Knee joint brace personalized design method based on 3D printing

Abstract

The invention relates to the technical field of knee joint brace design, and discloses a knee joint brace personalized design method based on 3D printing. The method comprises information of a skeleton structure, a soft tissue contour and a joint motion range. Determining a personalized anatomical feature set of the knee joint by combining a preset knee joint biomechanical knowledge base on the basis of the three-dimensional scanning data; according to the personalized anatomical feature set and the corresponding at least one design rule file, dividing the stereo scanning data into a macroscopic structure region and a microscopic detail region; the macrostructure area outputs skeleton frame data, and the data is transmitted to the microcosmic detail area for guiding soft tissue matching analysis; constructing a three-dimensional model of the knee joint brace based on a division result; and converting the three-dimensional model into a 3D printing instruction file which is used for driving 3D printing equipment to produce the personalized knee joint brace. According to the method, personalized manufacturing of the knee joint brace is achieved through multi-dimensional data fusion and structural collaborative design.

Description

Technical Field

[0001] This invention relates to the field of knee brace design technology, specifically a method for personalized knee brace design based on 3D printing. Background Technology

[0002] As a complex weight-bearing joint in the human body, the knee joint often requires braces for fixation, correction, or assisted movement when injured or dysfunctional. Traditional knee braces are designed and manufactured primarily by medical personnel manually measuring limb dimensions and determining the basic parameters of the brace based on experience. This method is limited by the subjectivity of manual operation, easily leading to measurement errors and discrepancies between the brace and the individual's knee joint anatomy.

[0003] While mass-produced standardized knee braces can meet some general needs, they are difficult to adapt to the different bone morphology, soft tissue thickness, and joint mobility characteristics of different patients. Some patients have limb morphology that differs significantly from the standard model due to knee joint deformities or anatomical variations after wound healing. When wearing standardized braces, problems such as local pressure and loosening are likely to occur, affecting the daily user experience.

[0004] With the development of digital technology, 3D scanning technology has been gradually applied to brace design, enabling the acquisition of three-dimensional morphological data of the knee joint area. However, existing 3D scanning-based design methods often treat skeletal structure and soft tissue contours as independent data, lacking consideration of their correlation. During model building, there is either an overemphasis on replicating the macroscopic skeletal framework while ignoring the dynamic changes in soft tissues, or an obsession with piling up microscopic details, resulting in insufficient overall structural stability of the brace.

[0005] Biomechanical properties are an important factor in determining the effectiveness of knee braces. In traditional designs, the analysis of knee biomechanics often relies on general databases, failing to incorporate individualized information such as the range of motion of the joint and muscle distribution. This makes it difficult for the brace to coordinate with the individual's biomechanical characteristics when assisting joint movement, and long-term wear may cause new limb discomfort.

[0006] 3D printing technology has made it possible to produce personalized braces, but in the current technology, the conversion of 3D models into printing instructions often results in mismatches between the printed braces and the soft tissue areas due to unreasonable model structure division, thus failing to fully realize the advantages of 3D printing in personalized manufacturing. Summary of the Invention

[0007] The purpose of this invention is to provide a method for personalized design of knee braces based on 3D printing, so as to solve the problems mentioned in the background art.

[0008] To achieve the above objectives, the present invention provides a method for personalized design of knee braces based on 3D printing, the method comprising: Three-dimensional scanning data of the patient's knee joint area is collected using a three-dimensional scanning device. The three-dimensional scanning data includes information on bone structure, soft tissue contours, and joint range of motion. Based on stereoscopic scan data and combined with a pre-set knee joint biomechanical knowledge base, a personalized anatomical feature set of the knee joint is determined. Based on a personalized anatomical feature set and at least one design rule file corresponding to that feature set, the stereoscopic scan data is divided into macroscopic structural regions and microscopic detail regions. Based on the division results of macroscopic structural regions and microscopic detail regions, a three-dimensional model of a knee brace is constructed from the stereoscopic scan data; The constructed 3D model is converted into a 3D printing instruction file, which is used to drive 3D printing equipment to produce personalized knee braces; In the process of dividing the macroscopic structural region and the microscopic detail region, the macroscopic structural region outputs skeletal framework data, which is then passed to the microscopic detail region to guide soft tissue matching analysis.

[0009] Preferably, the process of determining the personalized anatomical feature set of the knee joint includes: The stereoscopic scan data is input into a multi-feature recognition model, which outputs key anatomical landmarks in the scan data. Key anatomical landmarks are input into a pre-defined knee joint biomechanical knowledge base. Based on the association mapping relationship in the knowledge base, the set of biomechanical properties including all landmarks is identified. Based on the set of biomechanical characteristics, the range of joint movement angles and load distribution parameters are calculated, which are used for data interaction in subsequent 3D model construction.

[0010] Preferably, the process of dividing the macroscopic structural region and the microscopic detail region includes: Based on the personalized anatomical feature set, the design rule file corresponding to the feature set is called, and the intelligent analysis engine performs semantic parsing on the design rule file to output multiple design constraints. Based on design constraints, the stereoscopic scan data is segmented into macroscopic structural regions and microscopic detail regions. The macroscopic structural regions generate skeletal topology data, while the microscopic detail regions generate soft tissue deformation data. The skeletal topology data is directly input into the soft tissue deformation analysis module to optimize the segmentation results.

[0011] Preferably, the process of constructing a three-dimensional model of a knee brace from the stereoscopic scan data includes: Based on the division of macroscopic structural regions and microscopic detail regions, the preprocessed stereoscopic scan data is split into macroscopic structural subsets and microscopic detail subsets. The macroscopic structural subset and associated biomechanical parameters are input into the structural optimization model, and the microscopic detail subset is input into the detail generation model. By combining the outputs of the structural optimization model and the detail generation model, a complete 3D model of the knee brace is generated, and intermediate data for each model construction is recorded.

[0012] Preferably, the execution process of the structure optimization model and the detail generation model includes: The structural optimization model is established based on the geometric constraint algorithm. The structural optimization model first performs contour fitting on the macroscopic structural subset to obtain the brace support frame information. By comparing the matching degree between the support frame information and biomechanical parameters, it is determined whether the mechanical stability requirements are met. The detail generation model is used for surface texture analysis, enabling adaptive smoothing of subtle surfaces in the micro-detail subset. Support framework information is passed as input to the surface texture analysis to enhance data consistency.

[0013] Preferably, when performing contour fitting on the macroscopic structure subset, the process involves extracting microscopic detail enhancement regions from the initially divided macroscopic structure subset state. This is an iterative optimization process, and the specific steps of the iterative optimization include: Obtain a macroscopic structural subset of data. For each local unit in the macroscopic structural subset of data, calculate its rate of curvature change and density distribution index. When the rate of curvature change or density distribution index of any local unit exceeds a preset threshold, it is determined that the local unit has a geometric defect that needs to be finely reconstructed. Based on the identified local units of geometric defects, the corresponding micro-detail enhancement dataset is separated from the macro-structure subset dataset; Based on the geometric defect characteristics of different knee joint types, data extraction rules associated with the defect characteristics are established, and the geometric defect judgment results are directly input into the data extraction rule module to drive the separation operation.

[0014] Preferably, the specific steps of the iterative optimization further include: After separating the micro-detail enhancement dataset from the macro-structure subset, a data synchronization mechanism is established between the structure optimization model and the detail generation model. The structural optimization model is used to perform a frame evaluation on the separated region, and the coordinate range, size parameters and evaluation results of the separated region are passed to the detail generation model. After receiving the information from the structural optimization model, the detail generation model performs surface reconstruction on the micro-detail enhancement dataset, detects and generates specific surface optimization information, including texture coordinates, curvature change trends, and integration status with the supporting frame information. The surface optimization information is fed back to the frame evaluation module to complete closed-loop optimization.

[0015] Preferably, the specific steps of the iterative optimization further include: The generated surface optimization information is fed back to the structural optimization model in real time; During multiple data separation and model optimization processes, the optimization threshold and model configuration parameters are adjusted based on the deviation between each optimization result and the preset standard. The final generated complete 3D model information, key parameters, and optimization logs are output to the designer's end, and the full data record of each optimization process is stored. The model configuration parameter adjustment results are passed to the subsequent 3D model construction steps to update the processing logic.

[0016] Preferably, the specific process of separating the corresponding micro-detail enhancement dataset from the macro-structure subset includes: Identify local units with geometric defects in a macroscopic structural subset of data. These local units with geometric defects are composed of boundary voxels defined by a set of three-dimensional coordinate points. Based on the local unit of geometric defects, the parameters required to separate the micro-detail enhancement dataset are calculated, including the starting coordinate point, voxel width, and height dimension. Using the calculated separation parameters, the corresponding micro-detail enhancement dataset is separated from the macro-structure subset dataset through data pruning operations; Verify whether the separated micro-detail enhancement dataset completely covers the defect area and eliminate redundant data interference. The separation parameters are output to the verification module for automatic verification.

[0017] Preferably, when combining the output results of the structural optimization model and the detail generation model, a fusion weight allocation mechanism is adopted. Different weight coefficients are assigned according to the confidence scores of the structural optimization model and the detail generation model. The confidence scores are obtained through training with historical data and updated in real time. The weight coefficients are output to the model output integration module to coordinate the generation of the three-dimensional model.

[0018] Compared with the prior art, the beneficial effects of the present invention are: Three-dimensional scanning equipment is used to collect stereoscopic scan data of the patient's knee joint area, covering bone structure, soft tissue contours, and joint range of motion. This provides more comprehensive information, accurately reflecting the individual's knee joint anatomical features and movement status. Based on this stereoscopic scan data, combined with a pre-set knee joint biomechanical knowledge base, a personalized anatomical feature set is determined. This allows the design process to incorporate considerations of individual biomechanical characteristics, making the brace design more closely aligned with the patient's physical function and movement needs.

[0019] Based on personalized anatomical feature sets and corresponding design rule files, the 3D scan data is divided into macroscopic structural regions and microscopic detail regions, allowing for targeted processing of structural information at different levels. The skeletal framework data output from the macroscopic structural regions is transferred to the microscopic detail regions to guide soft tissue matching analysis. This ensures that the structural information of the skeleton and soft tissues are coordinated during model construction, preventing a disconnect between the two in the design process. This ensures that the overall structure of the brace not only meets the support requirements of the skeleton but also adapts to the morphology and dynamic changes of the soft tissues.

[0020] Based on the division of macroscopic structural regions and microscopic detail regions, a 3D model of the knee brace is constructed. This model simultaneously considers macroscopic structural stability and microscopic fit accuracy, presenting a structural form that better conforms to the individual knee joint morphology and movement characteristics. The constructed 3D model is converted into a 3D printing instruction file, driving 3D printing equipment to produce personalized knee braces. This allows the personalized features in the design to be directly translated into the attributes of the physical brace, ensuring a high degree of fit between the brace and the patient's knee joint area in terms of shape and structure. This results in a more natural contact with the body during wear, reducing local pressure or looseness caused by improper fit, and improving comfort and fit during use.

[0021] The entire design methodology, from data collection to final production, forms a coherent personalized process. Each step revolves around individual characteristics, enabling the final knee brace to better adapt to the anatomical differences, biomechanical characteristics, and movement needs of different patients, and to achieve a closer match with the individual in terms of form and function. Attached Figure Description

[0022] Figure 1 This is a schematic diagram illustrating the working principle of the 3D-printed personalized knee brace design method described in this invention. Figure 2 A flowchart for determining a personalized anatomical feature set for the knee joint; Figure 3 Flowchart for constructing a 3D model of a knee brace; Figure 4 A flowchart for optimizing data synchronization and closed-loop processing. Detailed Implementation

[0023] 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. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] See also Figure 1 This invention provides a method for personalized design of knee braces based on 3D printing, the method comprising: A 3D scanning device is used to acquire stereoscopic scan data of the patient's knee joint region. This data includes information on skeletal structure, soft tissue contours, and joint range of motion. Based on this data and a pre-defined knee joint biomechanical knowledge base, a personalized anatomical feature set for the knee joint is determined. According to this personalized anatomical feature set and at least one corresponding design rule file, the stereoscopic scan data is divided into macroscopic structural regions and microscopic detail regions. Based on the division of these regions, a 3D model of a knee brace is constructed. This 3D model is then converted into a 3D printing instruction file to drive the 3D printing equipment to produce the personalized knee brace. During the division of the macroscopic and microscopic detail regions, the macroscopic structural region outputs skeletal framework data, which is then transferred to the microscopic detail region to guide soft tissue matching analysis.

[0025] Example 1: See Figure 2 In determining the personalized anatomical feature set of the knee joint, stereoscopic scan data was acquired using a 3D scanning device, containing information on bone structure, soft tissue contours, and joint range of motion. This data is stored in point cloud or mesh format, covering the geometric morphology and dynamic features of the knee joint region. After acquisition, the data is input into a multi-feature recognition model for processing. The multi-feature recognition model employs a deep learning architecture, using convolutional neural networks and feature extraction algorithms to analyze the scan data and identify the spatial distribution of anatomical landmarks. Key anatomical landmarks output by the model include characteristic locations with clear biomechanical significance, such as the femoral condyle, tibial plateau, and patellar margin.

[0026] Key anatomical landmarks are then input into a pre-defined knee joint biomechanical knowledge base. This knowledge base uses a graph database structure and includes the topological relationships between anatomical landmarks, force transmission paths, and stress distribution patterns under typical movement modes. The knowledge base uses an association mapping mechanism to link the input landmarks with predefined biomechanical properties. For example, the radius of curvature of the femoral condyle is associated with the joint contact pressure distribution, and the tilt angle of the tibial plateau is associated with the weight-bearing transmission efficiency. Based on these mapping relationships, the system generates a set of biomechanical properties, stored in a structured data table format, including indicators such as stiffness coefficients, range of degrees of freedom, and ligament attachment point mechanical parameters.

[0027] Based on a set of biomechanical properties, the system derives the joint range of motion and weight-bearing distribution parameters through numerical calculations. The range of motion is determined by simulating the extreme positions of the knee joint during flexion, extension, and rotation, taking into account the elastic deformation of soft tissues and the sliding friction of the bone contact surface. The weight-bearing distribution parameters are estimated through finite element analysis, quantifying the pressure distribution of the knee joint in static standing and dynamic gait cycles into regionalized values. These parameters are stored in matrix form to guide structural optimization and material allocation in subsequent 3D model construction.

[0028] During the division of macroscopic structural regions and microscopic detail regions, the system invokes a design rule file that matches the personalized anatomical feature set. The design rule file is written in a semantic description language and includes geometric constraints for region division, material distribution gradient requirements, and mechanical property thresholds. The intelligent analysis engine parses the rule file, extracts key instructions, and converts them into executable computational tasks. During parsing, the engine uses natural language processing technology to identify logical relationships within the rules; for example, conditional statements such as "if the femoral condyle curvature is greater than the threshold, then increase the thickness of the lateral support" are converted into executable decision branches.

[0029] Based on the design constraints obtained from the analysis, the 3D scan data was segmented into macroscopic structural regions and microscopic detail regions. The macroscopic structural regions were divided according to the mechanical load-bearing requirements of the skeleton, using voxel clustering algorithms to identify high-density areas and generate skeletal topology data. This data was represented using non-uniform rational B-spline surfaces to accurately reconstruct the load-bearing framework of the skeleton. The microscopic detail regions focused on the morphological changes of soft tissue and skin surfaces, using curvature manifold analysis to identify regions with high gradient changes and generate soft tissue deformation data. This data included refined parameters such as epidermal texture features and subcutaneous fat thickness distribution.

[0030] Skeletal topology data is transmitted in real time to the soft tissue deformation analysis module via a data pipeline. This module employs an elastic deformation simulation algorithm to quantify the soft tissue displacement caused by skeletal movement into deformation field data. During the analysis, the module dynamically adjusts the vertex weights of the soft tissue deformation data to ensure that the generated microscopic detail regions maintain biomechanical consistency with macroscopic skeletal movement. For example, the skin stretching effect caused by changes in patellar trajectory is quantified as a vertex displacement vector and fed back into the surface reconstruction process of the microscopic detail regions.

[0031] The system improves the accuracy of region segmentation through an iterative optimization mechanism. In each iteration, the skeletal topology data of the macroscopic structural regions are compared with the soft tissue deformation data of the microscopic detail regions to detect continuity errors in the boundary regions. When abrupt changes in density or curvature between adjacent voxels are detected, the system automatically adjusts the segmentation threshold and recalculates the region segmentation results. The optimization process continues until the transition gradient between the macroscopic and microscopic regions meets the preset smoothness standard, ultimately outputting a hierarchical digital model of the knee joint.

[0032] During data interaction, the system employs a timestamp synchronization mechanism to ensure the timeliness of processing by each module. Updates to the coordinates of key anatomical landmarks trigger real-time queries of the biomechanical knowledge base, while changes in load-bearing distribution parameters directly feed back to the dynamic loading of the design rule file. This tightly coupled data flow design creates a closed-loop system for determining personalized anatomical feature sets and dividing regions, effectively capturing the correlation between knee joint morphology and function.

[0033] In terms of implementation details, the multi-feature recognition model employs a transfer learning strategy, adapting pre-trained weights to the anatomical variations of different patients. The knowledge base's association mapping incorporates fuzzy logic processing to accommodate noise and partial omissions in the scan data. The intelligent analysis engine is equipped with rule version management, allowing clinical experts to visually edit and simulate the effects of the designed rules. The entire implementation runs within a distributed computing framework, leveraging parallel computing to accelerate large-scale data processing while ensuring data recoverability after computational interruptions through a checkpoint mechanism.

[0034] The stereoscopic scan data undergoes coordinate system normalization during preprocessing, and all subsequent calculations are based on a standardized anatomical reference frame. Intermediate results generated during data partitioning employ an incremental storage strategy, retaining only the discrepancies to reduce memory usage. The system interface provides 3D visualization tools, allowing operators to review the region partitioning effect from any perspective and manually correct the automatically partitioned results using annotation tools. Corrections are recorded as new design rules and added to the knowledge base, forming a self-improvement mechanism for the system.

[0035] This implementation method employs a hierarchical data processing workflow to progressively transform raw scan information into a structured feature set that guides brace design. Automatic identification of anatomical landmarks reduces subjective bias from manual annotation, while the application of a biomechanical knowledge base transforms clinical experience into quantifiable design parameters. The dual-layer structure of region division balances overall mechanical performance with local fit accuracy, providing a differentiated data processing foundation for subsequent 3D model construction. The high degree of automation throughout the process significantly shortens the cycle from data acquisition to brace design, while standardized processing ensures the comparability of design results across different cases.

[0036] The system's log data includes processing time for each module, parameter adjustment records, and abnormal event reports. These logs are stored in a time-series database for analyzing process bottlenecks and optimizing computational resource allocation. User interface interactions are abstracted into a state machine model to ensure the traceability of user operations. For data security, role-based access control is employed to ensure encrypted protection of patient privacy information during transmission and storage. Network communication uses a checksum mechanism to prevent bit errors during data transmission from affecting computational accuracy.

[0037] The technical solution of this embodiment achieves automated conversion from knee joint scan data to design feature sets through the collaborative work of multiple modules. The digital representation of anatomical features provides precise input conditions for the design of personalized medical devices, while the hierarchical processing strategy balances the requirements of computational efficiency and model accuracy.

[0038] Example 2: See Figure 3 In the process of constructing the 3D model of the knee brace, the system first performs regionalization processing on the preprocessed stereoscopic scan data. Based on the previously defined macroscopic structural regions and microscopic detail regions, the scan data is systematically split into two logically independent but data-related subsets. The macroscopic structural subset mainly contains the geometric information of the skeletal framework, stored in the form of a high-precision triangular mesh, with each vertex coordinate accompanied by local curvature and thickness parameters. The microscopic detail subset records the texture features and dynamic deformation characteristics of the soft tissue surface, expressed using a combination of parametric surfaces and displacement fields.

[0039] The macroscopic structural subset and associated biomechanical parameters are input into a structural optimization model for processing. The biomechanical parameters are stored in JSON format and include key indicators such as joint range of motion limits, pressure distribution maps, and material stiffness coefficients. The structural optimization model employs a multi-objective optimization algorithm, simultaneously considering the brace's mechanical support performance and wearing comfort requirements. The model first performs topological analysis on the input skeletal frame to identify load-bearing areas requiring reinforcement and non-critical parts where weight reduction is appropriate. Subsequently, it simulates stress distribution under different load conditions using the finite element method, automatically adjusting the brace frame's thickness distribution and stiffener layout.

[0040] The structural optimization model employs a phased approach. The initial phase rapidly generates a coarse-grained model that meets basic mechanical requirements, serving as the foundation for subsequent refined optimization. The intermediate phase introduces constraint processing algorithms to ensure sufficient tolerance between the brace structure and the patient's anatomical morphology. The final phase utilizes locally refined meshes to specifically reinforce high-stress concentration areas, outputting support frame information that meets biomechanical requirements.

[0041] The micro-detail subset was fed in parallel into a detail generation model. This model focuses on processing the internal surface characteristics of the brace that come into contact with the patient's skin. The model first performs curvature analysis and contact pressure prediction on the soft tissue contour to identify sensitive areas requiring special treatment. Then, an adaptive subdivision algorithm is used to dynamically adjust the mesh density in high-pressure areas while maintaining the overall surface continuity. The detail generation model also integrates tribological parameters to automatically generate differentiated surface texture patterns based on the skin characteristics of different body parts.

[0042] The workflow of the detail generation model comprises two dimensions: geometric processing and physical simulation. In terms of geometric processing, the model preserves the natural morphological features of anatomical structures through feature line detection and surface parametric techniques. In terms of physical simulation, the model calculates the interaction between the brace and soft tissue, predicting potential changes in pressure distribution due to long-term wear.

[0043] The outputs of the structural optimization model and the detail generation model are integrated through a fusion weight allocation mechanism. The system assigns dynamic weight coefficients to each model, which are automatically adjusted based on the model's confidence score. The confidence score is calculated by analyzing the model's historical performance data and the degree of matching with the current input, and is updated in real time using a sliding window algorithm. The weight allocation process considers multiple influencing factors, including geometric accuracy, mechanical performance, and manufacturing feasibility.

[0044] During the model fusion phase, the system first performs a topological consistency check on the mesh structures output by the two models. When mismatched regions are found, local recalculation or interpolation smoothing is automatically triggered. The fusion algorithm preserves the main framework of the macroscopic structural model while accurately mapping the surface characteristics of the microscopic detail model to the corresponding regions. For transition regions with weight disputes, the system employs a progressive blending strategy to ensure a smooth transition of geometric features. The fused complete 3D model undergoes lightweight processing, removing redundant vertices and unseen surfaces to optimize data storage efficiency.

[0045] The system records intermediate data for each model build, including mesh models at each optimization stage, parameter adjustment records, and performance evaluation metrics. This data is managed using version control, allowing designers to backtrack to intermediate states at any step. The data storage format balances processing efficiency and storage space, and key parameters use differential encoding to reduce redundancy.

[0046] The quality verification of the 3D model employs a multi-level inspection mechanism. At the geometric level, the model's closure, normal consistency, and mesh quality are checked; at the functional level, the mechanical behavior of the supports under different motion states is simulated; and at the manufacturing level, the printability of the model and the requirements for the supporting structure are analyzed. Issues discovered during verification are automatically fed back to the corresponding processing module, triggering local recalculation or parameter adjustments. Verified models are converted into a lightweight intermediate format, ready for subsequent printing instruction generation.

[0047] In terms of system implementation, the model building module adopts a microservice architecture, with structure optimization and detail generation deployed as independent services. Services communicate asynchronously via message queues to ensure elastic scaling of computing resources. The task scheduler monitors the load status of each service and dynamically allocates computing tasks. The user interface provides real-time visualization capabilities, displaying model building progress and changes in key parameters. Interface operations are decoupled from backend computation, preventing complex calculations from blocking user interaction.

[0048] The data processing pipeline employs a fault-tolerant design, with checkpoints set for critical computational steps, allowing recovery from the most recent stable state after an abnormal interruption. The system resource monitoring module tracks CPU, memory, and video memory usage, automatically reducing computational precision or pausing secondary tasks when resources are insufficient. Network communication utilizes compressed transmission technology to reduce latency during large-scale data exchange. Database queries optimize the index structure, accelerating the retrieval and analysis of historical models.

[0049] The model building process supports manual intervention. Designers can pause automatic calculations at any stage to manually adjust parameters or directly modify the model geometry. Manual modification records are stored separately from automatic calculation logs, facilitating the analysis of potential improvements to the automated process. The system provides various editing tools, including local surface adjustment, feature line editing, and mechanical parameter overriding, with all modifications providing real-time visual feedback.

[0050] In terms of system performance, a multi-level parallel computing strategy is adopted. Macroscopic structure optimization utilizes multi-core CPUs for finite element analysis task decomposition, while microscopic detail generation leverages GPU-accelerated surface processing algorithms. Memory management employs an intelligent caching mechanism, retaining frequently accessed data in fast storage areas. Disk I / O is optimized through asynchronous writes to avoid the computation process being blocked by storage operations.

[0051] In terms of security measures, patient data is encrypted during transmission and storage. Access control is based on a role-based access control model to ensure that sensitive operations require authorization and authentication. System logs record the complete data flow trajectory, supporting security audits and compliance checks. Network communication uses two-way authentication to prevent man-in-the-middle attacks.

[0052] This implementation achieves a complete transformation from anatomical data to a printable model through the collaborative work of structural optimization and detail generation. The dual-model architecture balances overall performance and local adaptability, while a weighting mechanism prioritizes different design objectives. Complete intermediate data recording supports traceability of the design process, and multi-level verification ensures the reliability of the output model. The system implementation employs modern software engineering practices, ensuring processing efficiency while maintaining sufficient flexibility. The entire model building process strikes a balance between automation and manual control, providing technical support for the precise manufacturing of personalized braces.

[0053] Example 3: The construction of the structural optimization model is based on a geometric constraint algorithm system. This algorithm employs a hierarchical processing strategy when processing macroscopic structural subsets. In the initial stage, the input data undergoes noise reduction and smoothing to eliminate surface irregularities introduced during scanning. The processed data is converted into voxel representation for easier spatial topology analysis. The voxelization process uses an adaptive resolution strategy, automatically increasing sampling density in high-curvature regions. The structural optimization model identifies the mechanical key points of the skeletal structure through feature extraction algorithms; these key points constitute the basic positioning reference for the brace support frame.

[0054] The contour fitting process employs an iterative optimization method. The system first determines the initial support frame topology based on biomechanical parameters. The frame generation algorithm automatically plans the direction of stiffeners and the location of connection nodes based on the curvature distribution and thickness variations of the bone surface. During the mechanical stability assessment phase, the model performs a matching degree analysis between the generated support frame information and the input biomechanical parameters. The matching degree calculation considers multiple dimensions, including stiffness matching coefficient, stress dispersion efficiency, and weight distribution balance. When the main indicators fail to reach preset thresholds, the model triggers a frame reconstruction mechanism, adjusting key parameters and recalculating.

[0055] The mechanical stability assessment uses the following formula for quantitative analysis: ; in, This indicates the frame matching deviation value. The number of key assessment points. Let be the measured stiffness value at point i. This represents the theoretical stiffness value at the corresponding point. Let be the weight coefficient of the i-th point. This is the stress distribution equilibrium factor. This represents stress non-uniformity. The calculation result of this formula is used to determine whether frame structure adjustments are needed; when the deviation value exceeds a threshold, an optimization loop is triggered.

[0056] When processing micro-detail subsets, the detail generation model employs a multi-scale analysis approach. The model first parametrically represents the soft tissue surface, establishing a unified UV coordinate system. The surface texture analysis algorithm identifies sensitive areas requiring special processing by calculating the local rate of curvature change and normal direction distribution. Adaptive smoothing utilizes a physics-based diffusion equation to eliminate micro-scale irregularities while preserving key features. During processing, the model receives support framework information from the structural optimization model in real time, converting this information into geometric constraints to ensure that detail processing remains consistent with the overall structure.

[0057] The iterative optimization process of the macroscopic structural subset employs a dynamic data separation strategy. When evaluating the geometric quality of each local unit, the system calculates two key indicators: the rate of change of curvature reflects the degree of change in surface smoothness, and the density distribution indicator describes the spatial uniformity of voxel distribution. The evaluation algorithm establishes a moving analysis window in three-dimensional space, with the window size dynamically adjusted according to anatomical structural features. When the statistical value of any indicator within the window exceeds a preset threshold, the region is marked as a geometrically defective area requiring fine reconstruction. The threshold setting considers the normal physiological variation range of different knee joint locations to avoid ineffective reconstruction due to oversensitivity.

[0058] The separation of geometric defect regions employs precise boundary recognition technology. The system first performs morphological dilation on the defect region to ensure complete coverage of the problem area. Then, an edge detection algorithm is applied to determine the separation boundary, with the boundary curves satisfying the second-order geometric continuity requirement. Based on the boundary information, the data trimming module extracts a micro-detail enhancement dataset from the macroscopic structure subset. The extraction process preserves the topological connectivity of the original data, ensuring seamless integration into subsequent processing. The separation parameter calculation considers the anisotropic characteristics of three-dimensional space, setting higher separation precision in the principal mechanical directions.

[0059] The data extraction rule module establishes a classification mechanism based on knee joint type. The system maintains a feature rule base containing typical geometric defect patterns under different pathological conditions. The rule matching process employs fuzzy reasoning to handle common clinical variations. Once the geometric defect determination result is input into the rule module, it triggers the corresponding data processing flow. For example, edge enhancement is used for cartilage wear areas in osteoarthritis patients, while surface smoothing is emphasized for ligament attachment point abnormalities caused by sports injuries. The rule engine supports dynamic updates, allowing clinical experts to add new processing strategies or adjust existing parameters.

[0060] The data interaction between the structure optimization model and the detail generation model adopts an event-driven mechanism. When the macroscopic structure subset completes a specific processing stage, a standardized event message is generated, containing the coordinates of the processing area, feature parameters, and intermediate results. The message bus routes this information to subscribed detail generation model instances, triggering the corresponding micro-processing flow. Reverse communication follows the same pattern, ensuring that the processing progress of the two models remains synchronized. The data synchronization process employs optimistic concurrency control, merging versions based on timestamps when conflicts are resolved.

[0061] The framework evaluation module employs a multi-criteria decision-making approach, with evaluation indicators including geometric fit, mechanical performance, and production feasibility. The evaluation process generates detailed reports, highlighting key areas requiring special attention. Report data is converted into visual markers and overlaid on the 3D model, facilitating rapid problem identification by designers. The evaluation results, input into the detailed model generation process, guide the focus of surface reconstruction. For example, high-stress areas require finer surface subdivision, while major load-bearing components demand stricter dimensional tolerances.

[0062] Surface optimization information is generated using a parametric method, with the system recording the optimization type, intensity parameters, and influence range for each processed area. Texture coordinate processing considers the characteristics of subsequent manufacturing processes to ensure accurate reproduction of the generated surface pattern during printing. Curvature variation trend analysis establishes a mathematical model to predict wear patterns during long-term use of the support. Integration status checks with the support frame verify whether the micro-processing results maintain mechanical continuity with the overall structure; inconsistencies trigger local recalculation.

[0063] The geometric defect identification algorithm employs machine learning methods to improve accuracy. The system collects processing records from historical cases to train a classification model that identifies common defect patterns. During model inference, it combines low-level feature analysis with high-level semantic understanding to reduce the false positive rate. The judgment result includes a confidence score; low-confidence cases are automatically transferred to a manual review process. The algorithm is periodically retrained with new data to gradually adapt to novel variations encountered in clinical practice.

[0064] The processing of the micro-detail enhancement dataset employs a hierarchical refinement strategy. The base layer ensures the basic functionality of the geometry, the intermediate layers optimize parameters related to wearing comfort, and the surface layer focuses on aesthetics and tactile characteristics. Each processing level has independent quality control standards; only after passing the current level's verification can the next stage proceed. This progressive processing method effectively controls computational complexity and avoids unnecessary waste of precision.

[0065] Before data pruning, an integrity check is performed to verify the topological consistency and geometric validity of the input data. The pruning process employs a conservative strategy, preserving redundant data in transition areas for flexible adjustments in subsequent processing stages. The operation log records detailed pruning parameters and their impact range, supporting operation rollback and data reconstruction. The verification module uses a combination of sampling checks and global analysis to ensure that the separated dataset meets the requirements of subsequent processing.

[0066] Example 4: See Figure 4 In the personalized design process of knee braces, after separating the micro-detail enhancement dataset from the macro-structural subset, the system establishes a data synchronization mechanism between the structural optimization model and the detail generation model. This process can be clearly demonstrated through a specific clinical case: a 45-year-old male patient suffered from right knee instability due to sports injury. Three-dimensional scanning obtained stereoscopic data of his knee joint region with a resolution of 0.2 mm, covering the distal femur, proximal tibia, and surrounding soft tissues.

[0067] The data synchronization mechanism is first reflected in the structural optimization model's handling of separated regions. Taking the medial femoral condyle region of this patient as an example, the system identified three geometric defects in this region requiring fine-tuning, with dimensions of 8.7mm × 6.3mm, 5.2mm × 4.1mm, and 12.5mm × 9.8mm, respectively. When the structural optimization model performs a frame evaluation on these regions, it generates an evaluation report containing the following key parameters:

[0068] The evaluation report is transmitted to the detail generation model via a standardized data interface, triggering the corresponding surface reconstruction process. In this case, after receiving the coordinate range information of region 3 (X: 124.5-137.0mm, Y: 56.7-66.5mm, Z: 89.2-99.0mm), the detail generation model first performs data preprocessing on this region. The preprocessing steps include resampling the point cloud data, uniformly adjusting the density of the original scan points from 36 points per square millimeter to 25 points, and eliminating noise points generated during the scanning process.

[0069] The surface reconstruction algorithm employs an adaptive subdivision strategy, adjusting the processing intensity based on the received evaluation results. For the high stress concentration areas (regions 2 and 3) of the patient's medial femoral condyle, the algorithm automatically increases the mesh subdivision, reducing the average size of the triangular facets from the default 1.2 mm to 0.8 mm. When processing areas with excessive thickness change rates, the system activates the thickness compensation function, transforming abrupt thickness changes into smooth gradients by generating a progressive transition structure on the inner surface of the brace.

[0070] The process of generating surface optimization information integrates multi-source data. Taking the processing of region 1 as an example, the system combines curvature deviation data from the frame evaluation results with the normal direction information of the original scan points to generate a surface texture with a specific orientation. The texture units are elliptical, with the major axis aligned with the principal stress direction, and the unit spacing is dynamically adjusted according to the curvature gradient. This texture design ensures both the fit between the inner surface of the brace and the patient's anatomy and optimizes the pressure distribution during long-term wear.

[0071] The closed-loop optimization mechanism is reflected in the real-time feedback of surface optimization information to the frame evaluation module. In the above case, after the detail generation model completes the surface reconstruction of region 2, it packages the processing results into a feedback data package, containing the following key information: the reconstructed surface curvature distribution map, vertex displacement vector field, and texture unit density parameters. After receiving this data, the frame evaluation module recalculates the mechanical parameters of the region and finds that the stress concentration factor has decreased from the initial 2.3 to 1.9, approaching the safety threshold.

[0072] The data synchronization mechanism relies on a distributed messaging system. Clinical cases show that when handling a moderately complex case with 15 defective regions, the system exchanges an average of 83 messages per minute between the structural optimization model and the detailed generation model, peaking at 120 messages per minute. Message content is binary encoded, with an average message size of 4.7KB, containing core data such as coordinate transformation matrices, attribute bitmaps, and operational instructions. A priority mechanism is implemented in the message queue to ensure that synchronization requests for high-stress regions are processed first.

[0073] During multiple iterations of optimization, the system records a complete log of each data exchange. Taking the treatment data of this 45-year-old patient as an example, the system performed four iterations of optimization for the medial femoral condyle region. The intermediate data generated in each iteration included: the framework adjustment scheme output by the structural optimization model, the surface correction parameters generated by the detail generation model, the synchronization timestamps between the two models, and records of manual intervention (if any). This data is stored in time-series format, allowing designers to backtrack to the state at any optimization stage.

[0074] The specific implementation of data synchronization between models is also reflected in the anomaly handling mechanism. In clinical practice, when two models have significantly different treatment recommendations for the same area (such as opposite directions of curvature adjustment), the system automatically triggers a negotiation agreement. In the treatment of this patient, a similar situation occurred in the anterior proximal tibial region. The structural optimization model recommended increasing support stiffness, while the detail generation model, based on soft tissue adaptability considerations, recommended reducing contact pressure. The system resolved the discrepancy through the following steps: pausing the automatic optimization process, simultaneously displaying 3D renderings of both treatment options on the visualization interface, prompting the clinician to make an arbitration judgment, and ultimately adopting a hybrid approach that maintains the support structure while adding a buffer layer design.

[0075] The integrated status check of surface optimization information employs a hierarchical verification method. Taking the final optimization result of the above case as an example, the system performs three levels of checks: the geometric level verifies surface continuity and boundary fit, the physical level simulates contact pressure distribution under different buckling angles, and the manufacturing level analyzes the feasibility of 3D printing. Each check level generates a detailed pass / fail report, and items that fail the check are automatically given correction suggestions for the next round of optimization.

[0076] The time-sensitivity of data synchronization has demonstrated significant value in clinical applications. System monitoring shows that the average latency from the completion of framework evaluation by the structural optimization model to the return of surface optimization information by the detail generation model is 37 seconds, with 90% of requests completed within one minute. This near real-time interactive capability allows optimization of complex areas to be completed in a single treatment session, avoiding the cumbersome process of multiple scans and trial fittings required in traditional methods.

[0077] In terms of system implementation, inter-model communication utilizes a dedicated data channel based on the TCP / IP protocol, with transport layer encryption to ensure patient privacy and security. Network anomaly handling strategies include: automatic retry mechanism (up to 3 times), local caching of unsent data, and degradation processing (switching to batch processing mode when consecutive failures occur). Regarding hardware configuration, it is recommended to allocate an independent computing node to each model, interconnected via a high-speed local area network. Real-world testing shows that a dual-node configuration improves performance by 62% compared to a single-node configuration.

[0078] The clinician's user interface provides real-time visual monitoring of data synchronization. Interface elements include: data transfer status indicators between models, a 3D localization map of the currently processed area, a synchronization progress bar, and an anomaly alert area. Clinicians can pause the automatic synchronization process at any time to manually adjust parameters or switch processing modes. All operation records are merged and stored with the automatically generated synchronization log to form a complete design process document.

[0079] Data from 20 clinical cases using this approach showed that, through the collaborative work of the structural optimization model and the detail generation model, the average number of modifications to the final brace design was reduced from 6.2 times using the traditional method to 2.8 times. The initial adaptation period for patients was shortened, thanks to the improved design accuracy resulting from precise data synchronization between the models.

[0080] Example 5: In the personalized design process of knee braces, the real-time feedback mechanism of surface optimization information constitutes the core of the system's self-improvement. After the detailed generation model completes the reconstruction of the micro-surface, the generated surface optimization information is transmitted to the structural optimization model in real time through a dedicated data channel. This feedback is not a simple data transfer, but a composite information package containing multi-level processing results and correction suggestions. During transmission, the system intelligently compresses and prioritizes data packets to ensure that feedback information in critical areas is processed first. For example, optimization data in high-stress concentration areas or areas with frequent motion contact is marked as urgent and automatically placed at the top of the message queue.

[0081] After receiving feedback information, the structural optimization model initiates a multi-round evaluation and verification process. The model first parses the geometric correction parameters in the feedback data packet and compares them with the current frame structure. The comparison process employs a spatial mapping algorithm to convert the adjustments to micro-surfaces into equivalent corrections to the macro-frame. For example, if feedback indicates that a 0.5 mm buffer layer needs to be added to a certain contact area, the structural optimization model will adjust the thickness distribution and stiffener layout of the supporting structure in that area accordingly. This conversion is not a simple linear correspondence, but a nonlinear mapping that considers the material's mechanical properties and the overall structural influence.

[0082] During multiple data separation and model optimization processes, the system establishes a dynamic threshold adjustment mechanism. After each optimization cycle, the system automatically compares the deviation of the current result with the preset standard, and intelligently adjusts the judgment threshold and convergence conditions for the next round of optimization based on the distribution characteristics and trends of the deviation. Deviation analysis not only focuses on numerical differences but also examines the topological consistency of geometric features and the coordination of mechanical parameters. When the optimization effect in a specific region is found to be consistently unsatisfactory, the system gradually relaxes the convergence conditions for that region while strengthening the constraint strength of adjacent regions. This compensation strategy achieves a balance in overall performance.

[0083] The adjustment of model configuration parameters follows the principle of incremental optimization. The system maintains a parameter influence coefficient matrix, recording the weight of each configuration parameter's impact on the final design quality. After each optimization iteration, the system analyzes the correlation between parameter adjustments and effect improvement, and updates this influence coefficient matrix. Based on the updated matrix, a parameter adjustment scheme for the next round of optimization is generated. This data-driven parameter optimization method enables the system to gradually adapt to variations in anatomical characteristics and individualized needs of different patients.

[0084] The optimization process employs a hierarchical storage strategy for all data recording. Raw scan data, intermediate processing results, and the final design model are stored in different databases at different levels, linked and indexed using a unified timestamp and case number. Each optimization iteration generates a data package containing a complete processing log, recording the parameters and execution steps used to make model decisions. The log data is stored in a structured format, facilitating both automatic machine parsing and manual review and analysis. The system establishes an independent data version branch for each design task, allowing for easy rewind to any historical state.

[0085] The output information from the designer's end undergoes intelligent integration and visualization. The system automatically extracts key parameter change curves and important decision nodes during the optimization process, generating a concise design report. The report content is presented in a hierarchical manner: the first level presents the overall design quality and main performance indicators; the second level displays the optimization details of each functional area; and the third level provides a query interface for raw data and analysis methods. The visualization interface supports interactive browsing of the 3D model, allowing designers to view the structural features and parameter distribution of any cross-section through gesture operations.

[0086] The storage system is designed with long-term data management needs in mind. All data generated during each optimization process is archived in a standardized format, along with metadata description files recording the data's origin and processing history. Archived data utilizes differential storage technology, preserving only incremental changes relative to the previous version, significantly reducing storage space usage. The data retrieval system supports multi-condition queries, allowing clinicians to quickly locate target data based on various dimensions such as time range, anatomical region, and design parameters.

[0087] The results of model configuration parameter adjustments are transmitted to subsequent processing modules via a standardized interface. The interface data includes three parts: a description of the parameter changes, an assessment of the expected impact, and implementation recommendations. Upon receiving this data, subsequent modules first perform a compatibility check to confirm that the parameter adjustments will not cause conflicts or contradictions in the processing flow. After passing the check, each module updates its internal processing logic according to the new parameter configuration. This update can take effect immediately or be delayed until the start of the next processing task; the specific strategy is dynamically determined based on the importance and scope of the parameters.

[0088] The system's anomaly handling mechanism is implemented throughout the entire optimization process. When data anomalies or processing timeouts are detected, the system automatically initiates a diagnostic process to analyze the cause of the anomaly and attempt self-repair. Diagnostic information is displayed to designers in real time, along with various handling suggestions. For recurring anomalies of the same type, the system automatically adjusts relevant parameter thresholds or optimizes algorithm configurations to reduce the probability of subsequent occurrences. When a severe anomaly causes processing interruption, the system can automatically save the current state and resume execution from the breakpoint upon recovery.

[0089] Clinical decision support functionality is integrated into every stage of the optimization process. The system highlights key design points in the optimization results that require manual verification, such as parameter adjustments outside the norm or innovative structural solutions. The highlighted information includes detailed background explanations and reference cases to help designers quickly understand the essence of the problem. For optimization options with multiple solutions, the system will display a parallel comparison of the advantages and disadvantages of each option, but will not automatically make a final decision, reserving the decision-making authority of the clinician.

[0090] The hardware resource management module dynamically adjusts the computational load during the optimization process. The system monitors CPU, memory, and GPU memory usage in real time, automatically reducing the processing precision of non-critical tasks or pausing background analysis processes when resources are scarce. For time-consuming optimization iterations, the system supports distributed computing and breakpoint resumption, allowing tasks to be executed in parallel on multiple computing nodes or resumed midway after an interruption. The resource allocation strategy is automatically adjusted according to different optimization stages; for example, the initial coarse optimization stage prioritizes processing speed, while the later fine-tuning stage prioritizes computational precision.

[0091] The security audit function records all modifications to system configuration parameters. Whether the parameter adjustments are triggered by automatic optimization or manually entered by designers, detailed information such as the operator's identity, modification time, original parameter value, and new parameter value is recorded. The audit logs are stored using tamper-proof technology and support regular compliance checks and security vulnerability analysis. The parameter modification impact tracking function can display the cascading effects of specific adjustments on subsequent design processes, helping to understand the reasons for changes in system behavior.

[0092] The user interface design emphasizes seamless human-computer collaboration. Designers can intervene in the automated optimization process at any time to view the current status, adjust the processing direction, or modify specific parameters. Interface operation and background calculations employ an asynchronous interaction mode; user commands enter a priority queue and do not block ongoing optimization calculations. The system provides multiple view modes, allowing for a global overview of the optimization progress as well as in-depth analysis of the detailed processing of a specific area. The visual cue system highlights design changes and anomalies that require attention, but avoids excessive alerts that could cause interference.

[0093] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0094] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for personalized design of knee braces based on 3D printing, characterized in that, include: Three-dimensional scanning data of the patient's knee joint area is collected using a three-dimensional scanning device. The three-dimensional scanning data includes information on bone structure, soft tissue contours, and joint range of motion. Based on stereoscopic scan data and combined with a pre-set knee joint biomechanical knowledge base, a personalized anatomical feature set of the knee joint is determined. Based on a personalized anatomical feature set and at least one design rule file corresponding to that feature set, the stereoscopic scan data is divided into macroscopic structural regions and microscopic detail regions. Based on the division results of macroscopic structural regions and microscopic detail regions, a three-dimensional model of a knee brace is constructed from the stereoscopic scan data; The constructed 3D model is converted into a 3D printing instruction file, which is used to drive 3D printing equipment to produce personalized knee braces; In the process of dividing the macroscopic structural region and the microscopic detail region, the macroscopic structural region outputs skeletal framework data, which is then passed to the microscopic detail region to guide soft tissue matching analysis.

2. The method for personalized design of knee braces based on 3D printing as described in claim 1, characterized in that, The process of determining the personalized anatomical feature set of the knee joint includes: The stereoscopic scan data is input into a multi-feature recognition model, which outputs key anatomical landmarks in the scan data. Key anatomical landmarks are input into a pre-defined knee joint biomechanical knowledge base. Based on the association mapping relationship in the knowledge base, the set of biomechanical properties including all landmarks is identified. Based on the set of biomechanical characteristics, the range of joint movement angles and load distribution parameters are calculated, which are used for data interaction in subsequent 3D model construction.

3. The method for personalized design of knee braces based on 3D printing as described in claim 1, characterized in that, The process of dividing the macroscopic structural region and the microscopic detail region includes: Based on the personalized anatomical feature set, the design rule file corresponding to the feature set is called, and the intelligent analysis engine performs semantic parsing on the design rule file to output multiple design constraints. Based on design constraints, the stereoscopic scan data is segmented into macroscopic structural regions and microscopic detail regions. The macroscopic structural regions generate skeletal topology data, while the microscopic detail regions generate soft tissue deformation data. The skeletal topology data is directly input into the soft tissue deformation analysis module to optimize the segmentation results.

4. The method for personalized design of knee braces based on 3D printing as described in claim 1, characterized in that, The process of constructing a three-dimensional model of a knee brace from stereoscopic scan data includes: Based on the division of macroscopic structural regions and microscopic detail regions, the preprocessed stereoscopic scan data is split into macroscopic structural subsets and microscopic detail subsets. The macroscopic structural subset and associated biomechanical parameters are input into the structural optimization model, and the microscopic detail subset is input into the detail generation model. By combining the outputs of the structural optimization model and the detail generation model, a complete 3D model of the knee brace is generated, and intermediate data for each model construction is recorded.

5. The method for personalized design of knee braces based on 3D printing as described in claim 4, characterized in that, The execution process of the structure optimization model and the detail generation model includes: The structural optimization model is established based on the geometric constraint algorithm. The structural optimization model first performs contour fitting on the macroscopic structural subset to obtain the brace support frame information. By comparing the matching degree between the support frame information and biomechanical parameters, it is determined whether the mechanical stability requirements are met. The detail generation model is used for surface texture analysis, enabling adaptive smoothing of subtle surfaces in the micro-detail subset. Support framework information is passed as input to the surface texture analysis to enhance data consistency.

6. The method for personalized design of knee braces based on 3D printing as described in claim 5, characterized in that, When performing contour fitting on the macroscopic structure subset, the process involves extracting microscopic detail enhancement regions from the initial macroscopic structure subset state. This is an iterative optimization process, and the specific steps of the iterative optimization include: Obtain a macroscopic structural subset of data. For each local unit in the macroscopic structural subset of data, calculate its rate of curvature change and density distribution index. When the rate of curvature change or density distribution index of any local unit exceeds a preset threshold, it is determined that the local unit has a geometric defect that needs to be finely reconstructed. Based on the identified local units of geometric defects, the corresponding micro-detail enhancement dataset is separated from the macro-structure subset dataset; Based on the geometric defect characteristics of different knee joint types, data extraction rules associated with the defect characteristics are established, and the geometric defect judgment results are directly input into the data extraction rule module to drive the separation operation.

7. The method for personalized design of knee braces based on 3D printing as described in claim 6, characterized in that, The specific steps of the iterative optimization also include: After separating the micro-detail enhancement dataset from the macro-structure subset, a data synchronization mechanism is established between the structure optimization model and the detail generation model. The structural optimization model is used to perform a frame evaluation on the separated region, and the coordinate range, size parameters and evaluation results of the separated region are passed to the detail generation model. After receiving the information from the structural optimization model, the detail generation model performs surface reconstruction on the micro-detail enhancement dataset, detects and generates specific surface optimization information, including texture coordinates, curvature change trends, and integration status with the supporting frame information. The surface optimization information is fed back to the frame evaluation module to complete closed-loop optimization.

8. The method for personalized design of knee braces based on 3D printing as described in claim 7, characterized in that, The specific steps of the iterative optimization also include: The generated surface optimization information is fed back to the structural optimization model in real time; During multiple data separation and model optimization processes, the optimization threshold and model configuration parameters are adjusted based on the deviation between each optimization result and the preset standard. The final generated complete 3D model information, key parameters, and optimization logs are output to the designer's end, and the full data record of each optimization process is stored. The model configuration parameter adjustment results are passed to the subsequent 3D model construction steps to update the processing logic.

9. The method for personalized design of knee braces based on 3D printing as described in claim 6, characterized in that, The specific process of separating the corresponding micro-detail augmentation dataset from the macro-structure subset includes: Identify local units with geometric defects in a macroscopic structural subset of data. These local units with geometric defects are composed of boundary voxels defined by a set of three-dimensional coordinate points. Based on the local unit of geometric defects, the parameters required to separate the micro-detail enhancement dataset are calculated, including the starting coordinate point, voxel width, and height dimension. Using the calculated separation parameters, the corresponding micro-detail enhancement dataset is separated from the macro-structure subset dataset through data pruning operations; Verify whether the separated micro-detail enhancement dataset completely covers the defect area and eliminate redundant data interference. The separation parameters are output to the verification module for automatic verification.

10. The method for personalized design of knee braces based on 3D printing as described in claim 4, characterized in that, When combining the outputs of the structural optimization model and the detail generation model, a fusion weight allocation mechanism is adopted. Different weight coefficients are assigned according to the confidence scores of the structural optimization model and the detail generation model. The confidence scores are obtained through training with historical data and updated in real time. The weight coefficients are output to the model output integration module to coordinate the generation of the 3D model.

Images