Patient-specific finite element model-based vertebroplasty optimization method and related device

Abstract

The invention discloses a patient-specific finite element model-based vertebroplasty optimization method and a related device. The method comprises the following steps of: constructing a patient-specific three-dimensional spine model according to a vertebral bone structure and a vertebral soft tissue structure; discretizing the patient-specific three-dimensional spine model into tetrahedral finite element grids, endowing each finite element unit with bone material attributes according to pixel values of CT scanning images, and endowing intervertebral disc and ligament soft tissue structures with soft tissue material attributes; determining an optimal bone cement injection path and an optimal bone cement injection amount of the tetrahedral finite element mesh; and simulating the vertebroplasty in the tetrahedral finite element mesh, predicting the stress distribution of the vertebrae, the rigidity of the vertebrae and the stress change of the adjacent vertebrae after the operation, and presenting the simulation result of the simulated vertebroplasty in a three-dimensional visual form. According to the method, personalized optimization and prediction of preoperative simulation of the vertebroplasty are realized by constructing the patient-specific finite element model, and an efficient auxiliary reference operation scheme is provided for an operator.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence and medical surgery assistance technology, and specifically to a vertebroplasty optimization method and device, and a computing device based on a patient-specific finite element model. Background Art

[0002] Vertebroplasty is a common spinal surgery used to treat spinal compression fractures, osteoporosis and other diseases. This surgery involves injecting bone cement into the damaged vertebra to enhance its stability and strength.

[0003] However, the existing planning and optimization of vertebroplasty mainly relies on the doctor's experience and simple image analysis, and there are problems such as bone cement leakage, insufficient vertebral stability, and stress changes in adjacent vertebrae, which lead to poor surgical results and even complications. Although finite element analysis is widely used in the engineering field, no relevant applications in the medical field have been retrieved so far. In addition, the existing optimization methods for bone cement injection path and injection volume are also relatively rough, lacking accurate stress analysis and optimization algorithms. For example, if the doctor underestimates the injection path, it may cause bone cement to be concentrated in the front half of the vertebra, and the back half to be insufficiently supported, and the load-bearing capacity of the vertebra cannot be effectively restored. In addition, it is impossible to present three-dimensional visualization of simulation results, and it is impossible to intuitively display the stress distribution and deformation of the vertebra after surgery.

[0004] To solve the above problems, the present invention proposes a vertebroplasty optimization method based on a patient-specific finite element model. By constructing a patient-specific finite element model, personalized optimization of vertebroplasty is achieved and preoperative simulation is predicted, providing the surgeon with an efficient auxiliary reference surgical plan. Summary of the Invention

[0005] In view of the above problems, the present invention provides a vertebroplasty optimization method, apparatus, and computing device based on a patient-specific finite element model.

[0006] According to one aspect of the present invention, a method for optimizing vertebroplasty based on a patient-specific finite element model is provided, comprising: Acquiring a CT scan image and an X-ray scan image containing a target vertebra and its adjacent vertebrae; automatically segmenting the CT scan image and the X-ray scan image to extract the vertebral bone structure and vertebral soft tissue structure, respectively; and constructing a patient-specific three-dimensional spinal model based on the vertebral bone structure and vertebral soft tissue structure; discretizing the patient-specific three-dimensional spine model into a tetrahedral finite element mesh, assigning bone material properties to each finite element based on pixel values ​​of the CT scan image, and assigning soft tissue material properties to intervertebral disc and ligament soft tissue structures; Determining an optimal bone cement injection path of the tetrahedral finite element mesh by a density method, wherein the objective function of the optimal bone cement injection path is to minimize the maximum principal stress of the target vertebral body, and the constraints include the bone cement injection volume and the stability of the target vertebral body; Determining an optimal bone cement injection amount for the optimal bone cement injection path by taking minimizing the maximum principal stress of the target vertebra, maximizing the stiffness of the target vertebra, and minimizing the stress change of adjacent vertebrae as objective functions; According to the optimal bone cement injection path and the optimal bone cement injection amount, vertebroplasty is simulated in the tetrahedral finite element grid, the stress distribution of the vertebral body, the vertebral body stiffness and the stress changes of the adjacent vertebral bodies after surgery are predicted, and the simulation results of the simulated vertebroplasty are presented in a three-dimensional visualization form. The simulation results include bone cement distribution, stress cloud map and deformation map.

[0007] In an optional manner, the mapping relationship between the pixel values ​​of the CT scan image and the bone material properties is expressed as follows: in, For coordinates The elastic modulus of the bone material; It is the lower limit of elastic modulus of cancellous bone; is the upper limit of elastic modulus of cortical bone; is the curvature adjustment coefficient, ; is the bone density value corresponding to the turning point of bone density; The coordinates are The bone density value of the pixel point.

[0008] In an optional manner, the sensitivity formula of the objective function of minimizing the maximum principal stress of the target vertebral body is: in, Design variables for the unit; is the unit volume; is the von Mises equivalent stress; is the strain tensor; is the fourth-order elasticity tensor; is the spatial area of ​​the unit; is the maximum principal stress; It is a double dot product, used for tensor operations.

[0009] In an optional manner, constructing a patient-specific three-dimensional spine model based on the vertebral bone structure and the vertebral soft tissue structure further comprises: Performing surface reconstruction on the voxel data of the vertebral bone structure and the vertebral soft tissue structure according to the Marching Cubes algorithm to generate an initial surface mesh; The surface mesh data is smoothed and mesh simplified by a Laplacian Smoothing algorithm to obtain a surface mesh; A tetrahedral mesh model that meets the finite element analysis requirements of the surface mesh is generated using the tetrahedral mesh generation tool TetGen to obtain a patient-specific three-dimensional spinal finite element model.

[0010] In an optional manner, determining the optimal bone cement injection path of the tetrahedral finite element mesh by a density method further comprises: Assign a density variable to each tetrahedral finite element mesh, perform finite element analysis based on the current density distribution, and calculate the maximum principal stress of the target vertebra; Calculating the sensitivity of the maximum principal stress to each density variable, updating the density variable according to the sensitivity, checking whether the constraints of bone cement volume and vertebral stability are met, and adjusting the density variable if not met; The updated distribution of the density variable is converted into an actual bone cement injection path to obtain the optimal bone cement injection path of the tetrahedral finite element mesh.

[0011] In an optional manner, simulating vertebroplasty in the tetrahedral finite element mesh according to the optimal bone cement injection path and the optimal bone cement injection amount further comprises: The optimal injection path is converted into a spatial coordinate sequence, and a target unit set is marked in a finite element grid; the volume fraction of bone cement in each target unit is calculated based on the optimal injection volume, and the injection process is simulated by a progressive filling method; the elastic modulus of the unit injected with bone cement is linearly transitioned from the bone tissue value to the cement material value; The temperature field distribution is calculated based on the nonlinear transient heat conduction equation to simulate the exothermic process of bone cement hydration reaction. The Carreau model is used to describe the non-Newtonian fluid properties of bone cement during the flow phase, and the mesh deformation is handled according to the Lagrangian-Eulerian method. Stiffness degradation is initiated when the element equivalent plastic strain exceeds a preset threshold. The postoperative vertebral stiffness matrix was compared with the preoperative stiffness to obtain the stiffness recovery rate, and the pressure increment of adjacent intervertebral discs and the maximum shear strain of the annulus fibrosus were calculated. A particle system was used to simulate the cement diffusion process to simulate vertebroplasty, where the particle color mapped the injection time and the transparency channel of the stress contour was overlaid to show the bone cement distribution.

[0012] In an optional embodiment, during the simulated vertebroplasty procedure, the method further comprises: The curing behavior of bone cement is represented by a viscoelastic or viscoplastic constitutive model, wherein the constitutive model includes factors affecting the mechanical properties of bone cement such as the degree of hydration reaction of bone cement, temperature, and strain rate; The curing behavior of bone cement and the effects of bone cement injection on the internal microenvironment of the vertebral body were injected into the coupled biological model to predict the changes in bone tissue remodeling and long-term stability after bone cement injection. The effects of the internal microenvironment included changes in bone marrow pressure, stress and strain changes in bone cells, and microfractures. When determining the optimal bone cement injection volume, the vertebral compressive strength, fatigue life, and stress concentration of adjacent vertebral endplates were used as constraints of the optimization objective function.

[0013] In an optional manner, the method further includes: When converting the density variable distribution into the bone cement injection path, the density field is smoothed and threshold segmented to extract the bone cement injection channel; Eliminate sharp corners and thin connections in the geometry of the injection channel to reduce the risk of bone cement leakage; The manufacturability of the optimized bone cement injection path was evaluated based on the accessibility of the puncture needle, the range of injection angles, and the performance of the bone cement injection device.

[0014] According to another aspect of the present invention, there is provided a device for optimizing vertebroplasty based on a patient-specific finite element model, comprising: A spine model construction module is configured to obtain a CT scan image and an X-ray scan image containing a target vertebra and its adjacent vertebrae; automatically segment the CT scan image and the X-ray scan image to extract the vertebral bone structure and vertebral soft tissue structure, respectively; and construct a patient-specific three-dimensional spine model based on the vertebral bone structure and vertebral soft tissue structure. a finite element model preprocessing module for discretizing the patient-specific three-dimensional spinal model into a tetrahedral finite element mesh, assigning bone material properties to each finite element based on pixel values ​​of the CT scan image, and assigning soft tissue material properties to the intervertebral disc and ligament soft tissue structures; an injection path optimization module, configured to determine an optimal bone cement injection path of the tetrahedral finite element mesh by a density method, wherein the objective function of the optimal bone cement injection path is to minimize the maximum principal stress of the target vertebral body, and the constraints include the bone cement injection volume and the stability of the target vertebral body; an injection volume optimization module, configured to determine an optimal bone cement injection volume for the optimal bone cement injection path, taking minimizing the maximum principal stress of the target vertebra, maximizing the stiffness of the target vertebra, and minimizing the stress variation of adjacent vertebrae as objective functions; A vertebroplasty simulation module is used to simulate vertebroplasty in the tetrahedral finite element grid according to the optimal bone cement injection path and the optimal bone cement injection amount, predict the stress distribution of the vertebral body, the vertebral body stiffness and the stress changes of the adjacent vertebral bodies after surgery, and present the simulation results of the simulated vertebroplasty in a three-dimensional visual form, wherein the simulation results include bone cement distribution, stress cloud map and deformation map.

[0015] According to another aspect of the present invention, there is provided a computing device, comprising: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute operations corresponding to the above-mentioned vertebroplasty optimization method based on the patient-specific finite element model.

[0016] According to the solution provided by the present invention, a CT scan image and an X-ray scan image containing a target vertebra and its adjacent vertebrae are obtained; the CT scan image and the X-ray scan image are automatically segmented to extract the vertebral bone structure and the vertebral soft tissue structure respectively; a patient-specific three-dimensional spine model is constructed based on the vertebral bone structure and the vertebral soft tissue structure; the patient-specific three-dimensional spine model is discretized into a tetrahedral finite element mesh, and bone material properties are assigned to each finite element unit according to the pixel value of the CT scan image, and soft tissue material properties are assigned to the intervertebral disc and ligament soft tissue structure; the optimal bone cement injection path of the tetrahedral finite element mesh is determined by the density method, wherein the optimal bone cement The objective function of the injection path is to minimize the maximum principal stress of the target vertebra, and the constraints include the bone cement injection volume and the stability of the target vertebra; the optimal bone cement injection amount of the optimal bone cement injection path is determined by minimizing the maximum principal stress of the target vertebra, maximizing the stiffness of the target vertebra, and minimizing the stress change of the adjacent vertebrae as the objective function; according to the optimal bone cement injection path and the optimal bone cement injection amount, vertebroplasty is simulated in the tetrahedral finite element mesh to predict the stress distribution of the postoperative vertebra, the vertebral stiffness, and the stress change of the adjacent vertebrae after surgery, and the simulation results of the simulated vertebroplasty are presented in a three-dimensional visualization form, and the simulation results include bone cement distribution, stress cloud map, and deformation map. The present invention realizes personalized optimization of vertebroplasty and predicts preoperative simulation by constructing a patient-specific finite element model, providing the surgeon with an efficient auxiliary reference surgical plan. Specifically, a patient-specific three-dimensional spine model and a finite element model are constructed using the patient's own CT scan images and X-ray scan images, which more accurately reflect the patient's spinal structure and mechanical properties, and have high clinical application value. By assigning bone material properties to each finite element based on the pixel values ​​of the CT scan image and incorporating differences in bone density, the finite element model more realistically reflects the mechanical properties of the bone. Furthermore, soft tissue properties are assigned to the intervertebral disc and ligament soft tissue structures to ensure the accuracy of the overall mechanical properties of the spine. The optimal bone cement injection path is determined using a density method, with minimizing the maximum principal stress of the target vertebra as the optimization goal. This effectively reduces the risk of postoperative vertebral fractures and improves surgical safety. By minimizing the maximum principal stress of the target vertebra, maximizing the stiffness of the target vertebra, and minimizing the stress variation of adjacent vertebrae, a more comprehensive approach is used to find the optimal bone cement injection volume, thereby improving the stability of the target vertebra while reducing the risk of re-fracture of adjacent vertebrae. Vertebroplasty is simulated on a tetrahedral finite element mesh to predict the postoperative stress distribution, vertebral stiffness, and stress variations of adjacent vertebrae. The simulation results are presented in a 3D visualization, allowing for preoperative prediction of surgical outcomes and adjustment of surgical plans based on the simulation results, reducing surgical risks and improving both success rates and outcomes.

[0017] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are specifically listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present invention. The same reference symbols are used throughout the drawings to represent the same components. In the drawings: Figure 1 A schematic flow chart of a vertebroplasty optimization method based on a patient-specific finite element model according to an embodiment of the present invention is shown; Figure 2 A schematic diagram showing a comparison of postoperative images of a bone cement connection group and a non-connection group using X-ray films and CT images according to an embodiment of the present invention is shown; Figure 3 A schematic diagram showing the anatomical structure of the human sacrum and hip bone according to an embodiment of the present invention is shown; Figure 4 A schematic diagram showing an optimal bone cement injection path of a tetrahedral finite element mesh according to an embodiment of the present invention is shown; Figure 5 A schematic diagram of a framework of a vertebroplasty optimization device based on a patient-specific finite element model according to an embodiment of the present invention is shown; Figure 6 A schematic structural diagram of a computing device according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0019] Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0020] Figure 1 FIG. 4 is a flow chart of a vertebroplasty optimization method based on a patient-specific finite element model according to an embodiment of the present invention. Figure 1 As shown, the following steps are included: Step S101: Acquire a CT scan image and an X-ray scan image containing a target vertebra and its adjacent vertebrae; automatically segment the CT scan image and the X-ray scan image to extract the vertebral bone structure and the vertebral soft tissue structure respectively; and construct a patient-specific three-dimensional spinal model based on the vertebral bone structure and the vertebral soft tissue structure.

[0021] In this example, a 3D spinal model constructed from the patient's CT and X-ray images accurately reflects individual differences in bone morphology, density distribution, fracture patterns, and soft tissue status, compared to standardized, average spinal models. Vertebroplasty not only affects vertebral stability but also the surrounding intervertebral discs and ligamentous soft tissues. By simultaneously extracting both vertebral bone and soft tissue structures, a more comprehensive assessment of the overall impact of surgery on the spine is possible.

[0022] Specifically, a multi-slice spiral CT scanner was used to obtain Figure 2 The image shown shows an axial CT image of the target vertebra (e.g., L1) and its adjacent vertebrae (e.g., T12 and L2). Optionally, an X-ray scan can be combined to provide more comprehensive information about the entire spine. X-rays can be used to assist in confirming the relative position and overall curvature of the vertebrae. Image segmentation algorithms, such as threshold segmentation and region growing, are used to automatically identify and extract vertebral bone structures from the CT image. For example, a threshold segmentation algorithm sets a threshold range based on the CT value (HU value) and labels pixels within this range as bone tissue. Morphological operations (dilation and erosion) are then used to remove noise and voids from the segmented results, identifying and extracting soft tissue structures such as intervertebral discs and ligaments. Because soft tissue has low contrast in CT images and is difficult to segment, pre-built shape models of the intervertebral discs and ligaments are used to fit the models to the CT images, thereby achieving soft tissue segmentation. A 3D reconstruction algorithm, such as the Marching Cubes algorithm, is used to convert the segmented bone and soft tissue structures into 3D surface models. These 3D surface models are then integrated to form a complete spinal model.

[0023] For example, in planning vertebroplasty for a 65-year-old patient, a lumbar CT scan was performed to obtain images of the T12, L1, and L2 vertebrae. The bone structures of the T12, L1, and L2 vertebrae were segmented from the CT images, and the T12-L1 and L1-L2 intervertebral disc structures were extracted. The segmented bone and disc structures were converted into a 3D surface model using the Marching Cubes algorithm, and the bone and disc models were integrated into a complete spinal model. The spinal model was smoothed using the Laplacian Smoothing algorithm to remove surface noise and jagged edges. A mesh simplification algorithm was used to reduce the mesh size of the spinal model to 500,000 triangles to improve the efficiency of finite element analysis. Ultimately, a patient-specific 3D spinal model was obtained, which included information on the bone morphology of the T12, L1, and L2 vertebrae, intervertebral disc structure, and bone density, which was used to optimize the vertebroplasty plan.

[0024] In an optional manner, constructing a patient-specific three-dimensional spine model based on the vertebral bone structure and the vertebral soft tissue structure further comprises: Performing surface reconstruction on the voxel data of the vertebral bone structure and the vertebral soft tissue structure according to the Marching Cubes algorithm to generate an initial surface mesh; The surface mesh data is smoothed and mesh simplified by a Laplacian Smoothing algorithm to obtain a surface mesh; A tetrahedral mesh model that meets the finite element analysis requirements of the surface mesh is generated using the tetrahedral mesh generation tool TetGen to obtain a patient-specific three-dimensional spinal finite element model.

[0025] In this example, the Marching Cubes algorithm and the Laplacian Smoothing algorithm were used to accurately reconstruct the surface mesh of the vertebral bone and soft tissue structures. The TetGen tool was used to generate a tetrahedral mesh model that met the requirements of finite element analysis, ensuring the accuracy and reliability of the analysis results. The Laplacian Smoothing algorithm not only simplifies the mesh but also smoothes the surface mesh, reducing irregularities and sharp corners in the mesh and improving the smoothness and stability of the model.

[0026] Specifically, the Marching Cubes algorithm is used to reconstruct the surface of the voxel data of the vertebral bone and soft tissue structures to generate an initial surface mesh. The Marching Cubes algorithm generates a continuous surface mesh by extracting isosurfaces from the voxel data, ensuring surface continuity. The Laplacian Smoothing algorithm minimizes the mesh's Laplacian operator, reducing irregularities and sharp corners, and smoothing and simplifying the initial surface mesh. The TetGen tool generates a high-quality tetrahedral mesh from the surface mesh, ensuring mesh uniformity and the accuracy of the finite element analysis.

[0027] In an optional manner, the mapping relationship between the pixel values ​​of the CT scan image and the bone material properties is expressed as follows: in, For coordinates The elastic modulus of the bone material; It is the lower limit of elastic modulus of cancellous bone; is the upper limit of elastic modulus of cortical bone; is the curvature adjustment coefficient, ; is the bone density value corresponding to the turning point of bone density; The coordinates are The bone density value of the pixel point.

[0028] In this embodiment, the bone material properties (elastic modulus) of each finite element are determined using the pixel values ​​(HU values) of the CT scan image, enabling patient-specific material assignment and improving the accuracy of finite element analysis. By accounting for the influence of bone density (HU values), it is possible to distinguish between different degrees of osteoporosis and assign different elastic moduli to bones of varying densities. The sigmoid function ensures a continuous relationship between the elastic modulus and bone density, avoiding abrupt material property changes between regions of varying bone density, thereby improving the stability of the finite element analysis.

[0029] Step S102: discretize the patient-specific three-dimensional spinal model into a tetrahedral finite element mesh, assign bone material properties to each finite element according to the pixel value of the CT scan image, and assign soft tissue material properties to the intervertebral disc and ligament soft tissue structures.

[0030] In this embodiment, the pixel values ​​of CT scan images are directly converted into bone material properties (such as elastic modulus) within finite element models, more accurately reflecting the density distribution and mechanical properties of the individual patient's bones, thereby improving the accuracy of vertebroplasty simulations. Traditional finite element models typically use uniform or partitioned material properties, which cannot capture subtle density differences within the bone. Assigning material properties to the intervertebral disc and ligament soft tissue allows for a more comprehensive assessment of the impact of vertebroplasty on the overall stability and biomechanics of the spine. The finite element model forms the foundation for subsequent optimization of the bone cement injection path and injection volume. Only by establishing a solid foundation based on the distribution of material properties can the optimal treatment plan truly suited to the patient be determined.

[0031] Step S103, determining an optimal bone cement injection path of the tetrahedral finite element mesh by a density method, wherein the objective function of the optimal bone cement injection path is to minimize the maximum principal stress of the target vertebra, and the constraints include the bone cement injection volume and the stability of the target vertebra.

[0032] In this embodiment, the density method automatically searches for the optimal bone cement injection path, reducing manual intervention, minimizing the maximum principal stress of the target vertebra, and reducing the risk of vertebral fractures. By optimizing the bone cement injection path, stress distribution within the vertebra is more uniform, avoiding stress concentration. By constraining the bone cement injection volume and vertebral stability, stress is reduced without overusing bone cement or compromising overall vertebral stability. Furthermore, the injection path best suited to the patient's individual anatomy and bone density is identified, enabling personalized treatment.

[0033] Specifically, a density variable is assigned to each tetrahedral finite element mesh, and a finite element analysis is performed based on the current density distribution to calculate the maximum principal stress of the target vertebra. The sensitivity of the maximum principal stress to each density variable is calculated, and the density variable is updated based on the sensitivity. It is checked whether the constraints of bone cement volume and vertebral stability are met. If not, the density variable is adjusted. The distribution of the updated density variable is converted into an actual bone cement injection path to obtain the optimal bone cement injection path for the tetrahedral finite element mesh.

[0034] For example, initialize the density variable of all elements in a tetrahedral finite element mesh to 0.1. Apply a compressive load to the vertebral body and perform finite element analysis to determine the maximum principal stress of the target vertebral body to be 10 MPa. Calculate the sensitivity of the maximum principal stress to the density variable for each element (for example, the sensitivity of element A is -0.05 MPa / unit density, and the sensitivity of element B is -0.02 MPa / unit density). Update the density variable using gradient descent. If the step size is 0.1, the new density of element A is 0.1 + 0.1 × 0.05 = 0.105, and the new density of element B is 0.1 + 0.1 × 0.02 = 0.102. Calculate the bone cement volume for the current density distribution. Assume that the maximum allowable bone cement volume is 5 mL and the current volume is 4 mL, thus satisfying the volume constraint. Calculate the buckling load factor of the vertebral body. Assume that the minimum buckling load factor is set to 2.0 and the current value is 2.2, thus satisfying the stability constraint. Repeat the above steps until the maximum principal stress converges to a lower value. like Figure 4 As shown in the figure, the final density distribution was thresholded and the units (17 and 18) with density values ​​higher than 0.8 were considered as the bone cement injection area. This area was then smoothed to obtain the final bone cement injection path. The bone cement injection path optimized by the density method effectively reduces the maximum principal stress of the vertebral body, thereby reducing the risk of fracture and improving the efficacy of vertebroplasty, while satisfying the constraints of bone cement volume and vertebral stability.

[0035] Step S104 , determining the optimal bone cement injection amount of the optimal bone cement injection path with minimizing the maximum principal stress of the target vertebra, maximizing the stiffness of the target vertebra, and minimizing the stress variation of adjacent vertebrae as objective functions.

[0036] In this embodiment, minimizing stress changes in adjacent vertebrae reduces the risk of adjacent vertebral fractures caused by vertebroplasty. The injection of bone cement changes the stress distribution. If the stress changes in adjacent vertebrae are too large, it can easily lead to stress fractures in the vertebrae. Minimizing the principal stress and maximizing the stiffness simultaneously make the target vertebra more stable when bearing loads, with a more uniform stress distribution, thereby improving its fracture resistance. The amount of bone cement injected, obtained through multi-objective optimization, avoids excessive or insufficient injection, thereby improving the success rate of the operation and the quality of life of the patient.

[0037] For example, suppose the target vertebra is the L3 vertebra, and the injection path has been determined using the density method. The objective function constraints are: bone cement volume limit: 6 ml; vertebral compressive strength: >5 MPa; and stress concentration in adjacent vertebral endplates: <15 MPa. By simulating biological evolutionary processes (such as genetic evolution algorithms), the bone cement injection volume is continuously iterated and improved to find the optimal solution. In each iteration, the objective function values ​​corresponding to different injection volumes are evaluated, and selection, crossover, and mutation operations are performed based on fitness (the inverse of the objective function value). After multiple iterations, the optimal bone cement injection volume is found to be 4.5 ml. This injection volume minimizes the maximum principal stress and maximizes the stiffness of the L3 vertebra, while minimizing the stress change in the L2 and L4 vertebrae, while satisfying all constraints. The vertebral model after 4.5 ml of bone cement injection is visualized, showing the bone cement distribution, stress contours, and deformation maps. Based on these visualizations, the physician can assess the treatment outcome and make necessary adjustments.

[0038] In an optional manner, the sensitivity formula of the objective function of minimizing the maximum principal stress of the target vertebral body is: in, Design variables for the unit; is the unit volume; is the von Mises equivalent stress; is the strain tensor; is the fourth-order elasticity tensor; is the spatial area of ​​the unit; is the maximum principal stress; It is a double dot product, used for tensor operations.

[0039] In this embodiment, by introducing the von Mises equivalent stress ( ) and the maximum principal stress ( ), can more accurately evaluate the effect of bone cement injection on vertebral stress distribution, which helps to improve the success rate of vertebroplasty and the postoperative rehabilitation effect of patients. By calculating each unit design variable ( ) sensitivity to the maximum principal stress, the bone cement injection path and injection volume can be quickly adjusted to achieve the optimal stress distribution.

[0040] Step S105, based on the optimal bone cement injection path and the optimal bone cement injection amount, simulate vertebroplasty in the tetrahedral finite element mesh, predict the stress distribution of the vertebral body, the vertebral body stiffness and the stress changes of the adjacent vertebral bodies after surgery, and present the simulation results of the simulated vertebroplasty in a three-dimensional visual form, wherein the simulation results include bone cement distribution, stress cloud map and deformation map.

[0041] In this embodiment, the simulation process takes into account various physical phenomena, such as the non-Newtonian fluid properties of bone cement, the exothermic process of the hydration reaction, the curing behavior, and the impact of bone cement injection on the internal microenvironment of the vertebral body. The simulation results are closer to the actual situation. Therefore, by predicting the stress distribution of the vertebral body after bone cement injection, the stiffness recovery, and the stress changes of adjacent vertebrae, it helps doctors evaluate the treatment effect and further optimize the bone cement injection path and injection volume based on the simulation results to achieve the best treatment effect. The simulation results such as bone cement distribution, stress cloud map, and deformation map are presented through three-dimensional visualization, allowing doctors to more intuitively understand the condition of the vertebral body after surgery.

[0042] Specifically, the optimized bone cement injection path is converted into a coordinate sequence in three-dimensional space to determine the specific location for bone cement injection. The volume fraction of bone cement within each target cell is calculated based on the optimal bone cement injection volume to simulate the filling ratio during the injection process. A progressive filling method is used to simulate the injection process, gradually increasing the bone cement volume fraction within the target cell until the calculated value is reached. The material properties (such as the elastic modulus) of the cells filled with bone cement are linearly transitioned from the bone tissue value to the cement material value to reflect the reinforcing effect of bone cement. During the bone cement flow phase, the Carreau model is used to describe the non-Newtonian fluid properties of bone cement and simulate the flow and diffusion of bone cement within the vertebral body. The Lagrangian-Eulerian method is used to handle mesh deformation, and the temperature field distribution is calculated based on the nonlinear transient heat conduction equation to simulate the exothermic process of the bone cement hydration reaction. A viscoelastic or viscoplastic constitutive model is used to represent the curing behavior of bone cement. The constitutive model includes factors affecting the mechanical properties of bone cement, including the degree of bone cement hydration reaction, temperature, and strain rate. Stiffness degradation is initiated when the element equivalent plastic strain exceeds a preset threshold to simulate vertebral damage. By coupling the curing behavior of bone cement with the effects of cement injection on the vertebral microenvironment (e.g., changes in bone marrow pressure, stress and strain in osteocytes, and microfractures), a biological model was developed to predict the remodeling and long-term stability of vertebral bone tissue after cement injection. The stiffness recovery rate was calculated by comparing the postoperative vertebral stiffness matrix with the preoperative stiffness to evaluate the reinforcing effect of bone cement. The pressure increment of adjacent intervertebral discs and the maximum shear strain of the annulus fibrosus were calculated to assess the mechanical impact of cement injection on adjacent vertebrae. To determine the optimal cement injection volume, vertebral compressive strength, fatigue life, and stress concentration in adjacent vertebral endplates were used as constraints in the optimization objective function. A particle system was used to simulate the cement diffusion process to simulate vertebroplasty. Particle color mapped the injection time and displayed the order and extent of cement filling. Stress distribution within the vertebral body and adjacent vertebrae was visualized using stress contours, and the cement distribution was overlaid on a transparency channel to facilitate observation of the effect of cement on vertebral stress.

[0043] For example, a 70-year-old female patient suffered from L1 vertebral compression fracture due to osteoporosis and needed vertebroplasty. According to the above optimization method, the optimal bone cement injection path was determined to be two channels entering the vertebral body from the left and right posterior sides, and the optimal injection volume was 4 ml. The injection path was converted into a spatial coordinate sequence to determine the injection position. The progressive filling method was used to simulate the gradual injection process of bone cement and update the material properties. The hydration reaction exothermic and curing process of bone cement was simulated. The simulation results showed that the maximum principal stress of the L1 vertebra was significantly reduced after surgery, the stiffness recovery rate reached 80%, the pressure increment of adjacent intervertebral discs was within an acceptable range, and the maximum shear strain of the annulus fibrosus did not exceed the threshold. Figure 3 As shown in the figure, the visualization results show the distribution of bone cement inside the vertebral body, and the stress cloud map shows that the bone cement effectively disperses the stress of the vertebral body. Among them, the iliac crest 1 is used for surface positioning, the anterior superior iliac spine 2 is the protrusion of the front of the ilium, the posterior superior iliac spine 3 is the protrusion of the back of the ilium, located above the sacroiliac joint, the ischial tuberosity 4 is the protrusion of the ischium, the pubic symphysis 5 is the part where the two pubic bones are connected by fibrocartilage, the sacrum 6 is located at the bottom of the spine, formed by the fusion of five sacral vertebrae, and is a common puncture point for epidural anesthesia, the sacral angle (8, 9) is associated with the coccyx, the iliac crest 10 is a bony landmark used for surface positioning, the anterior superior iliac spine (11, 12, 13) is used for surface positioning, and the posterior superior iliac spine (14-1, 14-2) is located above the sacroiliac joint. The ischial tuberosity 15 is the main point that bears weight when sitting, and the pubic symphysis (7, 16) is the part connected by fibrocartilage. Through the simulation results, the doctor confirmed that the injection scheme can effectively enhance the stability of the vertebral body and has little effect on the adjacent vertebrae.

[0044] In an optional manner, the method further includes: When converting the density variable distribution into the bone cement injection path, the density field is smoothed and threshold segmented to extract the bone cement injection channel; Eliminate sharp corners and thin connections in the geometry of the injection channel to reduce the risk of bone cement leakage; The manufacturability of the optimized bone cement injection path was evaluated based on the accessibility of the puncture needle, the range of injection angles, and the performance of the bone cement injection device.

[0045] In this example, sharp corners and narrow connections are potential risk points for bone cement leakage. By smoothing the density field to remove these sharp corners and narrow connections, the extracted injection path is clearer and leakage into the spinal canal or surrounding soft tissue is avoided. The optimized injection path is evaluated based on the needle's accessibility, injection angle range, and the performance limitations of the bone cement injection device, avoiding the dilemma of a theoretically optimal solution being unattainable in practice.

[0046] For example, suppose the bone cement injection path optimized using the density method forms a "Y"-shaped structure within the vertebral body. However, from a posterolateral approach, the angle at which the needle reaches the two branches of this "Y" is too steep, making the surgical procedure difficult. A reachability analysis of the needle reveals that the current approach cannot safely reach the ends of both branches. By adjusting the optimization algorithm to add constraints on needle reachability, the generated injection path is more consistent with the actual surgical procedure (for example, adjusting the puncture entry point to a transpedicular approach to achieve a better angle).

[0047] According to the solution provided by the present invention, a CT scan image and an X-ray scan image containing a target vertebra and its adjacent vertebrae are obtained; the CT scan image and the X-ray scan image are automatically segmented to extract the vertebral bone structure and the vertebral soft tissue structure respectively; a patient-specific three-dimensional spine model is constructed based on the vertebral bone structure and the vertebral soft tissue structure; the patient-specific three-dimensional spine model is discretized into a tetrahedral finite element mesh, and bone material properties are assigned to each finite element unit according to the pixel value of the CT scan image, and soft tissue material properties are assigned to the intervertebral disc and ligament soft tissue structure; the optimal bone cement injection path of the tetrahedral finite element mesh is determined by the density method, wherein the optimal bone cement The objective function of the injection path is to minimize the maximum principal stress of the target vertebra, and the constraints include the bone cement injection volume and the stability of the target vertebra; the optimal bone cement injection amount of the optimal bone cement injection path is determined by minimizing the maximum principal stress of the target vertebra, maximizing the stiffness of the target vertebra, and minimizing the stress change of the adjacent vertebrae as the objective function; according to the optimal bone cement injection path and the optimal bone cement injection amount, vertebroplasty is simulated in the tetrahedral finite element mesh to predict the stress distribution of the postoperative vertebra, the vertebral stiffness, and the stress change of the adjacent vertebrae after surgery, and the simulation results of the simulated vertebroplasty are presented in a three-dimensional visualization form, and the simulation results include bone cement distribution, stress cloud map, and deformation map. The present invention realizes personalized optimization of vertebroplasty and predicts preoperative simulation by constructing a patient-specific finite element model, providing the surgeon with an efficient auxiliary reference surgical plan. Specifically, a patient-specific three-dimensional spine model and a finite element model are constructed using the patient's own CT scan images and X-ray scan images, which more accurately reflect the patient's spinal structure and mechanical properties, and have high clinical application value. By assigning bone material properties to each finite element based on the pixel values ​​of the CT scan image and incorporating differences in bone density, the finite element model more realistically reflects the mechanical properties of the bone. Furthermore, soft tissue properties are assigned to the intervertebral disc and ligament soft tissue structures to ensure the accuracy of the overall mechanical properties of the spine. The optimal bone cement injection path is determined using a density method, with minimizing the maximum principal stress of the target vertebra as the optimization goal. This effectively reduces the risk of postoperative vertebral fractures and improves surgical safety. By minimizing the maximum principal stress of the target vertebra, maximizing the stiffness of the target vertebra, and minimizing the stress variation of adjacent vertebrae, a more comprehensive approach is used to find the optimal bone cement injection volume, thereby improving the stability of the target vertebra while reducing the risk of re-fracture of adjacent vertebrae. Vertebroplasty is simulated on a tetrahedral finite element mesh to predict the postoperative stress distribution, vertebral stiffness, and stress variations of adjacent vertebrae. The simulation results are presented in a 3D visualization, allowing for preoperative prediction of surgical outcomes and adjustment of surgical plans based on the simulation results, reducing surgical risks and improving both success rates and outcomes.

[0048] Figure 5The schematic diagram of the framework of the vertebroplasty optimization device based on the patient-specific finite element model according to an embodiment of the present invention is shown. The vertebroplasty optimization device based on the patient-specific finite element model comprises: The spine model construction module 510 is configured to obtain a CT scan image and an X-ray scan image containing a target vertebra and its adjacent vertebrae; automatically segment the CT scan image and the X-ray scan image to extract the vertebral bone structure and the vertebral soft tissue structure; and construct a patient-specific three-dimensional spine model based on the vertebral bone structure and the vertebral soft tissue structure. a finite element model preprocessing module 520 for discretizing the patient-specific three-dimensional spinal model into a tetrahedral finite element mesh, assigning bone material properties to each finite element based on pixel values ​​of the CT scan image, and assigning soft tissue material properties to the intervertebral disc and ligament soft tissue structures; an injection path optimization module 530 for determining an optimal bone cement injection path of the tetrahedral finite element mesh by a density method, wherein the objective function of the optimal bone cement injection path is to minimize the maximum principal stress of the target vertebral body, and the constraints include the bone cement injection volume and the stability of the target vertebral body; An injection volume optimization module 540 is configured to determine an optimal bone cement injection volume for the optimal bone cement injection path, using minimizing the maximum principal stress of the target vertebra, maximizing the stiffness of the target vertebra, and minimizing the stress variation of adjacent vertebrae as objective functions; The vertebroplasty simulation module 550 is used to simulate vertebroplasty in the tetrahedral finite element mesh according to the optimal bone cement injection path and the optimal bone cement injection amount, predict the stress distribution of the vertebral body, the vertebral body stiffness and the stress changes of the adjacent vertebral bodies after surgery, and present the simulation results of the simulated vertebroplasty in a three-dimensional visual form. The simulation results include bone cement distribution, stress cloud map and deformation map.

[0049] Figure 6 The schematic diagram of the structure of the computing device embodiment of the present invention is shown. The specific embodiment of the present invention does not limit the specific implementation of the computing device.

[0050] like Figure 6 As shown, the computing device may include: a processor (processor) 602 , a communications interface (Communications Interface) 604 , a memory (memory) 606 , and a communication bus 608 .

[0051] Processor 602, communication interface 604, and memory 606 communicate with each other via a communication bus 608. Communication interface 604 is used to communicate with other devices, such as clients or other server network elements. Processor 602 is used to execute program 610, which may specifically perform the steps described in the embodiment of the vertebroplasty optimization method based on a patient-specific finite element model.

[0052] Specifically, the program 610 may include program codes, which include computer operation instructions.

[0053] Processor 602 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention. The one or more processors included in a computing device may be of the same type, such as one or more CPUs, or may be of different types, such as one or more CPUs and one or more ASICs.

[0054] The memory 606 is used to store the program 610. The memory 606 may include a high-speed RAM memory, or may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0055] According to the solution provided by the present invention, a CT scan image and an X-ray scan image containing a target vertebra and its adjacent vertebrae are obtained; the CT scan image and the X-ray scan image are automatically segmented to extract the vertebral bone structure and the vertebral soft tissue structure respectively; a patient-specific three-dimensional spine model is constructed based on the vertebral bone structure and the vertebral soft tissue structure; the patient-specific three-dimensional spine model is discretized into a tetrahedral finite element mesh, and bone material properties are assigned to each finite element unit according to the pixel value of the CT scan image, and soft tissue material properties are assigned to the intervertebral disc and ligament soft tissue structure; the optimal bone cement injection path of the tetrahedral finite element mesh is determined by the density method, wherein the optimal bone cement The objective function of the injection path is to minimize the maximum principal stress of the target vertebra, and the constraints include the bone cement injection volume and the stability of the target vertebra; the optimal bone cement injection amount of the optimal bone cement injection path is determined by minimizing the maximum principal stress of the target vertebra, maximizing the stiffness of the target vertebra, and minimizing the stress change of the adjacent vertebrae as the objective function; according to the optimal bone cement injection path and the optimal bone cement injection amount, vertebroplasty is simulated in the tetrahedral finite element mesh to predict the stress distribution of the postoperative vertebra, the vertebral stiffness, and the stress change of the adjacent vertebrae after surgery, and the simulation results of the simulated vertebroplasty are presented in a three-dimensional visualization form, and the simulation results include bone cement distribution, stress cloud map, and deformation map. The present invention realizes personalized optimization of vertebroplasty and predicts preoperative simulation by constructing a patient-specific finite element model, providing the surgeon with an efficient auxiliary reference surgical plan. Specifically, a patient-specific three-dimensional spine model and a finite element model are constructed using the patient's own CT scan images and X-ray scan images, which more accurately reflect the patient's spinal structure and mechanical properties, and have high clinical application value. By assigning bone material properties to each finite element based on the pixel values ​​of the CT scan image and incorporating differences in bone density, the finite element model more realistically reflects the mechanical properties of the bone. Furthermore, soft tissue properties are assigned to the intervertebral disc and ligament soft tissue structures to ensure the accuracy of the overall mechanical properties of the spine. The optimal bone cement injection path is determined using a density method, with minimizing the maximum principal stress of the target vertebra as the optimization goal. This effectively reduces the risk of postoperative vertebral fractures and improves surgical safety. By minimizing the maximum principal stress of the target vertebra, maximizing the stiffness of the target vertebra, and minimizing the stress variation of adjacent vertebrae, a more comprehensive approach is used to find the optimal bone cement injection volume, thereby improving the stability of the target vertebra while reducing the risk of re-fracture of adjacent vertebrae. Vertebroplasty is simulated on a tetrahedral finite element mesh to predict the postoperative stress distribution, vertebral stiffness, and stress variations of adjacent vertebrae. The simulation results are presented in a 3D visualization, allowing for preoperative prediction of surgical outcomes and adjustment of surgical plans based on the simulation results, reducing surgical risks and improving both success rates and outcomes.

[0056] Those skilled in the art will appreciate that the modules in the devices of the embodiments can be adaptively modified and deployed in one or more devices different from the embodiments. Modules, units, or components in the embodiments can be combined into a single module, unit, or component, and furthermore, they can be divided into multiple sub-modules, sub-units, or sub-components. All features disclosed in this specification (including the accompanying claims, abstract, and drawings), as well as all processes or units of any method or device disclosed therein, can be combined in any combination, unless at least some of such features and / or processes or units are mutually exclusive. Unless expressly stated otherwise, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) may be replaced by an alternative feature serving the same, equivalent, or similar purpose. Furthermore, those skilled in the art will appreciate that although some embodiments herein include certain features and not others included in other embodiments, combinations of features from different embodiments are intended to fall within the scope of the present invention and form different embodiments. For example, in the claims below, any of the claimed embodiments may be used in any combination. The present invention may be implemented by means of hardware comprising several distinct elements and by means of a suitably programmed computer. In a unit claim that lists several means, several of these means may be embodied by the same hardware item. Unless otherwise specified, the steps in the above embodiments should not be understood as limiting the order of execution.

Claims

1. A method for optimizing vertebroplasty based on a patient-specific finite element model, characterized in that: include: Acquiring a CT scan image and an X-ray scan image containing a target vertebra and its adjacent vertebrae; automatically segmenting the CT scan image and the X-ray scan image to extract the vertebral bone structure and vertebral soft tissue structure, respectively; and constructing a patient-specific three-dimensional spinal model based on the vertebral bone structure and vertebral soft tissue structure; discretizing the patient-specific three-dimensional spine model into a tetrahedral finite element mesh, assigning bone material properties to each finite element based on pixel values ​​of the CT scan image, and assigning soft tissue material properties to intervertebral disc and ligament soft tissue structures; Determining an optimal bone cement injection path of the tetrahedral finite element mesh by a density method, wherein the objective function of the optimal bone cement injection path is to minimize the maximum principal stress of the target vertebral body, and the constraints include the bone cement injection volume and the stability of the target vertebral body; Determining an optimal bone cement injection amount for the optimal bone cement injection path by taking minimizing the maximum principal stress of the target vertebra, maximizing the stiffness of the target vertebra, and minimizing the stress change of adjacent vertebrae as objective functions; According to the optimal bone cement injection path and the optimal bone cement injection amount, vertebroplasty is simulated in the tetrahedral finite element grid, the stress distribution of the vertebral body, the vertebral body stiffness and the stress changes of the adjacent vertebral bodies after surgery are predicted, and the simulation results of the simulated vertebroplasty are presented in a three-dimensional visualization form. The simulation results include bone cement distribution, stress cloud map and deformation map.

2. The method for optimizing vertebroplasty based on a patient-specific finite element model according to claim 1, wherein: The expression for the mapping relationship between the pixel value of the CT scan image and the bone material properties is: in, For coordinates The elastic modulus of the bone material; It is the lower limit of elastic modulus of cancellous bone; is the upper limit of elastic modulus of cortical bone; is the curvature adjustment coefficient, ; is the bone density value corresponding to the turning point of bone density; The coordinates are The bone density value of the pixel point.

3. The method for optimizing vertebroplasty based on a patient-specific finite element model according to claim 1, wherein: The sensitivity formula of the objective function of minimizing the maximum principal stress of the target vertebra is: in, Design variables for the unit; is the unit volume; is the von Mises equivalent stress; is the strain tensor; is the fourth-order elasticity tensor; is the spatial area of ​​the unit; is the maximum principal stress; It is a double dot product, used for tensor operations.

4. The method for optimizing vertebroplasty based on a patient-specific finite element model according to claim 1, wherein: The constructing of a patient-specific three-dimensional spine model based on the vertebral bone structure and the vertebral soft tissue structure further comprises: Performing surface reconstruction on the voxel data of the vertebral bone structure and the vertebral soft tissue structure according to the Marching Cubes algorithm to generate an initial surface mesh; The surface mesh data is smoothed and mesh simplified by a Laplacian Smoothing algorithm to obtain a surface mesh; A tetrahedral mesh model that meets the finite element analysis requirements of the surface mesh is generated using the tetrahedral mesh generation tool TetGen to obtain a patient-specific three-dimensional spinal finite element model.

5. The method for optimizing vertebroplasty based on a patient-specific finite element model according to claim 1, wherein: Determining the optimal bone cement injection path of the tetrahedral finite element grid by the density method further comprises: Assign a density variable to each tetrahedral finite element mesh, perform finite element analysis based on the current density distribution, and calculate the maximum principal stress of the target vertebra; Calculating the sensitivity of the maximum principal stress to each density variable, updating the density variable according to the sensitivity, checking whether the constraints of bone cement volume and vertebral stability are met, and adjusting the density variable if not met; The updated distribution of the density variable is converted into an actual bone cement injection path to obtain the optimal bone cement injection path of the tetrahedral finite element mesh.

6. The method for optimizing vertebroplasty based on a patient-specific finite element model according to claim 1, wherein: According to the optimal bone cement injection path and the optimal bone cement injection amount, simulating vertebroplasty in the tetrahedral finite element grid further comprises: The optimal injection path is converted into a spatial coordinate sequence, and a target unit set is marked in a finite element grid; the volume fraction of bone cement in each target unit is calculated based on the optimal injection volume, and the injection process is simulated by a progressive filling method; the elastic modulus of the unit injected with bone cement is linearly transitioned from the bone tissue value to the cement material value; The temperature field distribution is calculated based on the nonlinear transient heat conduction equation to simulate the exothermic process of bone cement hydration reaction. The Carreau model is used to describe the non-Newtonian fluid properties of bone cement during the flow phase, and the mesh deformation is handled according to the Lagrangian-Eulerian method. Stiffness degradation is initiated when the element equivalent plastic strain exceeds a preset threshold. The postoperative vertebral stiffness matrix was compared with the preoperative stiffness to obtain the stiffness recovery rate, and the pressure increment of adjacent intervertebral discs and the maximum shear strain of the annulus fibrosus were calculated. A particle system was used to simulate the cement diffusion process to simulate vertebroplasty, where the particle color mapped the injection time and the transparency channel of the stress contour was overlaid to show the bone cement distribution.

7. The method for optimizing vertebroplasty based on a patient-specific finite element model according to claim 1, wherein: During the simulated vertebroplasty procedure, the method further comprises: The curing behavior of bone cement is represented by a viscoelastic or viscoplastic constitutive model, wherein the constitutive model includes factors affecting the mechanical properties of bone cement such as the degree of hydration reaction of bone cement, temperature, and strain rate; The curing behavior of bone cement and the effects of bone cement injection on the internal microenvironment of the vertebral body were injected into the coupled biological model to predict the changes in bone tissue remodeling and long-term stability after bone cement injection. The effects of the internal microenvironment included changes in bone marrow pressure, stress and strain changes in bone cells, and microfractures. When determining the optimal bone cement injection volume, the vertebral compressive strength, fatigue life, and stress concentration of adjacent vertebral endplates were used as constraints of the optimization objective function.

8. The method for optimizing vertebroplasty based on a patient-specific finite element model according to claim 1, wherein: The method further comprises: When converting the density variable distribution into the bone cement injection path, the density field is smoothed and threshold segmented to extract the bone cement injection channel; Eliminate sharp corners and thin connections in the geometry of the injection channel to reduce the risk of bone cement leakage; The manufacturability of the optimized bone cement injection path was evaluated based on the accessibility of the puncture needle, the range of injection angles, and the performance of the bone cement injection device.

9. A vertebroplasty optimization device based on a patient-specific finite element model, characterized in that: Implementing the vertebroplasty optimization method based on a patient-specific finite element model as described in any one of claims 1 to 8, comprising: A spine model construction module is configured to obtain a CT scan image and an X-ray scan image containing a target vertebra and its adjacent vertebrae; automatically segment the CT scan image and the X-ray scan image to extract the vertebral bone structure and vertebral soft tissue structure, respectively; and construct a patient-specific three-dimensional spine model based on the vertebral bone structure and vertebral soft tissue structure. a finite element model preprocessing module for discretizing the patient-specific three-dimensional spinal model into a tetrahedral finite element mesh, assigning bone material properties to each finite element based on pixel values ​​of the CT scan image, and assigning soft tissue material properties to the intervertebral disc and ligament soft tissue structures; an injection path optimization module, configured to determine an optimal bone cement injection path of the tetrahedral finite element mesh by a density method, wherein the objective function of the optimal bone cement injection path is to minimize the maximum principal stress of the target vertebral body, and the constraints include the bone cement injection volume and the stability of the target vertebral body; an injection volume optimization module, configured to determine an optimal bone cement injection volume for the optimal bone cement injection path, taking minimizing the maximum principal stress of the target vertebra, maximizing the stiffness of the target vertebra, and minimizing the stress variation of adjacent vertebrae as objective functions; A vertebroplasty simulation module is used to simulate vertebroplasty in the tetrahedral finite element grid according to the optimal bone cement injection path and the optimal bone cement injection amount, predict the stress distribution of the vertebral body, the vertebral body stiffness and the stress changes of the adjacent vertebral bodies after surgery, and present the simulation results of the simulated vertebroplasty in a three-dimensional visual form, wherein the simulation results include bone cement distribution, stress cloud map and deformation map.

10. A computing device comprising: A processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the vertebroplasty optimization method based on a patient-specific finite element model according to any one of claims 1-8.

Images