LVAD multi-objective optimization method and system, and device, storage medium and product

Abstract

Disclosed in the present invention are an LVAD multi-objective optimization method and system, and a device, a storage medium and a product. The method comprises: constructing a three-dimensional model of the ventricles and aorta, and constructing a ventricular motion function; on the basis of each group of sample parameters, adding an LVAD structure to the three-dimensional model of the ventricles and aorta, so as to obtain a plurality of three-dimensional models of the ventricles and aorta including the LVAD structures; performing meshing on the three-dimensional models of the ventricles and aorta including the LVAD structures, so as to generate a mesh model; on the basis of the ventricular motion function, controlling ventricular contraction and rotational motions, so as to calculate platelet shear stress histories and platelet residence times; on the basis of the platelet shear stress history and platelet residence time under each group of sample parameters, constructing a Kriging model; and using an optimization algorithm to find the optimal group of structural parameters from prediction results of the Kriging model. The present invention uses a plurality of structural parameters as optimization variables, thereby effectively solving the problem of it not being possible to effectively reduce thrombosis after LVAD implantation.

Description

LVAD multi-objective optimization method, system, device, storage medium and product TECHNICAL FIELD

[0001] The present application belongs to the field of biomedical engineering, and particularly relates to a LVAD cannula structure and multi-objective optimization method, system, device, storage medium and product based on parameterized modeling. BACKGROUND

[0002] Heart failure is a common cardiovascular disease that affects the quality of life and life expectancy of millions of people. Left ventricular assisted device (LVAD) is a mechanical circulatory support device that can be used to treat end-stage heart failure patients and improve their survival rate and quality of life. However, LVAD still has some serious complications, such as stroke and device thrombosis, which limit its clinical effectiveness and safety. LVAD-related complications are related to a variety of risk factors, including patient factors, clinical factors, and hemodynamic factors. Hemodynamic factors mainly include adverse blood flow patterns, high shear stress, stagnant areas, and non-physiological inflow and outflow. Platelets are the main cellular components of thrombosis, which can sense abnormal hemodynamic stimuli and initiate the coagulation cascade. Structural parameters that directly affect LVAD hemodynamics include LVAD insertion position, insertion depth, insertion angle, cannula diameter, etc. There is currently no clear medical guidelines or recommendations for the above parameters.

[0003] Thrombosis is closely related to platelet shear stress history (SH) and platelet residence time (RT). Platelet shear stress history SH is equal to the integral of shear force experienced by platelets over time, which can evaluate the level of shear-induced platelet activation related to ventricular size; Platelet residence time RT refers to the time that platelets stay in the left ventricular area, and higher platelet shear stress history SH and longer platelet residence time RT will increase the risk of thrombosis. With the help of computed tomography (CT) images, some researchers have carried out ventricular reconstruction and non-steady-state computational fluid dynamics (CFD) of intraventricular blood flow patterns, and studied the influence of LVAD insertion position, insertion depth, insertion angle, cannula diameter, etc. on left ventricular blood flow pattern and platelet activation. However, research on a single factor cannot give the optimal solution combination for a specific target, such as the optimal solution combination of LVAD insertion position, insertion depth, insertion angle, and insertion diameter corresponding to the minimum platelet shear stress history SH or platelet residence time RT.

[0004] In the LVAD device, parameters related to the thrombus risk include: LVAD insertion position, insertion depth, insertion angle, cannula diameter, platelet shear stress history, platelet residence time, which can be quantified. The patent with the authorization announcement number CN114098692B discloses a left ventricular assist device implantation method based on blood flow distribution optimization, which specifically comprises: calculating the patient's cardiac output; by reconstructing the aorta and coronary vessels, the physiological parameters of the patient's heart are obtained; using the blood flow and resistance of the normal population, the blood flow velocity of the normal population is calculated; simulate the installation LAVD process, import the physiological parameters of the patient's heart, calculate the blood flow velocity of the patient; according to the error function, the optimal parameter value is obtained. After obtaining the optimal parameter value according to the above method, the optimal parameter value can be used to guide the implantation of LAVD, which can optimize the blood supply of each blood vessel after LAVD implantation, reduce the formation of vortex, and further reduce the formation of thrombus, thereby reducing the formation probability of complications such as pulmonary embolism and cerebral infarction. However, the method still has the following problems:

[0005] (1) It cannot automatically realize parameterized modeling of left ventricle and different LVAD input pipe shapes, insertion position and angle, and automatic mesh division; (2) It cannot realize simulation of ventricular motion, so it cannot consider the influence of blood flow rotation flow in the ventricle and the aorta on thrombus formation; (3) LVAD insertion position, insertion depth, insertion angle, and cannula diameter are important factors affecting thrombus, and the above method does not involve optimizing the above variables; (4) The above method predicts the risk of thrombus by analyzing the change of flow loss, however, the change of flow loss has little correlation with the risk of thrombus, and the main factors causing thrombus are platelet shear stress history (SH) and platelet residence time (RT), and the above method does not optimize the related parameters. SUMMARY

[0006] The purpose of the present application is to provide an LVAD multi-objective optimization method, system, device, storage medium and product, to solve the problem that the prior art does not optimize multiple important factors affecting thrombus at the same time, resulting in ineffective reduction of thrombus formation after LVAD implantation.

[0007] The present application solves the above technical problems by the following technical solutions: an LVAD multi-objective optimization method, comprising the following steps:

[0008] constructing a ventricular and aortic three-dimensional model according to a cardiac CT image;

[0009] constructing a ventricular motion function based on the ventricular and aortic three-dimensional model;

[0010] select different groups of sample parameters from the value range of the LVAD structure parameters, add the corresponding LVAD structure in the ventricle and aorta three-dimensional model according to each group of sample parameters, obtain the ventricle and aorta three-dimensional model containing the corresponding LVAD structure, and further obtain a plurality of ventricle and aorta three-dimensional models containing the corresponding LVAD structure;

[0011] grid division is performed on each ventricle and aorta three-dimensional model containing the LVAD structure to generate a corresponding network model;

[0012] based on the ventricular motion function, the ventricular contraction motion is controlled by controlling the motion of the ventricular wall surface node, and an angular velocity is applied to the ventricular wall surface to control the rotation motion of the ventricular wall surface, and the platelet shear stress history and the platelet residence time under each group of sample parameters are calculated;

[0013] a Kriging model is constructed according to the platelet shear stress history and the platelet residence time under different groups of sample parameters;

[0014] different groups of structure parameters of the LVAD structure are input into the Kriging model to obtain the platelet shear stress history and the platelet residence time under different groups of structure parameters;

[0015] An optimization algorithm is used to find the optimal group of structure parameters from the platelet shear stress history and the platelet residence time under different groups of structure parameters.

[0016] The present application takes the insertion position, insertion depth, insertion angle, and LVAD structure diameter as optimization variables, takes the platelet shear stress history and the platelet residence time as target parameters, combines the Kriging model with the optimization algorithm, and constructs a multi-objective optimization method of LVAD shape and position parameters with the lowest thrombosis risk of the LVAD, which can solve the optimal combination of the insertion position, insertion depth, insertion angle and structure diameter of the LVAD with the minimum thrombosis risk, provide guidance for preoperative planning of patients, and optimize the biocompatibility of LVAD treatment, effectively solving the problem of unable to effectively reduce thrombosis after LVAD implantation. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only one embodiment of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0018] Fig. 1 is a flow chart of the LVAD multi-objective optimization method in the embodiment of the present application;

[0019] Fig. 2 is a ventricle and aorta three-dimensional model diagram in the embodiment of the present application;

[0020] Fig. 3 is a flowchart of adding an LVAD structure in a three-dimensional model of a ventricle and an aorta in an embodiment of the present application;

[0021] Fig. 4 is a diagram of defining a position of an LVAD structure in an embodiment of the present application;

[0022] Fig. 5 is a diagram of angle positioning of an LVAD structure in an embodiment of the present application;

[0023] Fig. 6 is a diagram of meshing in an embodiment of the present application;

[0024] Fig. 7 is a flowchart of calculating a platelet shear stress history and a platelet residence time under each set of sample parameters in an embodiment of the present application;

[0025] Fig. 8 is a flowchart of constructing a Kriging model in an embodiment of the present application. DETAILED DESCRIPTION

[0026] The technical solutions in the present application will be described in detail below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0027] The technical solutions of the present application will be described in detail below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0028] EMBODIMENT

[0029] As shown in Fig. 1, the LVAD multi-objective optimization method provided in an embodiment of the present application includes the following steps:

[0030] Step S1: Constructing a three-dimensional model of a ventricle and an aorta according to a heart CT image.

[0031] Based on a computed tomography image of a patient's heart, a three-dimensional model of a ventricle and an aorta is constructed by a medical modeling software E3D, as shown in Fig. 2, and the three-dimensional model of the ventricle and the aorta is saved as an stl format file.

[0032] Step S2: Constructing a ventricle motion function based on the three-dimensional model of the ventricle and the aorta.

[0033] In the specific embodiment of the present application, the specific process of constructing the ventricular motion function is as follows: the ventricle is divided into 24 sections by two-dimensional speckle tracking echocardiography, a plurality of speckles are selected on each ventricular section, the motion trajectories of the speckles are obtained by tracking the changes of the speckle positions over time, the real-time motion and deformation of the myocardial tissue are reconstructed, and then the volume change amount (i.e. the contraction motion) and the angular velocity of the ventricle are obtained.

[0034] The contraction motion of the ventricle is generally considered to be approximately sinusoidal motion, so the ventricular motion function can be fitted according to the volume change amount of the ventricle. The ventricular motion function includes a ventricular contraction function and a ventricular expansion function, and the expression of the ventricular contraction function is as follows:

[0035] (1)

[0036] The expression of the ventricular expansion function is as follows:

[0037] (2)

[0038] wherein M(t) represents the volume change amount of the ventricle, i.e. the contraction amount or the expansion amount; t represents time; a represents a correction coefficient; and π represents a circular constant.

[0039] The ventricular contraction motion is realized by the ventricular contraction function, and at the same time, a rotational angular velocity is applied to the ventricular wall surface to realize the rotational motion or circumferential motion (Global Circumferential strain, GCS) of the ventricular wall surface, as shown in FIG. 1.

[0040] Step S3: Different groups of sample parameters are selected from the value range of the LVAD structure parameters, the corresponding LVAD structure is added to the ventricle and the aorta three-dimensional model according to each group of sample parameters, the ventricle and the aorta three-dimensional model containing the corresponding LVAD structure are obtained, and then a plurality of ventricle and aorta three-dimensional models containing the corresponding LVAD structure are obtained.

[0041] In the embodiment, the LVAD structure parameters include insertion position, insertion depth, insertion angle and insertion diameter, the value range of each structure parameter is determined according to medical guidelines, then in the Ansys Workbench software, different groups of sample parameters are selected from the value range of the LVAD structure parameters by using the Latin hypercube sampling method, each group of sample parameters includes specific values of the insertion position, the insertion depth, the insertion angle and the insertion diameter. One group of sample parameters corresponds to one LVAD structure, different groups of sample parameters correspond to different LVAD structures, and the number of groups of sample parameters selected by the Latin hypercube sampling method is the same as the number of LVAD structures. An LVAD structure is added to the three-dimensional model of the ventricle and the aorta, and then a three-dimensional model of the ventricle and the aorta containing the LVAD structure is obtained. Changing the sample parameters, another three-dimensional model of the ventricle and the aorta containing the LVAD structure is obtained, and then three-dimensional models of the ventricle and the aorta containing different LVAD structures are obtained.

[0042] In the specific embodiment of the application, as shown in FIG. 3, according to each group of sample parameters, the corresponding LVAD structure is added to the three-dimensional model of the ventricle and the aorta, and a three-dimensional model of the ventricle and the aorta containing the corresponding LVAD structure is obtained, including:

[0043] Step S3.1: The three-dimensional model of the ventricle and the aorta is exported from the medical modeling software E3D, and then the three-dimensional model of the ventricle and the aorta is imported into the UG NX software (Unigraphics NX).

[0044] Step S3.2: The surface of the ventricle is extracted from the three-dimensional model of the ventricle and the aorta, and the cannula region is divided at the apex.

[0045] Step S3.3: The reference point of the change of the LVAD structure cross-section primitive is determined, and a local polar coordinate system is constructed with the apex as the origin.

[0046] As shown in Figure 4, a local polar coordinate system (m, r) is constructed with the apex as the origin, and the center point f of the cross section of any LVAD structure on the ventricular surface can be represented by the local polar coordinate system as (mf, rf). The parametric design of the spatial relationship of the LVAD structure under the local polar coordinate system is shown in Figure 5. The three-dimensional coordinate system in Figure 5 has the apex as the origin, the direction perpendicular to the outer surface of the ventricle inward as the X axis, the tangential direction at any point on the outer surface of the ventricle as the Y axis, and the Y axis perpendicular to the central axis of the LVAD structure. According to the right-hand rule, the Z axis is determined, the incidence angle of the LVAD structure is θ (the angle between the LVAD structure and the Y direction), and the complex inclination angle is α (i.e. the angle between the LVAD structure and the Z direction). In order to realize the parametric modeling, the incidence angle θ and the complex inclination angle α under the three-dimensional coordinate system are converted to the local polar coordinate system to obtain the incidence angle θf and the complex inclination angle αf under the local polar coordinate system. Therefore, the geometric parameters of the LVAD structure can be represented as (mf, rf, θf, αf).

[0047] Step S3.4: Automatically construct an LVAD structure element in the cannula region according to each set of sample parameters.

[0048] Step S3.5: Set the complex inclination angle αf of the LVAD structure.

[0049] Step S3.6: Perform a Boolean operation on the LVAD structure and the imported three-dimensional model of the ventricle and aorta to obtain a three-dimensional model of the ventricle and aorta containing the LVAD structure.

[0050] Since the influence of the LVAD structure insertion segment on the thrombosis probability is not considered, the imported three-dimensional model of the ventricle and aorta is subjected to a Boolean operation with the LVAD structure drawn in UG NX software, so as to obtain a three-dimensional model of the ventricle and aorta containing the LVAD structure, and save it in stp universal format. In step S3.4, the sample parameters are changed, different LVAD structure elements are constructed, and different LVAD structures can be obtained.

[0051] The present application realizes the parametric modeling of the insertion position, insertion depth, insertion angle and insertion diameter of the LVAD structure through the secondary development function provided by UG NX software, and obtains a three-dimensional model of the ventricle and aorta containing the corresponding LVAD structure through Boolean operation.

[0052] Step S4: Meshing each three-dimensional model of the ventricle and aorta containing the LVAD structure to generate a corresponding network model.

[0053] Each three-dimensional model of the ventricle and aorta containing the LVAD structure is imported into ANSYS Workbench software for automatic meshing. Since the structure of the left ventricle and aorta is complex, unstructured meshing is adopted for the left ventricle and aorta containing the LVAD structure, and the meshing result is shown in Figure 6.

[0054] Step S5: Based on the ventricular motion function (formula (1) and formula (2)), the ventricular contraction motion is controlled by controlling the ventricular wall surface node motion, and at the same time, an angular velocity is applied to the ventricular wall surface to control the rotation motion of the ventricular wall surface, and the platelet shear stress history and the platelet residence time under each group of sample parameters are calculated.

[0055] Each mesh model is imported into the Ansys Fluent software, the ventricle is set as a motion area, the ventricular wall surface is set as a moving boundary, the ventricular motion function is constructed through a user-defined function, the ventricular contraction motion is controlled by controlling the ventricular wall surface node motion, and at the same time, an angular velocity is applied to the ventricular wall surface through the UDF tool (i.e. user-defined tool) to control the rotation motion of the ventricular wall surface, so that the contraction and rotation motion of the ventricle are realized, and the transportation and generation of each intermediate component in the thrombus generation process are simulated. During the contraction and rotation motion of the ventricle, the platelet shear stress history and the platelet residence time are calculated. A group of sample parameters corresponds to a network model, and a network model corresponds to a group of platelet shear stress history and platelet residence time, so that the platelet shear stress history and the platelet residence time under different groups of sample parameters are obtained.

[0056] The unsteady calculation method is used to calculate the blood flow in the ventricle. In one cardiac cycle, the ventricular systole period accounts for 0.3 seconds, and the ventricular diastole period accounts for 0.5 seconds. The contraction and rotation motion of the ventricle are defined through the user-defined function UDF, and the time step is uniformly set to 0.05. In the specific embodiment of the present application, as shown in FIG. 7, the platelet shear stress history and the platelet residence time under each group of sample parameters are calculated, including:

[0057] Step S5.1: Calculate the blood flow in the ventricle and the aorta at the current time step;

[0058] Step S5.2: Release the thrombus particles, and calculate the motion of the thrombus particles at the current time step;

[0059] Step S5.3: Move the network nodes in the arbitrary boundary and the fluid area through the DEFINE_GRID_MOTION macro in the user-defined function UDF; after the motion of the wall surface mesh nodes, the mesh is fairing to improve the mesh quality;

[0060] Step S5.4: Calculate the blood flow in the ventricle and the aorta and the motion of the thrombus particles at the next time step until all the thrombus particles leave the ventricle and the aorta;

[0061] Step S5.5: When the calculation is completed, the platelet shear stress history and the platelet residence time are counted.

[0062] The application realizes the whole process of automatic generation of sample parameters, parameterized modeling, mesh division, numerical calculation and optimization by using software and sampling methods, and greatly improves the calculation efficiency.

[0063] Step S6: constructing a Kriging model according to the platelet shear stress history and the platelet residence time under different groups of sample parameters.

[0064] The Kriging model is an unbiased estimation model for predicting the response of unknown test points from known test point information. The application can construct an unbiased estimation model of the response by using the numerical calculation results (i.e. the platelet shear stress history and the platelet residence time) of the known groups of sample parameters.

[0065] In the specific embodiment of the application, as shown in FIG. 8, the Kriging model is constructed according to the platelet shear stress history and the platelet residence time under different groups of sample parameters, including:

[0066] Step S6.1: constructing a sample data set according to the platelet shear stress history and the platelet residence time under different groups of sample parameters; wherein the sample data set includes multiple samples, each sample includes an input quantity and an output quantity, the input quantity is a group of sample parameters, and the output quantity is the platelet shear stress history and the platelet residence time under the group of sample parameters.

[0067] Before constructing the sample data set, each group of sample parameters is also normalized and abnormal values are removed.

[0068] Step S6.2: constructing the Kriging model, calculating the distance matrix between samples by using the Kriging model, and determining the spatial correlation between the input quantities.

[0069] According to the data distribution characteristics of the application, a polynomial Kriging model is selected.

[0070] Step S6.3: estimating the parameters of the Kriging model according to the spatial correlation between the input quantities.

[0071] Step S6.4: inputting the input quantity of the sample into the Kriging model, and outputting the result of the Kriging model.

[0072] Step S6.5: judging whether the Kriging model meets the accuracy requirement according to the output result of the Kriging model and the output quantity of the sample; if yes, outputting the Kriging model; if no, increasing the number of groups of sample parameters selected from the value range of the LVAD structure parameters, and returning to step S3 to execute steps S3-S5 to obtain the platelet shear stress history and the platelet residence time under more groups of sample parameters, so as to improve the Kriging model until the Kriging model meets the accuracy requirement.

[0073] The present embodiment first constructs an initial Kriging model according to the blood platelet shear stress history and the blood platelet residence time under 21 groups of sample parameters. The initial Kriging model does not meet the accuracy requirement, and therefore a more accurate Kriging model is obtained by using the blood platelet shear stress history and the blood platelet residence time under 61 groups of sample parameters. The blood flow motion, the blood platelet shear stress history and the blood platelet residence time in the ventricle can be obtained by using the Kriging model.

[0074] Step S7: input different groups of structure parameters of the LVAD structure into the Kriging model to obtain the blood platelet shear stress history and the blood platelet residence time under different groups of structure parameters.

[0075] After the Kriging model is constructed, different groups of structure parameters of the LVAD structure are obtained, the structure parameters are the same as the sample parameters, including the insertion position, the insertion depth, the insertion angle and the insertion diameter. Different groups of structure parameters are used as the input of the Kriging model to obtain the blood platelet shear stress history and the blood platelet residence time under different groups of structure parameters.

[0076] Step S8: an optimization algorithm is used to find the optimal group of structure parameters from the blood platelet shear stress history and the blood platelet residence time under different groups of structure parameters.

[0077] In this embodiment, the optimization algorithm is a multi-objective genetic algorithm, and the multi-objective genetic algorithm (reference: Zheng, A., Ma, H., Luo, X., et al. Multi-objective genetic algorithm based on Kriging [J]. Aviation Computing Technology, 2014(2).) is used to find the Pareto optimal solution set and the Pareto frontier in the blood platelet shear stress history and the blood platelet residence time under different groups of structure parameters output by the Kriging model, and finally one to multiple models are artificially obtained from the Pareto frontier as optimal models. The objectives of the multi-objective optimization problem are often of different dimensions and conflict with each other, and it is difficult to directly compare the objective values to determine the optimal solution as in the single-objective optimization problem. In this application, the Pareto optimal solution set is the set of all non-inferior solutions in the feasible range of the input quantity (i.e., different groups of structure parameters), and the Pareto optimal frontier is the set of blood platelet shear stress history and blood platelet residence time corresponding to the Pareto optimal solution set. The multi-objective genetic algorithm is a method for evaluating the quality of multi-objective algorithms, which can screen and retain non-dominated solutions by using fitness evaluation and selection mechanism, so as to find the Pareto optimal solution set and the Pareto frontier in the prediction results of the Kriging model, and finally obtain one to multiple models from the Pareto frontier as optimal models. Based on the CFD method, the optimal group structure parameters are calculated, the blood flow in the ventricle, the blood platelet shear stress history and the blood platelet residence time are analyzed, and an analysis report is given, which provides preoperative planning for LVAD implantation surgery for heart failure patients and optimizes the biocompatibility of LVAD treatment.

[0078] Embodiment

[0079] The LVAD multi-objective optimization system provided by the embodiment of the application comprises a first construction unit, a second construction unit, a third construction unit, a grid division unit, a first calculation unit, a fourth construction unit, a second calculation unit and an optimization unit.

[0080] The first construction unit is configured to construct a ventricle and aorta three-dimensional model according to a heart CT image; the second construction unit is configured to construct a ventricle motion function based on the ventricle and aorta three-dimensional model; the third construction unit is configured to select different groups of sample parameters from a value range of LVAD structure parameters, add corresponding LVAD structures in each group of sample parameters in the ventricle and aorta three-dimensional model to obtain a ventricle and aorta three-dimensional model containing corresponding LVAD structures, and further obtain a plurality of ventricle and aorta three-dimensional models containing corresponding LVAD structures; the grid division unit is configured to perform grid division on each ventricle and aorta three-dimensional model containing an LVAD structure to generate a corresponding network model; the first calculation unit is configured to control ventricle contraction motion by controlling ventricle wall surface node motion based on the ventricle motion function, and control the rotation motion of the ventricle wall surface by applying an angular velocity to the ventricle wall surface to calculate the platelet shear stress history and the platelet residence time under each group of sample parameters; the fourth construction unit is configured to construct a Kriging model according to the platelet shear stress history and the platelet residence time under different groups of sample parameters; the second calculation unit is configured to input different groups of structure parameters of the LVAD structure into the Kriging model to obtain the platelet shear stress history and the platelet residence time under different groups of structure parameters; and the optimization unit is configured to find the optimal group of structure parameters from the platelet shear stress history and the platelet residence time under different groups of structure parameters by using an optimization algorithm.

[0081] In some embodiments, the LVAD multi-objective optimization system can incorporate the method features of the LVAD multi-objective optimization method of any embodiment, and vice versa, which will not be repeated here.

[0082] Embodiments

[0083] The embodiments of the present application also provide an electronic device, which comprises a memory, a processor and a computer program / instruction stored in the memory, and the processor executes the computer program / instruction to implement the LVAD multi-objective optimization method in the embodiments of the present application.

[0084] Although not shown, the electronic device includes a processor, which can perform various appropriate operations and processes according to programs and / or data stored in a read-only memory (ROM) or programs and / or data loaded from a storage section into a random access memory (RAM). The processor can be a multi-core processor or can include multiple processors. In some embodiments, the processor can include a general-purpose main processor and one or more special-purpose coprocessors, such as a central processing unit, a graphics processing unit (GPU), a neural network processing unit (NPU), a digital signal processor (DSP), etc. In the RAM, there are also stored

[0085] Various programs and data required for operation of the device are stored. The processor, the ROM, and the RAM are connected to each other through a bus. An input / output (I / O) interface is also connected to the bus.

[0086] The above processor and memory are used together to execute programs / instructions stored in the memory, which, when executed by a computer, can implement the methods, steps, or functions described in the above embodiments.

[0087] Although not shown, the embodiments of the present application also provide a computer-readable storage medium having a computer program / instruction stored thereon, which, when executed by a processor, implements the LVAD multi-objective optimization method in the embodiments of the present application.

[0088] The storage medium of the embodiments of the present application includes permanent and non-permanent, removable and non-removable, and information storage can be realized by any method or technology. Examples of storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device.

[0089] The readable storage medium includes permanent and non-permanent, removable and non-removable media, and information storage can be realized by any method or technology. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. According to the definition herein, computer-readable media does not include transitory media such as modulated data signals and carriers.

[0090] Although not shown, the embodiments of the present application also provide a computer program product, comprising: computer programs / instructions, which, when executed by a processor, implement the LVAD multi-objective optimization method in the embodiments of the present application.

[0091] The above only describes the specific embodiments of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of changes or modifications within the technical range disclosed by the present application, which should be covered within the protection scope of the present application.

Claims

1. A LVAD multi-objective optimization method, characterized in that, The optimization method comprises the following steps: construct a three-dimensional model of a ventricle and an aorta according to a cardiac CT image; construct a ventricular motion function based on the three-dimensional model of the ventricle and the aorta; select different sets of sample parameters from a value range of LVAD structure parameters, add corresponding LVAD structures in the three-dimensional model of the ventricle and the aorta according to each set of sample parameters, obtain a three-dimensional model of the ventricle and the aorta containing corresponding LVAD structures, and further obtain a plurality of three-dimensional models of the ventricle and the aorta containing corresponding LVAD structures; divide a grid for each three-dimensional model of the ventricle and the aorta containing the LVAD structure to generate a corresponding network model; control the ventricular contraction motion by controlling the motion of the ventricular wall surface nodes based on the ventricular motion function, and control the rotation motion of the ventricular wall surface by applying an angular velocity to the ventricular wall surface to calculate the platelet shear stress history and the platelet residence time under each set of sample parameters; construct a Kriging model according to the platelet shear stress history and the platelet residence time under different sets of sample parameters; input different sets of structure parameters of the LVAD structure into the Kriging model to obtain the platelet shear stress history and the platelet residence time under different sets of structure parameters; find the optimal set of structure parameters from the platelet shear stress history and the platelet residence time under different sets of structure parameters by using an optimization algorithm.

2. The LVAD multi-objective optimization method of claim 1, wherein, construct a ventricular motion function based on the three-dimensional model of the ventricle and the aorta, comprising: track the change of a plurality of spot positions of different ventricular sections in the three-dimensional model of the ventricle and the aorta over time by using a two-dimensional speckle tracking echocardiography technology, and calculate the volume change of the ventricle and the angular velocity of the rotation of the ventricle; fit the ventricular motion function according to the volume change of the ventricle; wherein the ventricular motion function comprises a ventricular contraction function and a ventricular expansion function, and the expression of the ventricular contraction function is: ； the expression of the ventricular expansion function is: ； wherein M(t) represents the volume change of the ventricle, i.e. the contraction or expansion amount; t represents time; a represents a correction coefficient; and π represents a circular constant.

3. The LVAD multi-objective optimization method of claim 1, wherein, add corresponding LVAD structures in the three-dimensional model of the ventricle and the aorta according to each set of sample parameters, comprising: import the three-dimensional model of the ventricle and the aorta into UG NX software; extract the ventricular surface from the three-dimensional model of the ventricle and the aorta, and divide a cannula region at the apex; determine the change reference point of the LVAD structure section element, and construct a local polar coordinate system with the apex as the origin; construct an LVAD structure element in the cannula region according to each set of sample parameters; set the composite inclination angle of the LVAD structure; wherein the composite inclination angle refers to the included angle between the LVAD structure and the Z direction, and the Z direction refers to the spanwise direction; perform a Boolean operation on the LVAD structure and the imported three-dimensional model of the ventricle and the aorta to obtain a three-dimensional model of the ventricle and the aorta containing the LVAD structure.

4. The LVAD multi-objective optimization method of claim 1, wherein, calculate the platelet shear stress history and the platelet residence time under each set of sample parameters, comprising: calculate the blood flow in the ventricle and the aorta at the current time step; release thrombus particles and calculate the motion of the thrombus particles at the current time step; Mobile nodes in the ventricular wall surface and fluid region; Calculate blood flow and thrombus particle motion in the ventricle and aorta at the next time step until all thrombus particles leave the ventricle and aorta; Statistics of platelet shear stress history and platelet residence time.

5. The LVAD multi-objective optimization method of any one of claims 1-4, wherein, According to the platelet shear stress history and the platelet residence time under different groups of sample parameters, a Kriging model is constructed, including: According to the platelet shear stress history and the platelet residence time under different groups of sample parameters, a sample data set is constructed; wherein the sample data set includes multiple samples, each sample includes an input quantity and an output quantity, the input quantity is a group of sample parameters, and the output quantity is the platelet shear stress history and the platelet residence time under the group of sample parameters; A Kriging model is constructed, and the distance matrix between samples is calculated using the Kriging model to determine the spatial correlation between input quantities; According to the spatial correlation between the input quantities, the parameters of the Kriging model are estimated; The input quantity of the sample is input into the Kriging model, and whether the Kriging model meets the accuracy requirement is judged according to the output result of the Kriging model and the output quantity of the sample; if yes, output the Kriging model; if not, increase the number of groups of sample parameters selected from the value range of LVAD structure parameters until the Kriging model meets the accuracy requirement.

6. The LVAD multi-objective optimization method of claim 1, wherein, The optimization algorithm is a multi-objective genetic algorithm.

7. An LVAD multi-objective optimization system, characterized by, The system comprises: A first construction unit for constructing a ventricular and aortic three-dimensional model according to a cardiac CT image; A second construction unit for constructing a ventricular motion function based on the ventricular and aortic three-dimensional model; A third construction unit for selecting different groups of sample parameters from the value range of LVAD structure parameters, and adding corresponding LVAD structures in the ventricular and aortic three-dimensional model according to each group of sample parameters to obtain a ventricular and aortic three-dimensional model containing corresponding LVAD structures, and then obtain multiple ventricular and aortic three-dimensional models containing corresponding LVAD structures; A mesh division unit for mesh division of each ventricular and aortic three-dimensional model containing LVAD structures to generate a corresponding network model; A first calculation unit for controlling the ventricular contraction motion by controlling the motion of the ventricular wall surface node based on the ventricular motion function, and controlling the rotation motion of the ventricular wall surface by applying an angular velocity to the ventricular wall surface to calculate the platelet shear stress history and the platelet residence time under each group of sample parameters; A fourth construction unit for constructing a Kriging model according to the platelet shear stress history and the platelet residence time under different groups of sample parameters; A second calculation unit for inputting different groups of structure parameters of the LVAD structure into the Kriging model to obtain the platelet shear stress history and the platelet residence time under different groups of structure parameters; An optimization unit for finding the optimal group of structure parameters from the platelet shear stress history and the platelet residence time under different groups of structure parameters by using an optimization algorithm.

8. An electronic device comprising a memory, a processor, and a computer program / instructions stored on the memory, characterized in that, The processor executes the computer program / instructions to implement the LVAD multi-objective optimization method of any one of claims 1-6.

9. A computer readable storage medium having stored thereon computer programs / instructions, characterized in that, The computer program / instructions, when executed by the processor, implement the LVAD multi-objective optimization method of any one of claims 1-6.

10. A computer program product comprising computer programs / instructions, characterized in that, The computer program / instructions, when executed by the processor, implement the LVAD multi-objective optimization method of any one of claims 1-6.

Images