Artificial limb and residual limb coupling optimization method based on digital twinning

Abstract

The invention relates to an optimization method for coupling of an artificial limb and a residual limb based on digital twinning. Carrying out multi-modal data fusion on anatomical structure characteristics, functional attributes and stress distribution data of the skin, the soft tissue and the bone of the stump; building multi-layer digital stump relation modeling; defining a multi-scale contact space formed by a structure adaptation area, a force transmission area and an energy buffer area in the dynamic behavior of the residual limb and the artificial limb; simulating a shear force migration track, a stress concentration mode and soft tissue slippage response among micro-units of a coupling surface in dynamic interaction; constructing a physical partition relationship of the coordination domain, and assisting in optimizing an artificial limb interface structure and a material functional gradient; describing a minimally invasive area caused by wearing and using the artificial limb, and calculating a minimum tissue perturbation path; through deformation vector field analysis and thermal-mechanical coupling relation modeling, risk distribution of tissue fatigue induced by the prosthetic interface under long-time use is evaluated; and dynamic multi-scale coupling optimization and multi-objective optimization control are carried out to realize visual prediction and early warning of fatigue risks.

Description

Technical Field

[0001] The present invention relates to a method for optimizing the coupling between a prosthesis and a residual limb, and more particularly to a method for optimizing the coupling between a prosthesis and a residual limb based on digital twins. Background Art

[0002] The coupling relationship between prostheses and residual limbs has always been a key issue in the integration and intersection of biomechanical engineering, rehabilitation-assisted technology and intelligent manufacturing. The core lies in how to achieve a coordinated match between the prosthetic interface structure and the individual residual limb in terms of geometry, biomechanical properties and neuromuscular function. Most of the currently widely used optimization methods are still at the stage of static modeling and empirical debugging, and are unable to meet the real wearing needs of users under variable dynamic behaviors. For example, the technical solution disclosed in the Chinese patent CN118986601A, a method and device for optimizing the shape of a prosthetic socket, still has obvious deficiencies and structural limitations, especially in terms of adaptive control, dynamic prediction capabilities and user feedback fusion, which cannot meet the development needs of modern high-performance prosthetic systems.

[0003] Existing technologies mainly divide the residual limb surface into multiple regions, set extrusion centers, and apply simulated extrusion tests in multiple regions to obtain a series of static test indicators such as compression rate, suspension force gap, support force gap, and bone movement limit. These indicators are then used as objective functions for multi-objective optimization to ultimately obtain the shape parameters of the prosthetic socket. Although this method does achieve reasonable modeling of the local pressure tolerance of the residual limb and introduces optimization algorithms to improve the comfort and support of the cavity design, it still has the following technical drawbacks and shortcomings as a whole. First, the model construction of this scheme is entirely based on static residual limb geometry data and mechanical response results. It does not consider the tissue response characteristics of the wearer under dynamic behavior, nor does it introduce a modeling mechanism for multi-cycle motion simulation or behavioral scenario changes. As a result, the final optimization result can only be applied to the standing or initial wearing state, and cannot truly reflect the changes in shear stress, slip and tissue compression response during continuous movements such as walking, squatting, and climbing. This static deduction approach seriously limits its generalization ability in actual multi-scenario applications.

[0004] Secondly, the optimization objective function is based only on mechanical indicators such as compression rate and support gap, while ignoring important physiological factors such as heat accumulation, micro-slip trajectory, and the area of ​​local muscle function retention of the residual limb. These factors are often the core causes of user pain, pressure injuries, inflammation, and unstable wearing. Thirdly, the existing technology does not construct a multi-scale simulation space. All regional analyses are based on surface compression tests. It lacks the ability to divide the contact interface into biofunctional zones such as structural adaptation zones, force transmission zones, and energy buffer zones. It cannot support the setting of different structural strategies for different regions (such as elastic gradient adjustment or multi-material composite layout). This makes the optimization results tend to be homogenized and lacks the ability to respond to the complex mechanical functional zones of the human body in a personalized manner. Summary of the Invention

[0005] The purpose of the present invention is to provide an optimization method for coupling a prosthetic limb with a residual limb based on digital twins, thereby solving some of the drawbacks and deficiencies pointed out in the background technology.

[0006] The present invention solves the above-mentioned technical problems by adopting the following technical solutions: an optimization method for coupling a prosthetic limb with a residual limb based on digital twins, comprising: multimodal data fusion of the residual limb's anatomical structural characteristics, including skin, soft tissue, and bone, with functional attributes including residual muscle control capabilities, myoelectric response, and stress distribution data; constructing a multi-layer digital residual limb relationship model; introducing a functional residual mapping mechanism to calibrate the residual limb area responsible for force conduction and the area suitable for neural interface implantation;

[0007] Define the multi-scale contact space between the residual limb and the prosthesis during dynamic behavior, consisting of the structural adaptation zone, force transmission zone, and energy buffer zone; simulate the shear force migration trajectory, stress concentration pattern, and soft tissue slip response between the micro-units of the coupling surface during dynamic interaction; and construct the physical partition relationship of the coordination domain to assist in optimizing the prosthetic interface structure and material functional gradient.

[0008] Describe the micro-trauma areas caused by prostheses during wearing and use, and calculate the minimum tissue disturbance path, that is, the path that minimizes strain concentration or heat accumulation caused by the prosthesis on the tissue structure; through deformation vector field analysis and thermal-mechanical coupling relationship modeling, evaluate the risk distribution of tissue fatigue induced by prosthetic interfaces under long-term use.

[0009] Furthermore, in the simulated dynamic interaction, an individualized digital twin model of the residual limb including anatomical structure, functional attributes and stress response is constructed;

[0010] A multi-scale contact space is established between the residual limb twin model and the prosthetic limb digital model. The space is composed of a structural adaptation area, a force transmission area, and an energy buffer area. Based on dynamic behavior simulation, the contact space is divided into micro-units to simulate the shear force migration trajectory between each micro-unit.

[0011] Based on the force changes of micro-units, the stress concentration evolution pattern in the coupling area is identified; the soft tissue slip response of the contact interface is modeled and the slip amplitude map is output; according to the shear migration trajectory, stress concentration trend and slip tolerance, the structural parameters and material layout of the prosthetic interface are optimized.

[0012] Furthermore, the individualized residual limb twin model integrates the skin, muscle, and bone structure information of the residual limb, combines biomechanical properties with myoelectric response characteristics, and establishes a multi-layer simulation model integrating anatomy, function, and stress.

[0013] Furthermore, the multi-scale contact space is divided into:

[0014] Structural adaptation zone: makes the prosthesis fit the residual limb geometrically;

[0015] Force transfer area: used to transfer the movement intention of the residual limb to the prosthesis through force;

[0016] Energy buffer: used to absorb shear force and impact force and relieve tissue stress accumulation.

[0017] Furthermore, by constructing a micro-unit grid in the contact area between the residual limb and the prosthesis, the entire contact surface is divided into multiple discrete units, and a motion-driven timeline is introduced to simulate the dynamic flow path of shear force during human movement. The shear force state of each micro-unit at any moment is defined as a vector. To describe the migration trend and directionality of the shear force, the following shear force migration evolution function is constructed:

[0018]

[0019] in:

[0020] Indicates position in space The shear force migration sensitivity field intensity function at location and time t, i.e., the path aggregation trend of the shear force at this point; It represents the gradient in the spatial coordinate, reflecting the migration direction of the shear force along the position direction; Indicates the location The instantaneous rate of change of shear stress at time τ; It represents the slip tolerance tensor factor of the tissue at that location, which is related to the soft tissue type and physiological morphology; It represents the action impulse conversion factor of the micro-area unit under behavior drive, reflecting whether the point is in the high-frequency response path induced by behavior; It means integrating and superimposing the above factors in historical time to capture the cumulative migration trend of shear force on the path;

[0021] By using the output of the shear force transfer function and combining it with the nonlinear response properties of the tissue in the digital twin model, the present invention defines dynamic retention points, i.e., locations where stress cannot be effectively released and repeatedly compressed within a time series; these areas are marked as potential fatigue inducing points; the stress retention mechanism is based on the following logic: if a certain micro unit point of Local maxima appeared in multiple time periods and were not accompanied by shear stress relief processes;

[0022] At the same time, its tissue thermal decoupling parameters showed a continuous upward trend;

[0023] The system then marks this point as a stress-thermal coupling retention point, constituting a fatigue risk area;

[0024] Ultimately, the system outputs a multidimensional fatigue risk map, providing input for prosthetic interface design optimization, including buffer layer reconstruction, contact texture adjustment, and material elasticity redistribution.

[0025] Furthermore, the soft tissue slip response modeling includes:

[0026] Establishing the slip threshold between the skin and muscle layers;

[0027] Analyze the relative displacement distribution of the contact interface under different motion states;

[0028] The output slip amplitude map is used to evaluate the matching degree between the tissue friction area and the interface structure.

[0029] Furthermore, the optimization process is based on the shear force trajectory, stress concentration distribution and slip response results, and adopts a multi-objective evolutionary algorithm to jointly optimize the geometric structure, buffer material thickness and stiffness distribution of the prosthetic interface.

[0030] Furthermore, the method for evaluating the risk distribution of tissue fatigue induced by a prosthetic interface under long-term use includes:

[0031] S1. Establish a three-dimensional coupling model of the residual limb soft tissue and prosthetic interface in the digital twin model; perform deformation vector field analysis on the residual limb interface during prosthetic use, identify deformation concentration trajectory areas based on the user's repeated motion behavior, and mark potential fatigue-inducing points;

[0032] S2. Establish a thermal-mechanical coupling relationship model to simulate the accumulation and release of heat in the tissue during prosthetic wear; integrate the deformation trajectory with the thermal field change trend to generate a spatial distribution map of tissue fatigue risk;

[0033] S3. Optimize the prosthetic interface structure or material configuration according to the fatigue risk distribution map.

[0034] Furthermore, the deformation vector field analysis establishes a discrete grid of micro-units in the contact area, records the displacement direction and deformation amplitude of each grid unit in multi-cycle motion, and forms a continuous deformation trajectory path; the deformation trajectory is used to identify high-frequency repeated deformation areas as a basis for local fatigue risk assessment; the thermal-mechanical coupling relationship modeling integrates the thermal friction of the skin and prosthetic interface, the thermal conductivity of biological tissue and the thermal conductivity characteristics of the prosthetic material, and obtains the heat accumulation area through multi-cycle simulation.

[0035] Furthermore, according to the results of the fatigue risk distribution map, the structural parameters of the prosthetic interface are adjusted, including: buffer material distribution, regional elasticity control, contact texture design or thermal conductivity optimization.

[0036] The beneficial effects of this invention include: By integrating the residual limb's anatomical structure, myoelectric function, and biomechanical response, a multi-level digital twin model is established, enabling a shift in prosthetic design from standard templates to individually tailored fit, achieving optimal matching between the prosthetic interface and the residual limb in terms of morphology, mechanics, and function. The contact surface is divided into a structural adaptation zone, a force transmission zone, and an energy buffer zone. Dynamic simulation mechanisms such as shear force migration trajectory analysis, stress concentration identification, and slip response modeling are introduced to achieve multi-scale dynamic optimization of the coupled structure in real-world usage scenarios, improving prosthetic stability and physiological fit.

[0037] Through deformation vector field and thermomechanical coupling modeling, a tissue fatigue risk distribution map is formed, and potential high-risk areas for compression or micro-trauma are identified in advance, providing data support for interface material zoning design and structural buffer optimization, significantly improving wear safety and long-term tissue health. An evolutionary algorithm based on shear, thermal stress, and slip behavior feedback is used to achieve joint optimization of prosthetic interface geometry, material stiffness, and buffer structure, forming an automatic iterative update mechanism, significantly reducing manual trial and error design time and improving adaptation efficiency. Through soft tissue dynamic response simulation and heat diffusion channel design, local stress peaks and heat accumulation are effectively reduced, continuous wear time is extended, skin discomfort and pain probability are reduced, and the user's daily activity ability and quality of life are significantly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 This is the flowchart of the prosthetic and residual limb coupling optimization based on digital twins in the present invention.

[0039] Figure 2 This is a relationship diagram between the three types of functional areas and the optimized design functions of the prosthetic-stump coupling interface of the present invention.

[0040] Figure 3 This is a relationship diagram between the minimum tissue disturbance path modeling and closed-loop optimization of the prosthetic-stump interface of the present invention.

[0041] Figure 4This is a schematic diagram of the implementation process of the personalized prosthetic digital twin partitioning and structural optimization invented by the present invention.

[0042] Figure 5 This is the structural diagram of the prosthetic interface fatigue risk distribution assessment and multi-dimensional structure optimization system invented by the present invention. DETAILED DESCRIPTION

[0043] The following is a detailed description of the specific embodiments of the present invention with reference to the accompanying drawings.

[0044] Combined with attachment Figure 1 , obtain the multimodal raw data of the user's individualized residual limb, including the surface morphology data of the residual limb's skin layer, soft tissue thickness and elastic properties, and bone geometry and positioning data. The data can be collected by structured light scanning, CT image reconstruction and ultrasonic elastic imaging, and then the three types of anatomical structures are aligned in three-dimensional space through the registration algorithm to construct a multi-level residual limb digital model with clear anatomical hierarchy and coherent topological structure; at the same time, the response data of the user's residual muscle groups under different postures and movement intentions are obtained through the electromyographic signal acquisition device, and the residual control ability indicators of each muscle group, such as response strength, delay time and signal-to-noise ratio, are calculated by combining the movement information recorded by the wearable inertial sensor; in addition, in order to further realize physical and mechanical simulation, a somatosensory motion platform is used to induce the residual limb to produce typical activity states (such as walking, weight-bearing, flexion and extension, etc.), and the surface pressure sensor array is used to record the evolution of force in each area over time to obtain stress distribution and peak migration trajectory data; based on the anatomical structure, functional properties and stress evolution information obtained above, the system spatially and temporally transforms the multi-source heterogeneous data By integrating the physiological dimensions, a five-layer nested digital residual limb twin model is constructed, comprising the skin layer, soft tissue layer, bone structure layer, muscle activity layer, and stress response layer. A functional residual mapping mechanism is further introduced into this model. Based on the spatial distribution of myoelectric intensity and the inversion algorithm of the force conduction path, this mechanism evaluates the actual participation of each anatomical unit in different motion scenarios and compares it with the mechanical contribution under ideal conditions. This determines the functional residual value of each residual limb region, which is used to quantify the relative functional degradation or residual capacity of a region in actual use. Based on the coupling results of this functional residual value with the stress density distribution, the system generates a regional labeling map, which marks the key structural belts that primarily undertake force conduction tasks (such as the distal tibial load zone and residual tendon anchor points), as well as low shear stress and high biostability regions suitable for neural interface implantation or electrode placement. This zoning result can serve as a basis for subsequent prosthetic structure interface matching design and the docking of the neural-prosthetic system, thereby achieving personalized collaborative optimization of the human-machine interaction structure and the neural response system in prosthetic design.

[0045] Combined with attachment Figure 2Through dynamic behavior simulation, the contact interface between the residual limb and the prosthesis that produces coupling interaction during actual use is defined as a combination of three types of functional areas, namely the structural adaptation area, the force transmission area and the energy buffer area. The structural adaptation area is mainly used to achieve the fit between the geometric shape of the prosthesis and the skin surface of the residual limb. This area must ensure high surface fitting and contact stability. The force transmission area is the core path for the internal bones or muscles of the residual limb to transmit movement intentions and supporting forces through the prosthesis. It is mostly located at the tendon attachment points of the residual limb, the protruding areas of the bone surface and other parts with strong force feedback capabilities. The energy buffer area is used to absorb multiple forces during repeated movements or impacts. Residual energy, alleviate soft tissue shear damage, this area is usually distributed at the contact edge, where the soft tissue thickness is large or close to the area rich in blood vessels and nerves, and has the design requirements of strong buffering and high wear resistance; in order to simulate the response state of these three types of functional areas in actual dynamic behavior, the contact interface is further micro-unit meshed, and the residual limb prosthesis coupling surface is divided into high-density discrete units. The timing control parameters are introduced, and the user behavior data recorded in the twin model is used to drive the system for dynamic simulation. The shear stress, positive pressure, micro-slip vector and heat energy transfer of each unit in different postures are calculated in real time, and the coupling surface is tracked. The shear force migration trajectory is analyzed to analyze the flow trend of shear stress in time and space dimensions, and the micro-areas where stress retention still exists after multiple behavior cycles are identified to form stress concentration pattern recognition results; in addition, based on the tissue hierarchical structure and residual limb morphological response, a soft tissue slip response model is further introduced to determine whether the relative displacement between tissue layers in a specific area exceeds the threshold under repeated action conditions, thereby warning of potential wear risks; after fusion and cluster analysis of the above simulation data in the digital twin system, the spatial physical boundaries and behavioral response boundaries of the three functional areas of the coupling surface can be output, and a physical model of the coupling coordination domain can be constructed accordingly. The partition relationship diagram clearly defines the main functional role of each micro-area in actual use. As an important reference for subsequent prosthetic interface design, the system maps the partition to the CAD modeling environment and assigns corresponding material types and structural parameters based on the functional properties of different areas. For example, high-strength carbon fiber composite materials are used in the force transmission area to enhance the response speed, multi-layer variable modulus silicone is used in the buffer zone to improve the shock absorption capacity, and the elastic bonding layer and surface texture are optimized in the structural adaptation area, thereby forming a prosthetic interface structure with a continuous transition of material functional gradient, and ultimately achieving comprehensive optimization of coupling comfort, stability and physiological safety.

[0046] Combined with attachment Figure 3In order to solve the problem of chronic micro-trauma caused by prostheses to residual limb tissues during wearing and use, a modeling and evaluation mechanism of minimum tissue disturbance path is proposed to identify and minimize the impact of local high strain or heat accumulation caused by the prosthetic interface structure on biological tissues. In the constructed residual limb-prosthesis digital twin integrated model, the soft tissue in the coupling area is discretized by high-precision grid division, and the dynamic deformation data of each micro-unit is tracked and recorded during multiple simulations of typical residual limb usage movements (such as walking, standing up, crouching, turning, etc.) to form a continuous deformation vector field, where each vector represents the cumulative displacement and force direction of a specific micro-unit in the time series. This deformation field is used to evaluate the strain concentration trend of the tissue structure under multi-cycle stress. At the same time, in order to more realistically reflect the physiological response under long-term wearing conditions, the system further introduces a thermal-mechanical coupling modeling mechanism, and incorporates parameters such as frictional heat generation, tissue thermal conductivity, skin thermal conduction hysteresis, and thermal resistance characteristics of prosthetic materials into the model to simulate the heat accumulation in the residual limb contact area during repeated wearing and movement, with particular attention to the inability of heat to diffuse in time. the region and its corresponding soft tissue stress state; the system constructs a multi-dimensional energy consumption path map by fusing the above-mentioned mechanical deformation vector field and thermal energy conduction path data, and sets the optimization objective function on this basis, that is, to find the path with the smallest cumulative strain intensity and heat load on the soft tissue among all paths as the minimum tissue disturbance path. This path usually avoids bony protrusions, tissue junctions and vascular dense areas, so that it can be used to guide the design of optimization measures such as the shape of the prosthetic interface area, material stiffness distribution, and contact structure fine-tuning; in addition, the long-term dynamic evolution of the residual limb model is used to simulate the shear force migration, deformation residue and thermal stress hysteresis effect in the multi-day wearing scenario. According to the deformation history, temperature superposition effect and tissue stress recovery ability of each micro-unit, the system outputs a potential tissue fatigue risk distribution map. The map visually marks high-risk areas, providing an early warning basis for chronic compression injuries, skin damage, soft tissue atrophy and other problems that may occur in the later stage of prosthetic wearing, and supports updating the twin model and structural optimization suggestions based on feedback data to achieve preventive safety control of prosthetic design and continuous improvement of user comfort.

[0047] Example 1:

[0048] Combined with attachment Figure 4In this example, the user is a 35-year-old male, Mr. W, who lost his right lower leg in a traffic accident. He has been wearing a conventional prosthesis at the mid-tibia for four years. However, he has repeatedly experienced localized pain in the residual limb, skin redness and swelling, and loosening of the prosthesis. Doctors diagnosed this as chronic soft tissue fatigue and shear force accumulation caused by long-term wear. To address this, the team developed a digital twin-driven prosthetic interface adaptation solution for Mr. W. In the first step, a high-resolution CT scan was used to obtain the skeletal structure of Mr. W's right lower leg residual limb, and a near-infrared elastography was used to obtain the thickness distribution and elastic modulus of the soft tissue layer. At the same time, 3D surface scanning was used to reconstruct the skin surface geometry. These data were then fused into a three-layer residual limb structure model. To supplement functional information, an eight-channel surface electromyography sensor was used to collect electrical signals from Mr. W's residual biceps femoris, quadriceps femoris, and hamstring muscles during walking and squatting. The results showed that the maximum electromyographic peak was approximately 0.75mV, the response time delay was 70ms, and the signal stability index was 0.82. Comprehensively judged that the muscle group still had certain motor control potential. After completing the construction of the anatomical-functional fusion digital twin model of the residual limb, stress records between the residual limb and the prosthesis from historical wearing data were further introduced. It was found that the left posterior area of ​​the contact interface repeatedly showed local pressure peaks exceeding 120kPa during walking, accompanied by slippage exceeding 2.3mm, which was significantly higher than the tissue slip tolerance threshold (1.5mm) and was marked by the system as a high-risk stress accumulation area. In the second step, the team connected the twin residual limb model with the prosthetic limb structure model to be optimized, and established a multi-scale coupled contact space including a structural adaptation zone, a force transmission zone and an energy buffer zone. The adaptation zone covers the fitting surfaces corresponding to the front end of the tibia and the tail end of the fibula, the force transmission zone is concentrated in the mid-axis of the tibial section, and the energy buffer zone is located in the area with thicker soft tissue coverage on the inner side of the residual limb. The entire contact interface is divided into 2146 micro-unit grids, each with a side length of 3.5mm. By driving the twin model to perform a 10-minute walking motion simulation, the system records the shear stress vector field data and generates a shear force migration path map, identifying the main shear force channel migrating from the posterior and inferior to the front and superior during gait. The average shear stress is maintained at 37-54kPa in the first 3 minutes, and then gradually accumulates to more than 82kPa, far exceeding the residual limb tissue threshold. In the third step, based on these simulation data, the system analyzed a significant stress concentration pattern in the left-center region of the coupling surface. This shear stress retention time reached 6.8 seconds per cycle, significantly exceeding the 1.2-second average for normal regions. This pattern was accompanied by multiple amplified slip responses. Combined with the viscoelastic delay predicted in the soft tissue model, this region exhibited delayed displacement and rebound during each gait, creating a tissue strain trap and a potential focal point for future tissue fatigue and crush injury. The system output a slip amplitude map, marking this region as a high-slip zone in red.Finally, in the structural optimization stage, the system adjusted the buffer layer structure of the prosthetic interface according to the main shear force migration path and the stress concentration heat map, increased the thickness of the original silicone layer in the risk area from 2mm to 4mm, and added a directional microporous structure to disperse the shear stress. At the same time, the material used in this area was changed from a single thermoplastic elastomer to a composite double-layer buffer material (soft inner layer, moderate outer layer), and strengthened the carbon fiber support skeleton at the main force transmission path to optimize the overall stiffness continuous gradient. After re-simulation of the modified structure, it was shown that the maximum shear stress in the risk area decreased by 41.6%, the slip response was controlled within 1.1mm, the stress retention time was reduced to 1.9 seconds, and the fatigue risk assessment score was increased from the original 0.28 to 0.81. After wearing it, Mr. W reported that the wearing comfort was significantly improved, and he had no skin stress reaction after walking for 60 minutes continuously.

[0049] By integrating the anatomical structure information of the residual limb skin, muscles and bones, and combining the individual's biomechanical characteristics and electromyographic response behavior, a digital residual limb simulation system integrating anatomy, function and stress was established. CT scanning was used to reconstruct the three-dimensional structure of Mr. W's right lower leg stump bones, and the morphology of the middle residual end surface of the tibia and the lateral surface of the fibula were accurately extracted as the internal bone support layer. At the same time, the muscle layer thickness, arrangement direction and covering relationship were obtained in combination with MRI data to construct a deep soft tissue layer around the skeleton. The biomechanical parameters of different areas of the soft tissue were measured by near-infrared optical elastic imaging, including Young's modulus (ranging from 18 to 56 kPa), viscoelastic coefficient and shear delay response, and these parameters were injected into the model as muscle layer attribute input; the skin layer data was obtained through high-precision 3D structured light scanning for use The skin surface contour, capillary distribution density and tissue thickness changes were restored. The three types of structural data were spatially reconstructed and topologically bonded through voxel superposition and B-spline body registration algorithms to construct a complete three-layer structural basic model. Based on this structural model, electromyographic data were further introduced to establish a functional response layer. An 8-channel surface electromyograph was used to record the electrical activity data of Mr. W's residual biceps femoris, semitendinosus, gastrocnemius and other muscle groups during walking and stage-stepping movements. The activation threshold, peak delay and stability characteristics in the electromyographic waveform were systematically extracted, and these characteristics were converted into muscle group control capabilities through a neural network analysis module. The system uses the atlas to map the functional contribution of each muscle area under different actions, and maps it with the muscle anatomical layer to construct a distribution map of the functional contribution of each muscle area under different actions. This distribution map is used as the basis for weight calibration of the subsequent force transfer area. In addition, the system synchronously records the stress response behavior of the micro-units of each tissue layer in the dynamic gait simulation, and obtains the shear stress, compressive strain and residual deformation values ​​generated by each grid unit in the periodic motion through finite element calculation. Combined with the fatigue accumulation model of the tissue, a stress response distribution map is formed and superimposed on the structure-function model to form a system with anatomical geometry, neuromuscular control ability and mechanical mechanics response. The three-in-one digital twin model of the residual limb with high adaptability has the ability to evolve in time domain and can be periodically adaptively updated as the user's usage habits and tissue changes. In the case of Mr. W, the model was used to evaluate that the gastrocnemius area of ​​his residual limb has good force conduction ability and good myoelectric response. It is recommended that this area be used as the main force transmission path in the optimized design, while the medial side of the tibia is not suitable for the layout of the high contact pressure area due to obvious stress retention. Ultimately, the prosthetic interface structure is promoted to evolve towards a coupled form that is consistent with force, nerve and anatomy, effectively improving the structural matching and physiological compatibility, and ensuring that the prosthesis is both stable and safe under long-term wear.

[0050] Multi-scale contact space division and functional zoning modeling were performed on the coupled contact surface between the prosthesis and the residual limb. This zoning strategy, based on simulation data, functionally divides the entire contact interface into three sub-areas: structural adaptation area, force transmission area, and energy buffer area. On this basis, coordinated optimization of structure and material was carried out.The system constructs a complete three-dimensional coupling space between the personalized residual limb digital twin model established by Mr. W and the prosthetic limb digital interface model. In the preliminary dynamic simulation, by executing five typical gaits (including walking on flat ground, climbing a gentle slope, standing up, squatting, and turning), the contact behavior data of 2146 micro-units between the prosthetic interface and the residual limb are collected. According to the average fit, normal pressure stability and geometric matching error of each micro-unit during the simulation cycle, the area with good preliminary fit is marked as the structural adaptation area. For example, the fitting error between the front edge of the bottom of Mr. W's prosthesis and the main bearing surface of the tibial stump is controlled within ±1.4mm, and the contact pressure fluctuation is less than 5%. This area is included in the structural adaptation sub-area. The main function is to improve wearing stability and avoid interface displacement; then, the system calculates the consistency of the shear force vector of each micro-unit and the myoelectric control direction of the residual limb, and combines the overlap of the muscle activation channel map and the main migration path of the shear force to identify a main force transmission channel extending from the residual starting point of the gastrocnemius muscle to the carbon fiber transmission cavity of the prosthesis. The average shear stress of the micro-units on this path is 72kPa, and the direction deviation angle is less than 12°. It is defined as the force transmission area by the system. The design optimization focus of this area is to improve the response accuracy and reduce the force dissipation; in the above simulation, several impact dissipation areas are also identified, which are mainly concentrated in the area where the soft tissue covering thickness of the outer side of the residual limb exceeds 8mm, where the shear stress changes frequently but does not form an obvious concentration trend. At the same time, the thermal stress superposition exceeds that of other areas by about 21%, indicating that these areas are involved in a large amount of kinetic energy buffering and heat accumulation processes. The system thus marks them as energy buffer zones. Taking Mr. W as an example, the energy buffer zone is mainly concentrated at the junction of the inner upper edge of the prosthesis and the outer edge of the popliteal fossa of the residual limb. It is recommended to use locally embedded composite porous energy-absorbing materials in this area, and at the same time design a buffer layer separated from the prosthetic structure. After adding the buffer material to the simulation, the peak shear force of the same synchronic simulation decreased by 32.8%, and the regional temperature rise decreased by 1.6°C, proving the feasibility and effectiveness of the buffer zoning; the system outputs a complete three-region spatial distribution map and imports it into the interface structure CAD, adopting a local gradient hardness material strategy, and the structural adaptation zone is adopted. The device uses medium-hardness thermoplastic elastomer (Shore hardness 75A), a carbon fiber reinforced bracket integrated with a conductive electromyographic interface in the force transmission area, and a three-dimensional elastic mesh silicone and thermal insulation composite foam material for protection in the energy buffer zone. The overall interface structure has been verified by simulation again, and the shear force uniformity index has been increased from the original 0.67 to 0.89, the contact stability score has increased by 17.2%, and the thermal stress distribution has been significantly improved. Subsequently, Mr. W's walking distance wearing the new structure of the prosthesis has been extended to about 1.8 times, and there has been no redness, swelling, indentation, fatigue or pain. This fully verifies the engineering practicality, physiological safety and user experience improvement capabilities of the multi-scale contact space division strategy, and provides a standardized design template for the digital customization of prosthetic interfaces.

[0051] To further quantify and optimize the stress effects and fatigue risk control of prosthetic interfaces under long-term tissue use, a shear force transfer evolution function modeling approach was used to conduct a microscale dynamic analysis of the coupling region between the prosthesis and residual limb. A two-dimensional micro-cell grid was constructed at the contact interface between the residual limb and the prosthesis, dividing the entire coupling surface into 2146 tiny cells (with an average cell side length of 3.5 mm). A shear force vector recording unit was established on each cell. In the simulation system, Mr. W was driven to complete a continuous 60-minute gait cycle. The shear stress changes of each cell at each simulation moment were collected, and the path trend was modeled and quantified using the following newly constructed shear force transfer evolution function:

[0052]

[0053] The following parameter ranges are specifically set in the simulation platform to support example calculations:

[0054] The shear stress change rate in this unit is taken from the simulation data and is expressed in kPa / s. The observed change range during Mr. W’s gait is 5.6 to 27.3 kPa / s;

[0055] The slip tolerance tensor factor is set to 0.42 to 0.85 (dimensionless) based on the tissue structural properties of skin, fat, and muscle, and is assigned based on MRI imaging data and the physiological area of ​​the residual limb;

[0056] The motion impact conversion factor represents the intensity of participation of the area in the motion conduction chain. It is set to 0.3 to 1.2 based on the peak electromyographic value and the acceleration modulus. Mr. W can reach above 1.1 in the peak force range of the knee joint.

[0057] Taking micro-element point A (located at the medial lower edge of the tibia) on one of the main shear paths as an example, during the first 20 minutes of simulation, its average δ was 18.7 kPa / s, Γ was 0.67, and Ξ was 0.94. Substituting these values ​​into the formula, the integral output is:

[0058]

[0059] Then, the gradient of this value in the spatial direction is calculated, and combined with the evolution function value of the adjacent unit, the shear migration sensitivity field intensity Ψ(A, t = 1200s) of point A is obtained to be ≈ 3650 (the shear response intensity index after unit normalization), which belongs to the high-sensitivity shear force convergence area set by the system (the threshold is 3000); further, the system converts the shear migration sensitivity field intensity of point A into the shear force convergence area of ​​​​the system. If there are continuous local extreme values ​​and the area cannot release stress in the subsequent cycles (i.e., there is no downward trend of more than 40%), the point is marked as a stress retention point; the corresponding thermal decoupling parameter of the area (calculated by the local temperature accumulation rate + tissue thermal conductivity ratio) is 0.78 in the simulation (the system sets the threshold value to 0.65), indicating that the local area is in a state of dual accumulation of stress and heat for a long time, and the system marks the point as a stress-thermal coupling retention point; in the full contact surface micro-unit scanning, 19 such high-risk micro-areas were finally identified, most of which were concentrated under the residual limb. The edge of the sloped area fits the interface of the front end of the prosthesis. Combined with the fatigue risk map, the area is assigned a red mark by the system and fed back to the interface structure CAD model. The following design optimization measures are recommended: First, the thickness of the buffer layer in this area is increased from 2mm to 5mm, and the material is replaced from ordinary EVA to a three-layer foam structure (soft surface, energy-absorbing core, and heat-conducting middle). Second, a micro-wave texture is introduced to the contact surface to diffuse the shear force concentration. After optimization, the system re-simulated the same 60-minute action cycle, and the shear force change rate dropped to 12.4kPa / s, and the slip response dropped to 1.1mm. The peak value dropped to 2670, the risk level dropped from high to medium, the identification color changed from red to orange, and the thermal decoupling parameter dropped to 0.62, confirming that the tissue fatigue risk in this area was effectively controlled; after Mr. W actually wore the new interface, clinical observation and user feedback confirmed that the walking time could be increased from 30 minutes to more than 75 minutes, there was no fever, redness, swelling or indentation on the residual limb, and the muscle recovery was stable.

[0060] The soft tissue slip response modeling module was further implemented to solve the problems of skin wear, fatigue tearing and inflammation caused by the relative displacement between tissue layers during long-term wearing of prostheses. In this optimization process, magnetic resonance elastography (MRE) and photoacoustic tomography were used to scan the residual limb area of ​​Mr. W's right lower leg to obtain the tissue adhesion state and microstructural characteristics at different positions between the skin and muscles. After data processing, the system established a biological slip tolerance curve for each micro-region and constructed a slip threshold map between the skin and muscle layers. According to the tissue adhesion density and friction factor, the main slip threshold range of Mr. W's residual limb was set between 0.9mm and 1.6mm, of which the slip threshold of the outer edge of the gastrocnemius muscle was 0.95 mm, the slip threshold of the anterior tibia is increased to 1.45mm due to the thin skin and tight attachment of the fascia; then, the multi-scale simulation function of the residual limb twin model and the prosthetic interface model is used to drive Mr. W to perform multiple high shear stress action simulations in the simulation platform, including fast walking on a gentle slope, climbing stairs, and squatting and standing up. Each action is simulated for 10 minutes. In each action cycle, the system records the displacement difference between the skin layer and the muscle layer at the contact point, and outputs the relative displacement change value of the micro-area of ​​the entire contact interface, which is called the slip displacement distribution map; for example, in a continuous uphill gait, the residual limb tissue corresponding to the leading edge of the prosthesis shows a periodic slip peak of 2.1mm at the contact point C, and the slip beyond this area The threshold (1.2mm) is about 75%, and it has not been relieved after 30 cycles. The system marks this point as a slip overload risk point; further analysis shows that this point belongs to the sub-channel intersection area on the shear stress migration path, that is, the place where multiple weak shear forces converge. Although the pressure is low, the direction changes frequently, which can easily cause micro-tearing; the system generates a slip amplitude map based on this and combines historical data, marking the high slip area in red, reminding designers that the structural stiffness and friction characteristics at this location do not match, and buffer adjustment and material replacement are required; for this point, the system recommends optimization solutions including: replacing the interface material of this area from thermoplastic polyurethane (TPU) to coated silicone with a low friction coefficient, and introducing a microspherical elastic adsorption layer to make the tissue slip. The displacement is absorbed and buffered, and at the same time, variable microtexture is added to the surface of the prosthetic interface to destroy the consistency of the slip path, thereby inhibiting the repeated friction path; the newly designed structure is re-imported into the simulation platform to perform the same behavioral simulation, and the slip peak of contact point C drops to 0.88mm, which is controlled below the slip threshold (1.2mm) in this area. The displacement fluctuation amplitude decreases by 47%, and the periodic rebound phenomenon is significantly weakened. At the same time, the subjective feedback of the residual limb user is that the tingling sensation of the skin disappears during fast walking after wearing it, and the indentation is significantly reduced, indicating that the soft tissue slip response modeling mechanism is highly accurate and practical in terms of micro-area resolution, dynamic behavior mapping and structural adjustment guidance, and can significantly improve the wearing time of the prosthesis and the healthy life of the tissue.

[0061] Further implementation of the joint optimization process based on the multi-objective evolutionary algorithm, with the shear force trajectory, stress concentration distribution and soft tissue slip response results as input parameters, comprehensively considering the coupling efficiency, physiological safety and material adaptability of the prosthetic interface under complex dynamic behavior, and performing multi-dimensional collaborative optimization of the interface structure in terms of geometry, buffer layer thickness and material stiffness distribution. The system performs path energy field modeling on the shear force migration path obtained in the previous analysis, identifies the main channel and high shear stress concentration area, and outputs the stress risk map in combination with the stress retention point identification mechanism, and superimposes the slip amplitude map to generate a three-dimensional risk composite matrix R (x, y, z) as the environmental mapping basis of the optimization function; secondly, the system sets three main optimization goals: ① minimize the peak shear stress per unit area in the contact area (objective function F1), ② control the thermal-mechanical coupling energy in all potential retention point areas not to exceed the tissue threshold upper limit (F2), and ③ ensure The maximum displacement in each high-slip risk area is lower than its corresponding slip tolerance threshold (F3); based on this, a joint objective function F = w1·F1+w2·F2+w3·F3 is constructed, where the weights are set to w1=0.4, w2=0.35, and w3=0.25 due to the sensitivity of Mr. W's residual limb tissue. The NSGA-III multi-objective optimization algorithm is used to perform population evolution iteration. The initial population consists of 128 sets of parameters, covering the interface geometry edge fitting curvature (range: ±10° adjustment), buffer layer thickness (range: 2mm-7mm ), material elastic modulus (range: 15-85kPa), the system performs motion simulation in each generation and quickly evaluates the above objective function response through the digital twin model, calculates the fitness function and performs survival of the fittest cross-mutation. After 42 generations of evolution, the population converges and finally outputs the Pareto optimal solution set; in Mr. W's optimization results, the optimal solution increases the thickness of the buffer layer in the middle area of ​​the front edge of the prosthesis from 3mm to 5.5mm, and changes the original homogeneous structure to a central high-elastic area (modulus 45kPa) and a soft transition layer on the edge (modulus 28kPa) Coordinated distribution, while the leading edge geometric curvature was fine-tuned from 5.3° to 2.1° to reduce shear concentration. Simulation verification showed that the overall shear peak decreased by 38.6%, the energy of high-risk retention points decreased by 31%, and the slip peak was controlled within 0.92mm, meeting the safety threshold of all objective functions. Subsequently, the structural sample was manufactured through 3D printing and multi-material fusion process, and in the wearing experiment, Mr. W completed a 72-hour continuous use test without skin redness, indentation or tingling feedback, and the wearing stability score was improved from the original 72 to 91 (out of 100).

[0062] Example 2:

[0063] Combined with attachment Figure 5, a risk distribution assessment method for tissue fatigue induced by long-term use of prosthetic interfaces was implemented. This method comprehensively considers the tissue deformation history and thermal energy accumulation effect, combines simulation analysis with structural optimization feedback mechanism, and performs dynamic risk modeling and material adjustment on the prosthetic interface. In step S1, the team first reconstructed the three-dimensional model of Mr. W's residual limb soft tissue based on the high-resolution MRI image data of his right calf, and integrated the previously established prosthetic interface geometric model to complete the high-precision alignment of the residual limb and the prosthetic interface in the digital twin platform to form an interactive three-dimensional coupling model; then, dynamic gait simulation was performed in the model, including six types of common daily actions such as walking on flat ground, jogging uphill, squats and standing up. Each type of action was repeated 20 times, and the total simulation time was 60 minutes. During the simulation process, the system recorded the three-dimensional deformation vector of each micro-unit point at each moment, and generated deformation vector field data through the algorithm; Among them, when Mr. W was walking fast uphill, his tibia The residual limb tissue in the leading edge area showed a trend of continuous displacement focusing, and the cumulative offset of the deformation vector reached 3.2mm, which was much higher than the upper limit of the average deformation of the soft tissue in this area of ​​1.9mm. The system marked the trajectory aggregation area in this area as the potential fatigue inducing point P1; entering step S2, the system further started the thermal coupling modeling module to simulate the local temperature rise caused by friction and compression between the tissue and the interface during wearing, and combined the thermal conductivity of the tissue, the heat dissipation capacity of the skin and the blood flow data of the residual limb area to establish a heat diffusion and retention model. The results showed that after the 42nd minute of simulation, the temperature of the P1 area rose from the initial 36.5℃ to 39.2℃, with a temperature increase of 2.7℃, and at the time of receiving the The temperature remained above 38.7°C for the next 18 minutes, and the heat release efficiency was significantly lower than that of the surrounding areas. The system superimposed the deformation trajectory map of the P1 area and the thermal field response curve to generate a fatigue risk spatial distribution map, in which the point was marked as a red high-risk area; entering step S3, based on the risk map system analysis results, the team carried out structural and material configuration optimization for the interface area where point P1 is located, adjusted the thickness of the interface buffer structure, and adjusted the original 3.5mm silicone layer to a composite structure: the inner layer uses 5mm microporous thermal conductive silicone, the middle layer adds high thermal conductivity graphite sheets as heat diffusion elements, the outer layer is a shear-resistant polyurethane coating, and the angle curvature of the edge transition area of ​​the area is changed from 45° It was optimized to 30° to slow down the local compression gradient. After the system was re-simulated, the data showed that the maximum displacement of the deformation focus path in this area dropped to 1.8mm (a decrease of 43.8%), the peak temperature of heat accumulation dropped to 37.9℃ (a decrease of 1.3℃), and the heat retention time was shortened to less than 8 minutes; the risk distribution map changed from the original red risk to the orange controllable risk level, and the system risk score dropped from 0.84 to 0.36 (risk level 0 is the safest and 1 is the highest risk); Mr. W conducted a continuous 90-minute walking test in the actual test after wearing the optimized structure. There was no redness or tingling in the residual limb, the local temperature rise was controlled within the normal body temperature range, and the sense of structural adaptation was significantly improved.

[0064] A micro-area fatigue prediction mechanism for residual limbs based on deformation vector field analysis and thermal-mechanical coupling relationship modeling was introduced. By carefully tracking tissue deformation paths and heat retention behaviors, chronic pressure injury risk points were discovered in advance and assisted in iterative optimization of prosthetic structures. A micro-unit discrete grid was established in the contact area between Mr. W's residual limb and prosthesis, with a total area of ​​approximately 148 cm. 2, a total of 2124 quadrilateral micro-units were divided, and the side length of each unit was controlled between 3 and 4 mm. The system performed five types of high-frequency lower limb movement simulations on Mr. W through the digital twin drive platform (respectively, uphill brisk walking, standing up and down, squatting, striding, and static rotation). Each movement was repeated 30 cycles, and the total simulation time was 96 minutes; during this process, the system continuously recorded the three-dimensional displacement vector of each micro-unit in each movement cycle, and formed a time-continuous deformation vector field. For example, unit #0789 located at the anterior and inferior edge of the tibia showed a continuous deviation in the posterior and upward direction during uphill brisk walking, with an average displacement amplitude of 2.3 mm per cycle and a direction angle change of less than ±8°, forming an obvious linear deformation trajectory; the system executed the trajectory for all grid units The results of clustering of path stability and repetition amplitude showed that 16 areas formed high-frequency repetitive deformation trajectories, among which the P3 area (the inner front edge of the prosthesis) was the representative. The average repetitive offset trajectory of this area was 2.1mm, the fluctuation amplitude was controlled within 0.4mm, and the directional consistency score was 0.92 (full score 1.0), which was marked as a high-risk fatigue-inducing area; then, in order to evaluate the thermal stress coupling effect caused by mechanical deformation, a thermal-mechanical coupling model was further established in the twin model. This model comprehensively considered the contact friction coefficient between the skin and the prosthetic interface (the actual measurement in Mr. W's area was 0.43), tissue thermal conductivity (skin was 0.37W / m·K, fat layer was 0.21W / m·K), thermal conductivity of the prosthetic cushioning material (EVA foam), and thermal conductivity of the prosthetic cushioning material. The temperature response trajectory of the P3 area under continuous action was analyzed through multi-cycle simulation based on the heat accumulation curve per unit time. The simulation data showed that from the 45th minute, the temperature of the P3 area gradually increased from the initial 36.7°C to 39.1°C, and remained above 38.6°C for more than 20 minutes in the subsequent cycle, indicating that the area had typical thermal retention characteristics, which was highly consistent with the continuous deformation trajectory. The system merged the deformation focus path map of the P3 area with the heat accumulation thermodynamic map to generate a fatigue-induced risk superposition map. The risk index reached 0.87 (the threshold was 0.65), and it was listed as a first-level warning area. For this area, the optimization suggestions included replacing the original buffer structure of the area from a single-layer EVA material to a three-layer EVA material. The optimized structure is composed of a high-thermal-conductivity coating on the surface, a thermal-conductive elastic material containing graphite particles (thermal conductivity 0.75 W / m·K) in the middle layer, and a low-modulus silicone energy-absorbing layer on the bottom layer. A 0.5 mm concave buffer groove is introduced in the geometric structure to reduce strain concentration. Re-simulation after optimization showed that the maximum deformation amplitude of the P3 area dropped to 1.3 mm, the trajectory consistency fluctuation increased (the direction change range expanded to ±14°), the deformation trajectory tended to be discrete, the thermal peak temperature dropped to 37.8°C, the thermal retention time was shortened to less than 7 minutes, and the risk index dropped to 0.41, effectively escaping the high-risk level of fatigue. Finally, Mr. W wore the optimized structure for three consecutive days, walking for more than 90 minutes a day, without symptoms such as skin discoloration, blisters or tingling, and his gait posture remained stable.

[0065] After the system completed the generation of the fatigue risk spatial distribution map and identified the high-risk stress-heat superposition area, it further implemented a structural parameter adjustment mechanism based on risk map feedback, and intervened one by one around the four key optimization dimensions - cushioning material distribution, regional elasticity control, contact texture design and thermal conductivity optimization, so as to form a multi-level, physiological response matching prosthetic interface optimization structure. According to the fatigue inducing point of the P3 area identified in the digital twin model simulation, its spatial positioning is located at the junction of the prosthetic interface and the anterior inferior edge of Mr. W's tibial residual limb. The accumulated shear stress in this area frequently exceeds 80kPa during continuous gait, and the heat retention time exceeds 25 minutes. The system marks this area as a level 1 high risk; for this reason, in terms of cushioning material distribution, the system replaces the original 3mm single-layer EVA foam structure with a multi-layer partitioned structure, in which the central area uses 5mm high-elasticity silicone material (modulus of about 25kPa) as the main cushioning energy absorption layer, and the surrounding area gradually transitions to 3mm porous foam to achieve pressure release, and uses a dynamic thickness transition algorithm to make the interface hardness gradient change. Less than ±20% to ensure a fault-free pressure boundary; secondly, in terms of regional elasticity control, the system divides the entire contact surface into four elastic response zones, which are set as high elasticity, medium elasticity, buffering and soft elasticity. Among them, the P3 high strain area is classified into the soft elastic response zone, and its contact structure is given a local deformation rate of 0.37~0.45 (dynamically adjusted according to the simulation feedback curve) to avoid stress retention due to insufficient local tissue rebound ability; thirdly, in terms of contact texture design, in order to enhance the shear stress diffusion ability and reduce the repeatability of the skin slip path, the system introduces a random micro-texture array with a radius of 1.5mm on the surface of the fatigue risk area, and the distribution density is controlled at 25 textures / cm 2, the texture depth is 0.4mm. This design can effectively interrupt the continuous path of the main shear force channel in the simulation, making the stress distribution more uniform and the slip response reduced by 32%; finally, in terms of thermal conductivity optimization, the system combines the thermal retention data to introduce a 0.3mm thick graphite thermal conductive sheet into the middle of the risk area, which has a thermal conductivity of up to 8.5W / m·K, which can significantly increase the lateral diffusion rate of thermal energy. At the same time, a thermal conductive coating (such as a boron nitride composite resin coating) is applied to the surface of the original buffer structure, and its overall thermal conductivity is 0.78W / m·K, ensuring that heat is quickly dissipated to the prosthetic shell and avoiding local temperature rise of the residual limb; after optimization, the system is re-digitalized. A 72-minute gait simulation test was performed on the twin platform. The data showed that the maximum shear stress in the P3 area dropped to 54kPa, the peak temperature of heat accumulation dropped to 37.6°C, the tissue rebound rate improved by 16.4%, the slippage was controlled within 0.91mm, the system fatigue risk index dropped from 0.87 to 0.36, and the risk level changed from red to green. In application, Mr. W wore the optimized interface for three consecutive days, 90 minutes a day for exercise tests. His skin performance was stable, with no redness, swelling, blisters or pain. The symmetry of his exercise posture improved by about 14%, and the overall user satisfaction rating increased to 94 points (out of 100).

[0066] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the foregoing embodiments. The foregoing embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. The optimization method of coupling between prosthetic limb and residual limb based on digital twin is characterized by The following steps are involved: The multimodal data fusion is performed on the anatomical structural characteristics of the residual limb, including skin, soft tissue, and bone, and the functional properties of the residual muscle control ability, myoelectric response, and stress distribution data; a multi-layer digital residual limb relationship model is constructed; a functional residual mapping mechanism is introduced to calibrate the residual limb area responsible for force conduction and the area suitable for neural interface implantation; Define the multi-scale contact space between the residual limb and the prosthesis during dynamic behavior, consisting of the structural adaptation area, force transmission area, and energy buffer area; Simulate the shear force migration trajectory, stress concentration pattern, and soft tissue slip response between micro-units of the coupling surface during dynamic interaction; construct the physical partition relationship of the coordination domain to assist in optimizing the prosthetic interface structure and material functional gradient; Describe the microtrauma areas caused by prostheses during wearing and use, and calculate the path of minimum tissue disturbance, that is, the path that minimizes strain concentration or heat accumulation caused by the prosthesis on the tissue structure; Through deformation vector field analysis and thermal-mechanical coupling relationship modeling, the risk distribution of tissue fatigue induced by long-term use of prosthetic interfaces is evaluated.

2. The optimization method for coupling prosthetic and residual limbs based on digital twins according to claim 1 is characterized in that In the simulated dynamic interaction, a personalized digital twin model of the residual limb including anatomical structure, functional attributes and stress response is constructed; A multi-scale contact space is established between the residual limb twin model and the prosthetic limb digital model. The space is composed of a structural adaptation area, a force transmission area, and an energy buffer area. Based on dynamic behavior simulation, the contact space is divided into micro-units to simulate the shear force migration trajectory between each micro-unit. Based on the force changes of micro-units, the stress concentration evolution pattern in the coupling area is identified; the soft tissue slip response of the contact interface is modeled and the slip amplitude map is output; The structural parameters and material layout of the prosthetic interface are optimized based on the shear migration trajectory, stress concentration trend and slip tolerance.

3. The optimization method for coupling prosthetic limbs with residual limbs based on digital twins according to claim 2 is characterized in that The individualized residual limb twin model integrates the skin, muscle, and bone structure information of the residual limb, combines biomechanical properties with myoelectric response characteristics, and establishes a multi-layer simulation model integrating anatomy, function, and stress.

4. The optimization method for coupling prosthetic and residual limbs based on digital twins according to claim 3 is characterized in that The multi-scale contact space is divided into: Structural adaptation zone: makes the prosthesis fit the residual limb geometrically; Force transfer area: used to transfer the movement intention of the residual limb to the prosthesis through force; Energy buffer: used to absorb shear force and impact force and relieve tissue stress accumulation.

5. The optimization method for coupling prosthetic and residual limbs based on digital twins according to claim 4 is characterized in that The simulation of the shear force migration trajectory is based on micro-unit grid division. The twin model is driven by time-series residual limb motion data to continuously track the shear force change path in the contact space. The identification of the stress concentration pattern is based on a nonlinear tissue response model to monitor the stress retention and cumulative evolution of the micro-unit area under dynamic behavior.

6. The optimization method for coupling a prosthetic limb with a residual limb based on digital twinning according to claim 5 is characterized in that The soft tissue slip response modeling includes: Establishing the slip threshold between the skin and muscle layers; Analyze the relative displacement distribution of the contact interface under different motion states; The output slip amplitude map is used to evaluate the matching degree between the tissue friction area and the interface structure.

7. The optimization method for coupling prosthetic and residual limbs based on digital twins according to claim 6 is characterized in that The optimization process is based on shear force trajectory, stress concentration distribution and slip response results, and uses a multi-objective evolutionary algorithm to jointly optimize the geometric structure, buffer material thickness and stiffness distribution of the prosthetic interface.

8. The optimization method for coupling prosthetic and residual limbs based on digital twins according to claim 1 is characterized in that The method for evaluating the risk distribution of tissue fatigue induced by a prosthetic interface under long-term use includes: S1. Establish a three-dimensional coupling model of the residual limb soft tissue and prosthetic interface in the digital twin model; perform deformation vector field analysis on the residual limb interface during prosthetic use, identify deformation concentration trajectory areas based on the user's repeated motion behavior, and mark potential fatigue-inducing points; S2. Establish a thermal-mechanical coupling relationship model to simulate the accumulation and release of heat in the tissue during prosthetic wear; integrate the deformation trajectory with the thermal field change trend to generate a spatial distribution map of tissue fatigue risk; S3. Optimize the prosthetic interface structure or material configuration according to the fatigue risk distribution map.

9. The optimization method for coupling prosthetic and residual limbs based on digital twins according to claim 8, characterized in that The deformation vector field analysis establishes a discrete grid of micro-units in the contact area, records the displacement direction and deformation amplitude of each grid unit in multi-cycle motion, and forms a continuous deformation trajectory path; the deformation trajectory is used to identify high-frequency repeated deformation areas as a basis for local fatigue risk assessment; the thermal-mechanical coupling relationship modeling integrates the thermal friction of the skin-prosthesis interface, the thermal conductivity of biological tissue, and the thermal conductivity characteristics of the prosthetic material, and obtains the heat accumulation area through multi-cycle simulation.

10. The optimization method for coupling prosthetic limbs with residual limbs based on digital twins according to claim 9, characterized in that The structural parameters of the prosthetic interface are adjusted according to the results of the fatigue risk distribution map, including: buffer material distribution, regional elasticity control, contact texture design or thermal conductivity optimization.

Images