Aneurysm rupture risk prediction device, model training device and storage medium

Abstract

The embodiment of the invention discloses an aneurysm rupture risk prediction device, a model training device and a storage medium. The prediction device comprises a first three-dimensional model construction module used for constructing a three-dimensional fluid model and a three-dimensional solid model based on target artery three-dimensional imaging data; the first fluid-solid coupling simulation model building module is used for building a computational fluid dynamics simulation module based on a three-dimensional fluid model, building a static solid analysis simulation module based on a three-dimensional solid model and then building a one-way fluid-solid coupling simulation model; the first index parameter acquisition module is used for taking the blood flow velocity distribution in the target arterial blood vessel in space and time as the input of the fluid-solid coupling simulation model to obtain a hemodynamic index parameter and a solid mechanics index parameter; and the risk prediction module is used for predicting the rupture risk of the aneurysm based on the hemodynamic index parameters and the solid mechanics index parameters. According to the technical scheme, the prediction accuracy of the aneurysm rupture risk can be improved.

Description

Technical Field

[0001] The present invention relates to the field of medical technology, and in particular to an aneurysm rupture risk prediction device, an aneurysm rupture risk prediction model training device and a storage medium. Background Art

[0002] An aneurysm is a bulge or weak spot on the wall of a blood vessel. When an aneurysm ruptures, it causes bleeding in the cavity. In particular, once a cerebral aneurysm ruptures and causes intracranial bleeding, it may cause a stroke or even death. Currently, for cerebral aneurysms, after using imaging techniques such as magnetic resonance imaging (MRI) or computed tomography (CT) scanning to find the target aneurysm, hemodynamic analysis such as computational fluid dynamics (CFD) is usually used to evaluate the blood flow in the cerebral aneurysm and provide simulation results to help doctors make clinical decisions.

[0003] In addition, technicians in this field are also working on finding other solutions for predicting the risk of aneurysm exceptions. Summary of the invention

[0004] In view of this, the embodiments of the present application propose, on the one hand, an aneurysm rupture risk prediction device, and on the other hand, propose a training device and computer-readable storage medium for an aneurysm rupture risk prediction model to improve the prediction accuracy and reliability of aneurysm rupture risk.

[0005] An aneurysm rupture risk prediction device proposed in an embodiment of the present application includes: a first three-dimensional model construction module, which is used to construct a three-dimensional fluid model and a three-dimensional solid model of the target artery of the current patient based on the three-dimensional imaging data of the target artery of the current patient; a first fluid-solid coupling simulation model construction module, which is used to construct a computational fluid dynamics simulation module of the current patient based on the three-dimensional fluid model of the target artery, construct a static solid analysis simulation module of the current patient based on the three-dimensional solid model of the target artery, and construct a unidirectional fluid-solid coupling simulation model of the current patient based on the computational fluid dynamics simulation module and the static solid analysis simulation module; a first indicator parameter acquisition module, which is used to use the pre-acquired spatial and temporal blood flow velocity distribution in the target artery of the current patient as the input of the fluid-solid coupling simulation model to obtain the hemodynamic indicator parameters output by the computational fluid dynamics simulation module and the solid mechanics indicator parameters output by the static solid analysis simulation module; and a risk prediction module, which is used to predict the rupture risk of the aneurysm of the current patient based on the hemodynamic indicator parameters and the solid mechanics indicator parameters.

[0006] In one embodiment, the device further includes: a first blood flow velocity distribution acquisition module, which is used to obtain the Fourier series fitting value of the target arterial blood flow rate corresponding to the current patient that changes with time, and substitute the Fourier series fitting value that changes with time into the blood flow velocity model constructed based on the Womersley boundary condition to obtain the blood flow velocity distribution in space and time in the target arterial blood vessel.

[0007] In one embodiment, when there is ultrasound video or phase contrast imaging data within the target artery region of the current patient, the first blood flow velocity distribution acquisition module obtains the Fourier series fitting value of the target artery blood flow rate varying with time corresponding to the current patient based on the historical ultrasound video or phase contrast imaging data within the target artery region of the current patient; when there is no ultrasound video or phase contrast imaging data within the target artery region of the current patient, the first blood flow velocity distribution acquisition module obtains the Fourier series fitting average value or standard value of the target artery blood flow rate varying with time based on the Fourier series fitting values ​​of the target artery blood flow rate varying with time of different patients, and uses the Fourier series fitting average value or standard value of the target artery blood flow rate varying with time as the Fourier series fitting value of the target artery blood flow rate varying with time corresponding to the current patient.

[0008] A training device for an aneurysm rupture risk prediction model proposed in an embodiment of the present application includes: a data collection module, which is used to obtain historical aneurysm data groups that meet quantity requirements, each historical aneurysm data group includes the patient's historical target artery three-dimensional imaging data and corresponding historical aneurysm rupture risk data, and further, may also include relevant patient information; a second three-dimensional model construction module, which is used to construct, for each historical aneurysm data group, a corresponding target artery three-dimensional fluid model and a target artery three-dimensional solid model based on the historical target artery three-dimensional imaging data in the historical aneurysm data group; a second fluid-solid coupling simulation model construction module, which is used to construct a corresponding computational fluid dynamics simulation module based on the target artery three-dimensional fluid model constructed by the second three-dimensional model construction module, and to construct a corresponding static solid analysis simulation module based on the computational fluid dynamics simulation module and the static The solid analysis simulation module constructs a unidirectional fluid-solid coupling simulation model; the second index parameter acquisition module is used to use the pre-acquired spatial and temporal blood flow velocity distribution in the target artery corresponding to the historical aneurysm data group as the input of the fluid-solid coupling simulation model constructed by the second fluid-solid coupling simulation model construction module, and obtain the hemodynamic index parameters output by the computational fluid dynamics simulation module and the solid mechanics index parameters output by the static solid analysis simulation module; the training sample determination module is used to use the hemodynamic index parameters, the solid mechanics index parameters and even the corresponding relevant patient information as input training samples, and use the historical aneurysm rupture risk data in the historical aneurysm data group as output training samples; and the model training module is used to use the input training samples and the output training samples corresponding to each historical aneurysm data group to train an artificial intelligence network capable of machine learning to obtain an aneurysm rupture risk prediction model.

[0009] In one embodiment, the device further includes: a second blood flow velocity distribution acquisition module, which is used to obtain the Fourier series fitting value of the target arterial blood flow volume corresponding to the historical aneurysm data group that changes with time, and substitute the Fourier series fitting value that changes with time into the blood flow velocity model constructed based on the Womersley boundary condition to obtain the blood flow velocity distribution in space and time in the target arterial blood vessel.

[0010] In one embodiment, when the historical aneurysm data group further includes ultrasound video or phase contrast imaging data within the historical target artery region, the second blood flow velocity distribution acquisition module obtains the Fourier series fitting value of the target artery blood flow rate changing with time corresponding to the historical aneurysm data group based on the ultrasound video or phase contrast imaging data within the historical target artery region; when the historical aneurysm data group does not include ultrasound video or phase contrast imaging data within the historical target artery region, the Fourier series fitting average value or standard value of the target artery blood flow rate changing with time is obtained based on the Fourier series fitting value of the target artery blood flow rate changing with time of different patients, and the Fourier series fitting average value or standard value of the target artery blood flow rate changing with time is used as the Fourier series fitting value of the target artery blood flow rate changing with time corresponding to the historical aneurysm data group.

[0011] Another aneurysm rupture risk prediction device proposed in an embodiment of the present application includes: at least one memory and at least one processor, wherein: the at least one memory is used to store computer programs; the at least one processor is used to call the computer program stored in the at least one memory to perform the following operations: obtain the three-dimensional imaging data of the target artery of the current patient, and construct a three-dimensional fluid model and a three-dimensional solid model of the target artery of the current patient based on the three-dimensional imaging data of the target artery; construct a computational fluid dynamics simulation module of the current patient based on the three-dimensional fluid model of the target artery, construct a static solid analysis simulation module of the current patient based on the three-dimensional solid model of the target artery, and construct a unidirectional fluid-solid coupling simulation model of the current patient based on the computational fluid dynamics simulation module and the static solid analysis simulation module; use the pre-acquired spatial and temporal blood flow velocity distribution in the target artery of the current patient as the input of the fluid-solid coupling simulation model to obtain the hemodynamic index parameters output by the computational fluid dynamics simulation module and the solid mechanics index parameters output by the static solid analysis simulation module; predict the rupture risk of the aneurysm of the current patient based on the hemodynamic index parameters and the solid mechanics index parameters.

[0012] Another training device for an aneurysm rupture risk prediction model proposed in an embodiment of the present application includes: at least one memory and at least one processor, wherein: the at least one memory is used to store a computer program; the at least one processor is used to call the computer program stored in the at least one memory to perform the following operations:

[0013] Acquire historical aneurysm data sets that meet quantity requirements, each historical aneurysm data set including relevant patient information, historical target artery three-dimensional imaging data of the patient, and corresponding historical aneurysm rupture risk data;

[0014] For each historical aneurysm data group, a corresponding three-dimensional fluid model of the target artery and a three-dimensional solid model of the target artery are constructed based on the three-dimensional imaging data of the historical target artery in the historical aneurysm data group; a corresponding computational fluid dynamics simulation module is constructed based on the three-dimensional fluid model of the target artery, a corresponding static solid analysis simulation module is constructed based on the three-dimensional solid model of the target artery, and a one-way fluid-solid coupling simulation model is constructed based on the computational fluid dynamics simulation module and the static solid analysis simulation module; the pre-acquired spatial and temporal blood flow velocity distribution in the target artery corresponding to the historical aneurysm data group is used as the input of the fluid-solid coupling simulation model to obtain the hemodynamic index parameters output by the computational fluid dynamics simulation module and the solid mechanics index parameters output by the static solid analysis simulation module; the hemodynamic index parameters, the solid mechanics index parameters and the corresponding relevant patient information are used as input training samples, and the historical aneurysm rupture risk data in the historical aneurysm data group is used as output training samples;

[0015] An artificial intelligence network is trained using the input training samples and the output training samples corresponding to each historical aneurysm data group to obtain an aneurysm rupture risk prediction model.

[0016] A computer-readable storage medium is proposed in an embodiment of the present application, on which a computer program is stored; the computer program can be executed by a processor and implement the following operations: obtain three-dimensional imaging data of the target artery of the current patient, and construct a three-dimensional fluid model and a three-dimensional solid model of the target artery of the current patient based on the three-dimensional imaging data of the target artery; construct a computational fluid dynamics simulation module of the current patient based on the three-dimensional fluid model of the target artery, construct a static solid analysis simulation module of the current patient based on the three-dimensional solid model of the target artery, and construct a unidirectional fluid-solid coupling simulation model of the current patient based on the computational fluid dynamics simulation module and the static solid analysis simulation module; use the pre-acquired spatial and temporal blood flow velocity distribution in the target artery of the current patient as the input of the fluid-solid coupling simulation model to obtain the hemodynamic index parameters output by the computational fluid dynamics simulation module and the solid mechanics index parameters output by the static solid analysis simulation module; predict the rupture risk of the aneurysm of the current patient based on the hemodynamic index parameters and the solid mechanics index parameters.

[0017] A computer-readable storage medium is provided in an embodiment of the present application, on which a computer program is stored; the computer program can be executed by a processor and implement the following operations:

[0018] Acquire historical aneurysm data sets that meet quantity requirements, each historical aneurysm data set including relevant patient information, historical target artery three-dimensional imaging data of the patient, and corresponding historical aneurysm rupture risk data;

[0019] For each historical aneurysm data group, a corresponding three-dimensional fluid model of the target artery and a three-dimensional solid model of the target artery are constructed based on the three-dimensional imaging data of the historical target artery in the historical aneurysm data group; a corresponding computational fluid dynamics simulation module is constructed based on the three-dimensional fluid model of the target artery, a corresponding static solid analysis simulation module is constructed based on the three-dimensional solid model of the target artery, and a one-way fluid-solid coupling simulation model is constructed based on the computational fluid dynamics simulation module and the static solid analysis simulation module; the pre-acquired spatial and temporal blood flow velocity distribution in the target artery corresponding to the historical aneurysm data group is used as the input of the fluid-solid coupling simulation model to obtain the hemodynamic index parameters output by the computational fluid dynamics simulation module and the solid mechanics index parameters output by the static solid analysis simulation module; the hemodynamic index parameters, the solid mechanics index parameters and the corresponding relevant patient information are used as input training samples, and the historical aneurysm rupture risk data in the historical aneurysm data group is used as output training samples;

[0020] An artificial intelligence network is trained using the input training samples and the output training samples corresponding to each historical aneurysm data group to obtain an aneurysm rupture risk prediction model.

[0021] It can be seen from the above scheme that in the embodiment of the present application, since the hemodynamic index parameters and solid mechanics index parameters of the artery are taken into consideration at the same time, the accuracy and reliability of the prediction of the rupture risk of the aneurysm are greatly improved.

[0022] In addition, by combining the pre-trained aneurysm rupture risk prediction model, the prediction results can be made more intuitive and seamlessly integrated into the medical field. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings so that those skilled in the art can better understand the above and other features and advantages of the present invention. In the accompanying drawings:

[0024] Figure 1A is a schematic diagram of a bidirectional FSI in an embodiment of the present application, Figure 1B Schematic diagram of unidirectional FSI in an embodiment of the present application.

[0025] Figure 2 This is an exemplary flow chart of the method for predicting the risk of aneurysm rupture in an embodiment of the present application.

[0026] Figure 3A schematic diagram of Fourier series fitting in an example of this application.

[0027] Figure 4 Schematic diagram of velocity inlet boundary conditions for different Wo values ​​in an example of this application.

[0028] Figure 5 This is an exemplary flowchart of a method for training an aneurysm rupture risk prediction model in an embodiment of the present application.

[0029] Figure 6 This is an exemplary structural diagram of an aneurysm rupture risk prediction device in an embodiment of the present application.

[0030] Figure 7 This is an exemplary structural diagram of a training device for an aneurysm rupture risk prediction model in an embodiment of the present application.

[0031] Figure 8 Schematic diagram of the structure of another aneurysm rupture risk prediction device in an embodiment of the present application.

[0032] Fig. 9 Schematic diagram of the structure of another training device for an aneurysm rupture risk prediction model in an embodiment of the present application.

[0033] The reference numerals are as follows:

[0034]

[0035] DETAILED DESCRIPTION

[0036] In the embodiments of the present application, considering that only CFD is currently used to evaluate the blood flow in the target aneurysm, it may not be possible to capture all relevant factors, which will result in relatively low accuracy and reliability of its prediction. In addition, since the simulation results given by CFD based on the blood flow in the target aneurysm are not intuitive risk prediction results, doctors usually need to learn related knowledge to understand them, and cannot be seamlessly integrated into clinical applications.

[0037] In addition, considering that arteries such as cerebral arteries may deform under the action of the flow field, in order to capture as many relevant factors as possible, in the embodiments of the present application, in addition to using CFD, solid mechanics is also considered to be included in the risk prediction analysis of aneurysm rupture, that is, in the embodiments of the present application, two simulation modules, CFD and static solid analysis (SSA, Static Solid Analysis), are considered to be used simultaneously. Considering the interaction between blood flow movement and vascular deformation, fluid-structure interaction (FSI, Fluid-Structure Interaction) mechanics can be used to analyze arterial blood flow. In one embodiment, the SSA simulation module can be implemented by a finite element analysis (FEA, Finite Element Analysis) module.

[0038] However, considering that fluid-structure interaction is mainly used in engineering fields such as wing flutter, building sway, valve flutter and damping in MEMS, but has never been applied in the vascular field, detailed analysis and experimental research are needed to consider its feasibility.

[0039] First, considering that the typical application of FSI in engineering applications is bidirectional FSI, and the accuracy of bidirectional FSI is higher. However, because FSI uses multiple time steps (t i , i=1,2,3,…). If it is bidirectional FSI, then Figure 1A As shown in the figure, in each time step, the information of the two simulation modules, CFD and SSA, will be transferred to each other as boundary conditions, that is, the two will be iteratively completed. Considering the incompressible blood flow in the CFD simulation module, the nonlinear vascular characteristics in the SSA simulation module, and the complex three-dimensional geometry of the target aneurysm in vascular applications, this iterative process will be very unstable and will consume a lot of computing resources.

[0040] Therefore, the inventors of the present application have found through experimental research that, unlike cardiovascular vessels where the vessel wall deformation is significant, the aneurysm wall deformation of target arteries such as cerebral arteries is not large. For example, the deformation of the target aneurysm wall is usually less than 2% compared to the target aneurysm diameter. Therefore, the effect of the vessel deformation in SSA on the flow field in CFD simulation is negligible, because the flow field is the main factor of the aneurysm wall deformation. Therefore, Figure 1B As shown, it is only necessary to pass the information of the CFD simulation module to the SSA simulation module, and the information of the SSA simulation module does not need to be passed to the CFD simulation module, that is, a one-way FSI simulation model can be constructed in the embodiment of the present application.

[0041] By configuring the fluid-structure interaction in FSI as unidirectional, the following beneficial effects can be obtained: 1) the number of iterations between the CFD and SSA simulation modules can be reduced, thereby greatly reducing the total running time of FSI; 2) since the CFD and SSA simulation modules can be executed sequentially, the hardware requirements can be reduced; 3) using an automated meshing process to generate a conformal mesh for the fluid-structure interaction interface, reducing simulation errors.

[0042] Based on the one-way FSI simulation model, the fluid-solid coupling analysis of arterial blood flow, such as cerebral artery blood flow, the obtained hemodynamic index parameters and solid mechanics index parameters will be more accurate and reliable.

[0043] In addition, considering that directly judging the risk of aneurysm rupture based on hemodynamic index parameters and solid mechanics index parameters is more accurate and reliable than judging the risk of aneurysm rupture based only on hemodynamic index parameters of CFD, it is not an intuitive risk prediction result. Therefore, in order to provide a more intuitive risk prediction result, it is considered to pre-train an aneurysm rupture risk prediction model that can realize machine learning based on parameters such as hemodynamic index parameters and solid mechanics index parameters, and use the aneurysm rupture risk prediction model to directly output the corresponding aneurysm rupture risk prediction result.

[0044] Furthermore, considering that patient information such as different ages and genders may also affect the risk of aneurysm rupture, in the embodiments of the present application, demographic data may also be taken into consideration when training the aneurysm rupture risk prediction model, in order to provide a more comprehensive and accurate aneurysm rupture risk assessment.

[0045] In order to make the purpose, technical scheme and effect of the present invention more clearly understood, the specific implementation mode of the present invention is now described with reference to the accompanying drawings. In the drawings, the same reference numerals represent components with the same structure or similar structures but the same functions.

[0046] In this document, “exemplary” and “illustrative” mean “serving as an example, instance or illustration”, and any illustration or implementation described in this document as “exemplary” or “illustrative” should not be construed as a more preferred or more advantageous technical solution.

[0047] In order to simplify the drawings, each figure only schematically shows the parts related to the present invention, which do not represent the actual structure of the product.

[0048] In this article, "a" or "an" not only means "only one", but also means "more than one". In this article, "first", "second", etc. are only used to distinguish each other, rather than to indicate their importance and order.

[0049] Figure 2FIG. 1 is an exemplary flow chart of a method for predicting the risk of aneurysm rupture in an embodiment of the present application. Figure 2 As shown, the method may include the following steps:

[0050] Step S21 , obtaining the three-dimensional imaging data of the target artery of the current patient, and constructing a three-dimensional fluid model and a three-dimensional solid model of the target artery of the current patient based on the three-dimensional imaging data of the target artery.

[0051] In this step, the three-dimensional imaging data of the target artery may be a CT image, an MRI image, etc. In one embodiment, the three-dimensional imaging data of the target artery may be subjected to image processing such as image recognition and contour segmentation to obtain a three-dimensional fluid model and a three-dimensional solid model of the target artery. The fluid model refers to a blood flow model in the target artery, and the fixed model refers to a blood vessel model of the target artery.

[0052] In the embodiment of the present application, when constructing the three-dimensional solid model of the target artery, if the vascular wall thickness can be displayed in the three-dimensional imaging data of the target artery, for example, the vascular wall thickness can usually be displayed in the vascular wall MRI image, then the three-dimensional solid model can be constructed based on the vascular wall thickness in the three-dimensional imaging data. If the vascular wall thickness cannot be displayed in the three-dimensional imaging data of the target artery, a three-dimensional solid model of uniform thickness can be constructed based on the set thickness or the manually marked thickness.

[0053] Among them, using vascular wall MRI, the blood flow in the blood vessels and aneurysm areas will darken and the blood vessel walls will be highlighted, so that the thickness of blood vessels in different areas can be measured. This method is more significant when measuring the uneven thickness of the aneurysm wall, because the formation of aneurysms causes the proliferation and deformation of the blood vessel wall material. By measuring the uneven thickness of blood vessels and aneurysm walls, detailed modeling of the target aneurysm model can be obtained, which can not only improve the simulation accuracy, but also facilitate the study of the pathological causes of aneurysm formation, growth and rupture by integrating the different material properties of the proliferative tissue in the physical model.

[0054] Step S22, constructing a CFD simulation module of the current patient based on the three-dimensional fluid model of the target artery, constructing an SSA simulation module of the current patient based on the three-dimensional solid model of the target artery, and constructing a one-way FSI simulation model of the current patient based on the CFD simulation module and the SSA simulation module.

[0055] In this step, the corresponding CFD simulation module and SSA simulation module can be constructed by performing automatic meshing on the three-dimensional fluid model and the three-dimensional solid model, and adjusting the automatically divided meshes according to the pre-set mesh quality and quantity requirements. The automatic meshing procedure includes the following steps:

[0056] A. Automatically mesh the three-dimensional fluid model based on pre-selected parameters such as the maximum mesh size and the minimum mesh size to obtain the fluid domain mesh.

[0057] B. Automatically adjust and re-divide the fluid domain grid according to pre-defined grid quality (such as greater than 0.5) and quantity requirements (such as less than 2 million).

[0058] In this step B, after the mesh division is completed, the mesh quality and the total number of meshes of each mesh will be counted, and then the meshes with mesh quality lower than the minimum value will be adjusted and re-divided; when the total number of meshes exceeds the pre-selected value, the program will automatically increase the mesh size parameter and re-divide the mesh; when the quality of some meshes is lower than the minimum mesh quality value, the program will automatically reduce the mesh size parameter and re-divide the mesh; the above steps will be repeated until the minimum mesh quality value and the total number of meshes are met.

[0059] C. Divide the solid domain mesh of the three-dimensional solid model based on the boundary mesh of the fluid domain.

[0060] D. Automatically adjust and re-mesh the solid domain according to the pre-defined mesh quality (such as greater than 0.5).

[0061] The process in this step D is similar to the process of adjusting the mesh according to the mesh quality in the above step B.

[0062] Step S23, using the pre-acquired blood flow velocity distribution in space and time in the target artery of the current patient as the input of the FSI simulation model, to obtain the hemodynamic index parameters output by the CFD simulation module and the solid mechanics index parameters output by the SSA simulation module.

[0063] In one embodiment of this step, the spatial and temporal blood flow velocity distribution in the target artery of the current patient can be obtained by:

[0064] First, the blood velocity distribution in space and time in the target artery can be modeled based on the Womersley boundary condition. For example, for a three-dimensional fluid and solid model, the velocity inlet boundary condition of a blood vessel with an inlet radius of R can be defined as follows according to formula (1):

[0065]

[0066] Wherein, n is a natural number, ω is an angular frequency, and ν is the kinematic viscosity of blood. In practical applications, if the kinematic viscosity of the blood of the current patient can be measured, the kinematic viscosity of the blood of the current patient can be used, otherwise, the average value or standard value of the kinematic viscosity of the blood of different patients measured in advance can be used.

[0067] Based on this, the Womersley velocity boundary condition can be defined as follows according to equation (2):

[0068]

[0069] The Womersley velocity boundary condition shown in formula (2) can also be called the blood flow velocity model constructed based on the Womersley boundary condition. Where r is the radius corresponding to a certain position in the blood vessel, J0 and J1 are Bessel functions, Q0 and Q n is the Fourier series fitting value of blood flow changing with time.

[0070] In actual applications, when there is ultrasound video or phase contrast imaging data within the target artery region of the current patient, the Fourier series fitting value of the target artery blood flow rate changing with time corresponding to the current patient can be obtained based on the historical ultrasound video or phase contrast imaging data within the target artery region of the current patient; when there is no ultrasound video or phase contrast imaging data within the target artery region of the current patient, the Fourier series fitting average value or standard value of the target artery blood flow rate changing with time can be obtained based on the Fourier series fitting values ​​of the target artery blood flow rate changing with time of different patients obtained in advance, and the Fourier series fitting average value or standard value of the target artery blood flow rate changing with time is used as the Fourier series fitting value of the target artery blood flow rate changing with time corresponding to the current patient.

[0071] Figure 3 A schematic diagram of Fourier series fitting in an example is shown in FIG. Figure 3 In the figure, the horizontal axis is time (seconds, s) and the vertical axis is the maximum blood flow velocity (meters per second, m / s).

[0072] After obtaining the Fourier series fitting value of the target arterial blood flow rate corresponding to the current patient varying with time, the Fourier series fitting value varying with time is substituted into the blood flow velocity model constructed based on the Womersley boundary condition in formula (2), and the blood flow velocity distribution in space and time in the target artery can be obtained, that is, the velocity inlet boundary condition.

[0073] Figure 4 Graph 1 shows the velocity inlet boundary conditions for different Wo values ​​in an example of the present application. Figure 3 The horizontal axis in the middle is the speed u and the average speed for different Wo values Scaled Velocity, ), the ordinate is the relative radial position (Relative RadialPosition, r / R) between the radius r corresponding to a certain position in the blood vessel and the blood vessel radius R, such as Figure 3 As shown in the figure, different values ​​of Wo result in different boundary conditions at the velocity inlet. When Wo = 2, the shape is closest to the parabola P.

[0074] In one embodiment, if Wo takes the standard value, Q0 and Q n Standard values ​​are also taken, and because n and ω can take fixed values, after the vascular model of the current patient is determined, the vascular radius R at the entrance of the vascular model is a known value, and r is a value between 0 and R. Therefore, u(r, t) can be a predetermined standard target arterial blood flow velocity distribution in space and time. For different patients, it can be directly applied after determining its R value.

[0075] Of course, in another embodiment, Q0 and Q corresponding to the current patient can also be obtained each time. n , then the Q0 and Q n Substituting into the blood flow velocity model constructed based on the Womersley boundary condition described in formula (2), the blood flow velocity distribution in space and time in the target artery is obtained, that is, the velocity inlet boundary condition.

[0076] In one embodiment, the hemodynamic index parameters output by the CFD simulation module may include surface shear stress (Wall Shear Stress) and pressure (pressure), etc.; the solid mechanics index parameters output by the SSA simulation module may include: equivalent stress (Von Mises Stress) and deformation (Displacement), etc.

[0077] Step S24: predicting the rupture risk of the aneurysm of the current patient based on the hemodynamic index parameters and the solid mechanics index parameters.

[0078] In this step, there are many ways to implement it, and only a few of them are listed below.

[0079] The first method is to directly conduct mathematical analysis or empirical judgment based on these indicator parameters. This method requires the aneurysm rupture risk predictor to have a corresponding theoretical foundation. Furthermore, it can also be combined with demographic theory, such as age, gender and other information for comprehensive analysis and judgment.

[0080] The second method is to generate a simulation animation based on these index parameters and provide the simulation animation to the user for predicting the risk of aneurysm rupture. Furthermore, demographic theory, such as age, gender and other information, can be combined for comprehensive analysis and simulation.

[0081] The third method is to pre-train an aneurysm rupture risk prediction model, input the hemodynamic index parameters and the solid mechanics index parameters into the pre-trained aneurysm rupture risk prediction model, and obtain aneurysm rupture risk data output by the aneurysm rupture risk prediction model. The aneurysm rupture risk prediction model is trained by input training samples including historical hemodynamic index parameters and solid mechanics index parameters and output training samples including historical aneurysm rupture risk data.

[0082] Furthermore, when training the aneurysm rupture risk prediction model, demographics may be incorporated therein, that is, the input training sample may further include relevant patient information such as age, gender, etc. Accordingly, in this step, the hemodynamic index parameters, the solid mechanics index parameters, and relevant patient information of the current patient such as age, gender, etc. may be input into the pre-trained aneurysm rupture risk prediction model to obtain the aneurysm rupture risk data output by the aneurysm rupture risk prediction model.

[0083] The aneurysm rupture risk data may be a rupture risk probability value and / or a rupture state, etc. The data may be set according to actual needs.

[0084] Figure 5 FIG. 2 shows an exemplary flow chart of a method for training an aneurysm rupture risk prediction model in an embodiment of the present application. Figure 5 As shown, the method may include the following processing:

[0085] Step S51, obtaining historical aneurysm data sets that meet the quantity requirement, each historical aneurysm data set includes the patient's historical target artery three-dimensional imaging data and corresponding historical aneurysm rupture risk data, and further, may also include relevant patient information.

[0086] In this step, a large number of aneurysm patients' aneurysm data sets can be collected to ensure the diversity of aneurysm locations in the arterial vascular tree and cover a wide range of aneurysm types and sizes, thereby enhancing the versatility of the model. And by including aneurysms in different locations, the risk of aneurysm rupture in different regions can be better understood and more accurate predictions can be provided for individual cases.

[0087] Step S52: for each historical aneurysm data set, construct a corresponding target artery three-dimensional fluid model and a target artery three-dimensional solid model based on the historical target artery three-dimensional imaging data in the historical aneurysm data set.

[0088] The specific implementation process of step S52 may be similar to the implementation process described in step S21, and will not be described in detail here.

[0089] Step S53, constructing a corresponding CFD simulation module based on the three-dimensional fluid model of the target artery, constructing a corresponding SSA simulation module based on the three-dimensional solid model of the target artery, and constructing a one-way FSI simulation model based on the CFD simulation module and the SSA simulation module.

[0090] The specific implementation process of step S53 may be similar to the implementation process described in step S22, and will not be described in detail here.

[0091] Step S54, using the pre-acquired blood flow velocity distribution in space and time within the target artery corresponding to the historical aneurysm data set as the input of the FSI simulation model, to obtain the hemodynamic index parameters output by the CFD simulation module and the solid mechanics index parameters output by the SSA simulation module.

[0092] The specific implementation process of this step S54 may be similar to the implementation process described in step S23.

[0093] For example, in one embodiment, the spatial and temporal blood flow velocity distribution in the target artery corresponding to the historical aneurysm data set is obtained by:

[0094] Obtain the Fourier series fitting value of the target arterial blood flow rate corresponding to the historical aneurysm data group that varies with time, substitute the Fourier series fitting value that varies with time into the blood flow velocity model constructed based on the Womersley boundary condition, and obtain the blood flow velocity distribution in space and time in the target arterial blood vessel.

[0095] Wherein, when the historical aneurysm data group further includes ultrasound video or phase contrast imaging data within the historical target artery region, the Fourier series fitting value of the target artery blood flow rate varying with time corresponding to the historical aneurysm data group can be obtained based on the ultrasound video or phase contrast imaging data within the historical target artery region; when the historical aneurysm data group does not include ultrasound video or phase contrast imaging data within the historical target artery region, the Fourier series fitting average value or standard value of the target artery blood flow rate varying with time can be obtained based on the Fourier series fitting value of the target artery blood flow rate varying with time of different patients, and the Fourier series fitting average value or standard value of the target artery blood flow rate varying with time is used as the Fourier series fitting value of the target artery blood flow rate varying with time corresponding to the historical aneurysm data group.

[0096] Of course, in another embodiment, if Wo takes the standard value, Q0 and Q nStandard values ​​are also taken, and because n and ω can take fixed values, after the vascular model of the current patient is determined, the vascular radius R at the entrance of the vascular model is a known value, and r is a value between 0 and R. Therefore, u(r, t) can be a predetermined standard target arterial blood flow velocity distribution in space and time. For different patients, it can be directly applied after determining its R value.

[0097] Step S55, using the hemodynamic index parameters, the solid mechanics index parameters and even the corresponding relevant patient information as input training samples, and using the historical aneurysm rupture risk data in the historical aneurysm data group as output training samples.

[0098] Step S56, using the input training samples and the output training samples corresponding to each historical aneurysm data group to train an artificial intelligence network capable of machine learning, to obtain an aneurysm rupture risk prediction model.

[0099] Accordingly, Figure 2 In step S24, the Figure 5 The aneurysm rupture risk prediction model trained by the method shown predicts the aneurysm rupture risk of the current patient.

[0100] The above describes in detail the aneurysm rupture risk prediction method and the training method of the aneurysm rupture risk prediction model in the embodiment of the present application. The following describes in detail the aneurysm rupture risk prediction device and the training device of the aneurysm rupture risk prediction model in the embodiment of the present application. The aneurysm rupture risk prediction device and the training device of the aneurysm rupture risk prediction model in the embodiment of the present application can be used to implement the corresponding method in the embodiment of the present application. For details not disclosed in detail in the embodiment of the device of the present application, please refer to the corresponding description in the embodiment of the method of the present application, and will not be repeated here one by one.

[0101] Figure 6 : is an exemplary structural diagram of the aneurysm rupture risk prediction device in the embodiment of the present application. Figure 6 As shown, the device may include: a first three-dimensional model construction module 601, a first FSI simulation model construction module 602, a first indicator parameter acquisition module 603 and a risk prediction module 604.

[0102] The first three-dimensional model building module 601 is used to build a three-dimensional fluid model and a three-dimensional solid model of the target artery of the current patient based on the acquired three-dimensional imaging data of the target artery of the current patient.

[0103] The first FSI simulation model construction module 602 is used to construct the CFD simulation module of the current patient based on the three-dimensional fluid model of the target artery, to construct the SSA simulation module of the current patient based on the three-dimensional solid model of the target artery, and to construct the unidirectional FSI simulation model of the current patient based on the CFD simulation module and the SSA simulation module.

[0104] The first indicator parameter acquisition module 603 is used to use the pre-acquired spatial and temporal blood flow velocity distribution in the target artery of the current patient as the input of the FSI simulation model to obtain the hemodynamic indicator parameters output by the CFD simulation module and the solid mechanics indicator parameters output by the SSA simulation module.

[0105] The risk prediction module 604 is used to predict the rupture risk of the aneurysm of the current patient based on the hemodynamic index parameters and the solid mechanics index parameters.

[0106] In one embodiment, the risk prediction module 604 is used to input the hemodynamic indicator parameters, the solid mechanics indicator parameters and even the relevant patient information of the current patient into a pre-trained aneurysm rupture risk prediction model to obtain aneurysm rupture risk data output by the aneurysm rupture risk prediction model; the aneurysm rupture risk prediction model is trained by input training samples including historical hemodynamic indicator parameters and solid mechanics indicator parameters and output training samples including historical aneurysm rupture risk data.

[0107] In one embodiment, the device may further include: a first blood flow velocity distribution acquisition module 605, which is used to obtain the Fourier series fitting value of the target arterial blood flow rate corresponding to the current patient that changes with time, and substitute the Fourier series fitting value that changes with time into the blood flow velocity model constructed based on the Womersley boundary condition to obtain the blood flow velocity distribution in space and time in the target arterial blood vessel.

[0108] In one embodiment, the first blood flow velocity distribution acquisition module 605 may, when there is ultrasound video or phase contrast imaging data within the target artery region of the current patient, obtain the Fourier series fitting value of the target artery blood flow rate varying with time corresponding to the current patient based on the historical ultrasound video or phase contrast imaging data within the target artery region of the current patient; when there is no ultrasound video or phase contrast imaging data within the target artery region of the current patient, obtain the Fourier series fitting average value or standard value of the target artery blood flow rate varying with time based on the Fourier series fitting value of the target artery blood flow rate varying with time of different patients, and use the Fourier series fitting average value or standard value of the target artery blood flow rate varying with time as the Fourier series fitting value of the target artery blood flow rate varying with time corresponding to the current patient.

[0109] Figure 7 FIG. 1 is an exemplary structural diagram of a training device for an aneurysm rupture risk prediction model in an embodiment of the present application. Figure 7 As shown, the device may include: a data collection module 701, a second three-dimensional model construction module 702, a second FSI simulation model construction module 703, a second indicator parameter acquisition module 704, a training sample determination module 705 and a model training module 706.

[0110] The data collection module 701 is used to obtain historical aneurysm data sets that meet the quantity requirements, each of which includes the patient's historical target artery three-dimensional imaging data and corresponding historical aneurysm rupture risk data, and further, may include relevant patient information.

[0111] The second three-dimensional model building module 702 is used to build, for each historical aneurysm data group, a corresponding target artery three-dimensional fluid model and a target artery three-dimensional solid model based on the historical target artery three-dimensional imaging data in the historical aneurysm data group.

[0112] The second FSI simulation model construction module 703 is used to construct a corresponding CFD simulation module based on the three-dimensional fluid model of the target artery constructed by the second three-dimensional model construction module 702, to construct a corresponding SSA simulation module based on the three-dimensional solid model of the target artery constructed by the second three-dimensional model construction module 702, and to construct a unidirectional FSI simulation model based on the CFD simulation module and the SSA simulation module.

[0113] The second indicator parameter acquisition module 704 is used to use the pre-acquired spatial and temporal blood flow velocity distribution in the target artery corresponding to the historical aneurysm data group as the input of the FSI simulation model constructed by the second FSI simulation model construction module 703, and obtain the hemodynamic indicator parameters output by the CFD simulation module and the solid mechanics indicator parameters output by the SSA simulation module.

[0114] The training sample determination module 705 is used to use the hemodynamic index parameters, the solid mechanics index parameters and even the corresponding relevant patient information as input training samples, and use the historical aneurysm rupture risk data in the historical aneurysm data group as output training samples.

[0115] The model training module 706 is used to train an artificial intelligence network capable of machine learning using the input training samples and the output training samples corresponding to each historical aneurysm data group to obtain an aneurysm rupture risk prediction model.

[0116] In one embodiment, the device may further include: a second blood flow velocity distribution acquisition module 707, which is used to obtain the Fourier series fitting value of the target arterial blood flow volume corresponding to the historical aneurysm data group that varies with time, and substitute the Fourier series fitting value that varies with time into the blood flow velocity model constructed based on the Womersley boundary condition to obtain the blood flow velocity distribution in space and time in the target arterial blood vessel.

[0117] In one embodiment, when the historical aneurysm data group further includes ultrasound video or phase contrast imaging data within the historical target artery region, the second blood flow velocity distribution acquisition module 707 obtains the Fourier series fitting value of the target artery blood flow rate changing with time corresponding to the historical aneurysm data group based on the ultrasound video or phase contrast imaging data within the historical target artery region; when the historical aneurysm data group does not include ultrasound video or phase contrast imaging data within the historical target artery region, the Fourier series fitting average value or standard value of the target artery blood flow rate changing with time is obtained based on the Fourier series fitting value of the target artery blood flow rate changing with time of different patients, and the Fourier series fitting average value or standard value of the target artery blood flow rate changing with time is used as the Fourier series fitting value of the target artery blood flow rate changing with time corresponding to the historical aneurysm data group.

[0118] Figure 8 FIG. 1 is a schematic diagram of another device for predicting the risk of aneurysm rupture according to an embodiment of the present application. Figure 8As shown, the device may include: at least one first memory 81 and at least one first processor 82. In addition, some other components may also be included, such as communication ports, etc. These components communicate via a first bus 83.

[0119] Among them, at least one first memory 81 is used to store a computer program. In one embodiment, the computer program can be understood to include Figure 6 The modules of the aneurysm rupture risk prediction device are shown. In addition, at least one first memory 81 can also store an operating system, etc. The operating system includes but is not limited to: Android operating system, Symbian operating system, Windows operating system, Linux operating system, etc.

[0120] At least one first processor 82 is used to call the computer program stored in at least one first memory 81 to execute the aneurysm rupture risk prediction method described in the embodiment of the present application. The first processor 82 can be a CPU, a processing unit / module, an ASIC, a logic module or a programmable gate array, etc. It can receive and send data through the communication port.

[0121] Fig. 9 FIG. 1 is a schematic diagram of the structure of another training device for predicting the risk of aneurysm rupture in an embodiment of the present application. Fig. 9 As shown, the device may include: at least one second memory 91 and at least one second processor 92. In addition, some other components may be included, such as communication ports, etc. These components communicate via a second bus 93.

[0122] Among them, at least one second memory 91 is used to store a computer program. In one embodiment, the computer program can be understood to include Figure 7 The modules of the training device for the aneurysm rupture risk prediction model shown in the figure. In addition, at least one second memory 91 can also store an operating system, etc. The operating system includes but is not limited to: Android operating system, Symbian operating system, Windows operating system, Linux operating system, etc.

[0123] At least one second processor 92 is used to call the computer program stored in at least one second memory 91 to execute the training method of the aneurysm rupture risk prediction model described in the embodiment of the present application. The second processor 92 can be a CPU, a processing unit / module, an ASIC, a logic module or a programmable gate array, etc. It can receive and send data through the communication port.

[0124] It should be noted that not all steps and modules in the above processes and structure diagrams are necessary, and some steps or modules can be ignored according to actual needs. The execution order of each step is not fixed and can be adjusted as needed. The division of each module is only for the convenience of describing the functional division adopted. In actual implementation, a module can be implemented by multiple modules, and the functions of multiple modules can also be implemented by the same module. These modules can be located in the same device or in different devices.

[0125] It is understood that the hardware modules in the above-mentioned embodiments can be implemented mechanically or electronically. For example, a hardware module may include a specially designed permanent circuit or logic device (such as a dedicated processor, such as an FPGA or ASIC) for completing a specific operation. The hardware module may also include a programmable logic device or circuit (such as a general-purpose processor or other programmable processor) temporarily configured by software for performing a specific operation. As for whether to implement the hardware module in a mechanical manner, or using a dedicated permanent circuit, or using a temporarily configured circuit (such as configured by software), it can be determined based on cost and time considerations.

[0126] In addition, a computer program product is also provided in an embodiment of the present application, including a computer program, which can be executed by a processor and implement the aneurysm rupture risk prediction method or the aneurysm rupture risk prediction model training method described in the embodiment of the present application. A computer-readable storage medium is also provided in an embodiment of the present application, on which a computer program is stored, and the computer program can be executed by a processor and implement the aneurysm rupture risk prediction method or the aneurysm rupture risk prediction model training method described in the embodiment of the present application. Specifically, a system or device equipped with a storage medium can be provided, on which a software program code that implements the functions of any of the above-mentioned embodiments is stored, and the computer (or CPU or MPU) of the system or device reads and executes the program code stored in the storage medium. In addition, the operating system operating on the computer can also be used to complete part or all of the actual operations based on the instructions of the program code. The program code read from the storage medium can also be written to a memory provided in an expansion board inserted into the computer or to a memory provided in an expansion unit connected to the computer, and then the CPU installed on the expansion board or the expansion unit is made to perform part and all of the actual operations based on the instructions of the program code, thereby realizing the functions of any of the above-mentioned embodiments. Storage media implementations for providing program code include USB, floppy disk, hard disk, magneto-optical disk, optical disk (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), magnetic tape, non-volatile memory card and ROM, etc. Alternatively, the program code can be downloaded from a server computer via a communication network.

[0127] It can be seen from the above scheme that in the embodiment of the present application, since the hemodynamic index parameters and solid mechanics index parameters of the artery are taken into consideration at the same time, the accuracy and reliability of the prediction of the rupture risk of the aneurysm are greatly improved.

[0128] In addition, by combining the pre-trained aneurysm rupture risk prediction model, the prediction results can be made more intuitive and seamlessly integrated into the medical field.

[0129] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A device for predicting the risk of aneurysm rupture, characterized in that: include: A first three-dimensional model building module (601) is used to build a three-dimensional fluid model and a three-dimensional solid model of the target artery of the current patient based on the acquired three-dimensional imaging data of the target artery of the current patient; a first fluid-solid coupling simulation model construction module (602), configured to construct a computational fluid dynamics simulation module of the current patient based on the three-dimensional fluid model of the target artery, to construct a static solid analysis simulation module of the current patient based on the three-dimensional solid model of the target artery, and to construct a unidirectional fluid-solid coupling simulation model of the current patient based on the computational fluid dynamics simulation module and the static solid analysis simulation module; A first index parameter acquisition module (603) is used to use the previously acquired spatial and temporal blood flow velocity distribution in the target artery of the current patient as an input to the fluid-solid coupling simulation model to obtain the hemodynamic index parameters output by the computational fluid dynamics simulation module and the solid mechanics index parameters output by the static solid analysis simulation module; and The risk prediction module (604) is used to predict the rupture risk of the aneurysm of the current patient based on the hemodynamic index parameters and the solid mechanics index parameters.

2. The aneurysm rupture risk prediction device according to claim 1, characterized in that: The risk prediction module (604) inputs the hemodynamic index parameter, the solid mechanics index parameter and the relevant patient information of the current patient into a pre-trained aneurysm rupture risk prediction model to obtain aneurysm rupture risk data output by the aneurysm rupture risk prediction model; The aneurysm rupture risk prediction model is trained by input training samples including historical hemodynamic index parameters and solid mechanics index parameters and output training samples including historical aneurysm rupture risk data.

3. The aneurysm rupture risk prediction device according to claim 1 or 2, characterized in that: The device further comprises: The first blood flow velocity distribution acquisition module (605) is used to obtain the Fourier series fitting value of the target arterial blood flow volume corresponding to the current patient that varies with time, substitute the Fourier series fitting value that varies with time into the blood flow velocity model constructed based on the Womersley boundary condition, and obtain the blood flow velocity distribution in space and time in the target arterial blood vessel.

4. The aneurysm rupture risk prediction device according to claim 3, characterized in that: The first blood flow velocity distribution acquisition module (605) acquires the Fourier series fitting value of the target artery blood flow rate varying with time corresponding to the current patient based on the historical ultrasound video or phase contrast imaging data within the target artery region of the current patient when there is ultrasound video or phase contrast imaging data within the target artery region of the current patient; and acquires the Fourier series fitting average value or standard value of the target artery blood flow rate varying with time based on the Fourier series fitting values ​​of the target artery blood flow rate varying with time of different patients when there is no ultrasound video or phase contrast imaging data within the target artery region of the current patient, and uses the Fourier series fitting average value or standard value of the target artery blood flow rate varying with time as the Fourier series fitting value of the target artery blood flow rate varying with time corresponding to the current patient.

5. A training device for an aneurysm rupture risk prediction model, characterized in that: include: A data collection module (701) is used to obtain historical aneurysm data sets that meet the quantity requirement, each historical aneurysm data set includes the patient's historical target artery three-dimensional imaging data and the corresponding historical aneurysm rupture risk data, and further, may include relevant patient information; A second three-dimensional model building module (702) is used to build, for each historical aneurysm data group, a corresponding target artery three-dimensional fluid model and a target artery three-dimensional solid model based on the historical target artery three-dimensional imaging data in the historical aneurysm data group; A second fluid-solid coupling simulation model construction module (703) is used to construct a corresponding computational fluid dynamics simulation module based on the three-dimensional fluid model of the target artery constructed by the second three-dimensional model construction module (702), to construct a corresponding static solid analysis simulation module based on the three-dimensional solid model of the target artery constructed by the second three-dimensional model construction module (702), and to construct a unidirectional fluid-solid coupling simulation model based on the computational fluid dynamics simulation module and the static solid analysis simulation module; A second index parameter acquisition module (704) is used to use the pre-acquired blood flow velocity distribution in space and time in the target artery corresponding to the historical aneurysm data group as the input of the fluid-solid coupling simulation model constructed by the second fluid-solid coupling simulation model construction module (703), and obtain the hemodynamic index parameters output by the computational fluid dynamics simulation module and the solid mechanics index parameters output by the static solid analysis simulation module; A training sample determination module (705) is used to use the hemodynamic index parameter, the solid mechanics index parameter and even the corresponding relevant patient information as input training samples, and use the historical aneurysm rupture risk data in the historical aneurysm data group as output training samples; and The model training module (706) is used to train an artificial intelligence network capable of machine learning using the input training samples and the output training samples corresponding to each historical aneurysm data group to obtain an aneurysm rupture risk prediction model.

6. The training device for aneurysm rupture risk prediction model according to claim 5, characterized in that: The device further comprises: a second blood flow velocity distribution acquisition module (707), which is used to obtain the Fourier series fitting value of the target artery blood flow volume corresponding to the historical aneurysm data group that varies with time, and substitute the Fourier series fitting value that varies with time into the blood flow velocity model constructed based on the Womersley boundary condition to obtain the blood flow velocity distribution in space and time in the target artery.

7. The training device for the aneurysm rupture risk prediction model according to claim 6, characterized in that: When the historical aneurysm data group further includes ultrasound video or phase contrast imaging data in the historical target artery region, the second blood flow velocity distribution acquisition module (707) obtains the Fourier series fitting value of the target artery blood flow rate changing with time corresponding to the historical aneurysm data group based on the ultrasound video or phase contrast imaging data in the historical target artery region; when the historical aneurysm data group does not include ultrasound video or phase contrast imaging data in the historical target artery region, the Fourier series fitting average value or standard value of the target artery blood flow rate changing with time is obtained based on the Fourier series fitting value of the target artery blood flow rate changing with time of different patients, and the Fourier series fitting average value or standard value of the target artery blood flow rate changing with time is used as the Fourier series fitting value of the target artery blood flow rate changing with time corresponding to the historical aneurysm data group.

8. A device for predicting the risk of aneurysm rupture, characterized in that: include: At least one memory (81) and at least one processor (82), wherein: The at least one memory (81) is used to store a computer program; The at least one processor (82) is used to call the computer program stored in the at least one memory (81) to perform the following operations: Acquire the three-dimensional imaging data of the target artery of the current patient, and construct a three-dimensional fluid model and a three-dimensional solid model of the target artery of the current patient based on the three-dimensional imaging data of the target artery; constructing a computational fluid dynamics simulation module of the current patient based on the three-dimensional fluid model of the target artery, constructing a static solid analysis simulation module of the current patient based on the three-dimensional solid model of the target artery, and constructing a unidirectional fluid-solid coupling simulation model of the current patient based on the computational fluid dynamics simulation module and the static solid analysis simulation module; Using the previously acquired blood flow velocity distribution in space and time in the target artery of the current patient as the input of the fluid-solid coupling simulation model, to obtain the hemodynamic index parameters output by the computational fluid dynamics simulation module and the solid mechanics index parameters output by the static solid analysis simulation module; The risk of aneurysm rupture of the current patient is predicted based on the hemodynamic index parameters and the solid mechanics index parameters.

9. A training device for an aneurysm rupture risk prediction model, characterized in that: include: At least one memory (91) and at least one processor (92), wherein: The at least one memory (91) is used to store a computer program; The at least one processor (92) is used to call the computer program stored in the at least one memory (91) to perform the following operations: Acquire historical aneurysm data sets that meet quantity requirements, each historical aneurysm data set including relevant patient information, historical target artery three-dimensional imaging data of the patient, and corresponding historical aneurysm rupture risk data; For each historical aneurysm data group, a corresponding three-dimensional fluid model of the target artery and a three-dimensional solid model of the target artery are constructed based on the three-dimensional imaging data of the historical target artery in the historical aneurysm data group; a corresponding computational fluid dynamics simulation module is constructed based on the three-dimensional fluid model of the target artery, a corresponding static solid analysis simulation module is constructed based on the three-dimensional solid model of the target artery, and a one-way fluid-solid coupling simulation model is constructed based on the computational fluid dynamics simulation module and the static solid analysis simulation module; the pre-acquired spatial and temporal blood flow velocity distribution in the target artery corresponding to the historical aneurysm data group is used as the input of the fluid-solid coupling simulation model to obtain the hemodynamic index parameters output by the computational fluid dynamics simulation module and the solid mechanics index parameters output by the static solid analysis simulation module; the hemodynamic index parameters, the solid mechanics index parameters and the corresponding relevant patient information are used as input training samples, and the historical aneurysm rupture risk data in the historical aneurysm data group is used as output training samples; An artificial intelligence network is trained using the input training samples and the output training samples corresponding to each historical aneurysm data group to obtain an aneurysm rupture risk prediction model.

10. A first computer-readable storage medium having a computer program stored thereon; characterized in that: The computer program can be executed by a processor and implement the following operations: Acquire the three-dimensional imaging data of the target artery of the current patient, and construct a three-dimensional fluid model and a three-dimensional solid model of the target artery of the current patient based on the three-dimensional imaging data of the target artery; constructing a computational fluid dynamics simulation module of the current patient based on the three-dimensional fluid model of the target artery, constructing a static solid analysis simulation module of the current patient based on the three-dimensional solid model of the target artery, and constructing a unidirectional fluid-solid coupling simulation model of the current patient based on the computational fluid dynamics simulation module and the static solid analysis simulation module; Using the previously acquired blood flow velocity distribution in space and time in the target artery of the current patient as the input of the fluid-solid coupling simulation model, to obtain the hemodynamic index parameters output by the computational fluid dynamics simulation module and the solid mechanics index parameters output by the static solid analysis simulation module; The risk of aneurysm rupture of the current patient is predicted based on the hemodynamic index parameters and the solid mechanics index parameters.

11. A second computer-readable storage medium having a computer program stored thereon; characterized in that: The computer program can be executed by a processor and implement the following operations: Acquire historical aneurysm data sets that meet quantity requirements, each historical aneurysm data set including relevant patient information, historical target artery three-dimensional imaging data of the patient, and corresponding historical aneurysm rupture risk data; For each historical aneurysm data group, a corresponding three-dimensional fluid model of the target artery and a three-dimensional solid model of the target artery are constructed based on the three-dimensional imaging data of the historical target artery in the historical aneurysm data group; a corresponding computational fluid dynamics simulation module is constructed based on the three-dimensional fluid model of the target artery, a corresponding static solid analysis simulation module is constructed based on the three-dimensional solid model of the target artery, and a one-way fluid-solid coupling simulation model is constructed based on the computational fluid dynamics simulation module and the static solid analysis simulation module; the pre-acquired spatial and temporal blood flow velocity distribution in the target artery corresponding to the historical aneurysm data group is used as the input of the fluid-solid coupling simulation model to obtain the hemodynamic index parameters output by the computational fluid dynamics simulation module and the solid mechanics index parameters output by the static solid analysis simulation module; the hemodynamic index parameters, the solid mechanics index parameters and the corresponding relevant patient information are used as input training samples, and the historical aneurysm rupture risk data in the historical aneurysm data group is used as output training samples; An artificial intelligence network is trained using the input training samples and the output training samples corresponding to each historical aneurysm data group to obtain an aneurysm rupture risk prediction model.

Images