An orthopedic implant lattice design method, system and application

Abstract

The present application relates to the technical field of prosthesis lattice design, in particular to a kind of orthopedic implant lattice design method, system and application, by establishing parameterized lattice model sample set, and the polynomial regression model of elastic modulus and permeability coefficient is combined with the bayesian optimization framework with comprehensive score objective function as optimization goal, so that it can be iteratively optimized in parameter range efficiently, finally output optimal design parameter to make the mechanical property and mass transfer performance reach the best balance, fundamentally solve the problem that existing PEEK lattice implant design lacks double performance collaborative optimization, further improve lattice prosthesis postoperative recovery effect.

Description

Technical Field

[0001] This invention relates to the field of prosthesis lattice design technology, specifically to a method, system, and application for lattice design of orthopedic implants. Background Technology

[0002] Polyetheretherketone (PEEK), a high-performance thermoplastic engineering plastic, has been widely used in orthopedic implants, such as spinal fusion devices and artificial joint prostheses, due to its excellent biocompatibility, chemical resistance, radiation permeability, and elastic modulus similar to human bone. However, PEEK material itself is bioinert, and its dense structure makes it difficult to achieve ideal mechanical conduction and material exchange with bone tissue, which can easily lead to complications such as stress shielding at the implant-bone interface, bone resorption, and long-term loosening.

[0003] Triple Periodic Minimal Surface (TPMS) is a type of surface structure that is periodic in three independent directions and has an average curvature of zero, such as Diamond and Gyroid types. Compared with traditional pillar lattice structures, TPMS structures have advantages such as fully connected pores, high specific surface area, and no stress concentration cusps, which are more conducive to bone tissue ingrowth and body fluid transport. In recent years, TPMS porous structures have been widely regarded as the ideal bone replacement support structure for orthopedic implants.

[0004] However, orthopedic implants simultaneously perform the dual functions of transmitting mechanical loads and transporting metabolic substances from bone tissue within the body, as referenced... Figure 1 and Figure 2 The range of lattice structure parameters was divided into four levels by progressively increasing the range of parameters. The elastic modulus and permeability coefficients corresponding to different levels were obtained. Different design parameters correspond to different topological forms of the lattice structure, and their corresponding elastic modulus and permeability coefficients are also different. There are mutually reinforcing and mutually weakening relationships among the three parameters of the lattice structure. However, the existing TPMS lattice structure design for PEEK materials focuses on the influence law analysis of a single performance parameter and lacks the synergistic optimization design of mechanical properties and mass transfer properties, which leads to poor postoperative recovery. Summary of the Invention

[0005] The purpose of this invention is to provide a method, system and application for lattice design of orthopedic implants, and to solve the technical problem of poor performance of existing lattice prosthesis designs.

[0006] The solution of the present invention to the above-mentioned technical problems is as follows: A method for designing crystal lattices for orthopedic implants, comprising the following steps: Determine the crystal lattice structure of orthopedic implants and set the parameter range for crystal lattice structure design parameters; Based on the parameter range of the design parameters, different lattice structure models are established, and the elastic modulus and permeability coefficient of each lattice structure model are obtained through finite element simulation to obtain a model sample set. A multinomial regression model for predicting elastic modulus and permeability coefficient was constructed based on the model sample set. A comprehensive scoring objective function is constructed based on the aforementioned multinomial regression model combined with adjustable weights; Using the comprehensive scoring objective function as the optimization objective, a Gaussian process is used as the probabilistic surrogate model, and the expected improvement is used as the acquisition function. Bayesian multi-objective optimization iteration is performed until the convergence condition is met, and the optimal design parameters of the lattice structure model are obtained.

[0007] Further specifying, the lattice structure is a Diamond-type TPMS lattice topology, and the design parameters include wall thickness, mid-plane offset, and number of unit cells.

[0008] Further defining the implicit equation for the surface reference of the Diamond-type TPMS lattice topology is:

[0009] in, , and For the three-dimensional coordinates of the unit cell, For cell size, This is the mid-plane offset. The wall thickness is [not specified].

[0010] Further specifying, the acquisition of the elastic modulus and permeability coefficient of each lattice structure model through finite element simulation specifically involves: The elastic modulus of the crystal structure model was obtained through quasi-static compression simulation. The permeability coefficient of the crystal structure model is obtained through fluid simulation.

[0011] Further specifying, the construction of the polynomial regression model for predicting the elastic modulus and permeability coefficient based on the model sample set includes the following steps: Design parameters Convert to second-order polynomial eigenvectors :

[0012] in, , For wall thickness, This is the mid-plane offset. This represents the number of unit cell arrays; Construct a polynomial regression model to predict the elastic modulus and permeability coefficient:

[0013]

[0014] in, For the predicted elastic modulus, The elastic modulus prediction coefficient; The predicted permeability coefficient, For permeability coefficient prediction; Design parameters in the model sample set Substituting the corresponding elastic modulus and permeability coefficient into the polynomial regression model, we obtain... and The specific value.

[0015] Further specifying, the comprehensive scoring objective function is:

[0016]

[0017] in, As the weight for the elastic modulus score, As the weight for the penetration coefficient score, and All settings were determined based on the patient's bone quality; This is the normalization function.

[0018] Further defining the optimization objective as the comprehensive scoring objective function, using a Gaussian process as the probabilistic surrogate model, and employing the expected improvement as the acquisition function, Bayesian multi-objective optimization iterations are performed until the convergence condition is met. The optimal design parameters for this lattice structure model are obtained through the following steps: S1. Based on the parameter range of the design parameters and in conjunction with the polynomial regression model, calculate the comprehensive score of each lattice structure model to obtain the sample dataset. :

[0019] in, For the first Group design parameters; S2, using the current sample dataset To train the data, construct a probabilistic surrogate model for a Gaussian process. S3. Based on the parameter range of the design parameters, set multiple optimization design parameters. Substituting the values ​​into the Gaussian process probabilistic surrogate model yields the corresponding posterior distribution of the comprehensive score. ,in, To find the optimal design parameters The posterior mean, To find the optimal design parameters The posterior variance; S4. Based on the posterior distribution of the comprehensive score The expected improvement acquisition function is used to calculate each optimization design parameter. Expected improvement Solve Maximum Optimization Design Parameters As the first Next iteration sample design parameters ; S5. Calculate the sample design parameters. Overall rating And add to the sample dataset For the sample dataset renew; S6. Determine if the iterative convergence condition is met. If yes, proceed to step S7; otherwise, let... Repeat steps S2 to S6; S7. From the current sample dataset The design parameters corresponding to the highest comprehensive score are selected as the optimal design parameters.

[0020] Further specifying, the statement based on the posterior distribution of the comprehensive score... The expected improvement acquisition function is used to calculate each optimization design parameter. Expected improvement Specifically:

[0021] in, The cumulative distribution function of the standard normal distribution. Let be the probability density function. To find the optimal design parameters Overall score To find the optimal design parameters The posterior standard deviation.

[0022] A lattice design system for orthopedic implants, comprising: The parameter setting module is used to determine the crystal lattice structure of orthopedic implants and set the parameter range of the crystal lattice structure design parameters. The model sample set construction module is used to establish different lattice structure models based on the parameter range of the design parameters, and obtain the elastic modulus and permeability coefficient of each lattice structure model through finite element simulation to obtain the model sample set; The multinomial regression model building module is used to build multinomial regression models for predicting elastic modulus and permeability coefficient based on the model sample set. The comprehensive scoring objective function construction module is used to construct a comprehensive scoring objective function based on the multinomial regression model combined with adjustable weights. The optimal design parameter acquisition module is used to perform Bayesian multi-objective optimization iteration with the comprehensive scoring objective function as the optimization objective, Gaussian process as the probabilistic surrogate model, and expected improvement as the acquisition function, until the convergence condition is met, so as to obtain the optimal design parameters of the lattice structure model.

[0023] An orthopedic implant, wherein the interior and / or surface of the orthopedic implant is filled with a porous lattice structure, wherein the porous lattice structure is an orthopedic implant lattice structure constructed with optimal design parameters obtained by the above-mentioned orthopedic implant lattice design method, and is integrally formed by additive manufacturing process.

[0024] The beneficial effects of this invention are as follows: 1. This invention establishes a model sample set of orthopedic implant lattice structures and combines a multinomial regression model of elastic modulus and permeability coefficient with a Bayesian optimization framework with a comprehensive scoring objective function as the optimization target. This enables efficient iterative optimization within the design parameter range, ultimately outputting the optimal design parameters that achieve the best balance between mechanical properties and mass transfer properties. This fundamentally solves the problem of the lack of synergistic optimization of dual properties in existing PEEK lattice implant designs, further improving the postoperative recovery effect of lattice prostheses.

[0025] 2. This invention employs a computationally efficient multinomial regression model, which greatly reduces simulation costs and improves model stability and anti-interference capabilities under small sample conditions. By combining a Gaussian process with a Bayesian optimization strategy that aims to improve the acquisition function, it can intelligently balance exploration and utilization, and converge to the global optimum quickly with fewer iterations, thereby reducing design difficulty and improving design efficiency. Attached Figure Description

[0026] Figure 1 A graph showing the relationship between lattice structure design parameters and elastic model; Figure 2 A graph showing the relationship between lattice structure design parameters and permeability coefficient; Figure 3 This is a step diagram of the orthopedic implant lattice design method of the present invention; Figure 4 This is the Bayesian optimization convergence curve of the present invention; Figure 5 This is a schematic diagram of the dual-objective optimization results of the lattice structure of the present invention; Figure 6 This is a diagram of the lattice topology corresponding to the optimal design parameters of this invention; Figure 7 This is a displacement cloud diagram of the lattice topology corresponding to the optimal design parameters of this invention; Figure 8 This is a simulation stress distribution diagram of the lattice topology corresponding to the optimal design parameters of this invention. Figure 9 The fluid simulation diagram of the lattice topology corresponding to the optimal design parameters of this invention. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0028] Example 1 refer to Figure 3 This invention provides a method for designing crystal lattices for orthopedic implants, comprising the following steps: Determine the crystal lattice structure of orthopedic implants and set the parameter range for crystal lattice structure design parameters; Different lattice structure models were established based on the parameter range of the design parameters, and the elastic modulus and permeability coefficient of each lattice structure model were obtained through finite element simulation to obtain the model sample set. A multinomial regression model for predicting elastic modulus and permeability coefficient was constructed based on the model sample set. A comprehensive scoring objective function is constructed based on a multinomial regression model combined with adjustable weights; Using the comprehensive scoring objective function as the optimization objective, a Gaussian process as the probabilistic surrogate model, and the expected improvement as the acquisition function, Bayesian multi-objective optimization iterations are performed until the convergence condition is met, thus obtaining the optimal design parameters of the lattice structure model.

[0029] Specifically, since the diamond-type structure has a more uniform stress distribution and higher load-bearing efficiency, the preferred crystal structure is the diamond-type TPMS crystal topology. The design parameters of the crystal structure include wall thickness, mid-plane offset, and number of unit cells; among which, the wall thickness ranges from 0 to 0.4 mm, the mid-plane offset ranges from 0.1 mm to 0.4 mm, and the number of unit cells ranges from 1 to 5.

[0030] To further explain, the surface-based implicit equation for the Diamond-type TPMS lattice topology is:

[0031] in, , and For the three-dimensional coordinates of the unit cell, For cell size, This is the mid-plane offset. The wall thickness is [not specified].

[0032] To further explain, different lattice structure models are established based on the parameter range of the design parameters, and the elastic modulus and permeability coefficient of each lattice structure model are obtained through finite element simulation to obtain the model sample set. Specifically, multiple sets of original design parameters are preset based on the parameter range of the design parameters. Taking 16 sets of original design parameters as an example, refer to Table 1, so that 16 lattice structure models can be established.

[0033] Based on 16 crystal structure models, the corresponding elastic modulus and permeability coefficients were obtained through finite element simulation. The obtained elastic modulus and permeability coefficients are all simulation values. The model sample set is summarized as shown in Table 1. Table 1 Model Sample Set

[0034] The elastic modulus was calculated by extracting the force-displacement curve through quasi-static compression simulation. :

[0035] in, Stiffness, i.e., the slope of the curve, is expressed in N / mm. The initial length of the crystal lattice model is in mm; The initial cross-sectional area of ​​the crystal lattice structure model is in mm. 2 .

[0036] The lattice structure model was simulated using fluid dynamics. The fluid was set to an inlet velocity of 0.01 m / s, a boundary condition of 0.02 m / s, and a free outlet. A laminar flow model was used for steady-state simulation to extract the inlet pressure drop data. Darcy's law was then used to convert the pressure drop into the permeability coefficient. :

[0037] in, For pressure drop, Permeability coefficient, The viscosity is 1.005 mPa·s. Input speed, value 0.01 m / s; The length of the crystal lattice structure model is in mm.

[0038] To further explain, constructing a multinomial regression model for predicting the elastic modulus and permeability coefficient based on the model sample set includes the following steps: Design parameters Convert to second-order polynomial eigenvectors To capture nonlinear relationships, square terms and cross terms are introduced:

[0039] in, , For wall thickness, This is the mid-plane offset. This represents the number of unit cell arrays; Based on 16 sets of original design parameters in the model sample set and the elastic modulus and permeability coefficient obtained from simulation, a polynomial regression model capable of predicting the elastic modulus and permeability coefficient is constructed:

[0040]

[0041] in, For the predicted elastic modulus, The predictor coefficient for the elastic modulus; The predicted permeability coefficient, This is a predicted coefficient for the permeability coefficient; Design parameters in the model sample set Substituting the corresponding elastic modulus and permeability coefficient into the polynomial regression model, we obtain... and The specific value.

[0042] Based on 16 sets of original design parameters in the model sample set Substituting the elastic modulus and permeability coefficient obtained from the corresponding simulation into the polynomial regression model, we obtain the predicted permeability coefficient. and elastic modulus prediction coefficient The specific values ​​enable the prediction values ​​of the elastic modulus and permeability coefficient of the corresponding design parameters to be calculated based on any set of design parameters within the parameter range.

[0043] To further explain, the overall scoring objective function is:

[0044]

[0045] in, As the weight for the elastic modulus score, As the weight for the penetration coefficient score, and All settings are based on the patient's bone quality. For example, the weight of the elastic modulus score can be selected as 60% for patients with poor bone quality, and the weight of the permeability coefficient score can be selected as 55% for patients with normal bone quality. , Let's take an example to illustrate; This is a normalization function used to normalize the obtained values ​​respectively. and Map to the interval [0, 1].

[0046] To further explain, using the comprehensive scoring objective function as the optimization objective, employing a Gaussian process as the probabilistic surrogate model, and using the expected improvement as the acquisition function, Bayesian multi-objective optimization iterations are performed until the convergence condition is met. The optimal design parameters for this lattice structure model are obtained through the following steps: S1. Based on the parameter range of the design parameters and combined with a multinomial regression model, calculate the comprehensive score of each lattice structure model to obtain the sample dataset. :

[0047] in, For the first Group design parameters, refer to Table 1. That is, the first The original design parameters were set at this time. The value ranges from 1 to 16.

[0048] S2, using the current sample dataset To train the data, construct a probabilistic surrogate model for a Gaussian process. S3. Set multiple optimization design parameters based on the parameter range of the design parameters. Substituting the values ​​into the Gaussian process probabilistic surrogate model yields the corresponding posterior distribution of the comprehensive score. ,in, To find the optimal design parameters The posterior mean, To find the optimal design parameters The posterior variance; Among them, the Gaussian Process (GP) is based on the current sample dataset. Probabilistic modeling is used to evaluate the comprehensive score of any new design parameter, providing a data foundation for decision-making.

[0049] At this point, the optimal design parameters are sought. A parameter range space is constructed based on three dimensions: wall thickness, mid-plane offset, and number of unit cells, along with the ranges of each design parameter. Optimized design parameters are then obtained at any spatial location within this parameter range space. The optimization design parameters Substituting into the currently constructed Gaussian process probabilistic surrogate model, we obtain the posterior distribution of the comprehensive score. ; Optimize design parameters The number of values ​​must be at least two, and usually more, so that the current step will yield multiple corresponding posterior distributions of the comprehensive score. .

[0050] S4. Based on the posterior distribution of the comprehensive score The Expected Improvement (EI) acquisition function is used to calculate each optimization design parameter. Expected improvement Solve Maximum Optimization Design Parameters As the first Next iteration sample design parameters ; Specifically, calculate the expected improvement value of each design parameter. :

[0051] in, The cumulative distribution function of the standard normal distribution. Let be the probability density function. To find the optimal design parameters Overall score To find the optimal design parameters The posterior standard deviation.

[0052] At this point, the posterior distribution of multiple comprehensive scores is used. Obtain the corresponding multiple expected improvement values Subsequently, by optimizing the operator Find the value that increases the expectation Maximum Optimization Design Parameters , For ease of recording, the obtained expected increase value will be... Maximum Optimization Design Parameters Let these be the sample design parameters. During the first iteration, we obtain the value that increases the expected value. Largest sample design parameters .

[0053] S5. Calculate the sample design parameters Overall rating And add to the sample dataset For the sample dataset renew; Specifically, based on the sample design parameters wall thickness Mid-surface offset and cell array number The corresponding comprehensive score is calculated using the comprehensive scoring objective function. The obtained sample design parameters will then be... and overall score Substitute the sample dataset In the middle, referring to Table 1, the sample dataset after the first iteration The design parameters include 16 sets of original design parameters and 1 set of sample design parameters.

[0054] S6. Determine if the iterative convergence condition is met. If yes, proceed to step S7; otherwise, let... Repeat steps S2 to S6; For details, please refer to Figure 4 It can be seen that the highest comprehensive score no longer changes after more than 21 iterations. Therefore, to reduce computational difficulty and improve computational efficiency, the convergence condition is that the number of iterations reaches 21. When the number of iterations is satisfied, the sample dataset... The design parameters include 16 sets of original design parameters and 21 sets of sample design parameters.

[0055] Preferably, in order to improve the accuracy of the optimal design parameters, the following steps are included before performing step S7: Multiple random design parameters can be arbitrarily selected based on the parameter range. With 8 random design parameters Let's take an example to illustrate; calculate the random design parameters respectively. Corresponding overall score Then, the obtained 8 random design parameters and the corresponding 8 comprehensive scores Add to sample dataset For the sample dataset The update ensures that, after satisfying the iterative convergence condition, the final sample dataset... It includes 16 sets of original design parameters, 21 sets of sample design parameters, and 8 random design parameters.

[0056] S7. From the current sample dataset Select the design parameter corresponding to the highest comprehensive score as the optimal design parameter; Specifically, from the final sample dataset The design parameter corresponding to the highest comprehensive score is selected as the optimal design parameter. The optimal design parameter can be the original design parameter, the sample design parameter, or the random design parameter.

[0057] Through actual calculations, the optimal design parameters were found to be one of the sample design parameters: wall thickness of 0.15 mm, mid-surface offset of 0, and number of unit cells of 5; (Reference) Figure 5 Sixteen sets of original design parameters were used as the original samples, and the obtained optimal design parameters were added. The optimal design parameters can achieve the highest level of permeability performance in the model sample set while ensuring excellent mechanical load-bearing capacity. This breaks the bottleneck of the binary opposition between mechanics and permeability, and significantly improves the channel efficiency of bone tissue ingrowth while meeting the clinical mechanical standards of orthopedic implants.

[0058] The optimal design parameters correspond to the Diamond-type TPMS lattice topology as follows: Figure 6 As shown; displacement contour plots were obtained by simulating the lattice topology, with reference to... Figure 7 It can be seen that the force transmission path of this lattice topology is continuous and the deformation is uniform, the stress distribution is uniform and the compressive strength is strong, thereby improving the postoperative recovery effect.

[0059] For further explanation, please refer to Figure 8 Quasi-static compression simulation yielded an elastic modulus of 1152.79 MPa for the Diamond-type TPMS lattice topology corresponding to the optimal design parameters. The transition from high-stress nodes to low-stress porosity regions is continuous and smooth without sharp stress jumps, and the stress distribution exhibits no obvious discontinuities or localized weak areas. (Reference) Figure 9 The highest inlet pressure is approximately 3.25 Pa, the stable outlet pressure is approximately 0 Pa, and the total pressure drop is... The extremely low pressure drop level indicates minimal fluid flow resistance within the structure, resulting in excellent permeability and improved postoperative bone ingrowth. Fluid simulation revealed that the permeability coefficient of this lattice topology is 12.08 × 10⁻⁶. -8 m 2 It is suitable for the elastic modulus of human cancellous bone: 0.1GPa~1.5GPa, and the permeability coefficient is 0.4m. 2 ~15×10 -8 m 2 The orthopedic implant lattice model constructed by the optimal design parameters obtained through the orthopedic implant lattice design method provided in this embodiment is filled into the interior and / or surface of the orthopedic implant and integrally formed by additive manufacturing process. The resulting orthopedic implant replaces the human cancellous bone structure, perfectly adapts to the clinical needs of orthopedic implants, and achieves a synergistic match between mechanical support capacity and bone tissue fluid permeability.

[0060] Substituting the optimal design parameters into the constructed polynomial regression model, the predicted elastic modulus was calculated to be 1180 MPa with an error of 2.4%; the predicted permeability coefficient was 11.86 × 10⁻⁶. -8 m 2The error is 1.8%, meaning the polynomial regression model meets the computational requirements.

[0061] Example 2 Based on Example 1, this embodiment provides an orthopedic implant lattice design system, including: The parameter setting module is used to determine the crystal lattice structure of orthopedic implants and set the parameter range of the crystal lattice structure design parameters. The model sample set construction module is used to build different lattice structure models based on the parameter range of the design parameters, and obtain the elastic modulus and permeability coefficient of each lattice structure model through finite element simulation to obtain the model sample set; The multinomial regression model building module is used to build multinomial regression models for predicting elastic modulus and permeability coefficient based on the model sample set. The module for constructing the comprehensive scoring objective function is used to construct a comprehensive scoring objective function based on a multinomial regression model combined with adjustable weights. The optimal design parameter acquisition module is used to perform Bayesian multi-objective optimization iteration with the comprehensive scoring objective function as the optimization objective, Gaussian process as the probabilistic surrogate model, and expected improvement as the acquisition function, until the convergence condition is met, so as to obtain the optimal design parameters of the lattice structure model.

[0062] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the scope of the invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0063] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can be appropriately combined to form other embodiments that can be understood by those skilled in the art. The above content is only for illustrating the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made based on the technical concept proposed in this invention shall fall within the scope of protection of the claims of this invention.

[0064] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for designing crystal lattice for orthopedic implants, characterized in that, Includes the following steps: Determine the crystal lattice structure of orthopedic implants and set the parameter range for crystal lattice structure design parameters; Based on the parameter range of the design parameters, different lattice structure models are established, and the elastic modulus and permeability coefficient of each lattice structure model are obtained through finite element simulation to obtain a model sample set. A multinomial regression model for predicting elastic modulus and permeability coefficient was constructed based on the model sample set. A comprehensive scoring objective function is constructed based on the aforementioned multinomial regression model combined with adjustable weights; Using the comprehensive scoring objective function as the optimization objective, a Gaussian process is used as the probabilistic surrogate model, and the expected improvement is used as the acquisition function. Bayesian multi-objective optimization iteration is performed until the convergence condition is met, and the optimal design parameters of the lattice structure model are obtained.

2. The orthopedic implant lattice design method according to claim 1, characterized in that, The lattice structure is a Diamond-type TPMS lattice topology, and the design parameters include wall thickness, mid-plane offset, and number of unit cells.

3. The orthopedic implant lattice design method according to claim 2, characterized in that, The surface-based implicit equation for the Diamond-type TPMS lattice topology is: in, , and For the three-dimensional coordinates of the unit cell, For cell size, This is the mid-plane offset. The wall thickness is [not specified].

4. The orthopedic implant lattice design method according to claim 1, characterized in that, The specific steps for obtaining the elastic modulus and permeability coefficient of each lattice structure model through finite element simulation are as follows: The elastic modulus of the crystal structure model was obtained through quasi-static compression simulation. The permeability coefficient of the crystal structure model is obtained through fluid simulation.

5. The orthopedic implant lattice design method according to claim 1, characterized in that, The construction of the polynomial regression model for predicting the elastic modulus and permeability coefficient based on the model sample set includes the following steps: Design parameters Convert to second-order polynomial eigenvectors : in, , For wall thickness, This is the mid-plane offset. This represents the number of unit cell arrays; Construct a polynomial regression model to predict the elastic modulus and permeability coefficient: in, For the predicted elastic modulus, The elastic modulus prediction coefficient; The predicted permeability coefficient, For permeability coefficient prediction; Design parameters in the model sample set Substituting the corresponding elastic modulus and permeability coefficient into the polynomial regression model, we obtain... and The specific value.

6. The orthopedic implant lattice design method according to claim 5, characterized in that, The comprehensive scoring objective function is: in, As the weight for the elastic modulus score, As the weight for the penetration coefficient score, and All settings were determined based on the patient's bone quality; This is the normalization function.

7. The orthopedic implant lattice design method according to claim 6, characterized in that, Using the comprehensive scoring objective function as the optimization objective, a Gaussian process as the probabilistic surrogate model, and expected improvement as the acquisition function, Bayesian multi-objective optimization iterations are performed until the convergence condition is met. The optimal design parameters of the lattice structure model are obtained by the following steps: S1. Based on the parameter range of the design parameters and in conjunction with the polynomial regression model, calculate the comprehensive score of each lattice structure model to obtain the sample dataset. : in, For the first Group design parameters; S2, using the current sample dataset To train the data, construct a probabilistic surrogate model for a Gaussian process. S3. Based on the parameter range of the design parameters, set multiple optimization design parameters. Substituting the values ​​into the Gaussian process probabilistic surrogate model yields the corresponding posterior distribution of the comprehensive score. ,in, To find the optimal design parameters The posterior mean, To find the optimal design parameters The posterior variance; S4. Based on the posterior distribution of the comprehensive score The expected improvement acquisition function is used to calculate each optimization design parameter. Expected improvement Solve Maximum Optimization Design Parameters As the first Next iteration sample design parameters ; S5. Calculate the sample design parameters. Overall rating And add to the sample dataset For the sample dataset renew; S6. Determine if the iterative convergence condition is met. If yes, proceed to step S7; otherwise, let... Repeat steps S2 to S6; S7. From the current sample dataset The design parameters corresponding to the highest comprehensive score are selected as the optimal design parameters.

8. The orthopedic implant lattice design method according to claim 7, characterized in that, The posterior distribution based on comprehensive scoring The expected improvement acquisition function is used to calculate each optimization design parameter. Expected improvement Specifically: in, The cumulative distribution function of the standard normal distribution. Let be the probability density function. To find the optimal design parameters Overall score To find the optimal design parameters The posterior standard deviation.

9. A lattice design system for orthopedic implants, characterized in that, include: The parameter setting module is used to determine the crystal lattice structure of orthopedic implants and set the parameter range of the crystal lattice structure design parameters. The model sample set construction module is used to establish different lattice structure models based on the parameter range of the design parameters, and obtain the elastic modulus and permeability coefficient of each lattice structure model through finite element simulation to obtain the model sample set; The multinomial regression model building module is used to build multinomial regression models for predicting elastic modulus and permeability coefficient based on the model sample set. The comprehensive scoring objective function construction module is used to construct a comprehensive scoring objective function based on the multinomial regression model combined with adjustable weights. The optimal design parameter acquisition module is used to perform Bayesian multi-objective optimization iteration with the comprehensive scoring objective function as the optimization objective, Gaussian process as the probabilistic surrogate model, and expected improvement as the acquisition function, until the convergence condition is met, so as to obtain the optimal design parameters of the lattice structure model.

10. An orthopedic implant, characterized in that, The orthopedic implant is filled with a porous lattice structure inside and / or on the surface. The porous lattice structure is an orthopedic implant lattice structure constructed with the optimal design parameters obtained by the orthopedic implant lattice design method according to any one of claims 1 to 8, and is integrally formed by additive manufacturing process.

Images