this instability may be that the transition function width of the cellular electrophysiological model parameter mapping is not sufficient, or that the stability of the multi-physics coupling iteration is not sufficient. The next step is to determine which specific cause it is, which is essentially the process of deciding whether to increase the transition function width of the cellular electrophysiological model parameter mapping.

[0157] Specifically, determining whether the transition function width for the cell electrophysiology model parameter mapping needs to be increased includes:

[0158] The maximum gradient of the membrane potential after mapping the electrophysiological field equation is compared with the preset first gradient and the preset second gradient, respectively;

[0159] If the maximum gradient of the membrane potential after mapping the electrophysiological field equation is greater than the preset first gradient and less than or equal to the preset second gradient, then it is determined that the width of the transition function for mapping the cell electrophysiological model parameters needs to be increased.

[0160] If the maximum gradient of the membrane potential after mapping the electrophysiological field equation is greater than the preset second gradient, then it is determined that it is not necessary to increase the width of the transition function for mapping the cell electrophysiological model parameters.

[0161] Specifically, when the maximum gradient of the membrane potential after mapping the electrophysiological field equations is greater than the preset first gradient and less than or equal to the preset second gradient, it is determined that the reason for the non-compliance of the stability of the anisotropic properties of myocardial tissue is that the transition function width of the cellular electrophysiological model parameter mapping does not meet the requirements. Therefore, it is necessary to increase the transition function width of the cellular electrophysiological model parameter mapping. When the maximum gradient of the membrane potential after mapping the electrophysiological field equations is greater than the preset second gradient, it can be preliminarily determined that the stability of the multi-physics coupling iteration does not meet the requirements. Next, it is necessary to make a final determination on whether the stability of the multi-physics coupling iteration meets the requirements based on the force-fluid coupling residual, that is, to determine whether the reason for the non-compliance of the stability of the anisotropic properties of myocardial tissue is the non-compliance of the stability of the multi-physics coupling iteration.

[0162] It is understandable that the preset first gradient is less than the preset second gradient, and the three intervals divided by the preset first gradient and the preset second gradient correspond to three different cases:

[0163] The first interval is when the maximum gradient of the membrane potential after mapping the electrophysiological field equation is less than or equal to the preset first gradient. The corresponding situation is: the stability of the anisotropic properties of myocardial tissue meets the requirements. At this time, it is necessary to further determine whether the feature weights of the myocardial tissue structure texture meet the requirements.

[0164] The second interval is when the maximum gradient of the membrane potential after mapping the electrophysiological field equation is greater than the preset first gradient and less than or equal to the preset second gradient. The corresponding situation is: due to the inherent differences in the electrophysiological characteristics of cells in different regions of myocardial tissue, the transition function cannot adapt to this regional heterogeneity in the cell model parameter mapping, resulting in a significant abrupt change in the distribution of the membrane potential gradient after mapping the electrophysiological field equation. At this time, it is necessary to adjust the width of the transition function in the cell electrophysiological model parameter mapping.

[0165] The third interval is when the maximum gradient of the membrane potential after mapping the electrophysiological field equation is greater than the preset second gradient. The corresponding situation is that, due to the high-frequency oscillation phenomenon in the pressure feedback loop during the multi-physics coupling iteration process, the mechanical field pressure transmission signal is prone to superposition and conflict with other physical field steady-state response signals when synchronized. At this time, it is necessary to further determine whether the stability of the multi-physics coupling iteration meets the requirements.

[0166] Understandably, in the construction of a heart model based on multiphysics and geometric multiscales, the core logic of using a preset first gradient and a preset second gradient to characterize the stability of anisotropic properties of myocardial tissue is to transform the abstract stability of anisotropic properties into a quantifiable gradient range judgment by correlating the maximum membrane potential gradient with the stability of anisotropic properties of myocardial tissue. The first gradient serves as the dividing line between whether the stability of anisotropic properties of myocardial tissue meets the requirements, and its core function is to determine whether abrupt changes in membrane potential gradients are caused by regional electrophysiological heterogeneity. The second gradient serves as the dividing line between different causes of gradient anomalies. The preset first gradient and the preset second gradient can be set according to actual working conditions. The setting of the preset first gradient and the preset second gradient aims to ensure the stability and practicality of the heart model construction. Optionally, the preset first gradient and the preset second gradient are determined through a limited number of experiments by evaluating the effects of different gradients on the construction of the heart model. The determined preset first gradient and preset second gradient should satisfy the condition that they are neither too small nor cause excessive interference to the heart model construction process. For example, the preset first gradient is generally selected in the range of [190mV / mm, 210mV / mm], and the preset second gradient is generally selected in the range of [340mV / mm, 360mV / mm].

[0167] Preferably, the first gradient is 200mV / mm in a preferred embodiment, and the second gradient is 350mV / mm in a preferred embodiment.

[0168] Specifically, mV / mm is the unit of maximum membrane potential gradient, which means millivolts per millimeter, i.e., how many millivolts the voltage changes over a distance of one millimeter.

[0169] Specifically, the maximum gradient of the membrane potential after mapping the electrophysiological field equation is the maximum magnitude of the gradient obtained by spatially calculating the membrane potential distribution after mapping and numerically solving the cardiac electrophysiological field equation.

[0170] Specifically, the increase in the width of the transition function mapped by the cell electrophysiological model parameters is determined by the difference between the maximum gradient of the membrane potential after mapping by the electrophysiological field equation and the preset first gradient.

[0171] Specifically, when the difference between the maximum gradient of the membrane potential mapped by the electrophysiological field equation and the preset first gradient is within 30 mV / mm, the width of the transition function of the cell electrophysiological model parameter mapping is increased to 1.1 times the original value. When the difference between the maximum gradient of the membrane potential mapped by the electrophysiological field equation and the preset first gradient exceeds 30 mV / mm, in addition to increasing to 1.1 times the original value, the width of the transition function of the cell electrophysiological model parameter mapping is increased by 0.2 mm for every 10 mV / mm exceeding the original value. For example, when the difference between the maximum gradient of the membrane potential mapped by the electrophysiological field equation and the preset first gradient is 50 mV / mm, the current width of the transition function of the cell electrophysiological model parameter mapping is 1 mm, and the increased width of the transition function of the cell electrophysiological model parameter mapping is 1 × 1.1 + 0.2 × 2 = 1.5 mm.

[0172] In practice, the method of the present invention adjusts the width of the transition function for mapping the parameters of the cell electrophysiological model by setting a preset first gradient and a preset second gradient. Since there are inherent differences in the electrophysiological characteristics of cells in different regions of myocardial tissue, the transition function cannot adapt to this regional heterogeneity in the cell model parameter mapping, resulting in significant abrupt changes in the membrane potential gradient distribution after the electrophysiological field equation is mapped. Increasing the width of the transition function can effectively smooth the parameter mapping by optimizing the width of the transition function, making the changes in the membrane potential gradient between different regions smoother, thereby more realistically reflecting the electrophysiological characteristics and physiological differences of different regions of myocardial tissue, and further improving the stability of the heart model construction.

[0173] Please continue reading. Figure 4 As shown, it is a logical flowchart of the process of determining the pressure feedback relaxation factor in the coupling iteration of the heart model construction method based on multiphysics field and geometric multiscale in an embodiment of the present invention.

[0174] Specifically, the pressure feedback relaxation factor in the coupled iteration is determined based on the force-fluid coupling residual, including:

[0175] The force-fluid coupling residual is compared with the preset residual;

[0176] If the force-fluid coupling residual is less than or equal to the preset residual, then it is determined that the stability of the multiphysics coupling iteration meets the requirements, and it is not necessary to reduce the pressure feedback relaxation factor in the coupling iteration. It is also determined whether the transition function width of the cell electrophysiological model parameter mapping meets the requirements.

[0177] If the force-fluid coupling residual is greater than the preset residual, it is determined that the stability of the multiphysics coupling iteration does not meet the requirements, and the pressure feedback relaxation factor in the coupling iteration needs to be reduced.

[0178] When the force-fluid coupling residual is less than or equal to the preset residual, it is determined that the stability of the multi-physics coupling iteration meets the requirements. However, the stability of the anisotropic properties of myocardial tissue, which has been previously determined, does not meet the requirements. Therefore, it is necessary to further determine whether the transition function width of the cell electrophysiological model parameter mapping meets the requirements.

[0179] In practice, the transition function width of the cell electrophysiological model parameter mapping is compared with a predetermined transition function width threshold to determine whether the transition function width of the cell electrophysiological model parameter mapping meets the requirements. If the transition function width of the cell electrophysiological model parameter mapping is less than the predetermined transition function width threshold, it is determined that the transition function width of the cell electrophysiological model parameter mapping does not meet the requirements. The predetermined transition function width threshold is the average value of the transition function width of the cell electrophysiological model parameter mapping monitored in the previous three months of the historical cycle.

[0180] If the transition function width of the cell electrophysiological model parameter mapping does not meet the requirements, then increase the transition function width of the cell electrophysiological model parameter mapping; if the transition function width of the cell electrophysiological model parameter mapping meets the requirements, then re-acquire the maximum gradient of the membrane potential after the electrophysiological field equation mapping, and re-determine whether the stability of the anisotropic properties of myocardial tissue meets the requirements.

[0181] When the force-fluid coupling residual is greater than the preset residual, it can be determined that the reason why the stability of the anisotropic properties of myocardial tissue does not meet the requirements is that the stability of the multi-physics coupling iteration does not meet the requirements. Therefore, it is necessary to reduce the pressure feedback relaxation factor in the coupling iteration.

[0182] It is understandable that the two intervals of the preset residual division correspond to two different cases:

[0183] The first interval is when the force-fluid coupling residual is less than or equal to the preset residual. The corresponding situation is: the stability of the multi-physics coupling iteration is determined to meet the requirements. At this time, it is necessary to further determine whether the width of the transition function mapping of the cell electrophysiological model parameters meets the requirements.

[0184] The second interval is when the force-fluid coupling residual is greater than the preset residual. The corresponding situation is: due to the high-frequency oscillation phenomenon in the pressure feedback link during the multi-physics coupling iteration process, the mechanical field pressure transmission signal is prone to superposition and conflict with other physical field steady-state response signals when synchronized. At this time, it is necessary to reduce the pressure feedback relaxation factor in the coupling iteration.

[0185] Understandably, in the construction of a heart model based on multiphysics and geometric multiscale, the core logic of using preset residuals to characterize the multiphysics coupling iteration stability is to transform the abstract coupling iteration stability into a quantifiable oscillation amplitude judgment through the correlation between the force-fluid coupling residuals and the multiphysics coupling iteration stability. The preset residuals serve as the dividing line for distinguishing whether the coupling iteration stability meets the requirements. The preset residuals can be set according to actual working conditions. The setting of the preset residuals aims to ensure the stability and practicality of the heart model construction. Optionally, the preset residuals are determined through a limited number of experiments by evaluating the effects of different residuals on the construction of the heart model. The determined preset residuals should satisfy the condition that they are neither too small nor cause excessive interference to the heart model construction process. For example, the preset residuals are generally selected within the range of

[10] . -5 10 -3 ].

[0186] Preferably, the preferred embodiment of the preset residual is 10. -4 .

[0187] Specifically, the force-fluid coupling residual is the difference between the actual calculation result and the theoretical calculation result of the feedback iteration between the mechanical field and the fluid dynamics field.

[0188] Specifically, the reduction in the pressure feedback relaxation factor in the coupling iteration is determined by the difference between the force-flow coupling residual and the preset residual.

[0189] Specifically, when the difference between the force-fluid coupling residual and the preset residual is within 4×10 -5 When the pressure feedback relaxation factor in the coupled iteration is within a certain range, it decreases to 0.9 times its original value. When the difference between the force-fluid coupling residual and the preset residual exceeds 4 × 10⁻⁶, the pressure feedback relaxation factor in the coupled iteration is reduced to 0.9 times its original value. -5 At that time, based on reducing it to 0.9 times the original value, for every additional 2 × 10 -5 The pressure feedback relaxation factor in the coupling iteration is reduced by 0.01. For example, when the difference between the force-flow coupling residual and the preset residual is 6 × 10⁻⁶. -5 The current pressure feedback relaxation factor in the coupled iteration is 0.1, and the reduced pressure feedback relaxation factor in the coupled iteration is 0.1×0.9-0.01×1=0.08.

[0190] In practice, the method of the present invention adjusts the pressure feedback relaxation factor in the coupled iteration by setting a preset residual. Since the pressure feedback link has a high-frequency oscillation phenomenon in the multi-physics coupling iteration process, the mechanical field pressure transmission signal is prone to superposition and conflict with the steady-state response signal of other physical fields when synchronized. By reducing the pressure feedback relaxation factor, the accumulation of force-fluid coupling residual caused by pressure fluctuation can be reduced, so that the pressure feedback response and the feedback iteration of other physical fields are close to the residual control standard, reducing the coupling imbalance caused by residual exceeding the threshold, promoting the convergence of multi-physics feedback interaction towards a low residual steady state, and further improving the stability of the heart model construction.

[0191] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A method for constructing a cardiac model based on multi-physics and geometric multiscale, characterized in that, The method comprises the following steps: segmenting and reconstructing a three-dimensional model of a heart image to generate an initial heart geometric model, and assigning anisotropic properties to the initial heart geometric model based on a regularization algorithm to obtain a myocardial tissue structure; embedding a cell electrophysiological model into a computational grid of the myocardial tissue structure through parameter mapping to output a heart model with multi-scale geometric characteristics; respectively constructing an electrophysiological field equation, a mechanical field equation, and a cardiac fluid dynamics field equation, inputting the membrane potential activation time output by the electrophysiological field into the mechanical field equation to obtain a heart wall deformation, and inputting the heart wall deformation into the cardiac fluid dynamics field equation to obtain a blood flow; applying a load to the myocardial wall according to the pressure generated by the blood flow, and inputting the load into the mechanical equation in reverse to realize coupling between multiple physical fields; determining whether the stability of the heart model construction meets the requirements based on the structural integrity of the initial heart geometric model; if the stability of the heart model construction does not meet the requirements, determining whether the feature weight of the myocardial tissue structure texture needs to be increased; if the feature weight of the myocardial tissue structure texture does not need to be increased, determining whether the stability of the anisotropic properties of the myocardial tissue meets the requirements based on the maximum gradient of the membrane potential after the electrophysiological field equation mapping; if the stability of the anisotropic properties of the myocardial tissue does not meet the requirements, determining whether the transition function width of the cell electrophysiological model parameter mapping needs to be increased; if the transition function width of the cell electrophysiological model parameter mapping does not need to be increased, determining a pressure feedback relaxation factor in the coupling iteration based on a force flow coupling residual error; determining the pressure feedback relaxation factor in the coupling iteration based on the force flow coupling residual error, comprising: comparing the force flow coupling residual error with a preset residual error; if the force flow coupling residual error is less than or equal to the preset residual error, determining that the multiple physical field coupling iteration stability meets the requirements, the pressure feedback relaxation factor in the coupling iteration does not need to be reduced, and whether the transition function width of the cell electrophysiological model parameter mapping meets the requirements; if the force flow coupling residual error is greater than the preset residual error, determining that the multiple physical field coupling iteration stability does not meet the requirements, and the pressure feedback relaxation factor in the coupling iteration needs to be reduced; the feature weight of the myocardial tissue structure texture is a dimensionless parameter for quantitatively expressing the importance of local texture features in a myocardial image; the transition function width of the cell electrophysiological model parameter mapping is a spatial distance of a smooth transition region set to avoid discontinuous mutation of model parameters in space when mapping cell-level electrophysiological dynamics model parameters from a local to a tissue scale grid; the pressure feedback relaxation factor in the coupling iteration is an important parameter for controlling the update amplitude of the pressure load feedback from the fluid dynamics field to the mechanical field in the iterative solution of the fluid dynamics field and the mechanical field coupling. 2.The method of claim 1, wherein, determining whether the stability of the heart model construction meets the requirements based on the structural integrity of the initial heart geometric model, comprising: comparing the structural integrity of the initial heart geometric model with a preset second integrity rate; If the structural integrity of the initial heart geometric model is greater than the preset second integrity, it is determined that the stability of the heart model construction meets the requirements. If the structural integrity of the initial heart geometric model is less than or equal to the preset second integrity, it is determined that the stability of the heart model construction does not meet the requirements, and whether the feature weight of the myocardial tissue structure texture needs to be increased is determined based on the structural integrity of the initial heart geometric model. 3.The method of claim 2, wherein, The determination of whether the feature weight of the myocardial tissue structure texture needs to be increased comprises: Comparing the structural integrity of the initial heart geometric model with a preset first integrity and the preset second integrity respectively; If the structural integrity of the initial heart geometric model is less than or equal to the preset first integrity, it is determined that the feature weight of the myocardial tissue structure texture needs to be increased. If the structural integrity of the initial heart geometric model is greater than the preset first integrity and less than or equal to the preset second integrity, it is determined that the feature weight of the myocardial tissue structure texture does not need to be increased.

4. The multi-physics and geometric multi-scale based cardiac model construction method of claim 3, wherein, The increase range of the feature weight of the myocardial tissue structure texture is determined by the difference between the preset first integrity and the structural integrity of the initial heart geometric model.

5. The multi-physics and geometric multi-scale based cardiac model construction method of claim 4, wherein, The determination of whether the stability of the anisotropy property of the myocardial tissue meets the requirements based on the maximum gradient of the membrane potential after the mapping of the electrophysiological field equation comprises: Comparing the maximum gradient of the membrane potential after the mapping of the electrophysiological field equation with a preset first gradient; If the maximum gradient of the membrane potential after the mapping of the electrophysiological field equation is less than or equal to the preset first gradient, it is determined that the stability of the anisotropy property of the myocardial tissue meets the requirements, and whether the feature weight of the myocardial tissue structure texture meets the requirements is determined. If the maximum gradient of the membrane potential after the mapping of the electrophysiological field equation is greater than the preset first gradient, it is determined that the stability of the anisotropy property of the myocardial tissue does not meet the requirements, and whether the transition function width of the cell electrophysiological model parameter mapping needs to be increased is determined based on the maximum gradient of the membrane potential after the mapping of the electrophysiological field equation. 6.The method of claim 5, wherein, The determination of whether the transition function width of the cell electrophysiological model parameter mapping needs to be increased comprises: Comparing the maximum gradient of the membrane potential after the mapping of the electrophysiological field equation with the preset first gradient and a preset second gradient respectively; If the maximum gradient of the membrane potential after the mapping of the electrophysiological field equation is greater than the preset first gradient and less than or equal to the preset second gradient, it is determined that the transition function width of the cell electrophysiological model parameter mapping needs to be increased. If the maximum gradient of the membrane potential after the mapping of the electrophysiological field equation is greater than the preset second gradient, it is determined that the transition function width of the cell electrophysiological model parameter mapping does not need to be increased.

7. The multi-physics and geometric multi-scale based heart model construction method of claim 6, wherein, The increase range of the transition function width of the cell electrophysiological model parameter mapping is determined by the difference between the maximum gradient of the membrane potential after the mapping of the electrophysiological field equation and the preset first gradient.

8. The multi-physics and geometric multi-scale based heart model construction method of claim 7, wherein, The force flow coupling residual is the difference between the actual calculation result and the theoretical calculation result of the feedback iteration between the mechanical field and the fluid dynamics field.

9. The multi-physics and geometric multi-scale based heart model construction method of claim 8, wherein, The decrease range of the pressure feedback relaxation factor in the coupling iteration is determined by the difference between the force flow coupling residual and a preset residual.

Images