Coronary artery lesion evaluation and intervention simulation system based on fluid-solid coupling and working method thereof

Abstract

A fluid-structure interaction-based coronary artery lesion assessment and intervention simulation system and its working method involve computational hemodynamics and biomechanics. It acquires coronary medical imaging data of hospitalized patients, reconstructs a three-dimensional model of the patient's coronary arteries, and determines whether stenosis or risk exists. Using patient-specific flow velocity and pressure as boundary conditions, it solves the Navier-Stokes equations and the structural mechanics equations of the vessel wall, achieving bidirectional fluid-structure interaction to obtain key parameters of the flow field and coronary artery wall. Healthy coronary arteries can be simulated to identify potential risk areas for preventative measures. If plaques are present, a plaque formation stenosis model is constructed in the target coronary artery segment, and the degree of lesion risk is determined based on the coupling results. When stent intervention is required, the stent model can be selected based on the stenosis simulation and implanted into the model. The stent intervention effect is evaluated based on the coupling results, and the rationality of the stent's location, size, structure, and materials is assessed, providing a certain simulation basis for coronary artery diagnosis and treatment.

Description

Technical Field

[0001] This invention relates to the fields of computational hemodynamics and biomechanics, specifically providing a coronary artery lesion assessment and interventional simulation system based on fluid-structure interaction and its working method. Background Technology

[0002] Coronary atherosclerosis is one of the important pathological bases leading to coronary heart disease. Atherosclerotic plaques cause varying degrees of narrowing or even occlusion of the blood vessel lumen, thus significantly altering the local hemodynamic environment. Prolonged exposure to an abnormal hemodynamic environment further promotes plaque growth and increases instability, leading to serious cardiovascular diseases such as acute myocardial infarction. Coronary angiography, CTA angiography, and other medical imaging results, as well as invasive physiological parameter measurements, remain the main methods for assessing the degree of lesions. However, these methods are costly, lack quantitative analysis of hemodynamic changes, and are difficult to comprehensively reflect the true blood flow status of the lesion area, nor can they predict the specific impact of blood flow on the coronary artery wall.

[0003] With the development of computational fluid dynamics (CFD) technology, researchers are attempting to reconstruct three-dimensional models using medical imaging data and perform hemodynamic simulations to analyze flow field parameters and reveal the mechanisms of coronary artery stenosis. In real physiological environments, vascular compliance in the healthy stage is the basis for risk prediction; the interaction between the elastic deformation of the vessel wall and plaque in the diseased stage determines the degree of disease progression; the confinement of stents in the vessel after intervention significantly alters the internal flow field and mechanical environment. Therefore, treating the vessel as a rigid structure and ignoring its dynamic evolution and fluid-structure interaction effects weakens the accuracy of coronary artery simulation. Currently, more and more scholars are beginning to consider the interaction between blood flow and the coronary artery wall and are conducting fluid-structure interaction simulations. However, most studies are usually limited to biomechanical simulations under a single morphology of specific stenosis or stent intervention, with limited research on personalized healthy coronary artery risk prediction and comparison of interventional effects. Furthermore, the models, vascular plaque material parameters, and results analysis are often simplified, limiting their reference value in coronary heart disease prevention and personalized interventional programs. Summary of the Invention

[0004] To address the problems existing in the prior art, this invention provides a fluid-structure interaction-based coronary artery lesion assessment and intervention simulation system and its working method, providing simulation references for predicting coronary artery risk, assessing lesion risk, and evaluating the effectiveness and rationality of stent intervention for different patients. The system constructs a healthy model of the target coronary artery segment or an image-based stenosis model, as well as a model of stent placement in the stenotic segment when intervention is required, based on patient imaging information, enabling fluid-structure interaction simulations for different patients at different pathological stages. Through this system, not only can potential risk areas in healthy blood vessels be preliminarily identified and the degree of plaque damage to vascular function assessed, but it can also provide simulation references for predicting the effectiveness of stent intervention, thus providing a certain simulation basis for the early prevention, accurate diagnosis, and personalized treatment planning of coronary heart disease.

[0005] The technical solution adopted in this invention is: a working method of a coronary artery lesion assessment and interventional simulation system based on fluid-structure interaction, comprising the following steps:

[0006] S1. Acquire coronary CTA image data of the patient, perform vascular segmentation and structural identification of the coronary arteries, identify and screen out coronary arteries with health risks or plaques, and use software to reconstruct a three-dimensional geometric model of the patient's specific coronary arteries.

[0007] S1.1 Acquire coronary artery image data of patients through CTA technology and output it as a DICOM format file, which contains all slices and scan parameters;

[0008] S1.2. Read the patient's file into the medical image processing software for preprocessing, including noise reduction, contrast enhancement, and image registration;

[0009] S1.3. Based on coronary artery medical imaging, determine whether there is any risk or significant stenosis in the coronary arteries, and classify them into two situations:

[0010] Coronary artery health risks: No obvious stenosis in the coronary arteries, but early symptoms of stenosis where the outer diameter of a certain segment is slightly larger than adjacent segments; a segment of the coronary artery exhibiting a bifurcation angle greater than 60° or excessive curvature (k greater than 0.08 mm). -1 Abnormal geometric features, such as those described above, are classified as coronary arteries at risk of health problems.

[0011] Stenotic plaque coronary artery: Imaging shows stenosis in the coronary artery with atherosclerotic plaques. The stenotic segment is marked and classified as a stenotic coronary artery.

[0012] S1.4. Use the software's segmentation tool to segment the coronary artery, extract the target coronary artery segment and its branch structure, including at least two adjacent branches; then import the STL file into the reverse engineering surface processing software for surface repair and adjustment, and finally import it into the 3D modeling software to generate facets and apply the coronary artery wall thickness according to the patient's image data to obtain the coronary artery wall model.

[0013] The computational domain is divided into two parts: the solid domain and the fluid domain.

[0014] Coronary artery health risk: The obtained coronary artery wall is directly used as the risk solid domain, and its volume is extracted to obtain the risk fluid domain;

[0015] Stenotic plaque coronary artery: Measure the length and width of the stenotic segment in the CTA image and use them as the major and minor axes of an ellipse. Rotate to obtain an ellipsoid, construct the ellipsoidal plaque entity, and determine the location and extent of the plaque based on the image. Merge the plaque entity into the stenotic location of the coronary artery. Treat the plaque and coronary artery wall as the stenotic solid domain, and extract the corresponding stenotic fluid domain by volume extraction in 3D modeling software.

[0016] S2. Measure specific flow velocity and pressure in the patient's coronary artery as boundary conditions, add hyperelastic models of the patient's vessel wall and different plaques, and perform bidirectional fluid-structure interaction numerical simulation of the coronary artery;

[0017] S2.1. Import the obtained model into numerical simulation software for mesh generation. Both the fluid domain and the solid domain meshes are discretized using unstructured tetrahedral elements. Five wall-attached boundary layers are arranged near the vessel wall in the fluid domain, and a progressive refinement strategy is adopted for the boundary layers. The quality of the mesh generation is higher than 0.7, and the element skewness is less than 0.85.

[0018] S2.2 Calculation of the fluid domain:

[0019] Boundary condition settings for the fluid domain:

[0020] Flow velocity and pressure measurement: Using a specific flow velocity curve within one cycle of the target coronary artery segment as the inlet boundary condition and the pressure curve as the outlet boundary condition, the flow velocity and pressure curves are imported into mathematical modeling software to fit a Fourier series polynomial, which is then applied to the inlet and outlet of the target coronary artery.

[0021]

[0022] V(t) and P(t) are the instantaneous velocity or pressure at time t; n is the Fourier order; a0 is the average velocity or pressure over one period; a n and b n These are the Fourier coefficients of the cosine and sine terms of the nth harmonic, respectively. Angular frequency;

[0023] Viscosity measurement: Blood viscosity was fitted using a non-Newtonian Carreau model via mathematical modeling software.

[0024]

[0025] in Blood viscosity; Shear strain rate; The blood viscosity at which the shear rate is lowest; λ is the viscosity at the highest shear rate; λ is the relaxation time; n is the flow index.

[0026] The calculation process of the fluid domain: After the mesh is generated, the fluid domain is imported into the fluid calculation module. The laminar flow model is adopted, and the fluid domain is calculated by solving the continuity equation and the Navier-Stokes equation.

[0027] S2.3, Solid Domain Calculation

[0028] The Mooney-Rivlin hyperelastic model was applied to the coronary artery wall, and the same hyperelastic model was used for plaques in stenotic coronary arteries. The constitutive equations are as follows:

[0029]

[0030] Where I1 is the first stress invariant, I2 is the second stress invariant, and c 10 c 01 c 20 c 11 and c 02 is the hyperelastic parameter of the material's deformation characteristics, D is the incompressible parameter, and J is the elastic volume ratio;

[0031] Import the solid domain calculation module, solve the hyperelastic strain energy function in the solid domain, and solve the stress-strain by differentiating the strain energy function with respect to the deformation gradient.

[0032] S2.4, Fluid-structure Interaction

[0033] A coupling surface is set at the fluid-structure interaction (FSI) interface using a software system coupling module to satisfy the conditions of displacement continuity and stress equilibrium. Within each time step, the fluid transfers interfacial pressure and shear stress to the solid. After calculating deformation, the solid feeds back the interfacial displacement or velocity to the fluid. This iterative data exchange continues until the interfacial residuals converge, achieving bidirectional fluid-structure interaction. The pressure-velocity coupling is solved using the SIMPLE algorithm. To ensure computational convergence, the residual convergence criterion is set to 10. -4 The time step was set to 0.001, and one cardiac cycle is 1 second. To ensure calculation stability, the result of the third cycle was used as the basis for analysis.

[0034] S3. Healthy coronary arteries are simulated using fluid-structure interaction to obtain blood flow field, vessel wall stress parameters (wall shear stress, particle residence time, oscillation coefficient), VMS stress, and vessel wall pulsation morphology. Potential risk areas are identified based on hemodynamic indicators and vessel morphology.

[0035] The results were imported into visualization post-processing software to extract the coupling results of the third cycle, which served as a reference for identifying healthy coronary artery risk and assessing the risk of stenotic coronary artery.

[0036] S3.1, For coronary arteries with health risks:

[0037] Extracting flow field data and wall shear stress data from the fluid domain, and setting a threshold.

[0038] Flow field parameters: The flow field results are extracted and a trace map of the systolic peak is generated in the software. The instantaneous turbulent area ratio R is introduced on the key section perpendicular to the vessel wall.

[0039] R=A d / A×100%

[0040] A d R represents the area of ​​the region in the characteristic section where the velocity vector deflection angle is greater than 30°, and A is the area of ​​the section; if R exceeds its risk threshold, the result of prevention and warning is output.

[0041] Wall shear stress parameters: Time-averaged wall shear stress (TAWSS) reflects the long-term shear stress level experienced by the vessel wall; the oscillatory shear index describes the degree of change in the direction of blood flow shear stress during the cardiac cycle.

[0042]

[0043] The relative residence time reflects the time that blood particles spend near the blood vessel wall:

[0044]

[0045] The distribution of risk areas under different indicators is visualized and observed, and the area of ​​the risk area under different indicators is recorded. Now, a coefficient C is introduced to include all indicators in the scope of consideration:

[0046] C = (AT + AO + AR - AG) / A × 100%

[0047] AT represents the area of ​​regions with a TAWSS value less than 0.4 Pa, AO represents the area of ​​regions with an OSI value greater than 0.2, and AR represents the area of ​​regions with an RRT value greater than 10 Pa. -1 In the region, AG is the area of ​​the overlapping part of the three indicators AT, AO and AR, and A is the total wall area. If the coefficient C exceeds its risk threshold, the result of prevention and early warning will be output.

[0048] VMS stress index: Extracts the overall VMS stress of the coronary artery wall from the coronary solid domain results. If the VMS stress exceeds its risk threshold, a prevention and warning result is output.

[0049] If any of the above-mentioned flow field indicators, wall shear stress indicators, and VMS stress indicators exceed the threshold, a preventive warning will be issued.

[0050] S3.3, Stenotic plaque coronary artery:

[0051] In the fluid domain, the flow field is analyzed, a cross section is established at the narrowest point, and the degree of stenosis is determined by the ratio of the cross section area to the coronary artery diameter. When the degree of stenosis is greater than the stenosis intervention threshold, the result of interventional treatment is directly output. When the degree of stenosis is less than the stenosis intervention threshold, the relationship between the instantaneous turbulent area ratio R of the key cross section and the turbulent area ratio intervention threshold is further determined. When R exceeds the threshold, the result of interventional treatment is output.

[0052] When the coefficient C in the wall shear stress index exceeds its intervention threshold, the result of interventional treatment is output.

[0053] Stress and deformation data are extracted from the solid domain. The ratio of the average stress on the plaque edge to the average stress on the inner wall of the coronary artery is the degree of stress concentration. When the ratio exceeds the intervention threshold, the result of interventional treatment is output.

[0054] When the ratio of the average displacement of the coronary artery wall on the plaque side to the value on the opposite side exceeds its intervention threshold, the result of the interventional treatment is output.

[0055] S4. Based on the situation in step S3 where stent intervention is required, a three-dimensional model of the selected stent is established before the intervention and implanted into the coronary artery model for fluid-structure interaction simulation.

[0056] S3.1 Selecting the stent model: By examining the location and length of the stenotic area using CTA image data, and measuring the diameter of the healthy coronary arteries at both ends of the stenotic segment, the length and diameter of the stent are initially determined; the length of the stent is fine-tuned based on the abnormal flow field and stress areas in the coronary artery coupling simulation results to ensure complete coverage of the abnormal areas; the material of the stent is then selected based on the stress magnitude and curvature of the stenotic coronary artery; thus, a pre-selected stent is obtained.

[0057] S3.2 After adding the pre-selected stent to the coronary artery, perform fluid-structure interaction simulation to remove the stenotic segment, establish a complete three-dimensional model of the pre-selected stent, and bend the stent to the same curvature as the coronary artery; determine the implantation location, obtain the pre-selected stent intervention model through virtual implantation, and extract the volume to obtain the fluid domain;

[0058] S3.3 Import into the numerical simulation software, set specific boundary conditions in the fluid domain, input the elastic modulus and Poisson's ratio of the support, apply them to the solid domain, solve the continuity equation and Navier-Stokes equation for the fluid domain, solve the elastoplastic equation for the solid domain, and realize data transfer through the preset fluid-solid interface in each time step to perform fluid-structure interaction simulation.

[0059] S5. Before the interventional procedure, a three-dimensional model of the selected stent is established and implanted into the coronary artery model for fluid-structure interaction simulation. The simulation results are compared with those of the stenosis model for various parameters.

[0060] S5.1 Extract the fluid domain results and compare the peak flow field traces and the velocities of key sections during the contraction period with the results of the narrow model: the wall stress analysis coefficient C exceeds its evaluation threshold, and the support is reselected to repeat the fluid-structure interaction simulation process;

[0061] S5.2 Extract the Von-Mises stress distribution of the stent in the solid domain during the cardiac cycle, monitor the stress concentration at the stent connecting ribs and support peaks, and when the maximum stress value is higher than its evaluation threshold, it indicates that the stent material or structure selection is unreasonable and the stent needs to be reselected and the fluid-structure interaction simulation process repeated.

[0062] S5.3, The displacement restriction ratio M is the change in radial displacement of the coronary artery wall at the distal or proximal end of the stent segment compared to the change in coronary artery wall displacement at the stent segment:

[0063] M = (DR - DS) / DR × 100%

[0064] Wherein, DR is the average radial displacement of the adjacent structurally normal proximal or distal healthy blood vessel segment during the cycle, and DS is the average radial displacement of the stent intervention segment; if the displacement limitation ratio exceeds its evaluation threshold, the stent is reselected and the fluid-structure interaction simulation process is repeated.

[0065] If the wall stress analysis coefficient, Von-Mises stress, and displacement confinement ratio all meet the requirements, then the support is selected.

[0066] A coronary artery lesion assessment and interventional simulation system based on fluid-structure interaction includes an image acquisition and 3D modeling module, a fluid-structure interaction simulation configuration module, a healthy coronary artery risk identification module, a stenotic coronary artery lesion assessment module, and an interventional effect simulation assessment module.

[0067] The image acquisition and 3D modeling module is used to acquire patient coronary CTA or MRI image data, segment and identify coronary arteries, determine the health status of blood vessels and plaque conditions, and reconstruct a patient-specific 3D geometric model of the coronary artery based on the image data, while generating a solid domain coronary artery wall model and a fluid domain model.

[0068] The fluid-structure interaction simulation configuration module is used to measure specific flow velocities and pressures in the patient's coronary arteries as boundary conditions, configure the non-Newtonian fluid properties of blood, the hyperelastic material model of the vessel wall and plaque, divide the fluid domain and solid domain meshes, and complete the pre-setup for bidirectional fluid-structure interaction numerical simulation.

[0069] The healthy coronary artery risk identification module is used to perform bidirectional fluid-structure interaction simulation on healthy coronary arteries, extract data such as blood flow field, vessel wall stress parameters, VMS stress and vessel wall pulsation morphology, identify potential risk areas by combining hemodynamic indicators and vessel morphology, and quantitatively evaluate parameters and visualize dangerous areas.

[0070] The stenotic coronary artery lesion assessment module is used to perform bidirectional fluid-structure interaction simulation of stenotic coronary arteries, based on the flow field simulation results, stress and deformation of solid domain plaques, and related indicators.

[0071] The interventional effect simulation evaluation module is used to establish a three-dimensional model of the selected stent and implant it into the coronary artery model to perform fluid-structure interaction simulation, and compare the parameters of the stenosis model with the healthy / stenosis model.

[0072] The modules described above work together to achieve the above-described working method.

[0073] The beneficial effects of this invention are as follows: This invention enables multi-dimensional quantitative analysis of coronary hemodynamics and vessel wall mechanical properties, allowing for the early identification of potentially high-risk coronary artery areas such as pre-stenosis and geometrical abnormalities, avoiding missed diagnoses and delayed interventions. It also establishes standardized interventional assessment indicators for plaque stenosis lesions, providing objective data support for clinical diagnosis and treatment decisions. Furthermore, through preoperative virtual stent implantation and fluid-structure interaction simulation iteration optimization, it accurately predicts stent compatibility, mechanical stability, and blood flow improvement effects, avoiding clinical risks such as improper stent selection, stress concentration, and restricted vessel displacement, significantly improving the accuracy of interventional procedures. Attached Figure Description

[0074] Figure 1 This is a flowchart of a coronary artery lesion assessment and interventional simulation system based on fluid-structure interaction and its working method, as described in this invention.

[0075] Figure 2 This is a schematic diagram of the local coronary artery modeling in the embodiment, including part of the left anterior descending branch and the left circumflex branch. (a) shows the CTA image data segmentation process, and (b), (c), and (d) are three-dimensional reconstruction models of healthy coronary arteries, stenotic coronary arteries, and stent-interventional coronary arteries, respectively.

[0076] Figure 3 The diagrams show the streamlines and velocity distributions within the coronary arteries obtained after fluid-structure interaction simulations of three models. (a), (b), and (c) are the front and rear views of the traces and velocity distributions of the local coronary arteries at peak systolic velocity under the morphology of healthy coronary arteries, stenotic coronary arteries, and stent-interventional coronary arteries, respectively.

[0077] Figure 4 The following are visualization diagrams of stress analysis and danger zones under different coronary artery segments in this embodiment. (a1), (a2), and (a3) ​​are TAWSS diagrams of three models: healthy coronary artery, stenotic coronary artery, and stent-interventional coronary artery. (b1), (b2), and (b3) are danger zones with TAWSS less than 0.4 Pa. (c1), (c2), and (c3) are areas showing excessively long particle residence time with RRT greater than 10 Pa⁻¹.

[0078] Figure 5In the diagram, (a1), (a2), and (a3) ​​represent the deformation of the coronary artery wall in healthy coronary arteries, narrowed coronary arteries, and stent-interventional coronary arteries, respectively. (b1), (b2), and (b3) represent their internal stress diagrams. Detailed Implementation

[0079] To make the objectives, technical solutions, and advantages of the present invention clearer, the relevant technical solutions will be described in detail below with reference to the accompanying drawings. The embodiments described are only some embodiments of the present invention and not all embodiments.

[0080] Detailed steps:

[0081] Step 1: Acquire coronary artery imaging data of the patient using CTA technology, outputting it as a DICOM format file. The file must contain all slices and scan parameters. Read the patient's file into medical image processing software (such as MIMICS) for preprocessing, including noise reduction, contrast enhancement, and image registration to improve the identifiability of the coronary arteries. Based on the coronary artery imaging, determine whether there is any risk or significant stenosis in the coronary arteries, and categorize it into two situations:

[0082] Scenario 1: No obvious stenosis in the coronary artery, but early symptoms of stenosis where the outer diameter of a certain part is slightly larger than that of the adjacent segment; abnormal geometric features such as a bifurcation angle greater than 60° or excessive curvature in a certain segment of the coronary artery. In these cases, there is no stenosis in the coronary artery, but there is a certain risk. The risk area is marked and classified as a healthy coronary artery.

[0083] Scenario 2: Imaging shows coronary artery stenosis with atherosclerotic plaques. The stenotic segment is marked and classified as a stenotic coronary artery.

[0084] The coronary arteries are then segmented using the software's in-system segmentation tool to extract the target coronary artery segment and its branch structures, including at least two adjacent branches. The STL file is then imported into reverse engineering surface processing software (such as GeomagicWrap) for surface repair and adjustment. Finally, it is imported into 3D modeling software (such as Spaceclaim) to generate facets and apply a coronary artery wall thickness, typically 0.4 mm, based on the patient's image data, resulting in a coronary artery wall model. The computational domain is divided into two parts: a solid domain and a fluid domain.

[0085] Healthy coronary arteries: No stenosis is observed, so the obtained coronary artery wall is directly used as a solid domain for volume extraction to obtain a fluid domain;

[0086] Stenotic coronary arteries: Most current studies simulate stenosis by reducing the diameter of the coronary arteries. This invention achieves accurate model building by manually adding plaques. The length and width of the stenosis are measured in CTA images and used as the major and minor axes of an ellipse. An ellipsoid is obtained by rotation, and an ellipsoidal plaque entity is constructed. The location and extent of the plaque are determined based on the images, and the plaque entity is merged into the location of the coronary artery stenosis. Redundant entities are deleted to obtain the patient's coronary artery with the specific plaque added, i.e., the solid domain. The corresponding stenotic fluid domain is obtained by volume extraction in 3D modeling software.

[0087] Step 2: Import the obtained model into numerical simulation software (such as ANSYS, COMSOL) for mesh generation in preparation for finite element simulation. Both the fluid and solid domains are discretized using unstructured tetrahedral elements. In the fluid domain, five wall-attached boundary layers are placed near the vessel wall to improve the velocity gradient resolution in the near-wall region. A progressive refinement strategy is used for the boundary layers to ensure the accuracy of wall shear stress calculation. Furthermore, the mesh quality should be higher than 0.7, and the element skewness should be less than 0.85.

[0088] I. Fluid Calculation

[0089] 1) Boundary condition settings

[0090] Because there are significant individual differences among patients in terms of coronary flow velocity, blood pressure, and blood viscosity, most studies use uniform boundary conditions and blood viscosity, which can mask the true hemodynamic characteristics of patients and lead to significant deviations in simulation results.

[0091] Flow velocity and pressure measurement: Patients often have conditions such as excessively high blood flow velocity or hypertension. Therefore, it is necessary to obtain a specific flow velocity curve of the target coronary artery segment within one cycle through Doppler guidewire measurement or CTA estimation as the inlet boundary condition for simulation, and to measure the pressure curve through a pressure guidewire as the outlet boundary condition. The obtained curves are then imported into mathematical modeling software (such as MATLAB) to fit a Fourier series polynomial, which is applied to the inlet and outlet of the target coronary artery.

[0092]

[0093] V(t) and P(t) are the instantaneous velocity or pressure at time t; n is the Fourier order; a0 is the average velocity or pressure over one period; a n and b n These are the Fourier coefficients of the cosine and sine terms of the nth harmonic, respectively. It is the angular frequency; its order and coefficients are determined based on the fitting results to ensure that the difference from the actual measurement curve is less than 5%.

[0094] Viscosity Measurement: Blood viscosity is also an important parameter affecting the results. Previously, blood viscosity was mostly set to a constant value or a non-Newtonian model with the same parameters. This invention measures the blood viscosity of patients at different shear rates using a rheometer, and fits the results using a non-Newtonian Carreau model through mathematical modeling software.

[0095]

[0096] in Blood viscosity; Shear strain rate; The blood viscosity at which the shear rate is lowest; λ is the viscosity at the highest shear rate; λ is the relaxation time; n is the flow index; set the parameters according to the fitting results.

[0097] 2) Calculation process

[0098] After the mesh is generated, the fluid domain is imported into a fluid calculation module (such as ANSYS Fluent). Since the coronary artery is a small-diameter pipe with a Reynolds number generally not greater than 1000, a laminar flow model is adopted, and the fluid domain is calculated by solving the continuity equation and the Navier-Stokes equation.

[0099] II. Solid Domain Calculation

[0100] 1) Solid domain material settings

[0101] The coronary artery wall is a hyperelastic material. Treating the solid domain as a rigid or linearly elastic material does not reflect real physiological conditions. This invention uses a 5-parameter Mooney-Rivlin hyperelastic model to better reflect the material properties of the blood vessel wall. The plaque in the stenosis model uses the same hyperelastic model, and its constitutive equation is:

[0102]

[0103] Where I1 is the first stress invariant, I2 is the second stress invariant, and c 10 c 01 c 20 c 11 and c 02 is the hyperelastic parameter describing the deformation properties of the material, D is the incompressible parameter describing the compressibility, and J is the elastic volume ratio. The parameters are obtained by fitting multiaxial tensile test data of blood vessel walls and plaques.

[0104] 2) Calculation process

[0105] Import the solid domain calculation module (such as ANSYS Transient Structural), solve the hyperelastic strain energy function of the solid domain, and solve the stress and strain by differentiating the deformation gradient with respect to the strain energy function.

[0106] III. Fluid-structure interaction

[0107] In reality, the deformation of the coronary artery wall significantly alters the local blood flow and wall stress distribution, while changes in blood flow, in turn, affect the motion and deformation of the coronary artery wall. Therefore, fluid-structure interaction becomes a necessary path to achieve accurate coronary artery simulation.

[0108] A coupling surface is set at the fluid-structure interaction (FSI) using a software system coupling module (such as ANSYS System Coupling) to satisfy the conditions of displacement continuity and stress equilibrium. Within each time step, the fluid transfers interfacial pressure and shear stress to the solid. After calculating deformation, the solid feeds back the interfacial displacement or velocity to the fluid. This iterative data exchange continues until the interfacial residuals converge, achieving bidirectional fluid-structure interaction. The pressure-velocity coupling is solved using the SIMPLE algorithm. To ensure computational convergence, the residual convergence criterion is set to 10. -4 The time step was set to 0.001, and one cardiac cycle is 1 second. To ensure calculation stability, the result of the third cycle was used as the basis for analysis.

[0109] Step 3: Import the results into visualization post-processing software (such as CFDPOST, Tecplot) to extract the coupling results of the third cycle, which can be used as a reference for the identification of healthy coronary artery risk and the assessment of stenotic coronary artery risk.

[0110] Healthy coronary arteries:

[0111] 1) Flow field data and wall shear stress data need to be extracted from the fluid domain.

[0112] Flow field parameters: Extract the flow field results and generate a trace plot of the contraction peak in the software, focusing on observing anomalies.

[0113] At bifurcation points and in areas of coronary artery abnormality in imaging, it is important to check for eddies, backflow, and secondary flows. Since these turbulent flows are direct factors affecting coronary artery health, a critical cross-section perpendicular to the vessel wall is created in areas of significant turbulence, introducing an instantaneous turbulence area ratio R.

[0114] R=A d / A×100%

[0115] A d Let A be the area of ​​the region in the characteristic cross-section where the velocity vector deflection angle is greater than 30°. If the calculated R is greater than 20%, it indicates that there is strong dynamic disturbance in the coronary artery at that location, which is prone to inducing endothelial cell damage.

[0116] Wall shear stress parameters:

[0117] Wall shear stress represents the magnitude of the tangential stress exerted on the vessel wall by blood flow at a specific moment. However, data from a single moment cannot reflect the stress over the entire cycle. Therefore, time-averaged wall shear stress (TAWSS) should be calculated to reflect the long-term shear stress level on the vessel wall. A TAWSS value less than 0.4 Pa is considered a low-shear region and is relatively dangerous. The oscillatory shear index (OSI) is calculated to describe the degree of change in the direction of blood flow shear stress during the cardiac cycle.

[0118]

[0119] For patients with coronary artery disease, an OSI value greater than 0.2 indicates that the blood flow in this area is in an oscillatory state, which can easily lead to endothelial cell dysfunction and promote the occurrence of atherosclerosis.

[0120] The relative residence time (RRT) reflects the time blood particles spend near the vessel wall.

[0121]

[0122] For the inner wall of the coronary artery, a pressure greater than 10 Pa is required. -1 The area indicated by the symbol represents an increased time that red blood cells and impurities in the blood remain in this area, increasing the likelihood of lipid deposition and inflammatory responses, and further promoting plaque formation.

[0123] Based on this, the distribution of risk areas under different indicators is visualized and observed, and the area of ​​risk areas under different indicators is recorded. In the past, a single indicator was often used to judge dangerous areas. Now, a coefficient C is introduced to include all indicators in the scope of consideration.

[0124] C=(A T +A O +A R -A G ) / A×100%

[0125] A T For the area of ​​TAWSS less than 0.4 Pa, A O A represents the area of ​​the region with an OSI value greater than 0.2. R RRT greater than 10 Pa -1 Area A G The area of ​​the overlapping part of the three indicators is C, and A is the total wall area. When C is greater than 15%, the induced patch growth range is large and timely prevention is necessary.

[0126] 2) VMS stress (Von Mises stress) is a method that equates complex principal stresses in three directions to...

[0127] A comprehensive index of a single scalar is used to assess whether a material yields or fails under multi-directional stress. In coronary artery simulation, it can visually quantify the mechanical load exerted by blood flow pressure on the vessel wall. It extracts the overall VMS stress of the coronary artery wall from the solid domain results of healthy coronary arteries. According to studies of normal coronary arteries, stress within the range of 0.03-0.2 MPa is considered normal; exceeding this range indicates abnormal stress, suggesting that the mechanical load on the coronary artery wall exceeds normal values. If any of the above indicators exceeds the threshold, preventative treatment is necessary.

[0128] Stenotic coronary arteries:

[0129] 1) In fluid domain flow field analysis, a cross-section needs to be established at the narrowest point. The degree of stenosis is determined by the ratio of the cross-sectional area to the coronary artery diameter. When the stenosis is greater than 70%, interventional treatment should be considered directly. When it is less than 70%, the instantaneous turbulent area ratio R of the critical cross-section should be considered. If R is greater than 50%, it indicates that the stenosis has caused large-area turbulence, and stent intervention should be performed promptly. Since plaque has already formed, a coefficient C of 10% in the wall shear stress index indicates a high risk of plaque growth, and a coefficient greater than 15% requires stent intervention to reduce the probability of continued plaque growth.

[0130] 2) In the solid domain, the presence of plaque leads to significant VMS stress concentration and restricted coronary artery movement, further exacerbating plaque growth. Therefore, the results primarily extract stress and deformation data. The ratio of the average stress on the plaque edge to the average stress on the inner wall of the coronary artery indicates the degree of stress concentration. A ratio greater than 4 indicates that the stress concentration caused by the plaque has a significant impact on the coronary artery, making plaque rupture and growth more likely. Similarly, a ratio of the average displacement of the coronary artery wall on the plaque side to that on the opposite side greater than 1.5 indicates significant abnormal deformation on the plaque side, which, under long-term pulsation, will lead to decreased elasticity of the coronary artery wall and weakened blood transfusion capacity. Based on these indicators, the degree of stenosis in the current coronary artery can be assessed, and when a certain indicator exceeds the corresponding threshold, stent intervention should be considered.

[0131] Step 4: Based on the situation requiring stent intervention in Step 3, a three-dimensional model of the selected stent is established before the intervention and implanted into the coronary artery model for fluid-structure interaction simulation. The intervention effect is evaluated by comparing the simulation results of various parameters with the stenosis model, and the rationality of the stent's position, size, structure and materials are assessed.

[0132] Stent selection: First, the location and length of the stenotic area are determined using CTA imaging data, and the diameters of the healthy coronary arteries at both ends of the stenotic segment are measured to initially determine the length and diameter of the stent. Then, the stent length is fine-tuned based on the abnormal flow field and stress influence areas identified in the coronary artery coupling simulation results to ensure complete coverage of the abnormal areas. Next, the stent material is selected based on the stress magnitude and curvature of the stenotic coronary artery. If the coronary artery stress is less than 0.2 MPa, a cobalt-chromium alloy can be used; if it is greater than 0.2 MPa, a higher-strength stainless steel material is considered; and if the curvature is greater than 0.1, a nickel-titanium alloy with a higher fatigue limit should be considered.

[0133] Although a stent model has been initially selected, the actual effect is unknown. Therefore, it is necessary to insert the stent into the coronary artery before performing fluid-structure interaction (FSI) simulation. The stenotic segment of the patient's coronary artery is removed, and a complete 3D model of the stent is created in 3D modeling software. The stent is bent to the same curvature as the coronary artery. After determining the implantation location, a stent intervention model is obtained through virtual implantation, and the volume is extracted to obtain the fluid domain. This is imported into numerical simulation software, and specific boundary conditions are set in the fluid domain. The stent material is generally an isotropic elastoplastic material. The elastic modulus and Poisson's ratio of the stent are input and applied to the solid domain. The continuity equation and Navier-Stokes equations are solved for the fluid domain, and the elastoplastic equations are solved for the solid domain. Data transfer is achieved through a pre-set fluid-solid interface at each time step to perform FSI simulation.

[0134] The fluid domain results were extracted, and the peak flow field traces and velocities at key sections during systole were compared with the stenosis model results. If the proportion of turbulent flow area R was significantly reduced and not higher than 5%, it indicated that the intervention improved coronary blood flow, the velocity distribution was more uniform, and the stent position and size were more reasonable. In terms of wall shear stress, due to the adhesion between the stent implantation and the vessel wall, a certain danger zone will be formed around the stent, which is a normal intervention phenomenon. A wall stress analysis coefficient C of less than 15% is within the normal acceptable range. When it is greater than 15%, it indicates that the size of the interventional stent may be too large, causing the danger zone area to exceed the threshold, and the stent diameter needs to be reconsidered.

[0135] The Von-Mises stress distribution of the stent in the solid domain during the cardiac cycle is extracted, with a focus on monitoring stress concentration at the stent connectors and support peaks. When the maximum stress value is below 80% of the yield strength of the selected material (e.g., cobalt-chromium alloy, approximately 600 MPa), the risk of fatigue fracture under long-term pulsating circulation can be avoided. When the stress value exceeds this threshold, it indicates that the stent material or structure selection is unreasonable and a new stent needs to be chosen. Secondly, the displacement restriction ratio M is introduced. By comparing the changes in radial displacement of the coronary artery wall at the distal or proximal end of the stent segment with the changes in coronary artery wall displacement at the stent segment, the degree of stent constraint on the natural vasomotor function of the blood vessel is quantified.

[0136] M=(D R-D S ) / D R ×100%

[0137] Where D R D represents the average radial displacement of the structurally normal proximal or distal healthy vessel segment adjacent to the stent implantation segment during the cycle. S This represents the average radial displacement of the stent segment. Typically, the displacement restriction ratio should not exceed 30% to prevent restenosis at the stent edge due to loss of mechanical compliance in the coronary artery. Based on these indicators, when the stent's own stress is within safe fatigue limits and the stent segment displacement restriction ratio is less than 30%, the selection of the stent can be deemed clinically reasonable. Otherwise, a new stent should be selected, and the fluid-structure interaction simulation process repeated.

[0138] Example 1

[0139] See Figure 1 The flowchart illustrates the working method of the coronary artery lesion assessment and interventional simulation system based on fluid-structure interaction of this invention, including:

[0140] S1. Acquire the patient's coronary CTA image data, perform vascular segmentation and structural identification of the coronary arteries, determine whether they are healthy or have plaques, and use software to reconstruct a patient-specific three-dimensional geometric model of the coronary arteries.

[0141] The DICOM image file from the CTA scan of the patient in Example 1 was read in and preprocessed, including denoising, contrast enhancement, and image registration, to improve the identifiability of the coronary arteries. Subsequently, medical image processing software (such as MIMICS) was used to segment the coronary arteries and extract the target coronary artery segment and its branch structures.

[0142] Based on this, the blood vessel wall region is analyzed to identify the presence, location, morphology, and extent of atherosclerotic plaques, thereby determining whether the coronary arteries are healthy or diseased. See also Figure 2 Finally, the segmented vascular contour data is imported into 3D reconstruction software. Through surface reconstruction and smoothing, a patient-specific 3D geometric model of the coronary artery is established, and the corresponding fluid domain is extracted.

[0143] S2. Measure specific flow velocity and pressure in the patient's coronary artery as boundary conditions, add hyperelastic models of the patient's vessel wall and different plaques, and perform bidirectional fluid-structure interaction numerical simulation of the coronary artery.

[0144] Boundary conditions are used to describe the hemodynamic characteristics of the coronary artery 3D model at the inlet and outlet boundaries, mainly including parameters such as blood flow velocity and pressure distribution. These boundary conditions are affected by the individual physiological state of the patient, and cardiac blood flow may vary significantly under different physiological states. Therefore, each patient's specific physiological parameters should be measured individually as boundary conditions for that patient's simulation.

[0145] Then, mathematical modeling software was used to transform the measured periodic curves into multi-order Fourier series polynomials, which were imported into simulation software (such as Ansys, COMSOL, etc.) as boundary conditions for the fluid domain. Appropriate hyperelastic models and parameters were selected based on the patient's vascular condition. The Navier-Stokes equations and hyperelastic nonlinear structure equations were solved for both the fluid and solid domains, and the fluid-solid data transmission surface was coupled. At least three cardiac cycles were calculated, and the results of the last cycle were used for analysis.

[0146] S3. Healthy coronary arteries are simulated using fluid-structure interaction to obtain blood flow field, vessel wall stress parameters (wall shear stress, particle residence time, oscillation coefficient), VMS stress, and vessel wall pulsation morphology. Potential risk areas are identified based on hemodynamic indicators and vessel morphology.

[0147] The "healthy coronary artery" model, extracted from relatively healthy patient imaging data, focuses on segments more prone to plaque formation, primarily to mitigate related risks. Blood flow field data is extracted from the calculation results and visualized using simulation post-processing software to generate contour maps.

[0148] See the cloud map of the flow field trace in healthy coronary arteries. Figure 3 Based on this, it can be determined that there is some turbulent flow in the coronary artery, but the calculated instantaneous turbulent area ratio R is 8.2%, which does not exceed 20%, and the proportion is relatively small and safe; the visualization cloud map of its stress-related indicators can be found in [reference needed]. Figure 4 As can be seen, some areas of the coronary artery in the embodiment are in an abnormal environment of low shear stress and high particle residence time, but the calculated coefficient C value is 12.5%, which does not exceed 15% and is within an acceptable range; see also Figure 5 It can be determined that the deformation and stress of the coronary artery wall are within the normal range, and overall, only disease prevention is needed.

[0149] S4. For stenotic coronary arteries, the impact of plaques on the coronary arteries is determined based on the flow field simulation results, the stress and deformation of solid-domain plaques, and other indicators. This helps to assess the risk level of the lesion, provide a certain reference for whether to insert a stent, and allow for the selection of a suitable stent model based on the simulation results.

[0150] Based on the difference in brightness between plaques and blood vessels, plaque information was obtained from the patient's CTA file by setting a threshold. The plaque was approximately 9 mm long and consisted mostly of a lipid core. This core was manually added to the model to obtain a patient-specific three-dimensional model of coronary artery stenosis. The fluid domain was extracted, and the plaque material properties in the solid domain were set according to the plaque type.

[0151] The results of fluid-structure interaction are then visualized, see reference. Figure 3-5Severe turbulence was observed in the flow field of the narrowed coronary artery. The calculated instantaneous turbulent area ratio R was 55.8%, greater than 50%, and the influence extended to the distal bifurcation. From a stress perspective, there was a certain risk downstream of the plaque. The calculated wall stress analysis coefficient C was 8%, less than 10%, but located directly below the plaque, which could lead to further plaque development. From the perspective of the coronary artery wall, severe deformation anomalies and stress concentration were observed in the plaque area. The wall stress around the plaque was approximately 5 times that in the center; a ratio exceeding 4 times indicates severe stress concentration. Furthermore, the ratio of the average displacement of the coronary artery wall on the plaque side to that on the opposite side was 1.8, greater than 1.5. Considering all these factors, an interventional stent should be considered. Based on the tube diameter and the extent of plaque influence, a coronary stent with a length of 13 mm and an inner diameter of 4 mm, made of cobalt-based alloy, was selected.

[0152] S5. Before the interventional procedure, a three-dimensional model of the selected stent is established and implanted into the coronary artery model for fluid-structure interaction simulation. The interventional effect is evaluated by comparing the simulation results of various parameters with those of the stenosis model, and the rationality of the stent's position, size, structure and materials are assessed to provide simulation basis for actual diagnosis and treatment.

[0153] The stenotic segment of the patient's coronary artery was removed, and the stent was bent to match the coronary artery. The implantation position was determined so that the top of the stent was 8 mm from the coronary artery inlet. After implantation, the volume was extracted again to form a fluid domain, and the material properties of the cobalt-based alloy stent were applied to the stent. The elastic modulus was approximately 243 GPa, the Poisson's ratio was 0.3, and the yield strength was approximately 900 MPa. Then, fluid-structure interaction simulation was performed.

[0154] To visualize the results, see [link / reference]. Figure 3-5 The flow field traces tend to be stable, and the turbulent flow area ratio R is 3.8%, which is less than 5%, indicating a good intervention effect and a relatively uniform velocity distribution. However, the stress analysis cloud map shows that there are many low shear stress areas around the stent, which can be determined to be caused by the stent compressing the coronary artery wall. The wall stress analysis coefficient C is about 13%, which is less than 15%. In terms of deformation, the presence of the stent restricts the movement of the coronary artery to a certain extent, but the calculated displacement restriction ratio M is about 22.6%, which is less than 30%, which is within an acceptable range. There is a certain stress concentration inside, and the stress at the stent edge is about 1-2 times that of the nearby area, which is relatively small. Moreover, the maximum stress value at the stent connecting rib and the support peak is 452 MPa, which is less than 80% of the stent yield strength. Overall, the simulation results of this stent during intervention are good, and its use can be considered in combination with clinical experience.

[0155] The above description is only for illustrating several preferred embodiments of the present invention and their technical principles. Those skilled in the art will understand that the scope of protection of the present invention is not limited to the specific technical solutions described in the above embodiments. Any technical solutions formed by equivalent substitutions, changes, or combinations of related technical features without departing from the technical concept of the present invention should be considered to fall within the scope of protection of the present invention.

Claims

1. A working method for a coronary artery lesion assessment and interventional simulation system based on fluid-structure interaction, characterized in that: Includes the following steps: S1. Acquire patient coronary CTA image data, perform vascular segmentation and structural identification of coronary arteries, identify and screen out coronary arteries with health risks and stenotic plaques, and use software to reconstruct a three-dimensional geometric model of the patient-specific coronary arteries; S2. Measure specific flow velocity and pressure in the patient's coronary artery as boundary conditions, add hyperelastic models of the patient's vessel wall and different plaques, and perform bidirectional fluid-structure interaction numerical simulation of the coronary artery; S2.

1. Import the obtained model into numerical simulation software for mesh generation. Both the fluid domain and the solid domain meshes are discretized using unstructured tetrahedral elements. Five wall-attached boundary layers are arranged near the vessel wall in the fluid domain, and a progressive refinement strategy is adopted for the boundary layers. The quality of the mesh generation is higher than 0.7, and the element skewness is less than 0.

85. S2.2 Calculation of the fluid domain: Boundary condition settings for the fluid domain: Flow velocity and pressure measurement: Using a specific flow velocity curve within one cycle of the target coronary artery segment as the inlet boundary condition and the pressure curve as the outlet boundary condition, the flow velocity and pressure curves are imported into mathematical modeling software to fit a Fourier series polynomial, which is then applied to the inlet and outlet of the target coronary artery. ； V(t) and P(t) are the instantaneous velocity or pressure at time t; n is the Fourier order; a0 is the average velocity or pressure over one period; a n and b n These are the Fourier coefficients of the cosine and sine terms of the nth harmonic, respectively. Angular frequency; Viscosity measurement: Blood viscosity was fitted using a non-Newtonian Carreau model via mathematical modeling software. ； in Blood viscosity; Shear strain rate; The blood viscosity at which the shear rate is lowest; It is the viscosity at the highest shear rate; λ is the relaxation time; n is the liquidity index; The calculation process of the fluid domain: After the mesh is generated, the fluid domain is imported into the fluid calculation module. The laminar flow model is adopted, and the fluid domain is calculated by solving the continuity equation and the Navier-Stokes equation. S2.3, Solid Domain Calculation The Mooney-Rivlin hyperelastic model was applied to the coronary artery wall, and the same hyperelastic model was used for plaques in stenotic coronary arteries. The constitutive equations are as follows: ； Where I1 is the first stress invariant, I2 is the second stress invariant, and c 10 c 01 c 20 c 11 and c 02 is the hyperelastic parameter of the material's deformation characteristics, D is the incompressible parameter, and J is the elastic volume ratio; Import the solid domain calculation module, solve the hyperelastic strain energy function in the solid domain, and solve the stress-strain by differentiating the strain energy function with respect to the deformation gradient. S2.4, Fluid-structure Interaction A coupling surface is set at the fluid-solid interface through the software system coupling module to satisfy the conditions of displacement continuity and stress balance. In each time step, the fluid transfers the interface pressure and shear stress to the solid. After the solid calculates the deformation, it feeds back the interface displacement or velocity to the fluid. The data is exchanged iteratively until the interface residual converges. S3. Healthy coronary arteries are simulated using fluid-structure interaction to obtain blood flow field, vessel wall stress parameters, VMS stress, and vessel wall pulsation morphology. Potential risk areas are identified based on hemodynamic indicators and vessel morphology. S3.1, For coronary arteries with health risks: Flow field data and wall shear stress data are extracted from the fluid domain. Flow field parameters: The extracted flow field results are used to generate a trace map of the contraction peak in the software. The instantaneous turbulent area ratio R is introduced on the key section perpendicular to the blood vessel wall. R=A d / A×100%; A d R represents the area of ​​the region in the characteristic section where the velocity vector deflection angle is greater than 30°, and A is the area of ​​the section; if R exceeds its risk threshold, the result of prevention and warning is output. Wall shear stress parameters: Time-averaged wall shear stress (TAWSS) reflects the long-term shear stress level experienced by the vessel wall; the oscillatory shear index describes the degree of change in the direction of blood flow shear stress during the cardiac cycle. ； The relative residence time reflects the time that blood particles spend near the blood vessel wall: ； The distribution of risk areas under different indicators is visualized and observed, and the area of ​​the risk area under different indicators is recorded. Now, a coefficient C is introduced to include all indicators in the scope of consideration: C = (AT + AO + AR - AG) / A × 100%; AT represents the area of ​​regions with a TAWSS value less than 0.4 Pa, AO represents the area of ​​regions with an OSI value greater than 0.2, and AR represents the area of ​​regions with an RRT value greater than 10 Pa. -1 In the region, AG is the area of ​​the overlapping part of the three indicators AT, AO and AR, and A is the total wall area. If the coefficient C exceeds its risk threshold, the result of prevention and early warning will be output. VMS stress index: Extract the overall VMS stress of the coronary artery wall from the coronary solid domain results. If the VMS stress exceeds its risk threshold, a prevention and warning result is output. If any of the above-mentioned flow field indicators, wall shear stress indicators, and VMS stress indicators exceed the threshold, a preventive warning will be issued. S3.2, Stenotic plaque coronary artery: In the fluid domain, the flow field is analyzed, a cross section is established at the narrowest point, and the degree of stenosis is determined by the ratio of the cross section area to the coronary artery diameter. When the degree of stenosis is greater than the stenosis intervention threshold, the result of interventional treatment is directly output. When the degree of stenosis is less than the stenosis intervention threshold, the relationship between the instantaneous turbulent area ratio R of the key cross section and the turbulent area ratio intervention threshold is further determined. When R exceeds the threshold, the result of interventional treatment is output. When the coefficient C in the wall shear stress index exceeds its intervention threshold, the result of interventional treatment is output. Stress and deformation data are extracted from the solid domain. The ratio of the average stress on the plaque edge to the average stress on the inner wall of the coronary artery is the degree of stress concentration. When the ratio exceeds the intervention threshold, the result of interventional treatment is output. When the ratio of the average displacement of the coronary artery wall on the plaque side to the value on the opposite side exceeds its intervention threshold, the result of the interventional treatment is output. S4. Based on the situation in step S3 where stent intervention is required, a three-dimensional model of the selected stent is established before the intervention and implanted into the coronary artery model for fluid-structure interaction simulation. S5. Before the interventional procedure, a three-dimensional model of the selected stent is established and implanted into the coronary artery model for fluid-structure interaction simulation. The simulation results are compared with those of the stenosis model for various parameters. S5.1 Extract the fluid domain results and compare the peak flow field traces and the velocities of key sections during the contraction period with the results of the narrow model: the wall stress analysis coefficient C exceeds its evaluation threshold, and the support is reselected to repeat the fluid-structure interaction simulation process; S5.2 Extract the Von-Mises stress distribution of the stent in the solid domain during the cardiac cycle, monitor the stress concentration at the stent connecting ribs and support peaks, and when the maximum stress value is higher than its evaluation threshold, it indicates that the stent material or structure selection is unreasonable and the stent needs to be reselected and the fluid-structure interaction simulation process repeated. S5.3, The displacement restriction ratio M is the change in radial displacement of the coronary artery wall at the distal or proximal end of the stent segment compared to the change in coronary artery wall displacement at the stent segment: M = (DR - DS) / DR × 100%; Wherein, DR is the average radial displacement of the adjacent structurally normal proximal or distal healthy blood vessel segment during the cycle, and DS is the average radial displacement of the stent intervention segment; if the displacement limitation ratio exceeds its evaluation threshold, the stent is reselected and the fluid-structure interaction simulation process is repeated. If the wall stress analysis coefficient, Von-Mises stress, and displacement confinement ratio all meet the requirements, then the support is selected.

2. The working method of the coronary artery lesion assessment and interventional simulation system based on fluid-structure interaction according to claim 1, characterized in that, Step S1 specifically includes the following sub-steps: S1.1 Acquire coronary artery image data of patients through CTA technology and output it as a DICOM format file, which contains all slices and scan parameters; S1.

2. Read the patient's file into the medical image processing software for preprocessing, including noise reduction, contrast enhancement, and image registration; S1.

3. Based on coronary artery medical imaging, determine whether there is any risk or significant stenosis in the coronary arteries, and classify them into two situations: Coronary arteries at health risk: Coronary arteries without significant stenosis, but with a portion having a larger outer diameter than adjacent segments, are classified as coronary arteries at health risk; a segment of the coronary artery exhibits a bifurcation angle greater than 60° or a curvature k greater than 0.08 mm. -1 Abnormal geometric features are classified as coronary arteries at health risk. Stenotic plaque coronary artery: Imaging shows stenosis in the coronary artery with atherosclerotic plaques. The stenotic segment is marked and classified as a stenotic coronary artery. S1.

4. Use the software's segmentation tool to segment the coronary artery, extract the target coronary artery segment and its branch structure, including at least two adjacent branches; then import the STL file into the reverse engineering surface processing software for surface repair and adjustment, and finally import it into the 3D modeling software to generate facets and apply the coronary artery wall thickness according to the patient's image data to obtain the coronary artery wall model. The computational domain is divided into two parts: the solid domain and the fluid domain. Coronary artery health risk: The obtained coronary artery wall is directly used as the risk solid domain, and its volume is extracted to obtain the risk fluid domain; Stenotic plaque coronary artery: The length and width of the stenotic segment are measured in CTA images and used as the major and minor axes of an ellipse. An ellipsoid is obtained by rotation, and an ellipsoidal plaque entity is constructed. The location and extent of the plaque are determined based on the images, and the plaque entity is merged into the stenotic location of the coronary artery. The plaque and the coronary artery wall are regarded as the stenotic solid domain, and the corresponding stenotic fluid domain is obtained by volume extraction in 3D modeling software.

3. The working method of the coronary artery lesion assessment and interventional simulation system based on fluid-structure interaction according to claim 1, characterized in that, Step S4 specifically includes the following sub-steps: S4.1 Selecting the stent model: By examining the location and length of the stenotic area using CTA image data, and measuring the diameter of the healthy coronary arteries at both ends of the stenotic segment, the length and diameter of the stent are initially determined; the length of the stent is fine-tuned based on the abnormal flow field and stress areas in the coronary artery coupling simulation results to ensure complete coverage of the abnormal areas; the material of the stent is then selected based on the stress magnitude and curvature of the stenotic coronary artery; thus, a pre-selected stent is obtained. S4.2 After adding the pre-selected stent to the coronary artery, perform fluid-structure interaction simulation to remove the stenotic segment, establish a complete three-dimensional model of the pre-selected stent, and bend the stent to the same curvature as the coronary artery; determine the implantation location, obtain the pre-selected stent intervention model through virtual implantation, and extract the volume to obtain the fluid domain; S4.3 Import the data into the numerical simulation software, set specific boundary conditions in the fluid domain, input the elastic modulus and Poisson's ratio of the support, apply them to the solid domain, solve the continuity equation and Navier-Stokes equation for the fluid domain, solve the elastoplastic equation for the solid domain, and realize data transfer through the preset fluid-solid interface in each time step to perform fluid-structure interaction simulation.

4. A coronary artery lesion assessment and interventional simulation system based on fluid-structure interaction, characterized in that, The system includes an image acquisition and 3D modeling module, a fluid-structure interaction simulation configuration module, a healthy coronary artery risk identification module, a stenotic coronary artery lesion assessment module, and an interventional effect simulation assessment module. The image acquisition and 3D modeling module is used to acquire patient coronary CTA or MRI image data, segment and identify coronary arteries, determine the health status of blood vessels and plaque conditions, and reconstruct a patient-specific 3D geometric model of the coronary artery based on the image data, while generating a solid domain coronary artery wall model and a fluid domain model. The fluid-structure interaction simulation configuration module is used to measure specific flow velocities and pressures in the patient's coronary arteries as boundary conditions, configure the non-Newtonian fluid properties of blood, the hyperelastic material model of the vessel wall and plaque, divide the fluid domain and solid domain meshes, and complete the pre-setup for bidirectional fluid-structure interaction numerical simulation. The healthy coronary artery risk identification module is used to perform bidirectional fluid-structure interaction simulation on healthy coronary arteries, extract data such as blood flow field, vessel wall stress parameters, VMS stress and vessel wall pulsation morphology, identify potential risk areas by combining hemodynamic indicators and vessel morphology, and quantitatively evaluate parameters and visualize dangerous areas. The stenotic coronary artery lesion assessment module is used to perform bidirectional fluid-structure interaction simulation of stenotic coronary arteries, based on the flow field simulation results, stress and deformation of solid domain plaques, and related indicators. The interventional effect simulation evaluation module is used to establish a three-dimensional model of the selected stent and implant it into the coronary artery model to perform fluid-structure interaction simulation, and compare the parameters of the stenosis model with the healthy / stenosis model. The above modules work together to implement the working method described in claims 1-3.

Images