A 3D printing trajectory control method for bone joint surface reconstruction

Abstract

The present invention relates to a 3D printing trajectory control method for bone joint surface reconstruction. First, an algorithm based on deep learning is used to identify and classify the damage in the three-dimensional model of bone joints obtained by medical imaging technology, and a repair path is generated according to the type and severity of the damage; wherein the algorithm based on deep learning includes a convolutional neural network (CNN) architecture, which identifies and classifies bone joint damage from the three-dimensional model and evaluates the type and severity of the damage; according to the type and severity of the damage, the 3D printing trajectory is designed using image processing and computational geometry technology, and the repair path is optimized using a dynamic programming method; then the 3D printing path is optimized using a bionic algorithm, the print head speed and movement trajectory are adjusted in real time, and the material deposition is controlled and optimized; finally, an intelligent algorithm is developed to select printing materials and parameters, and a database is established to recommend the best materials and printing parameters according to the repair requirements.

Description

Technical Field

[0001] The invention relates to a 3D printing trajectory control method for reconstruction, and in particular to a 3D printing trajectory control method for reconstruction of a bone joint surface. Background Art

[0002] When analyzing the existing 3D printing trajectory control methods for bone joint surface reconstruction, such as the technology described in Chinese invention patent No. 2016103730876, it can be seen that it has many advantages, such as using CT / MRI tomographic image data of the patient's limbs for three-dimensional reconstruction, using reverse engineering technology and computer three-dimensional space registration technology to accurately obtain the preoperative defect area morphology, and the high matching of personalized implants and direct molding through 3D printing. However, although these technologies are quite mature in many aspects, there are still some potential defects and limitations: First, regarding the accuracy of data acquisition and processing, although CT / MRI provides a reliable data source, these medical image data need to be processed complexly before they can be converted into models suitable for 3D printing. This process involves a large number of data parsing, filtering and conversion steps, and each step will introduce errors. For example, the resolution and contrast of the image data and the slight movement of the patient during the scanning process will affect the accuracy of the final model. In addition, medical image data usually requires professional radiologists and engineers to interpret and process, which not only increases the time cost, but also may cause differences due to the subjective judgment of professionals. Secondly, although reverse engineering and computer 3D spatial registration technology can improve the design accuracy of prosthetic implants, these technologies often rely on advanced software and hardware support, which means high technical equipment investment and operating costs, especially in primary medical institutions with limited resources, and the application of this technology may be limited.

[0003] Thirdly, although personalized implant design can provide extremely high anatomical matching, highly personalized design also means that each implant requires an independent design and printing process, which is a considerable challenge in terms of production efficiency. The implant design of each patient needs to be carried out separately, which is not only time-consuming, but also difficult to achieve the economic benefits of mass production. In addition, highly personalized designs also require multiple adjustments and tests to ensure the best matching and functionality of the implant, which further increases the complexity and cost of production. In addition, although 3D printing provides the possibility of manufacturing implants with complex structures, the types and properties of printing materials are still one of the limiting factors. Although there are many 3D printing materials available for biomedical use on the market, the mechanical properties, biocompatibility, and long-term biostability of each material still need to be further studied and verified. The properties of these materials directly affect the safety and effectiveness of the implant, so the selection and verification of suitable printing materials is an issue that cannot be ignored. Finally, although 3D printing technology provides unprecedented flexibility for the manufacture of complex implant structures, precision control in the printing process, optimization of support structures, and post-processing processes are also technical difficulties that need to be overcome in the implementation of the technology. The precision in the printing process directly affects the surface smoothness, structural strength, and final functional performance of the implant. In addition, post-processing processes such as polishing, disinfection and sterilization are also crucial to the safety and effectiveness of implants. The optimization of these steps also requires a lot of technical research and experience accumulation. Summary of the invention

[0004] The purpose of the present invention is to provide a 3D printing trajectory control method for bone joint surface reconstruction, thereby solving some of the drawbacks and shortcomings pointed out in the background technology.

[0005] The present invention solves the above-mentioned technical problems by adopting the following technical solutions: firstly, an algorithm based on deep learning is used to identify and classify the damage in the three-dimensional model of the bone and joint obtained by medical imaging technology, and a repair path is generated according to the damage type and severity;

[0006] The deep learning-based algorithm includes a convolutional neural network (CNN) architecture, which identifies and classifies different types of bone and joint injuries including cracks, fractures, and fragments from a three-dimensional model generated by fusing multimodal medical imaging data including X-rays, CT, and MRI, and assesses the type and severity of the injury;

[0007] According to the type and severity of the injury, the 3D printing trajectory is designed using image processing and computational geometry techniques, and the repair path is optimized using dynamic programming methods, while controlling the balance between the mechanical strength and biocompatibility of the structure;

[0008] Then, the 3D printing path is optimized using a bionic algorithm to simulate the natural growth path of bones. By designing a path optimization model based on a genetic algorithm, the print head speed and movement trajectory are adjusted in real time during the 3D printing process to control and optimize material deposition.

[0009] Finally, an intelligent algorithm is developed to select printing materials and parameters based on bone and joint bearing capacity and biocompatibility requirements. By establishing a database containing a variety of biocompatible materials and their properties, the optimal materials and printing parameters are recommended according to the repair requirements to control the accuracy of the printing process and the quality of the repair.

[0010] Furthermore, the deep convolutional network (CNN) architecture is expressed by the function:

[0011]

[0012] To process 3D volume data and optimize the printing path; Represents a three-dimensional coordinate point, and the material deposition density at this coordinate point is given by Function control, where is the integration variable, represents the standard deviation of the normal distribution;

[0013] Then an algorithm is used to convert the damage classification information into 3D printing instructions. The algorithm is defined as:

[0014]

[0015] Converts injury information from medical imaging into printing instructions, where Indicates the severity of the injury. Represents the damaged area, The integral in the function expresses the change in energy distribution within the damage area at a given damage severity;

[0016] Finally, a real-time path optimization system is introduced to dynamically adjust the 3D printing parameters according to the damage characteristics identified from the 3D model, through the formula:

[0017]

[0018] in and Represents the adjustment factor, which is related to the printing speed and material flow rate. Indicates the number of levels to be adjusted. The summation and logarithmic parts of the formula calculate the cumulative effect of the adjustments at each level.

[0019] Furthermore, the process of designing the 3D printing trajectory is as follows:

[0020] S1. First, the image segmentation technology based on deep learning is used to extract the contour of the damaged area, and the damage type is identified by feature extraction technology. The composite function combining image segmentation and feature extraction is defined as:

[0021]

[0022] It is used to extract features from 3D image data and generate models suitable for 3D printing. represents the spatial coordinates in the image data, is the sigmoid function, which is used to enhance the sensitivity of feature extraction. is the threshold parameter, which is adjusted by To optimize the sensitivity of feature extraction;

[0023] S2. Then, parametric 3D reconstruction technology is used through the function:

[0024]

[0025] in represents the control parameters and variables in the model, the formula Through the Gaussian function and the sine function A combination of , providing dynamic geometric adjustment for different types and degrees of damage;

[0026] S3. Further, the dynamic programming algorithm is applied to optimize the printing path, through the integral formula:

[0027]

[0028] Calculate the optimal deposition path, where and Represents the factors that affect the printing trajectory, which are printing speed and material flow rate. The formula within Partially represents the variation and distribution of material deposition during the printing process;

[0029] S4. Finally, conduct simulation tests on mechanical strength and biocompatibility, applying the formula:

[0030]

[0031] To predict mechanical strength and biocompatibility, Represents material properties, formula In It is an adjustment factor that adjusts the overall strength model according to the characteristics of the material.

[0032] Furthermore, the method for optimizing the 3D printing path using a bionic algorithm includes:

[0033] First, a bionic algorithm is used to simulate the natural growth mechanism of human bones to generate a natural and biomechanical 3D printing path;

[0034] Then a path optimization model based on genetic algorithm was used to adapt to the complex shapes and irregular surfaces of bone joints;

[0035] Adjust the printing parameter system in real time according to the material deposition and inter-layer bonding quality monitored during the printing process;

[0036] Finally, the technology of controlling the material deposition process is used to integrate the rheological model for the properties of orthopedic materials and optimize the material deposition strategy for orthopedic applications.

[0037] Furthermore, the bionic algorithm is used to simulate the natural growth mechanism of human bones:

[0038] S1. First, the printing path generation module based on the bionic algorithm is adopted, and the composite functional analysis formula is used:

[0039]

[0040] Dynamically generate 3D printing paths that conform to bone growth mechanisms. is the spatial coordinate, indicating the position in three-dimensional space, represents the time variable, is a Gaussian function used to simulate the distribution of growth factors that diffuse outward from the center of the injury. Represents the standard deviation of the distribution and controls the breadth of the distribution;

[0041] S2. In addition, by combining finite element analysis FEA and computational fluid dynamics CFD, a calculation model including surface integral is used:

[0042]

[0043] to predict and verify the mechanical properties of the printed path, represents the force applied on the boundary, where represents the magnitude of stress, represents the mechanical coefficient of the material, is a function The derivative with respect to the boundary normal direction represents the rate of change of material displacement at the boundary, Indicated by the parameter Defined geometric boundaries, integral Along the border Integrate the above expression to calculate the total effect over the entire boundary.

[0044] Furthermore, the genetic algorithm-based path optimization model dynamically optimizes the 3D printing path through bionic simulation and mathematical modeling formulas:

[0045]

[0046] in is the spatial coordinate, representing the position in the 3D printing space, is the time variable, indicating the change of the function over time, is an infinite series, indicating that as the complexity of time and space variables increases, the path adjustment strategy becomes more and more sophisticated. It is a periodic function, which means that the path selection has periodic adjustment, which is related to the print head movement and material deposition cycle;

[0047] A geometric adaptability algorithm is then used to ensure that 3D printing can replicate the complex geometry and surface irregularities of the joint:

[0048]

[0049] and Represents parameters related to the geometry, is an exponential decay function, Emphasize The effect decreases rapidly with the increase of the material properties and the changes in the material state during the printing process.

[0050] Furthermore, the system for adjusting printing parameters in real time:

[0051] S1. First, a high-resolution sensing system is integrated into the 3D printer, including optical sensors, infrared cameras and pressure sensors, to monitor the material deposition of each layer and the interlayer adhesion quality in real time, using the formula:

[0052]

[0053] is the spatial coordinate, indicating the position of the print head in 3D space, is the time variable, the internal integral Indicates the time from the start of printing to The material deposition concentration;

[0054] S2. Apply data processing algorithms and machine learning techniques to analyze and identify printing defects and inconsistencies using the collected data. The prediction model is based on the formula:

[0055]

[0056] Optimize printing parameter settings, It is an adjustable parameter during the printing process. Represents the contribution and impact of different printing parameters on printing quality;

[0057] S3, based on the feedback provided by the machine learning model, adjust the printing speed, printing temperature, and nozzle movement rate in real time, and use a closed-loop feedback control system to ensure that the adjustment of printing parameters can be carried out continuously, using the formula:

[0058]

[0059] Indicates the parameters during printing. is with The coefficient of correlation.

[0060] Further, the controlling material deposition process comprises:

[0061] S1. Build a high-precision rheological model to simulate the behavior of biocompatible materials during 3D printing, and optimize the fluidity and curing time of the material according to the load and stress distribution unique to bone joints, using the formula:

[0062]

[0063] in Indicates the properties of the material or the position of the print head, is the time variable, the internal integral Describes the time-varying influence factor of the material fluidity, which is related to the viscosity or temperature of the material. represents how this fluidity is adjusted based on position or material properties;

[0064] S2. Then, the real-time material flow control system is used to adjust the printing parameters in real time to adapt to the areas with different bone density, optimize the smoothness of the joint surface and the accuracy of the joint part, through the formula:

[0065]

[0066] and Indicates the printing speed and the amount of extruded material. is the time variable, The expression in Simulate how printing parameters change dynamically over time and printing conditions;

[0067] S3. Finally, the interlayer bonding and structural integrity optimization technology is used to optimize the interlayer bonding to ensure that the printed structure can biomechanically withstand the pressure and movement in actual use. The formula is:

[0068]

[0069] represents the properties related to interlayer adhesion, represents the temperature, In Describe how different interlayer parameters affect the overall adhesion.

[0070] Furthermore, the control of the accuracy and repair quality during the printing process is achieved by: firstly, optimizing the 3D printing path by using data analysis and machine learning algorithms, and processing the bone structure data obtained from medical images by functions:

[0071]

[0072] represents the coordinates in three-dimensional space, is the exponential function of time, function Describes the attenuation effect from a certain point to the center point, which is related to the density of material deposition;

[0073] Next, develop a personalized printing material and parameter recommendation system to recommend the best printing material and parameters based on the specific bone and joint injury type and location, as well as the patient's physiological characteristics, using the formula:

[0074]

[0075] and Represent the elastic modulus and melting point of the material, respectively. quantify material strength and flexibility;

[0076] Finally, by establishing a database containing a variety of biocompatible materials and their properties, an intelligent matching algorithm is used, formula:

[0077]

[0078] Represents a metric related to material performance, Each of represents the contribution of material properties, Represents the complexity and variability of different attributes.

[0079] Beneficial effects of the present invention:

[0080] 1. By using the patient's own CT / MRI data, the present invention is able to create a highly personalized implant design. This method allows the implant to be highly matched to the patient's original joint surface in terms of anatomical structure and function, thereby providing a more natural joint movement and reducing friction between the implant and the native bone.

[0081] 2. By using reverse engineering and computer 3D spatial registration technology, the shape and size of the defect area can be accurately determined. Such accuracy ensures that the implant can perfectly adapt to the patient's joint defect and improve the success rate of the operation.

[0082] 3. Fast design and manufacturing process: Compared with traditional surgical implant production methods, 3D printing technology significantly shortens the time from design to manufacturing. Through direct printing, complex personalized implants can be manufactured from data obtained from medical imaging within a few hours, while traditional methods take several weeks.

[0083] 4. Reduce surgical risks and recovery time: Due to the high personalization and precise matching of the implant, the surgical process is simpler, which can reduce the risks during surgery and the postoperative recovery time. Patients can expect a faster recovery process and fewer postoperative complications. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] Figure 1 This is a flow chart of a 3D printing trajectory control method for bone joint surface reconstruction according to the present invention.

[0085] Figure 2 This is a module diagram of the deep convolutional network (CNN) architecture construction process of the present invention.

[0086] Figure 3 A 3D printing trajectory flow chart is designed for the present invention.

[0087] Figure 4 This is a flow chart of the method for optimizing the 3D printing path using a bionic algorithm according to the present invention.

[0088] Figure 5 This is a flow chart of the present invention using a bionic algorithm to simulate the natural growth mechanism of human bones. DETAILED DESCRIPTION

[0089] The specific implementation modes of the present invention will be described in detail below in conjunction with the accompanying drawings.

[0090] Combined with Figure 1 , a 3D printing trajectory control method for bone joint surface reconstruction includes the following steps: first, a convolutional neural network (CNN) is used to analyze and process detailed three-dimensional models obtained by fusion of multiple medical imaging sources such as X-rays, CT, and MRI. These models are enhanced by specific image processing algorithms to ensure high-resolution representation of the damaged area. CNN can identify various types of damage in these models, including cracks, fractures, and fragments, and classify and assess the severity of these damages.

[0091] After the deep learning algorithm identifies and classifies the injury, the next step is to use image processing and computational geometry techniques to accurately design the 3D printing trajectory for the repair. This includes applying dynamic programming methods to optimize the trajectory so that it not only follows the natural shape of the bone, but also takes into account the forces that must be carried and the necessary biocompatibility. The dynamic programming algorithm optimizes material usage and time efficiency by calculating the shortest path from the starting point to the target area, while ensuring that the newly generated bone joint part has sufficient structural strength and biological function.

[0092] In addition, to further improve printing accuracy and efficiency, the method also integrates bionic algorithms to simulate the natural growth path of bones. These algorithms refer to biological growth principles, including branching growth patterns, to optimize material deposition paths in a natural and uniform manner. The path optimization model based on genetic algorithms is used to adjust the print head speed and movement trajectory in real time. Such real-time adjustments dynamically optimize parameters based on feedback during the printing process to ensure the quality and accuracy of the final product.

[0093] Finally, the printing materials and parameters are selected through an intelligent algorithm, which is connected to a database that records the detailed properties of various biocompatible materials, including mechanical strength, elasticity, and biodegradation rate. The intelligent algorithm analyzes and recommends the most suitable materials and printing parameters based on specific bone and joint reconstruction needs, including load bearing capacity and biocompatibility. This is based not only on the physical and chemical properties of the material, but also takes into account the patient's specific medical and physiological needs to ensure the highest standards of printing process and results.

[0094] Embodiment 1:

[0095] Combined with Figure 2 In this embodiment, a deep convolutional network (CNN) architecture is used to process the three-dimensional model of bone joints obtained through medical imaging technology. A three-dimensional model is generated from the patient's X-ray, CT and MRI data to identify the damaged area. The CNN architecture is expressed by the function:

[0096]

[0097] To process 3D volume data and optimize the printing path. Represents a three-dimensional coordinate point, Represents the standard deviation of the normal distribution, set .

[0098] Set the three-dimensional coordinate point to , substitute into the formula to calculate:

[0099]

[0100] Substituting into the formula:

[0101]

[0102] Using numerical integration method to calculate, we get:

[0103]

[0104] Indicates that at the coordinate point ( ) has a material deposition density of 0.456.

[0105] The embodiment then converts the damage classification information into 3D printing instructions using the following algorithm:

[0106]

[0107] Indicates the severity of the injury. Represents the damaged area, The function is used to describe the change in energy distribution in the injury area under a given injury severity. Set the patient's joint injury severity 5 (value range is 1 to 10), the damaged area 2cm² (range 0.1 to 5cm²).

[0108] Substitute specific values:

[0109]

[0110] The part that calculates the integral:

[0111]

[0112] Assume that the numerical integration method (Simpson integration method) is used to obtain:

[0113]

[0114] Substitute into the function :

[0115]

[0116] This result indicates that when the damage severity is 5 and the damage area is 2cm², the energy distribution in the damage area changes by 1.635.

[0117] The last embodiment introduces a real-time path optimization system to dynamically adjust 3D printing parameters to ensure the accuracy of the printing process and the quality of the repair. The calculation is performed using the following formula:

[0118]

[0119] in, and Represents the adjustment factor, which is related to the printing speed and material flow rate. Indicates the number of levels to adjust.

[0120] In order to ensure the best material deposition and path for each layer during the printing process, a real-time path optimization system is used. During the printing process, the following adjustment factors and number of layers are determined after the damage features are identified through the 3D model:

[0121] Print speed adjustment factor =1.2 (value range is 1.0 to 2.0);

[0122] Material flow adjustment factor =1.1 (value range is 1.0 to 2.0);

[0123] Adjust the number of levels =5 (value range is 1 to 10).

[0124] Substitute specific values:

[0125]

[0126] Calculate the value of each term step by step:

[0127] for :

[0128]

[0129] for :

[0130]

[0131] for :

[0132]

[0133] for :

[0134]

[0135] for :

[0136]

[0137] Add these values ​​together:

[0138]

[0139] Based on the calculation results, the real-time path optimization system determines that the adjustment coefficient of printing speed and material flow is 0.9826. When printing each layer, the system will dynamically adjust the printing speed and material flow to meet this optimization coefficient, thereby ensuring uniform material deposition and strong adhesion between layers.

[0140] Set the initial printing speed of each layer to 20mm / s and the material flow rate to 1.5mm³ / s. After applying the optimization factor 0.9826:

[0141] New printing speed = 20mm / s×0.9826≈19.65mm / s

[0142] New material flow rate = 1.5 mm³ / s × 0.9826 ≈ 1.4739 mm³ / s

[0143] Make the printing process more precise, especially when dealing with complex geometries and joints that require high-strength bonding, ensuring the stability and durability of the printed structure.

[0144] Embodiment 2:

[0145] The embodiment uses deep learning-based image segmentation technology to extract the contour of the damaged area and uses feature extraction technology to identify the damage type. The specific composite function used is:

[0146]

[0147] in:

[0148] Represents the spatial coordinates in the image data;

[0149] It is the sigmoid function, which is used to enhance the sensitivity of feature extraction;

[0150] is the threshold parameter, which is adjusted by To optimize the sensitivity of feature extraction, set .

[0151] Set the three-dimensional coordinate point to be processed , calculate the eigenvalue of this point.

[0152] Calculate the distance from the origin:

[0153]

[0154] Substituting into the function:

[0155]

[0156] Calculate the value of each part:

[0157]

[0158]

[0159] Compute the value of a composite function:

[0160]

[0161] Through the above calculations, the characteristic value of each point can be extracted and the damage type and severity can be identified.

[0162] Next, the example uses parametric 3D reconstruction technology to accurately construct the 3D printing trajectory, using the function:

[0163]

[0164] in:

[0165] is the control parameter, which is related to the deposition rate and density of the material;

[0166] and are the variables in the model, which are respectively related to the spatial dimensions of the three-dimensional model;

[0167] function Through the Gaussian function and the sine function The combination provides dynamic geometric adjustment for different types and degrees of damage.

[0168] Setting the damage area requires special adjustments to match the natural shape of the bone. Selecting parameters , and The ranges are arrive Within this range, calculate The value of determines the specific adjustment of the print track.

[0169] set up ,calculate:

[0170]

[0171] Specific calculation:

[0172]

[0173]

[0174]

[0175] The results showed that The material deposition density at the coordinate point should be adjusted to 0.428 to ensure consistency with the surrounding structure and the stability of the overall structure. Using this method, the printing parameters of each layer can be dynamically adjusted throughout the 3D printing process to ensure that each part is optimized according to the severity and location of the damage.

[0176] Combined with Figure 3 , the embodiment further applies a dynamic programming algorithm to optimize the 3D printing path to ensure optimal material deposition and printing efficiency. The optimal deposition path is calculated by the integral formula:

[0177]

[0178] in:

[0179] and Represents the factors that affect the print trajectory, is the printing speed, the value range is 1.0 to 2.0, is the material flow rate, ranging from 1.0 to 2.0;

[0180] In the formula The section represents the variation and distribution of material deposition during the printing process.

[0181] set up , ,calculate:

[0182]

[0183] First calculate :

[0184]

[0185] Compute the integral:

[0186]

[0187] This integral is in standard form, representing the arc length integral from 0 to 1, and results in :

[0188]

[0189] therefore:

[0190]

[0191] According to the calculation results, the adjustment factor of the optimal deposition path is 1.79. This means that during the printing process, material deposition should be optimized according to this factor to ensure uniform distribution and optimal bonding.

[0192] The initial material flow rate is set to 1.5 mm³ / s, and the new material flow rate is obtained through the optimized adjustment coefficient:

[0193]

[0194] This adjustment ensures accurate deposition of material during printing, especially in complex geometries and areas requiring high-strength bonding. The printing speed and material flow rate are optimally adjusted through the path optimized by this dynamic programming algorithm.

[0195] The final example performs simulation tests on mechanical strength and biocompatibility to ensure the stability and safety of the printed structure, using the following formula to predict:

[0196]

[0197] in:

[0198] Indicates material properties, with a value range of 0 to 0.9;

[0199] In the formula It is an adjustment factor that adjusts the overall strength model according to the characteristics of the material.

[0200] The material used is set to have characteristics , calculate the adjustment factor:

[0201]

[0202] Calculate the integral part:

[0203]

[0204] By numerical integration method (Gaussian integration), we get:

[0205]

[0206] Substitute into the formula :

[0207]

[0208] According to the calculation results, the predicted value of mechanical strength and biocompatibility is 0.860, indicating that the printed bone joint structure has high reliability and safety in load-bearing capacity and biocompatibility.

[0209] Embodiment 3:

[0210] Combined with Figure 4 The embodiment adopts a printing path generation module based on a bionic algorithm, and the composite functional analysis formula used is as follows:

[0211]

[0212] in:

[0213] is the spatial coordinate, indicating the position in three-dimensional space;

[0214] is a time variable;

[0215] is a Gaussian function used to simulate the distribution of growth factors diffusing outward from the center of the injury;

[0216] Represents the standard deviation of the distribution, controls the breadth of the distribution, and ranges from 0.5 to 2.0.

[0217] set up The specific three-dimensional coordinate points are , time variable .

[0218] Calculate the Gaussian function part:

[0219]

[0220] Calculate the internal gradient function part:

[0221]

[0222] Compute the extrinsic gradient:

[0223]

[0224] therefore:

[0225]

[0226] Through the above calculations, the characteristic values ​​of each point in the printing path can be extracted. These characteristic values ​​are used to dynamically generate the 3D printing path to ensure that the path conforms to the natural growth mechanism of the bone.

[0227] The embodiment further combines finite element analysis FEA and computational fluid dynamics CFD to verify the mechanical properties of the printing path. The following calculation model is used:

[0228]

[0229] Parameter definition:

[0230] represents the force applied on the boundary;

[0231] Indicates the magnitude of stress;

[0232] Indicates the mechanical coefficient of the material, ranging from 0.5 to 2.0;

[0233] is a function The derivative with respect to the boundary normal direction represents the rate of change of material displacement at the boundary;

[0234] Indicated by the parameter Defined geometric boundaries, integral Along the border Integrate the above expression to calculate the total effect over the entire boundary.

[0235] Set the material parameters as: , , the geometric boundary defined is a circle with a radius of 1.

[0236] Compute the force applied on the boundary:

[0237]

[0238] Calculate the rate of change of material displacement at the boundary:

[0239] Set displacement function for ,in is the polar angle.

[0240]

[0241] Compute the surface integral:

[0242]

[0243] Setting Integration Boundaries is the circumference, then the integral result is:

[0244]

[0245] Specific calculation:

[0246]

[0247]

[0248] therefore,

[0249]

[0250] According to the calculation results, the mechanical performance index of the printing path is 3.08. This index is used to verify the stability and durability of the printed structure in actual use.

[0251] Embodiment 4:

[0252] Combined with Figure 5 The embodiment uses a path optimization model based on a genetic algorithm to dynamically optimize the 3D printing path through bionic simulation and mathematical modeling formulas to ensure the stability and biomechanical compatibility of the printed structure. The following formula is used:

[0253]

[0254] in:

[0255] is the spatial coordinate, indicating the position in the 3D printing space;

[0256] is the time variable, indicating the change of the function over time;

[0257] is an infinite series, indicating that as the complexity of time and space variables increases, the path adjustment strategy becomes more and more sophisticated;

[0258] It is a periodic function, which means that the path selection has periodic adjustment, which is related to the print head movement and material deposition cycle.

[0259] Set the specific three-dimensional coordinate point to and time variables .

[0260] Evaluate the first few terms of an infinite series to approximate the result:

[0261]

[0262] Calculate the value of each term:

[0263] Item 1:

[0264]

[0265] Item 2:

[0266]

[0267] Item 3:

[0268]

[0269] Add the values ​​of the first three items:

[0270]

[0271] By calculating the path optimization value, the printing path can be adjusted to make it more in line with the natural growth rules of bones.

[0272] The embodiment further ensures that 3D printing can replicate the complex geometry and surface irregularities of the joint through a geometric adaptability algorithm. The algorithm is as follows:

[0273]

[0274] in:

[0275] and Represents parameters related to geometry;

[0276] is an exponential decay function, Emphasize The increase of ,the influence decreases rapidly, which is related to the changes of the material properties and the material state during the printing process;

[0277] Reflects the complexity and irregularity of geometric shapes.

[0278] set up (value range is 0.5 to 3.0); (The value range is 0.1 to 2.0).

[0279] Calculate the integral part:

[0280]

[0281] Calculate the inner exponential decay function and sine function part:

[0282]

[0283]

[0284] The integral expression becomes:

[0285]

[0286] The calculation results are:

[0287]

[0288] The calculated geometric adaptability algorithm values ​​can be used to adjust the 3D printing path to better adapt to the complex geometry and surface irregularities of the joint.

[0289] Embodiment 5:

[0290] The embodiment integrates a high-resolution sensing system in the 3D printer, including an optical sensor, an infrared camera, and a pressure sensor, to monitor the material deposition of each layer and the interlayer bonding quality in real time to ensure the stability and functionality of the printed structure. The following formula is used to monitor and adjust the printing parameters:

[0291]

[0292] in:

[0293] is the spatial coordinate, indicating the position of the print head in 3D space;

[0294] is a time variable;

[0295] Indicates the time from the start of printing to The material deposition concentration.

[0296] Set the space coordinate point to:

[0297] (value range is 0.5 to 2.0);

[0298] (value range is 0.5 to 2.0);

[0299] (The value range is 0.5 to 2.0).

[0300] Set the current time variable is 2.0, and compute the integral inside:

[0301]

[0302] because is a constant and can be extracted:

[0303]

[0304] Compute the external derivative:

[0305]

[0306] The derivative of the material deposition concentration at the current time point is zero, indicating that the rate of change of material deposition at this time point is zero, indicating that the material has been fully deposited or has reached a state of equilibrium. The calculated results can be used to adjust the printing parameters in real time to ensure the uniformity of material deposition and the quality of interlayer adhesion.

[0307] The embodiment then applies data processing algorithms and machine learning techniques to analyze and identify printing defects and inconsistencies through the collected data, optimize the printing parameter settings, and adopt the following prediction model:

[0308]

[0309] in:

[0310] are the adjustable parameters during the printing process, which are printing speed, material flow rate and temperature;

[0311] Represents the contribution and impact of different printing parameters on print quality.

[0312] Set the initial values ​​of the following parameters collected:

[0313] (value range is 1.0 to 2.0);

[0314] (value range is 1.0 to 2.0);

[0315] (The value range is 0.5 to 1.5).

[0316] Evaluate the first few terms of an infinite series to approximate the result:

[0317]

[0318] Calculate the value of each term:

[0319] Item 1:

[0320]

[0321] Item 2:

[0322]

[0323] Item 3:

[0324]

[0325] Add the values ​​of the first three items:

[0326]

[0327] The calculated results can be used to optimize printing parameter settings and reduce printing defects and inconsistencies.

[0328] The embodiment further adjusts the printing parameters in real time based on the feedback provided by the machine learning model to ensure the continuity and high quality of the printing process. The following formula is used to adjust the printing speed, printing temperature and nozzle movement rate in real time:

[0329]

[0330] in:

[0331] Indicates the parameters in the printing process;

[0332] is with The coefficient of correlation ranges from 0.1 to 1.0;

[0333] Setting the print speed , selected is 0.5.

[0334] Compute the integral of the interior:

[0335]

[0336] Set the current printing speed , calculate the integral:

[0337]

[0338] By numerical integration method (Simpson integration method), the calculated result is approximately 0.782.

[0339] Calculation formula :

[0340]

[0341] Based on the calculated results, the printing parameters can be adjusted in real time to ensure the continuity and high quality of the printing process.

[0342] Embodiment 6:

[0343] The following formula is used to construct a high-precision rheological model to simulate the behavior of biocompatible materials during 3D printing:

[0344]

[0345] in:

[0346] Indicates the properties of the material or the position of the print head;

[0347] is a time variable;

[0348] Internal integration Describes the factors that influence the fluidity of materials over time, which are related to the viscosity or temperature of the material;

[0349] External partial derivatives Indicates how this fluidity is adjusted based on position or material properties.

[0350] set up (value range is 0.5 to 2.0); current time (value range is 0.1 to 2.0), calculate the internal integral:

[0351]

[0352] By numerical integration method (Simpson integration method), the calculated result is approximately 1.253.

[0353] Compute the external partial derivatives:

[0354]

[0355] set up :

[0356]

[0357] The calculated results can be used to optimize the material's fluidity and curing time, ensuring uniform deposition of each layer of material and good adhesion between layers.

[0358] The embodiment further utilizes a real-time material flow control system to adjust printing parameters in real time to adapt to areas with different bone densities and optimize the smoothness of the joint surface and the accuracy of the joint part, using the following formula:

[0359]

[0360] in:

[0361] Indicates the printing speed, the value range is 10 to 20 mm / s;

[0362] Indicates the amount of extruded material, ranging from 1.0 to 2.0 mm³ / s;

[0363] is a time variable;

[0364] expression Simulate how printing parameters change dynamically over time and printing conditions.

[0365] Set the current printing speed mm / s, amount of extruded material mm³ / s, and the current time .

[0366] Evaluate the inner expression:

[0367]

[0368] Compute the external derivative:

[0369]

[0370] The derivative result is zero, indicating that at the current time point When the printing speed and the amount of extruded material are balanced, no further adjustment is required. Through the real-time material flow control system, the printing parameters can be dynamically adjusted during the printing process to meet the needs of different bone density areas.

[0371] The final example uses interlayer bonding and structural integrity optimization technology to optimize interlayer bonding to ensure that the printed structure can biomechanically withstand the pressure and movement in actual use. The following formula is used to optimize interlayer bonding:

[0372]

[0373] in:

[0374] Represents temperature, ranging from 0.1 to 2.0;

[0375] expression Describe how different interlayer parameters affect the overall adhesion.

[0376] set up is 1.5, and the value of the interlayer adhesion optimization formula is calculated.

[0377] The part that computes the infinite series:

[0378]

[0379] set up , calculate the first 3 terms to approximate the result.

[0380] Item 1:

[0381]

[0382] Item 2:

[0383]

[0384] Item 3:

[0385]

[0386] Add the first three terms:

[0387]

[0388] Compute the external logarithmic function:

[0389]

[0390] Based on the calculated results, the interlayer bonding parameters were adjusted to optimize the overall adhesion and mechanical properties of the printed structure.

[0391] Embodiment 7:

[0392] The embodiment uses data analysis and machine learning algorithms to optimize the 3D printing path, and achieves printing accuracy and repair quality by processing bone structure data obtained from medical images, using the following formula:

[0393]

[0394] in:

[0395] Represents the coordinates in three-dimensional space, with a value range of -10 to 10 mm;

[0396] It is a time, ranging from 0 to 5s.

[0397] Exponential function in function Describes the attenuation effect from a point to a center point, which is related to the density of material deposition.

[0398] The setting is obtained from the CT scan specific points , current time s.

[0399] Evaluate the inner expression:

[0400]

[0401] Points calculation:

[0402]

[0403] Set current time seconds, the following integral needs to be calculated:

[0404]

[0405] Compute the integral:

[0406] Use Simpson's integral method to calculate the integral , using the following formula:

[0407]

[0408] in Is an even number.

[0409] The integration interval Divide The calculation is performed on subintervals.

[0410] Calculation step length

[0411]

[0412] Calculate the function value of each node

[0413]

[0414] Calculate each node Function value at:

[0415]

[0416]

[0417]

[0418]

[0419]

[0420] Substituting into Simpson's formula:

[0421]

[0422]

[0423]

[0424]

[0425]

[0426] Therefore, by using Simpson's integral method, we get:

[0427]

[0428] The calculated results can generate an optimized printing path to ensure uniform and precise material deposition for each layer.

[0429] The embodiment further develops a personalized printing material and parameter recommendation system to recommend the best printing material and parameters based on the specific bone and joint injury type and location, as well as the patient's physiological characteristics. The following formula is used:

[0430]

[0431] in:

[0432] and They represent the elastic modulus (MPa) and melting point (°C) of the material, with values ​​ranging from 1 to 10 and 100 to 300 respectively;

[0433] molecular Quantify the strength and flexibility of a material.

[0434] Set the selected materials and parameter metrics as follows:

[0435] Represents the elastic modulus of the material, with a value of 5Mpa;

[0436] Represents the melting point of the material, which is 200°C.

[0437] Calculating molecules :

[0438]

[0439] Calculate the denominator :

[0440]

[0441] Calculate recommended values :

[0442]

[0443] The calculation results can be used to recommend the best printing materials and parameters to ensure that the strength and flexibility of the printed structure meet the patient's needs.

[0444] The final embodiment establishes a database containing a variety of biocompatible materials and their properties, and uses an intelligent matching algorithm to select suitable materials, using the following algorithm:

[0445]

[0446] in:

[0447] Represents a metric related to material properties, which is the elastic modulus (in GPa), ranging from 0.5 GPa to 5 GPa;

[0448] Each term in the function represents the contribution of material properties, and Represents the complexity and variability of different attributes.

[0449] It is necessary to select a material suitable for manufacturing joint implants, and the elastic modulus of the material is known. GPa.

[0450] The first 5 items were selected to simplify calculations and improve efficiency:

[0451]

[0452] Calculate the values:

[0453]

[0454]

[0455]

[0456]

[0457] Through the above calculation results, the material with a score of 7.526 was selected, which reflects the optimal contribution of this material property to the production of shoulder implants.

[0458] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A 3D printing trajectory control method for bone joint surface reconstruction, characterized in that The following steps are involved: First, a deep learning-based algorithm is used to identify and classify injuries in the three-dimensional bone and joint models obtained through medical imaging technology, and a repair path is generated based on the injury type and severity. The deep learning-based algorithm includes a convolutional neural network (CNN) architecture, which identifies and classifies different types of bone and joint injuries including cracks, fractures, and fragments from a three-dimensional model generated by fusing multimodal medical imaging data including X-rays, CT, and MRI, and assesses the type and severity of the injury; According to the type and severity of the injury, the 3D printing trajectory is designed using image processing and computational geometry techniques, and the repair path is optimized using dynamic programming methods, while controlling the balance between the mechanical strength and biocompatibility of the structure; Then, the 3D printing path is optimized using a bionic algorithm to simulate the natural growth path of bones. By designing a path optimization model based on a genetic algorithm, the print head speed and movement trajectory are adjusted in real time during the 3D printing process to control and optimize material deposition. Finally, an intelligent algorithm is developed to select printing materials and parameters based on bone and joint bearing capacity and biocompatibility requirements. By establishing a database containing a variety of biocompatible materials and their properties, the optimal materials and printing parameters are recommended according to the repair requirements to control the accuracy of the printing process and the quality of the repair.

2. A bone joint surface reconstruction 3D printing trajectory control method according to claim 1, characterized in that The convolutional neural network CNN architecture is expressed by a function: , To process 3D volume data and optimize the printing path; Represents a three-dimensional coordinate point, and the material deposition density at this coordinate point is given by Function control, where is the integration variable, represents the standard deviation of the normal distribution; Then an algorithm is used to convert the damage classification information into 3D printing instructions. The algorithm is defined as: , Converts injury information from medical imaging into printing instructions, where Indicates the severity of the injury. Represents the damaged area, The integral in the function expresses the change in energy distribution within the damage area at a given damage severity; Finally, a real-time path optimization system is introduced to dynamically adjust the 3D printing parameters according to the damage characteristics identified from the 3D model, through the formula: , in and Represents the adjustment factor, which is related to the printing speed and material flow rate. Indicates the number of levels to be adjusted. The summation and logarithmic parts of the formula calculate the cumulative effect of the adjustments at each level.

3. A bone joint surface reconstruction 3D printing trajectory control method according to claim 1, characterized in that The process of designing the 3D printing trajectory is as follows: S1. First, the image segmentation technology based on deep learning is used to extract the contour of the damaged area, and the damage type is identified by feature extraction technology. The composite function combining image segmentation and feature extraction is defined as: , It is used to extract features from 3D image data and generate models suitable for 3D printing. represents the spatial coordinates in the image data, is the sigmoid function, which is used to enhance the sensitivity of feature extraction. is the threshold parameter, which is adjusted by To optimize the sensitivity of feature extraction; S2. Then, parametric 3D reconstruction technology is used through the function: , in represents the control parameters and variables in the model, the formula Through the Gaussian function and the sine function A combination of , providing dynamic geometric adjustment for different types and degrees of damage; S3. Further, the dynamic programming algorithm is applied to optimize the printing path, through the integral formula: , Calculate the optimal deposition path, where and Represents the factors that affect the printing trajectory, which are printing speed and material flow rate. The formula within Partially represents the variation and distribution of material deposition during the printing process; S4. Finally, conduct simulation tests on mechanical strength and biocompatibility, applying the formula: , To predict mechanical strength and biocompatibility, Represents material properties, formula In It is an adjustment factor that adjusts the overall strength model according to the characteristics of the material.

4. A bone joint surface reconstruction 3D printing trajectory control method according to claim 1, characterized in that The method for optimizing the 3D printing path using a bionic algorithm comprises: First, a bionic algorithm is used to simulate the natural growth mechanism of human bones to generate a natural and biomechanical 3D printing path; Then a path optimization model based on genetic algorithm was used to adapt to the complex shapes and irregular surfaces of bone joints; Adjust the printing parameter system in real time according to the material deposition and inter-layer bonding quality monitored during the printing process; Finally, the technology of controlling the material deposition process is used to integrate the rheological model for the properties of orthopedic materials and optimize the material deposition strategy for orthopedic applications.

5. A bone joint surface reconstruction 3D printing trajectory control method according to claim 4, characterized in that The bionic algorithm is used to simulate the natural growth mechanism of human bones: S1. First, the printing path generation module based on the bionic algorithm is adopted, and the composite functional analysis formula is used: , Dynamically generate 3D printing paths that conform to bone growth mechanisms. is the spatial coordinate, indicating the position in three-dimensional space, represents the time variable, is a Gaussian function used to simulate the distribution of growth factors that diffuse outward from the center of the injury. Represents the standard deviation of the distribution and controls the breadth of the distribution; S2. In addition, by combining finite element analysis FEA and computational fluid dynamics CFD, a calculation model including surface integral is used: , to predict and verify the mechanical properties of the printed path, represents the force applied on the boundary, where represents the magnitude of stress, represents the mechanical coefficient of the material, is a function The derivative with respect to the boundary normal direction represents the rate of change of material displacement at the boundary, Indicated by the parameter Defined geometric boundaries, integral Along the border Integrate the above expression to calculate the total effect over the entire boundary.

6. A bone joint surface reconstruction 3D printing trajectory control method according to claim 4, characterized in that The genetic algorithm-based path optimization model dynamically optimizes the 3D printing path through bionic simulation and mathematical modeling formulas: , in is the spatial coordinate, representing the position in the 3D printing space, is the time variable, indicating the change of the function over time, is an infinite series, indicating that as the complexity of time and space variables increases, the path adjustment strategy becomes more and more sophisticated. It is a periodic function, which means that the path selection has periodic adjustment, which is related to the print head movement and material deposition cycle; A geometric adaptability algorithm is then used to ensure that 3D printing can replicate the complex geometry and surface irregularities of the joint: , and Represents parameters related to the geometry, is an exponential decay function, Emphasize The effect decreases rapidly with the increase of the material properties and the changes in the material state during the printing process.

7. A bone joint surface reconstruction 3D printing trajectory control method according to claim 4, characterized in that The real-time adjustment system for printing parameters: S1. First, a high-resolution sensing system is integrated into the 3D printer, including optical sensors, infrared cameras and pressure sensors, to monitor the material deposition of each layer and the interlayer adhesion quality in real time, using the formula: , is the spatial coordinate, indicating the position of the print head in 3D space, is the time variable, the internal integral Indicates the time from the start of printing to The concentration of material deposition; S2. Apply data processing algorithms and machine learning techniques to analyze and identify printing defects and inconsistencies using the collected data. The prediction model is based on the formula: , Optimize printing parameter settings, It is an adjustable parameter during the printing process. Represents the contribution and impact of different printing parameters on printing quality; S3, based on the feedback provided by the machine learning model, adjust the printing speed, printing temperature, and nozzle movement rate in real time, and use a closed-loop feedback control system to ensure that the adjustment of printing parameters can be carried out continuously, using the formula: , Indicates the parameters during printing. is with The coefficient of correlation.

8. A method for controlling the trajectory of 3D printing of bone joint surface reconstruction according to claim 4, characterized in that The controlling material deposition process comprises: S1. Build a high-precision rheological model to simulate the behavior of biocompatible materials during 3D printing, and optimize the fluidity and curing time of the material according to the load and stress distribution unique to bone joints, using the formula: , in Indicates the properties of the material or the position of the print head, is the time variable, the internal integral Describes the time-varying influence factor of the material fluidity, which is related to the viscosity or temperature of the material. represents how this fluidity is adjusted based on position or material properties; S2. Then, the real-time material flow control system is used to adjust the printing parameters in real time to adapt to the areas with different bone density, optimize the smoothness of the joint surface and the accuracy of the joint part, through the formula: , and Indicates the printing speed and the amount of extruded material. is the time variable, The expression in Simulate how printing parameters change dynamically over time and printing conditions; S3. Finally, the interlayer bonding and structural integrity optimization technology is used to optimize the interlayer bonding to ensure that the printed structure can withstand the pressure and movement in actual use in biomechanics. The formula is used: , represents the properties related to interlayer adhesion, In Describe how different interlayer parameters affect the overall adhesion.

9. A bone joint surface reconstruction 3D printing trajectory control method according to claim 1, characterized in that The control of the accuracy and repair quality during the printing process is achieved by: first, using data analysis and machine learning algorithms to optimize the 3D printing path, and processing the bone structure data obtained from medical images through functions: , represents the coordinates in three-dimensional space, is the exponential function of time, function Describes the attenuation effect from a certain point to the center point, which is related to the density of material deposition; Next, develop a personalized printing material and parameter recommendation system to recommend the best printing material and parameters based on the specific bone and joint injury type and location, as well as the patient's physiological characteristics, using the formula: , and Represent the elastic modulus and melting point of the material, respectively. quantify material strength and flexibility; Finally, by establishing a database containing a variety of biocompatible materials and their properties, an intelligent matching algorithm is used, formula: , Represents a metric related to material performance, Each of represents the contribution of material properties, Represents the complexity and variability of different attributes.

Images