Orthopedic insole parameterization design method based on biomechanical data and rule base

By employing a parametric design method for orthotic insoles based on biomechanical data and a rule base, the problem of the lack of quantitative standards and verification methods in existing orthotic insole designs is solved, achieving personalized and reliable corrective effects and improving manufacturing efficiency and safety.

CN121365563AActive Publication Date: 2026-01-20XUZHOU MEDICAL UNIVERSITY

Patent Information

Application Number
CN202511939200.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-12-22
Publication Date
2026-01-20
Estimated Expiration
2045-12-22

Smart Images

  • Figure CN121365563A_ABST
    Figure CN121365563A_ABST
Patent Text Reader

Abstract

The invention belongs to the technical field of parameterization design, and discloses an orthopedic insole parameterization design method based on biomechanical data and a rule library, which comprises the following steps: by acquiring high-precision three-dimensional scanning data of a foot below a knee joint of a child, carrying out denoising processing and three-dimensional reconstruction on a foot point cloud to obtain a three-dimensional scanning data of the foot below the knee joint of the child; geometrical characteristic parameters and mechanical distribution characteristics of the foot are identified, and regional function segmentation is carried out on the foot according to the geometrical characteristic parameters and the mechanical distribution characteristics; the foot correction demand degree is calculated by analyzing an arch flatness quantitative index and a heel posture deviation characteristic, so that a personalized design parameter range is screened out; by fusing foot geometric features, simulation analysis results and material mechanical properties, an insole design scheme is constructed and optimized, a correction effect is predicted and compared with the simulation analysis results for analysis, and a design precision evaluation result is obtained; the method is applied to a personalized insole design scheme; according to the invention, accurate identification and personalized correction of the child foot problems are realized.
Need to check novelty before this filing date? Find Prior Art

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of parametric design, and more particularly, to an orthopedic insole parametric design method based on biomechanical data and a rule base. BACKGROUND

[0002] Flexible flatfoot in children is a common developmental disease of the foot, and its core feature is that the arch collapses under weight, and recovers when not under weight. The incidence rate of children over 6 years old is as high as 23.4%-59.98%, and the incidence rate of males is significantly higher than that of females. Epidemiological studies have shown that the disease is positively correlated with childhood obesity (the obesity rate of Chinese children has risen from 1.2% to 23.4% over the past 34 years) and a decrease in daily walking activities. Long-term non-intervention can easily cause foot tendon stretching and tearing, lower limb compensatory pain (hips, knees), and even affect gait development and overall body posture.

[0003] At present, orthopedic insoles are the core means of conservative treatment for flexible flatfoot in children, which play a correcting role by supporting the arch and optimizing the force line of the sole. However, the existing technical solutions have obvious defects: Traditional hand-made solutions rely on experience: generally use plaster or foam box molding methods, the entire process relies on the subjective judgment and experience of orthopedic therapists for modification, lacks quantitative standards, resulting in low individualization and poor consistency, and the effects of insoles made by different orthopedic therapists differ significantly.

[0004] Lack of verification of digital parameters: although existing digital technologies (such as CAD design) have improved production efficiency to some extent, the biomechanical effects of key design parameters (such as arch height, support area, and medial wedge height) lack systematic research, and cannot accurately match the development characteristics and individual differences of children's feet, affecting the correcting effect.

[0005] Single verification method: existing evaluations mostly rely on clinical subjective feedback or simple questionnaires, lack of objective pressure distribution monitoring and finite element simulation analysis support, and it is difficult to scientifically quantify the actual biomechanical effect of the correcting insole, and cannot ensure the safety and effectiveness of its long-term use.

[0006] In view of this, the present application proposes an orthopedic insole parametric design method based on biomechanical data and a rule base to solve the above problems. SUMMARY

[0007] In order to overcome the above-mentioned defects of the prior art, in order to achieve the above-mentioned purpose, the present application provides the following technical solutions: The orthopedic insole parametric design method based on biomechanical data and a rule base comprises: Step S1: High-precision three-dimensional scanning of the child's foot below the knee joint, collecting foot point cloud data and reconstructing a three-dimensional foot model based on it; based on the three-dimensional foot model, performing foot mechanics analysis to obtain foot correction requirement data; Step S2: Constructing a parameterized design candidate scheme library, performing parameter rule matching and scheme screening according to the foot correction requirement data to obtain an individualized insole design scheme, and configuring material attributes to obtain insole material parameter data; Step S3: Based on the obtained individualized insole design scheme, constructing a biomechanical simulation model, performing multi-posture mechanical simulation analysis and correction effect evaluation based on it to obtain design effect prediction results; and based on it, performing parameter iterative optimization on the individualized insole design scheme; obtaining an optimized insole design scheme; Step S4: According to the optimized insole design scheme, intelligent manufacturing and processing are performed to obtain a physical orthopedic insole, and pressure distribution testing is performed on the physical orthopedic insole; comparing and analyzing the pressure distribution test results with the corresponding simulation analysis results to obtain design precision evaluation results; and applying it to the individualized insole design scheme.

[0008] Further, the process of obtaining foot correction requirement data includes: A predetermined three-dimensional scanning device is used to perform high-precision data scanning on the child's foot below the knee joint to obtain foot point cloud data, and the obtained foot point cloud data is subjected to denoising filtering and data cleaning to obtain purified point cloud data; Based on the purified point cloud data, missing area identification and compensation are performed to obtain complete foot point cloud data, and based on it, a triangular mesh reconstruction is performed to obtain a three-dimensional mesh model of the foot; and based on the three-dimensional mesh model of the foot, feature parameter extraction is performed to obtain a set of foot geometric feature parameters; Based on the set of foot geometric feature parameters, an index quantification calculation is performed to obtain a foot arch flatness quantification index; and foot posture deviation data is simultaneously obtained; According to the foot posture deviation data and the foot arch flatness quantification index, foot type classification is performed, and based on the classification result, correction requirement degree evaluation is performed to obtain foot correction requirement data.

[0009] Further, the process of obtaining the foot arch flatness quantification index includes: The corresponding three-dimensional mesh model of the foot is subjected to region segmentation, and the foot region is divided into three functional regions of forefoot region, midfoot region and hindfoot region, and the area area distribution data of each functional region is obtained respectively; Based on the area area distribution, foot arch flatness calculation is performed to obtain a foot arch index value; and based on it, grade division is performed to obtain a foot arch flatness quantification index.

[0010] Further, the process of obtaining complete foot point cloud data includes: The spatial continuity of the purified point cloud data is detected, the foot scanning blind area and the data missing area are identified and marked, and the potential missing area is obtained; Based on the prior knowledge of foot anatomy, a shape prediction model is constructed to calculate the geometric shape of the potential missing area. Radial basis function interpolation method is used to compensate and reconstruct the point cloud data in the data missing area. The point cloud data after compensation and reconstruction is optimized for smoothness, and the complete foot point cloud data is obtained.

[0011] Further, the process of obtaining the insole material parameter data includes: Based on the orthogonal experimental design theory, a three-factor three-level parameterized combination matrix is constructed to obtain the insole parameter combination design space, and the constraint boundary is set to obtain the feasible parameter range data. Based on the feasible parameter range data, a standardized parameter combination is generated to obtain a parameterized design candidate scheme library. The parameterized design candidate scheme library is composed of several basic design candidate schemes; According to the foot correction requirement data, a parameter intelligent matching decision rule is constructed to obtain a multi-level matching rule tree. The multi-level matching rule tree is optimized with adaptive weight coefficients to obtain an adaptive matching rule set; Based on the adaptive matching rule set, the basic design candidate schemes are adaptively scored and sorted, and the best matching basic design candidate scheme is selected based on the adaptive scoring and sorting, which is marked as the personalized insole design scheme; Obtain the child's weight and activity data, and select the EVA material hardness intelligently based on it to obtain the material hardness parameter. Based on the personalized insole design scheme and the material hardness parameter, a complete insole design scheme is generated. The material mechanics attribute of the complete insole design scheme is configured to obtain the insole material parameter data.

[0012] Further, the process of obtaining the optimized insole design scheme includes: Based on the foot three-dimensional grid model and the personalized insole design scheme, a geometric assembly relationship is established to obtain the foot-insole contact interface geometry data. According to the insole material parameter data, the hyperelastic material constitutive relationship is set to obtain the material attribute configuration parameter; Based on the foot three-dimensional grid model and the foot-insole contact interface geometry data, respectively, the finite element grid is automatically divided to obtain the finite element grid model. Combined with the material attribute configuration parameter, a foot-insole-shoe multi-body coupling simulation model is established. The model is optimized to obtain a biomechanical simulation model. The biomechanical simulation model is simulated under multiple load conditions to obtain multi-posture simulation parameters; Based on the multi-posture simulation parameters, the stress concentration degree is analyzed to obtain the stress distribution evaluation result. The stress distribution evaluation result is quantitatively evaluated for correction effect to obtain the design effect prediction result; Based on the design effect prediction result, design defect recognition and parameter sensitivity analysis are performed to obtain the parameter optimization direction; and based on the same, parameter iteration update is performed on the individualized insole design scheme to obtain an optimized insole design scheme.

[0013] Further, the construction process of the biomechanical simulation model comprises: The tetrahedral mesh of the foot three-dimensional grid model is automatically divided, and the mesh density gradient control is set to obtain the foot finite element grid; Based on the individualized insole geometric design scheme, the insole geometric model is obtained, and the hexahedral dominant hybrid mesh is divided, and the mesh is encrypted in the key stress area to obtain the insole finite element grid; The contact constraint relationship between the foot bottom and the insole contact surface is established to obtain the contact constraint configuration data; Based on the foot finite element grid, the insole finite element grid and the contact constraint configuration data, the model geometry is assembled to obtain the foot-insole-shoe multi-body coupled simulation model; The boundary conditions and load conditions of the foot-insole-shoe multi-body coupled simulation model are set, and the grid quality is checked and the element distortion is optimized to obtain the biomechanical simulation model.

[0014] Further, the process of obtaining the design effect prediction result comprises: The peak pressure of the static pressure distribution data in the multi-posture simulation parameters is identified and regionally divided to obtain the high pressure area distribution atlas; The dynamic mechanical response data in the multi-posture simulation parameters are extracted to obtain the key phase mechanical data; Based on the obtained high pressure area distribution atlas and key phase mechanical data, the correction effect is evaluated, the index quantization index of foot pressure distribution, maximum von Mises stress and arch strain energy is analyzed, and the multi-dimensional correction effect score is obtained; and the design effect prediction result is generated according to the multi-dimensional correction effect score.

[0015] Further, the design precision evaluation result is obtained; and the process of applying it to the individualized insole design scheme comprises: The intelligent planning of the manufacturing process path is performed on the optimized insole design scheme to obtain the numerical control machining trajectory data; the cutting parameter optimization configuration is performed according to the material characteristics and the machining precision requirement to obtain the manufacturing process parameter set; Based on the numerical control machining trajectory data and the manufacturing process parameter set, the numerical control machining program file is generated; and the five-axis linkage numerical control milling is performed according to the numerical control machining program file to obtain the insole rough machining product; the precise polishing and surface finishing treatment is performed on the insole rough machining product to obtain the solid orthopedic insole; The thin film pressure sensor array is embedded in a physical orthopedic insole, and based on the embedded physical orthopedic insole, a dynamic test of actual wearing of children is carried out, and the plantar pressure time series data under various motion states are collected in real time to obtain the measured pressure distribution data; The measured pressure distribution data and the simulation mechanical response data are established in space-time correspondence to obtain a comparative analysis data set; Based on the comparative analysis data set, an accuracy evaluation index is calculated to obtain a design accuracy evaluation result; the accuracy evaluation index includes a pressure distribution correlation index, a peak value prediction accuracy index, and a region prediction accuracy index; According to the design accuracy evaluation result, an error source analysis and a parameter correction strategy are formulated to obtain feedback optimization parameters; and the feedback optimization parameters are applied to the individualized insole design scheme.

[0016] Further, the process of obtaining the design accuracy evaluation result includes: Based on the comparative analysis data set, a Pearson correlation coefficient of the measured value and the simulation value is calculated to obtain the pressure distribution correlation index; The measured pressure peak value and the simulation pressure peak value are subjected to error statistical analysis, and the mean absolute error and the root mean square error are calculated to obtain the peak value prediction accuracy index; The prediction accuracy of the biomechanical simulation model for the high-pressure region is evaluated by a receiver operating characteristic curve analysis method to obtain the region prediction accuracy index; A multi-dimensional accuracy evaluation model is constructed to comprehensively weight and score the pressure distribution correlation index, the peak value prediction accuracy index, and the region prediction accuracy index to obtain the design accuracy evaluation result.

[0017] The technical effects and advantages of the orthopedic insole parameterized design method based on biomechanical data and a rule base include: 1. Solving the "experience dependence" problem of the traditional manual scheme, realizing standardization and precision: the traditional manual scheme completely depends on the personal experience of an orthopedist, and is highly subjective and has poor consistency; the present application realizes objective quantitative collection of foot data through high-precision three-dimensional scanning, and combines parameterized design rules to completely eliminate the dependence on manual experience, and significantly improves the accuracy of diagnostic data and the repeatability of design schemes.

[0018] 2. Solving the "poor adaptability" problem of the existing digital scheme, realizing full-type individualization: the existing digital scheme mostly uses limited fixed parameters, which is difficult to cope with complex clinical conditions of flat feet combined with hallux valgus; the present application constructs a parameterized design candidate scheme library based on orthogonal experimental design, which can accurately cover all common types of mild, moderate and severe flat feet combined with different severity of hallux valgus, and truly realizes one insole for one person and precise matching.

[0019] 3. Innovatively introduce simulation and verification links to ensure the safety and effectiveness of correction: the traditional scheme lacks objective verification means; the application introduces finite element simulation after design, pre-evaluates the biomechanical effect, and verifies the actual effect after the physical production through the pressure monitoring system, forming a complete closed-loop design-verification-optimization process, significantly improving the scientificity and reliability of the correction scheme.

[0020] 4. Greatly improve the production efficiency and clinical generalizability: the traditional manual scheme takes several days to produce; the application can be completed within several hours from scanning to finished product production, with an efficiency improvement of more than 80%. At the same time, the technical process is highly standardized (equipment, software, parameter settings are all quantified), which reduces the technical threshold of the operator and is easy to promote and apply in medical institutions and rehabilitation equipment enterprises at all levels, and has important social and economic benefits. BRIEF DESCRIPTION OF DRAWINGS

[0021] Figure 1 It is a schematic diagram of the parameterized design method of the orthopedic shoe pad based on biomechanical data and rule base of the application. DETAILED DESCRIPTION

[0022] The technical solutions in the embodiments of the application will be described clearly and completely below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, not all. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the application.

[0023] Embodiment 1 Please refer to Figure 1 The parameterized design method of the orthopedic shoe pad based on biomechanical data and rule base of the embodiment includes: Step S1: High-precision three-dimensional scanning of the foot below the child's knee joint, collecting foot point cloud data and reconstructing a three-dimensional foot model based on it; performing foot mechanics analysis based on the three-dimensional foot model to obtain foot correction requirement data; Step S2: Construct a parameterized design candidate scheme library, perform parameter rule matching and scheme screening according to the foot correction requirement data, obtain an individualized shoe pad design scheme, and configure material properties to obtain shoe pad material parameter data; Step S3: Based on the obtained individualized shoe pad design scheme, a biomechanical simulation model is constructed, and based on it, multi-posture mechanical simulation analysis and correction effect evaluation are performed to obtain design effect prediction results; and based on it, parameter iterative optimization is performed on the individualized shoe pad design scheme; an optimized shoe pad design scheme is obtained. Step S4: intelligent manufacturing processing is performed according to the optimized insole design scheme, an entity orthopedic insole is obtained, and pressure distribution testing is performed on the entity orthopedic insole; the pressure distribution testing result is compared with the corresponding simulation analysis result, a design precision evaluation result is obtained, and the design precision evaluation result is applied to the individualized insole design scheme.

[0024] It needs to be further explained that, in the specific implementation process, the foot correction requirement data acquisition process includes: A preset three-dimensional scanning device is used to perform high-precision data scanning on the foot below the knee joint of the child, foot point cloud data is obtained, and the obtained foot point cloud data is subjected to denoising filtering and data cleaning to obtain purified point cloud data; Based on the purified point cloud data, missing area identification and compensation are performed to obtain complete foot point cloud data, and based on the complete foot point cloud data, triangular mesh reconstruction is performed to obtain a foot three-dimensional mesh model; and based on the foot three-dimensional mesh model, feature parameter extraction is performed to obtain a foot geometric feature parameter set; Based on the foot geometric feature parameter set, an index quantization calculation is performed to obtain a foot arch flatness quantization index; and foot posture deviation data is synchronously obtained. According to the foot posture deviation data and the foot arch flatness quantization index, foot type classification is performed, and based on the classification result, correction requirement degree evaluation is performed to obtain foot correction requirement data.

[0025] In this embodiment, first, the child's foot is placed in a preset three-dimensional scanning device (such as a Starlex lower limb three-dimensional scanner); and it is ensured that the foot is in a natural relaxed state; during the scanning process, multiple orientations are included, such as a front view angle, a left 45-degree angle, a right 45-degree angle, and a foot bottom view angle; the foot area below the knee joint of the child is scanned, and the scanning range includes key areas such as the foot bottom, the foot back, and the foot side; foot point cloud data is obtained; the collected foot point cloud data is subjected to denoising filtering processing by using Gaussian filtering, wherein the filter kernel size of the Gaussian filtering is dynamically adjusted according to the point cloud density of the foot point cloud data; after the filtering is completed, the filtered foot point cloud data is subjected to data cleaning, which refers to detecting outlier points in the foot point cloud data by statistical outlier detection, and the outlier points are removed to obtain purified point cloud data; for example, the average distance of each point to its K nearest neighbors is calculated, when the average distance of a certain point exceeds the mean value plus 2 times the standard deviation, the point is marked as an outlier and removed, and K is an integer.

[0026] Then, the purified point cloud data is subjected to spatial continuity detection, missing area caused by scanning angle limitation or foot occlusion is identified, a morphology prediction model is constructed based on prior knowledge of foot anatomy, and the missing area is subjected to point cloud data compensation reconstruction by using a radial basis function interpolation method; the compensated and reconstructed point cloud data is subjected to smoothness optimization processing to obtain complete foot point cloud data; Then, when the triangular mesh reconstruction is performed based on the complete foot point cloud data, first, the complete foot point cloud data is preprocessed, including resampling and normal vector estimation. The resampling process ensures the uniform distribution of point cloud density, avoiding the problem of too large or too small triangles in high curvature areas. The normal vector estimation fits a local plane in the neighborhood of each point cloud data point by using principal component analysis method, and calculates the plane normal vector as the normal vector of the point. Then, the maximum minimum angle principle is followed to ensure that the generated triangular mesh is as close to an equilateral triangle as possible. An incremental insertion strategy is used to insert point clouds one by one and update the triangulation structure to generate a triangular network model. The triangular network model contains several triangular facets, and the facet size is adaptively adjusted according to the geometric complexity of different regions of the foot: smaller facets are used in high curvature regions such as toes and arches to capture detailed features, and larger facets are used in relatively flat regions such as the plantar surface to improve computational efficiency. Topological checking and repair are performed on the generated triangular network model to obtain a three-dimensional mesh model of the foot.

[0027] Anatomical feature point recognition is performed on the constructed three-dimensional mesh model of the foot. The feature point recognition process first obtains the principal curvature and average curvature of each mesh vertex in the three-dimensional mesh model of the foot. Potential anatomical landmark points are identified through curvature extreme value detection. Then, feature point screening is performed in combination with prior knowledge of foot anatomy. The main recognized feature points include the first metatarsophalangeal joint, the fifth metatarsophalangeal joint, the scaphoid node, the posterior end of the calcaneus, the highest point of the arch, the lowest point of the sole, and other key anatomical landmarks. Based on the identified set of key feature points of the foot, geometric parameters are extracted, including foot length (distance from the posterior end of the calcaneus to the longest toe end), foot width (width from the first metatarsal head to the fifth metatarsal head), arch height (vertical distance from the scaphoid node to the ground), arch length (distance from the anterior end of the calcaneus to the first metatarsophalangeal joint), and calcaneal angle. These parameters constitute the foot geometric feature parameter set. Furthermore, the obtained three-dimensional mesh model of the foot is subjected to region segmentation, and the segmentation boundary is determined according to the foot length ratio to divide the foot region into several functional regions. The area distribution data of each functional region is obtained, and the arch index AI is calculated based on it. The arch flatness quantization index is obtained through level quantization based on it. Meanwhile, a rear high-definition camera is used to capture the heel posture, and the child keeps a natural standing posture during shooting. The camera is located behind the heel, and the lens optical axis is parallel to the ground. After image acquisition, the heel contour is extracted. First, the image is preprocessed, including grayscale, Gaussian filter denoising, and histogram equalization to enhance contrast. Then, the Canny edge detection algorithm is used to extract the heel contour, and the detection parameters are adaptively adjusted according to the image characteristics. After contour extraction, the least squares method is used to fit the heel axis, which represents the main direction of the heel. The heel valgus angle is defined as the angle between the heel axis and the vertical reference line. A positive value indicates valgus, and a negative value indicates varus. When classifying and evaluating foot types based on the arch flatness quantitative indicator and heel posture deviation data, a multi-factor comprehensive evaluation model is used. The multi-factor comprehensive evaluation model first classifies according to the arch index value, and then refines the classification combined with the heel posture deviation data. Foot type classification includes normal foot type, flat foot, high arch foot, varus foot, valgus foot, and corresponding compound foot types. For each foot type, there are corresponding correction requirement parameters, including required support strength, correction angle, and key correction area. Through table lookup matching and parameter interpolation, personalized foot correction requirement data is generated, providing accurate input conditions for subsequent insole parameterized design.

[0028] It should be further explained that in the specific implementation process, the arch flatness quantitative indicator acquisition process includes: The corresponding foot three-dimensional mesh model is regionally segmented, and the foot region is divided into three functional regions: forefoot region, midfoot region, and hindfoot region. The area distribution data of each functional region is obtained. Based on the area distribution, the arch flatness is calculated to obtain the arch index value. Based on this, the arch flatness quantitative indicator is obtained.

[0029] In this embodiment, the foot three-dimensional mesh model is regionally segmented using an adaptive segmentation method based on anatomical knowledge. The segmentation process first determines the segmentation boundary according to the foot length ratio: the forefoot region from the toe tip to the first metatarsal head line; the midfoot region from the first metatarsal head line to the rear of the highest arch point; the hindfoot region from the rear of the highest arch point to the heel. The segmentation method establishes a segmentation plane in three-dimensional space, uses the least squares method to fit the best segmentation plane through the landmark points, and then judges the region attribution according to the relative position of the grid vertex and the segmentation plane. For vertices near the segmentation boundary, a distance weighting method is used for smooth attribution to avoid region discontinuity caused by hard boundaries. The projection area of each region is calculated, with the projection direction being vertically downward (negative Z direction). The total area is calculated using the triangular facet projection area accumulation method. The projection areas of the forefoot region, midfoot region, and hindfoot region are obtained. The specific values vary with the child's age and foot size. Based on the area distribution data of the region, the arch index value AI is calculated, which is a key indicator for evaluating the degree of flat foot; the calculation formula is: AI = midfoot area / (forefoot area + midfoot area + hindfoot area); then, according to the obtained arch index value AI, and combined with the clinical standard, it is divided into normal foot arch, mild flat foot, moderate to severe flat foot three grades, to obtain the flat foot arch quantification index.

[0030] It needs to be further explained that in the specific implementation process, the acquisition process of the complete foot point cloud data includes: The spatial continuity of the purified point cloud data is detected, and the foot scanning blind area and the data missing area are identified and marked to obtain the potential missing area; Based on the prior knowledge of foot anatomy, a shape prediction model is constructed to calculate the geometric shape of the potential missing area; and a radial basis function interpolation method is used to compensate and reconstruct the point cloud data in the data missing area; the smoothed and optimized point cloud data after compensation and reconstruction is obtained.

[0031] In this embodiment, the eight-neighbor connectivity analysis method is used to detect the spatial continuity of the purified point cloud data, the neighborhood density of each point cloud data point in its adjacent neighborhood is analyzed, and the difference between the neighborhood densities is identified, the neighborhood area with a difference less than the neighborhood density threshold is marked as a potential missing area; through connectivity analysis, the scanning blind area inside the arch, the toe gap and the rear side of the heel are identified; based on the prior knowledge of foot anatomy, a statistical shape morphological prediction model is established using a large amount of anatomical data of normal children's feet; the morphological prediction model integrates the standard morphological data and growth and development rules of children's feet in different age groups, including the statistical distribution of typical geometric features of each region of the foot, such as arch curvature, toe bone length ratio, heel width and other parameters; the morphological prediction model uses principal component analysis method to extract the main variation mode of foot morphology, establishes a foot morphology space, and then based on the foot morphology space, according to the approximate range and shape characteristics of the known foot point cloud data in the potential missing area, radial basis function interpolation method is used for compensation and reconstruction, and the corresponding compensation and reconstruction mathematical formula is: ; In the formula, represents the point (x, y, z) in the three-dimensional space; The result of compensation and reconstruction; p(x, y, z) represents a polynomial term, which is used to ensure the consistency of interpolation; represents the point cloud data point that needs to be compensated and reconstructed, with coordinates (x, y, z); ; represents the i-th known point cloud data point, i = 1, 2, …, n, n is an integer, representing the total number of known point cloud data points; represents the radial basis function; The interpolation weight coefficient is represented; after the compensation reconstruction is completed, the compensation point cloud data is obtained, and it is merged with the original purification point cloud data to obtain complete foot point cloud data; and in the compensation reconstruction process, the smooth connection with the surrounding point cloud is preferentially ensured to avoid the generation of sharp changes or unreasonable geometric shapes; the point cloud data after the compensation reconstruction is subjected to smoothness optimization processing, and the Laplace smoothing algorithm is adopted to eliminate the geometric discontinuity that may be generated in the reconstruction process, so as to ensure the spatial consistency and smoothness of the entire foot point cloud.

[0032] It needs to be further explained that, in the specific implementation process, the acquisition process of the insole material parameter data includes: Based on the orthogonal test design theory, a three-factor three-level parameterized combination matrix is constructed to obtain an insole parameter combination design space, and a constraint boundary is set to obtain feasible parameter range data; and based on the feasible parameter range data, a standardized parameter combination is generated to obtain a parameterized design candidate scheme library; the parameterized design candidate scheme library is composed of a plurality of basic design candidate schemes; According to the foot correction requirement data, a parameter intelligent matching decision rule is constructed to obtain a multi-level matching rule tree; the multi-level matching rule tree is subjected to weight coefficient adaptive optimization to obtain an adaptive matching rule set; Based on the adaptive matching rule set, the basic design candidate schemes are subjected to adaptive scoring and sorting, and the best matching basic design candidate scheme is selected based on the adaptive scoring and sorting and is marked as a personalized insole design scheme; The child weight and activity data are obtained, and based thereon, EVA material hardness intelligent selection is performed to obtain material hardness parameters; based on the personalized insole design scheme and the material hardness parameters, a complete insole design scheme is generated; the complete insole design scheme is subjected to material mechanics attribute configuration to obtain insole material parameter data.

[0033] In this embodiment, first determine the key geometric parameters that affect the correction effect of insole, mainly including arch lifting height, arch support area and heel medial wedge height and other factors; And through the use of L27 orthogonal table for parameter combination design, and design three levels for each factor (for example: arch lifting height: 4mm (suitable for mild support), 7mm (moderate support), 10mm (strong support)); Through the parameter combination of L27 orthogonal table, generate several different parameter combinations, and form the insole parameter combination design space, ensure the uniform distribution and independence of each factor level; Set the constraint boundary for the obtained insole parameter combination design space; According to the characteristics of children's foot development and insole manufacturing process Set the constraint condition; The constraint condition includes that the arch support height should not exceed the reference ratio of the natural height of the arch, in order to avoid excessive correction; The combination of calcaneal wedge angle and forefoot inclination angle should ensure the overall balance of the insole; The maximum thickness of the insole should not exceed 20mm, in order to ensure the adaptability in the standard shoes; Through constraint screening, eliminate the parameter combinations that do not meet the conditions, and obtain several groups of effective combination of feasible parameter range data; According to the feasible parameter range data, generate standardized parameter combinations; For each feasible parameter combination, perform three-dimensional geometric modeling to generate the corresponding insole geometry; The modeling process uses parameterized CAD technology to automatically generate a three-dimensional solid model of the insole by controlling key geometric parameters; Each design scheme contains complete geometric information, such as surface curvature distribution, volume, center of gravity position, etc.; Verify the geometric reasonableness of the generated basic design candidate scheme set, the inspection items include: the smoothness of the connection between the arch lifting and the wedge structure, the coordination between the support area and the insole boundary, the rationality of the overall thickness distribution. The manufacturing process compatibility check mainly targets CNC carving and 3D printing two main manufacturing methods, to evaluate the processability of parameter combination. The inspection standards include: the minimum feature size is not less than the CNC tool limit, the overhang angle is not more than the 3D printing support limit, the surface curvature changes continuously and smoothly (post-processing process requirement); Ergonomic check to ensure that the design conforms to the principles of human foot biomechanics; The verified schemes are retained in the candidate scheme library to form the final parameterized design candidate scheme library, which contains several verified basic design candidate schemes; When constructing the multi-level parameter matching rule tree based on foot correction requirement data, the matching rules are organized in a decision tree structure; the first layer of the multi-level parameter matching rule tree branches according to foot type classification, including main branches such as normal foot, flat foot, and high arch foot; the second layer is further subdivided according to the correction degree, such as mild correction, moderate correction, and severe correction; the third layer performs final matching according to specific geometric parameter requirements; each rule node contains a judgment condition and a corresponding parameter recommendation value, the judgment condition is based on foot geometric feature parameters, and the parameter recommendation value corresponds to the value range of the insole design parameter; the construction of the rule tree is based on a large number of clinical cases and expert experience, and the rule patterns are extracted from historical data through machine learning methods; when adaptively optimizing the weight coefficients of the matching rule set, a weight adjustment mechanism based on feedback learning is adopted; the initial weights are set according to expert experience, and then dynamically adjusted through feedback data of actual use effect; the objective function of weight optimization considers evaluation indexes of correction effect, comfort, and adaptability, and iteratively optimizes the weight coefficients through gradient descent algorithm, so that the matching rules can more accurately reflect the actual requirements; after weight optimization is completed, the adaptive matching rule set is obtained.

[0034] When scoring the parameterized design candidate scheme library according to the adaptive matching rule set, a multi-factor weighted scoring model is adopted. The calculation formula of the scoring model is: Score = W1 x Correction + W2 x Comfort + W3 x Stability + W4 x Adaptability; wherein Score represents the comprehensive adaptability score of the corresponding basic design candidate scheme, Correction represents the correction effect score, Comfort represents the comfort score, Stability represents the stability score, Adaptability represents the adaptability score, W1, W2, W3, and W4 are the corresponding weight coefficients, and W1 + W2 + W3 + W4 = 1; each score is scored on a scale of 0-100, and the score is determined according to the matching degree of the design scheme and individual requirements; the correction effect score can be used to show the degree of coincidence between the design parameters and the correction requirements; the comfort score can be used to show the uniformity of pressure distribution and contact area of the insole; the stability score can be used to show the structural stability and deformation resistance of the insole; the adaptability score is used to show the compatibility of the insole with commonly used footwear; based on the comprehensive adaptability score of the scheme, the schemes are sorted and selected, and the scheme with the highest score is selected as the individualized insole design scheme; the top 3 high-score schemes can also be selected for further optimization and selection.

[0035] The child's weight and activity intensity data are acquired, and an EVA material hardness selection decision model is established. The weight data is directly measured, and the activity intensity data is obtained through questionnaire survey and activity scale evaluation, which is divided into three levels: low intensity (mainly daily walking), medium intensity (including running and sports activities), and high intensity (frequently participating in competitive sports). The decision model considers both weight and activity intensity, and establishes a two-dimensional decision matrix. The basic decision rule is: weight < 20 kg and low activity intensity, select EVA hardness 65 (provide soft support); weight 20-40 kg or medium activity intensity, select EVA hardness 75 (balance support and comfort); weight > 40 kg or high activity intensity, select EVA hardness 80 (provide firm support); the model also considers adjustment rules for special cases: for children with light weight but high activity intensity, the hardness is appropriately increased to cope with impact load; for children with heavy weight but sensitive feet, the hardness can be appropriately reduced to improve comfort; the final material hardness parameter not only includes the numerical value, but also includes the hardness tolerance range and quality control requirements, to ensure the consistency of material performance in the manufacturing process.

[0036] Based on the individualized insole design scheme and material hardness parameters, a complete set of material mechanics attribute parameters is configured; according to the selected EVA hardness level, the corresponding mechanics attribute parameters are obtained from the material database, including elastic modulus, Poisson's ratio and density, etc.; and data integration is carried out to obtain a complete insole design scheme containing information such as geometry, size specification, material specification, surface treatment, etc.; and dynamic mechanics performance parameters are configured, including compression permanent deformation rate, resilience, fatigue resistance, etc., which are crucial for predicting the long-term performance of the insole; the insole material parameter data is obtained; among them, the insole material parameter data also includes parameters related to manufacturing process, such as processing temperature range, tool selection suggestion, surface treatment requirement, etc., providing comprehensive technical support for subsequent manufacturing process. By establishing a complete material parameter database, it is ensured that the design scheme can be accurately converted into specific manufacturing instructions and quality control standards.

[0037] It needs to be further explained that in the specific implementation process, the optimization of the insole design scheme acquisition process includes: Based on the three-dimensional mesh model of the foot and the individualized insole design scheme, the geometric assembly relationship is established, and the foot-insole contact interface geometric data is obtained; the hyperelastic material constitutive relationship is set according to the insole material parameter data, and the material attribute configuration parameters are obtained; Based on the three-dimensional mesh model of the foot and the foot-insole contact interface geometric data, the finite element mesh is automatically divided, and the finite element mesh model is obtained; the foot-insole-shoe multi-body coupling simulation model is established combined with the material attribute configuration parameters; and the model is optimized to obtain the biomechanics simulation model; and the biomechanics simulation model is simulated under multiple working conditions to obtain the multi-posture simulation parameters; Based on the multi-pose simulation parameters, stress concentration degree analysis is performed to obtain stress distribution evaluation results; the stress distribution evaluation results are subjected to quantitative evaluation of correction effect to obtain design effect prediction results; Based on the design effect prediction results, design defect identification and parameter sensitivity analysis are performed to obtain parameter optimization direction; and based on the same, parameter iterative updating is performed on the individualized insole design scheme to obtain an optimized insole design scheme.

[0038] In this embodiment, when establishing the geometric assembly relationship based on the foot three-dimensional mesh model and the individualized insole design scheme, first, the coordinate system is unified, and the foot three-dimensional mesh model and the three-dimensional model of the insole are placed in the same three-dimensional coordinate system. The assembly process is realized through feature matching and surface fitting to ensure the contact relationship of the insole and the foot bottom forming the force; the key of the assembly is to determine the contact area and the contact state of the foot bottom and the insole surface; including the complete contact area, the partial contact area and the non-contact area; the foot-insole contact interface geometric data includes contact area, contact pressure distribution estimation, contact normal vector and other information; When setting the hyperelastic material constitutive relationship according to the insole material parameter data; according to the elastic modulus, Poisson's ratio and other parameters in the insole material parameter data, the Mooney-Rivlin or Neo-Hookean hyperelastic material constitutive relationship is set in the finite element software, and the nonlinear mechanical behavior of the EVA material is described by using the hyperelastic material model to define the nonlinear stress-strain response characteristics of the material, and the complete material attribute configuration parameters are formed.

[0039] Then the foot three-dimensional mesh model is discretized by using tetrahedral mesh elements (such as Tet4 or Tet10 elements), and the foot complex geometry is discretized, and denser grids are set in the high stress areas such as the heel and the forefoot to improve the calculation accuracy. The hexahedral dominant hybrid mesh division strategy is used for the insole geometric model, the hexahedral element (such as Hex8 element) is used in the regular area to improve the calculation efficiency, and the grid density gradient control strategy is set in the mesh division process, and the mechanical importance and geometric complexity of different areas of the foot are controlled in stages; such as using tetrahedral elements in complex geometric areas to ensure grid quality, and performing grid encryption processing in key stress areas such as the arch support area and the heel contact area to ensure the calculation accuracy of the stress gradient of the area with large stress gradient, and generating a foot finite element grid model meeting the quality requirements (such as element shape ratio, Jacobian determinant and other indicators); When the personalized insole design scheme is used to obtain the insole geometric model, the geometric parameters in the parameterized design scheme are converted into a three-dimensional CAD model, including the accurate geometric description of the functional partitions such as the arch support area, the calcaneal wedge area, and the forefoot cushioning area; the model generation adopts NURBS surface modeling to ensure smooth transition and overall geometric continuity between the areas; when the hexahedral dominant hybrid mesh is divided for the insole geometric model, the block mapping mesh division strategy is adopted, the insole geometry is first divided into several regular topological blocks, and structured hexahedral mesh is used inside each block; for the transition area with complex geometry, tetrahedral mesh is used for filling to form a hexahedral-tetrahedral hybrid mesh structure; the generation of the hexahedral mesh adopts a combination of the sweep algorithm and the mapping algorithm, the sweep algorithm is suitable for areas with stretching characteristics, and the mapping algorithm is suitable for regular quadrilateral areas; when the mesh is encrypted in the key stress area, the focus is on the arch support point, the calcaneal contact area, the forefoot support area and other high stress areas, and the mesh density is adaptively improved; the mesh encryption adopts a gradual transition strategy to avoid stress concentration artifacts caused by sudden changes in mesh density; finally, the insole finite element mesh model is generated.

[0040] When the contact constraint relationship is established between the foot bottom and the insole contact surface, the geometric identification and pairing of the contact surface are first performed, and the corresponding relationship between the foot bottom surface and the insole upper surface is determined through the surface projection algorithm; the nearest point projection method is used for contact surface pairing, and for each node of the foot bottom surface, the shortest distance and the projection point position to the insole surface are calculated. The master-slave surface contact algorithm is used to establish the contact constraint relationship, and the foot bottom surface is defined as the master contact surface and the insole upper surface is defined as the slave contact surface; the selection of master and slave surfaces is based on mesh density and surface curvature, and the surface with relatively coarse mesh density or smaller curvature is usually selected as the master surface; the mathematical description of the contact constraint uses the penalty function method, and based on these parameters, the contact constraint configuration data including the complete parameter set of the master-slave surface pairing information, the penalty function, etc. is obtained, providing a numerical basis for contact analysis.

[0041] When the model geometry is assembled based on the foot finite element mesh model, the insole finite element mesh model and the contact constraint configuration data, a hierarchical assembly strategy is used to ensure the correct positioning and constraint relationship of each component; the assembly process first establishes a global coordinate system, and places the foot finite element mesh model and the insole finite element mesh model in a unified three-dimensional space coordinate system; the positioning of the foot model is based on the anatomical landmark points to ensure that the foot is in a natural standing posture; the positioning of the insole model is achieved through feature point matching to ensure that the insole and the foot bottom form a reasonable initial contact state; the model assembly also includes the establishment of the constraint relationship of the shoe sole and the ground, and the shoe sole surface is subjected to fixed constraints to simulate the supporting effect of the ground; the connection relationship between the components is realized through node sharing, constraint equations or contact constraints to ensure the correct force transmission path. The foot-insole-shoe multi-body coupled simulation model after assembly contains three main components: foot, insole and shoe sole; When setting the boundary conditions and load conditions for the foot-insole-shoe multi-body coupling simulation model, the boundary conditions include fixed constraints and symmetry constraints; the fixed constraints are applied to the upper surface of the foot to simulate the fixed support of the lower leg; the symmetry constraints are applied to the symmetry surface of the model to reduce the calculation scale; the load conditions include two types of body weight load and dynamic load; the body weight load is applied according to the actual weight of the child and the contact area distribution of the foot; the dynamic load simulates the load change in the gait process, including the load distribution of typical gait phases such as heel landing, plantar flat support, and toe propulsion; the application of load uses a time history function control to simulate the dynamic characteristics of the real gait; After the settings are completed, a multi-level quality evaluation system is used to ensure the numerical reliability of the model; the grid quality check includes geometric quality check and topological quality check; the geometric quality check mainly evaluates the shape quality of the element, including aspect ratio, skewness, and warping degree; the topological quality check mainly checks the correctness of the grid connection, including node sharing check, element normal check, and boundary integrity check; for elements that do not meet the quality standards, local grid optimization algorithm is used for correction; element distortion optimization uses a combination of node position adjustment and element redivision; node position adjustment is realized by Laplace smoothing algorithm; for severely distorted elements, local redivision method is used to delete the distorted elements and generate high-quality grids; the optimization process is iterated until all elements meet the quality standards; the final biomechanical simulation model has good grid quality, regular element shape, and correct connection relationship, which can ensure the numerical accuracy and convergence of finite element calculation, and provide a reliable calculation model basis for accurately predicting the biomechanical performance of the insole.

[0042] When analyzing the stress concentration degree according to the multi-attitude simulation parameters, von Mises stress is used as the evaluation index to identify the high stress areas in the foot and insole; the stress concentration degree is defined as the ratio of the local maximum stress to the average stress; the area with a ratio greater than the preset threshold is considered to have stress concentration; the foot pressure distribution evaluation is realized by extracting the contact pressure data of the foot-insole contact interface; the uniformity of the pressure distribution and the position of the peak pressure are analyzed; the uniformity of the pressure distribution is evaluated by the coefficient of variation; a coefficient of variation less than the preset variation threshold indicates a more uniform pressure distribution; the stress distribution evaluation results include the maximum von Mises stress value, the stress concentration area distribution, the foot pressure peak, and the uniformity of the pressure distribution, etc. When quantitatively evaluating the correction effect based on the stress distribution evaluation results, a multi-dimensional evaluation system is established, including arch strain energy, foot pressure redistribution degree, high pressure area change, etc.; the correction effect evaluation uses comparative analysis with the basic state (without insole); the improvement degree of each index is calculated to form a quantitative design effect prediction result.

[0043] When performing defect identification and parameter sensitivity analysis on the predicted design effect, the first step is to establish effect evaluation criteria, including specific indicator limits such as pressure reduction rate, stress concentration elimination rate, and arch support effect improvement rate. Aspects that do not meet the standards are identified as design defects and require parameter adjustments. The parameter sensitivity analysis uses a single-factor variation method, changing the design parameters one by one and observing the degree of impact on the simulation results to identify the key parameters with the greatest impact on the corrective effect. The results of the sensitivity analysis are used to determine the priority and adjustment direction of parameter optimization. When iterating and updating the personalized insole design scheme according to the parameter optimization direction, a gradient optimization algorithm is used to adjust the design parameters along the direction of effect improvement. The step size of parameter adjustment is determined based on the sensitivity analysis results; parameters with high sensitivity use smaller step sizes, and parameters with low sensitivity can use larger step sizes. The iteration process is usually carried out for 3-5 rounds until the design effect reaches the preset standard or the improvement is less than 5%, at which point the final optimized insole design scheme is obtained.

[0044] It should be further explained that, in the specific implementation process, the process of obtaining the design effect prediction results includes: Peak pressure and region division are performed on the static pressure distribution data within the multi-pose simulation parameters to obtain a high-pressure zone distribution map. The key phases of the gait cycle are extracted from the dynamic mechanical response data within the multi-posture simulation parameters to obtain the key phase mechanical data; Based on the obtained high-pressure zone distribution map and key temporal mechanical data, the correction effect is evaluated. The quantitative indicators of plantar pressure distribution, maximum von Mises stress and arch strain energy are analyzed to obtain a multi-dimensional correction effect score. The design effect prediction result is generated based on the multi-dimensional correction effect score.

[0045] In this embodiment, specifically, when identifying peak pressure and dividing regions in the static pressure distribution data within the multi-pose simulation parameters, the plantar contact area is first divided into 9 functional regions according to anatomical standards, including the big toe region, other toe regions, the first metatarsal head region, the second-third metatarsal head region, the fourth-fifth metatarsal head region, the medial midfoot region, the lateral midfoot region, the medial heel region, and the lateral heel region. The pressure data within each region is continuously distributed through node pressure interpolation, and the local extremum search algorithm is used to identify the pressure peak points of each region. The peak pressure identification standard is: the pressure value within the region is greater than a preset multiple of the average pressure of the region and the difference between the peak pressure of the region and the peak pressure of the adjacent region is greater than a preset pressure threshold. After region division, the average pressure, peak pressure, contact area, and pressure integral value of each region are statistically analyzed to form a standardized pressure distribution feature vector. The high-pressure zone determination standard is that the pressure value exceeds a preset multiple threshold of the average pressure obtained by dividing the body weight by the total plantar contact area. Through threshold segmentation and connected region analysis, continuous high-pressure zones are identified and their area proportions are calculated. A high-pressure zone distribution map is generated.

[0046] The dynamic mechanical response data in the multi-attitude simulation parameters are subjected to gait cycle key phase extraction, and the typical gait events are identified based on the time domain characteristics of the plantar total pressure curve; the gait cycle division adopts double criteria of pressure change rate and pressure amplitude, and four key phase points of heel landing, plantar flat, heel lifting and toe lifting are identified; the heel landing phase is identified by the sharp rise of the rear foot area pressure, and the determination standard is that the pressure change rate is greater than the preset change rate threshold and the duration exceeds the preset time threshold; the plantar flat phase is identified by the relatively stable full foot pressure; the heel lifting phase is identified by the rapid decline of the rear foot area pressure; the extracted key phase mechanical data include parameters such as pressure distribution state, pressure center position, total pressure size and regional pressure proportion at each phase point.

[0047] When the obtained high pressure area distribution map and key phase mechanical data are used for correction effect evaluation, a multi-dimensional evaluation index system is established to comprehensively evaluate the static and dynamic correction effects of the insole; the static correction effect evaluation focuses on the improvement of the uniformity of the pressure distribution, and calculates the variation coefficient of the pressure distribution, the reduction amplitude of the peak pressure and the reduction proportion of the high pressure area; the dynamic correction effect evaluation focuses on the pressure transfer mode and timing characteristics in the gait process, and analyzes the smoothness of the pressure center trajectory, the rationality of the pressure distribution at each phase and the dynamic response of the arch support; the quantitative score of the correction effect adopts weighted comprehensive evaluation to obtain the multi-dimensional correction effect score; the design effect prediction result is generated according to the multi-dimensional correction effect score, which needs to be combined with the pre-set score level standard; for example, more than 90 points is excellent, indicating that the correction effect is significant; 75-90 points is good, indicating that the correction effect is obvious; 60-75 points is qualified, indicating that the correction effect is general; less than 60 points is unqualified and needs to be redesigned; the design effect prediction result not only includes the overall score, but also provides detailed analysis and improvement suggestions for each index.

[0048] It needs to be further explained that, in the specific implementation process, the design precision evaluation result is obtained; and it is applied to the process of individualized insole design scheme, which includes: The optimized insole design scheme is subjected to intelligent planning of manufacturing process path to obtain numerical control machining trajectory data; the cutting parameter optimization configuration is carried out according to the material characteristics and machining precision requirements to obtain a manufacturing process parameter set; The numerical control machining program file is generated based on the numerical control machining trajectory data and the manufacturing process parameter set; and the five-axis linkage numerical control milling is carried out according to the numerical control machining program file to obtain the insole rough machining product; the insole rough machining product is subjected to precise polishing and surface finishing treatment to obtain the physical orthopedic insole; The thin film pressure sensor array is embedded in the physical orthopedic insole, and based on the embedded physical orthopedic insole, the actual wearing dynamic test of children is carried out, and the plantar pressure time series data under various motion states are collected in real time to obtain the measured pressure distribution data; The measured pressure distribution data and the simulated mechanical response data are established in space-time correspondence to obtain a comparative analysis data set; Based on the comparative analysis data set, precision evaluation indexes are calculated to obtain design precision evaluation results; the precision evaluation indexes include pressure distribution correlation indexes, peak prediction precision indexes, and regional prediction precision indexes; According to the design precision evaluation results, error source analysis and parameter correction strategy are made to obtain feedback optimization parameters; and the feedback optimization parameters are applied to the individualized insole design scheme.

[0049] In this embodiment, when the optimized insole design scheme is manufactured, first, the insole three-dimensional CAD model is imported into professional CAM software for process analysis, the geometric features such as grooves, bevels, fillets, etc. in the model are identified, and the optimal machining direction and fixture positioning scheme are determined; the process path planning adopts a layered processing strategy, the insole model is divided into multiple processing layers according to the Z-axis direction, the machining path generation adopts a method combining the contour line strategy and the area filling strategy, the contour line path is used for flat areas to ensure the surface quality, and the area filling path is used for steep areas to improve the processing efficiency; the tool path optimization adopts a genetic algorithm for global optimization, and the objective function comprehensively considers the processing time, surface quality, tool wear and other factors; the generated numerical control machining track data contains complete tool path coordinates, feed speed, spindle speed and cooling liquid control instructions; when the cutting parameter optimization configuration is carried out according to the EVA material characteristics and the processing precision requirements, the thermoplasticity and viscoelasticity characteristics of the EVA material are fully considered; the optimized cutting parameters include spindle speed, feed speed control, cutting depth and tool selection.

[0050] When generating a numerical control machining program file based on numerical control machining trajectory data and a set of manufacturing process parameters, a complete machining program is written using standard G code and M code; the program structure includes three parts: the program header (workpiece coordinate system setting, tool calling), the main machining program (rough machining, semi-finishing machining, and finishing machining), and the program tail (tool return, spindle stop, and coolant shutdown); real-time monitoring instructions are embedded in the machining program, including spindle load monitoring, temperature monitoring, and vibration monitoring, to ensure the stability of the machining process. According to the numerical control program code, five-axis linkage numerical control milling machining is performed, and after the machining is completed, size detection is performed, and when precise polishing and surface finishing treatment of the insole rough machining product is performed, a gradual polishing process is adopted, from coarse grinding (80-120 mesh sandpaper) to fine grinding (600-1000 mesh sandpaper) to gradually improve the surface quality; during the polishing process, the pressure and temperature are controlled to avoid material deformation caused by excessive heating, and a solid orthopedic insole is obtained.

[0051] A thin film pressure sensor array is embedded in a solid orthopedic insole, and based on the embedded sensor, a dynamic test of children's actual wear is performed, including static standing, slow walking, normal walking, fast walking, and light running in five motion states, to obtain sufficient data samples; the real-time collected plantar pressure time series data is processed through digital filtering and noise elimination to extract effective pressure distribution information.

[0052] When establishing the space-time correspondence relationship between the measured pressure distribution data and the simulated mechanical response data, first, time synchronization calibration is performed to align the time axes of the measured data and the simulation data to the same gait cycle reference system; the space correspondence relationship is established through the mapping of sensor positions and simulation grid nodes, and the nearest neighbor interpolation method is used to interpolate discrete sensor data onto continuous simulation grids; during comparative analysis, the effects of measurement error and simulation error are considered, and statistical methods are used for data processing and analysis; the generated comparative analysis dataset includes measured pressure values and simulation prediction values at corresponding times and positions, providing a direct data basis for accuracy evaluation. When calculating the accuracy evaluation indicators based on the comparative analysis dataset, a variety of statistical indicators are used to comprehensively evaluate the prediction accuracy. The pressure distribution correlation index is obtained by calculating the Pearson correlation coefficient of the measured value and the simulation value, reflecting the linear correlation degree of the two; the peak prediction accuracy index is obtained by calculating the relative error of the peak pressure, evaluating the accuracy of extreme value prediction; the regional prediction accuracy index is obtained by analyzing the pressure prediction error of each anatomical region, evaluating the reliability of spatial distribution prediction. The design accuracy evaluation result integrates these three indicators, and the overall accuracy score is calculated using a weighted average method.

[0053] When error source analysis and parameter correction strategy are made according to the design accuracy evaluation results, the main factors affecting the prediction accuracy are identified by using the error analysis method; the error source analysis includes three aspects of model error (geometric simplification, material parameter deviation), numerical error (grid discretization, algorithm convergence) and measurement error (sensor accuracy, environmental interference); the relative contribution of each type of error source is determined through error decomposition and sensitivity analysis, and an error correction model is established; the parameter correction strategy is made according to the error source analysis results, including material parameter correction, geometric model correction, boundary condition optimization and other specific measures; the generation of feedback optimization parameters uses machine learning method, which trains the error prediction model through historical data to realize intelligent correction of parameters; when the feedback optimization parameters are applied to the individualized insole design scheme, the incremental updating strategy is used for hierarchical correction according to the error size and confidence to ensure the reliability of the correction effect. Through this closed-loop feedback optimization mechanism, the system can continuously improve the design accuracy and product quality, and realize the continuous improvement and optimization of the design scheme.

[0054] It needs to be further explained that in the specific implementation process, the acquisition process of the design accuracy evaluation results includes: The Pearson correlation coefficient of the measured value and the simulation value is calculated based on the comparative analysis data set to obtain the pressure distribution correlation index; The error statistical analysis is performed on the measured pressure peak value and the simulation pressure peak value, and the mean absolute error and the root mean square error are calculated to obtain the peak value prediction accuracy index; The prediction accuracy of the biomechanical simulation model for the high pressure area is evaluated by the receiver operating characteristic curve analysis method to obtain the area prediction accuracy index; A multi-dimensional precision evaluation model is constructed to comprehensively weight and score the pressure distribution correlation index, the peak value prediction accuracy index and the area prediction accuracy index to obtain the design accuracy evaluation results.

[0055] In this embodiment, when calculating the Pearson correlation coefficient of the measured value and the simulation value based on the comparative analysis data set, first, the measured pressure data and the simulation pressure data are preprocessed, including abnormal value detection, data smoothing and standardization processing; then, the correlation coefficient of the pressure distribution in the comparative analysis data set and the pressure distribution result in the simulation result is calculated based on the Pearson correlation coefficient, and is recorded as the pressure distribution correlation index; When performing error statistical analysis on the measured pressure peak value and the simulation pressure peak value, first, identify the pressure peak value points in each gait cycle, including the full foot pressure peak value and the regional pressure peak value; calculate the mean absolute error and the root mean square error to quantify the peak value prediction accuracy; the peak value prediction accuracy index also includes relative error analysis, which calculates the distribution characteristics and statistical parameters of the relative error to evaluate the systematicness and randomness of the prediction deviation.

[0056] When evaluating the prediction accuracy of the biomechanical simulation model for the high pressure area by the receiver operating characteristic curve (ROC) analysis method, first, a binary classification standard for the high pressure area is established, and the area with a pressure value exceeding the threshold value is defined as the high pressure area, and the remaining area is defined as the normal pressure area; the ROC analysis compares the simulation results with the measured results, and the true positive rate (sensitivity) and false positive rate (1-specificity) are calculated under different decision thresholds; the true positive rate represents the proportion of correctly identified high pressure areas, and the false positive rate represents the proportion of incorrectly identified normal areas as high pressure areas; by changing the decision threshold, an ROC curve is drawn, and the area under the curve (AUC) is used as a quantitative indicator of prediction accuracy to obtain the regional prediction accuracy indicator; the regional prediction accuracy indicator also includes sub-regional prediction accuracy analysis, focusing on evaluating the prediction reliability of key areas such as the metatarsal head area, the arch area, and the heel area.

[0057] When constructing the multi-dimensional accuracy evaluation model, the pressure distribution correlation index, the peak value prediction accuracy index, and the regional prediction accuracy index are comprehensively weighted and scored, and the weight coefficients are determined according to the clinical importance and application requirements; the design accuracy evaluation result is obtained; wherein, the grade division standard of the design accuracy evaluation result is: 90 points and above for excellent, indicating that the simulation prediction is highly reliable; 80-90 points for good, indicating that the simulation prediction is basically reliable; 70-80 points for qualified, indicating that the simulation prediction has certain reference value; 70 points and below for unqualified, indicating that the simulation prediction has insufficient reliability, and the model or method needs to be improved; based on the accuracy evaluation result, a detailed analysis report is also generated, pointing out the weak links and improvement direction of the prediction accuracy, providing a scientific basis for model optimization and design improvement.

[0058] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify, replace or change some technical features of the technical solutions described in the foregoing embodiments. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0059] The formulas in the present specification are dimensionless numerical calculations, and the formulas are obtained by software simulation of a large amount of data to obtain a formula closest to the real situation. The preset parameters and threshold values in the formula are set by those skilled in the art according to the actual situation.

[0060] Although embodiments of the present application have been shown and described, those skilled in the art can understand that various changes, modifications, replacements and variations can be made to these embodiments without departing from the principles and spirit of the present application, and the scope of the present application is defined by the claims and their equivalents.

Claims

1. A parametric design method for orthopedic insoles based on biomechanical data and a rule base, characterized in that, include: Step S1: Perform a high-precision 3D scan of the child's foot below the knee joint, collect foot point cloud data, and reconstruct a 3D foot model based on the data; perform foot biomechanical analysis based on the 3D foot model to obtain foot correction requirement data; Step S2: Construct a parametric design candidate solution library, perform parameter rule matching and solution screening based on foot correction requirement data to obtain personalized insole design solutions, and configure their material properties to obtain insole material parameter data; Step S3: Based on the obtained personalized insole design scheme, construct a biomechanical simulation model, and conduct multi-posture mechanical simulation analysis and correction effect evaluation to obtain the design effect prediction results; And based on this, the parameters of the personalized insole design scheme are iteratively optimized; Optimized insole design solution obtained; Step S4: Based on the optimized insole design, perform intelligent manufacturing to obtain a solid orthotic insole, and conduct pressure distribution testing on the solid orthotic insole; compare and analyze the pressure distribution test results with the corresponding simulation analysis results to obtain the design accuracy evaluation results; And it was applied to personalized insole design solutions.

2. The parametric design method for orthopedic insoles based on biomechanical data and rule base as described in claim 1, characterized in that, The process of obtaining foot correction requirement data includes: A pre-set 3D scanning device is used to perform high-precision data scanning on the child's feet below the knee joint to obtain foot point cloud data. The obtained foot point cloud data is then subjected to noise reduction filtering and data cleaning to obtain purified point cloud data. Based on the purified point cloud data, missing regions are identified and compensated to obtain complete foot point cloud data, and triangular mesh reconstruction is performed on it to obtain a three-dimensional foot mesh model; then feature parameters are extracted based on the three-dimensional foot mesh model to obtain a set of foot geometric feature parameters. Based on the set of geometric feature parameters of the foot, quantitative calculation of the index is performed to obtain the quantitative index of arch flatness; and heel posture deviation data is acquired simultaneously. Foot type classification is performed based on heel posture deviation data and arch flatness quantification index, and the degree of correction needs is assessed based on the classification results to obtain foot correction needs data.

3. The parametric design method for orthopedic insoles based on biomechanical data and rule base according to claim 2, characterized in that, The process of obtaining the quantitative indicators for flat foot arches includes: The corresponding three-dimensional mesh model of the foot is segmented into three functional areas: the forefoot area, the midfoot area, and the hindfoot area. The area distribution data of each functional area is obtained. Based on the area distribution, the arch flatness is calculated to obtain the arch index value. Based on the index value, the arch flatness is classified to obtain a quantitative index.

4. The parametric design method for orthopedic insoles based on biomechanical data and rule base according to claim 2, characterized in that, The process of acquiring complete foot point cloud data includes: Spatial continuity detection is performed on the purification point cloud data to identify and mark the blind spots and missing data areas of foot scanning, thereby obtaining potential missing areas; A morphological prediction model is constructed based on prior knowledge of foot anatomy to calculate the geometric shape of potential missing areas. Radial basis function interpolation is then used to compensate and reconstruct point cloud data for missing data areas. The point cloud data after compensation and reconstruction is then optimized for smoothness to obtain complete foot point cloud data.

5. The parametric design method for orthopedic insoles based on biomechanical data and rule base according to claim 2, characterized in that, The process of obtaining insole material parameter data includes: Based on orthogonal experimental design theory, a three-factor, three-level parametric combination matrix is ​​constructed to obtain the insole parameter combination design space, and constraint boundaries are set to obtain feasible parameter range data; based on the feasible parameter range data, standardized parameter combinations are generated to obtain a parametric design candidate scheme library; the parametric design candidate scheme library consists of several basic design candidate schemes; Based on foot correction demand data, parameter intelligent matching decision rules are constructed to obtain a multi-level matching rule tree; the weight coefficients of the multi-level matching rule tree are adaptively optimized to obtain an adaptive matching rule set. Based on the adaptive matching rule set, the candidate basic design schemes are ranked by adaptive scoring, and the best matching candidate basic design scheme is selected based on the adaptive scoring ranking and marked as the personalized insole design scheme. The system acquires children's weight and activity data, and intelligently selects the hardness of EVA material based on this data to obtain material hardness parameters. It then generates a complete insole design scheme based on the personalized insole design scheme and the material hardness parameters. Finally, it configures the material mechanical properties of the complete insole design scheme to obtain insole material parameter data.

6. The parametric design method for orthopedic insoles based on biomechanical data and rule base according to claim 5, characterized in that, The process of obtaining optimized insole design solutions includes: Based on the three-dimensional mesh model of the foot and the personalized insole design scheme, the geometric assembly relationship is established to obtain the geometric data of the foot-insole contact interface; the constitutive relationship of the hyperelastic material is set according to the material parameter data of the insole to obtain the material property configuration parameters. Finite element meshes were automatically generated based on the three-dimensional mesh model of the foot and the geometric data of the foot-insole contact interface to obtain a finite element mesh model. A multi-body coupled simulation model of foot-insole-shoe was established by combining material property configuration parameters. The model was then optimized to obtain a biomechanical simulation model. Multi-condition load simulation was performed on the biomechanical simulation model to obtain multi-posture simulation parameters. Stress concentration analysis is performed based on multi-posture simulation parameters to obtain stress distribution evaluation results; the correction effect of the stress distribution evaluation results is quantitatively evaluated to obtain design effect prediction results; Based on the design effect prediction results, design defects are identified and parameter sensitivity analysis is performed to obtain parameter optimization directions; and based on this, the parameters of the personalized insole design scheme are iteratively updated to obtain the optimized insole design scheme.

7. The parametric design method for orthopedic insoles based on biomechanical data and rule base as described in claim 6, characterized in that, The process of constructing a biomechanical simulation model includes: The three-dimensional mesh model of the foot is automatically generated into a tetrahedral mesh, and the mesh density gradient is set to obtain the finite element mesh of the foot. Based on the personalized insole geometric design scheme, the insole geometric model is obtained, and hexahedral dominant hybrid mesh is generated. Mesh refinement is performed in key stress areas to obtain the insole finite element mesh. Establish contact constraint relationships between the sole and the insole to obtain contact constraint configuration data; Based on the finite element mesh of the foot, the finite element mesh of the insole, and the contact constraint configuration data, the model geometry is assembled to obtain a multi-body coupled simulation model of foot-insole-shoe. Boundary and load conditions were set for the foot-insole-shoe multibody coupling simulation model; mesh quality was checked and element distortion was optimized to obtain the biomechanical simulation model.

8. The parametric design method for orthopedic insoles based on biomechanical data and rule base according to claim 6, characterized in that, The process of obtaining the design effect prediction results includes: Peak pressure and region division are performed on the static pressure distribution data within the multi-pose simulation parameters to obtain a high-pressure zone distribution map. The key phases of the gait cycle are extracted from the dynamic mechanical response data within the multi-posture simulation parameters to obtain the key phase mechanical data; Based on the obtained high-pressure zone distribution map and key temporal mechanical data, the correction effect is evaluated. The quantitative results of the indicators of plantar pressure distribution, maximum von Mises stress and arch strain energy are analyzed to obtain a multi-dimensional correction effect score. The design effect prediction results are generated based on the multi-dimensional correction effect score.

9. The parametric design method for orthopedic insoles based on biomechanical data and rule base according to claim 6, characterized in that, Obtain the design accuracy assessment results; The process of applying this to personalized insole design solutions includes: Intelligent planning of manufacturing process paths is performed on the optimized insole design to obtain CNC machining trajectory data; cutting parameters are optimized and configured according to material properties and machining accuracy requirements to obtain a set of manufacturing process parameters; A CNC machining program file is generated based on the CNC machining trajectory data and manufacturing process parameter set; and five-axis linkage CNC milling is performed according to the CNC machining program file to obtain a rough insole product; the rough insole product is then precision polished and surface-finished to obtain a solid orthopedic insole. A thin-film pressure sensor array was embedded in a solid orthotic insole, and dynamic tests were conducted on children wearing the insole. Real-time data of plantar pressure under various movement states were collected to obtain measured pressure distribution data. A spatiotemporal correspondence was established between the measured pressure distribution data and the simulated mechanical response data to obtain a comparative analysis dataset. Accuracy evaluation indicators are calculated based on the comparative analysis dataset to obtain the design accuracy evaluation results; the accuracy evaluation indicators include pressure distribution correlation indicators, peak prediction accuracy indicators, and regional prediction accuracy indicators. Based on the design accuracy assessment results, error source analysis and parameter correction strategies are formulated to obtain feedback optimization parameters; and the feedback optimization parameters are applied to personalized insole design schemes.

10. The parametric design method for orthopedic insoles based on biomechanical data and rule base according to claim 9, characterized in that, The process of obtaining the design accuracy assessment results includes: Based on the comparative analysis dataset, the Pearson correlation coefficient between the measured and simulated values ​​was calculated to obtain the pressure distribution correlation index. Statistical analysis of the error between the measured peak pressure and the simulated peak pressure was performed, and the mean absolute error and root mean square error were calculated to obtain the peak prediction accuracy index. The accuracy of the biomechanical simulation model in predicting high-pressure areas was evaluated by the receiver operating characteristic curve analysis method, and the regional prediction accuracy index was obtained. A multi-dimensional accuracy assessment model was constructed, and the pressure distribution correlation index, peak prediction accuracy index, and regional prediction accuracy index were comprehensively weighted and scored to obtain the design accuracy assessment results.

Citation Information

Patent Citations

  • Insole production system based on finite element mechanical analysis

    CN105956349A

  • Testing method based on landing movement foot area and shoe sole finite element modeling

    CN109662392A

  • Personalized orthopedic insole design and rapid mechanical simulation method

    CN117408073A

  • Insole design method, device and equipment

    CN118364526A

  • Correcting insole intelligent management system for flatfoot treatment

    CN120046481A

Cited By

  • Footwear design method based on gait biomechanics and footwear

    CN121598453A

  • Inside and outside overturning method and system of insole three-dimensional model based on dynamic control point correction

    CN121600231A

  • Method and system for modifying a 3d model of an insole based on dynamic control points

    CN121600231B