Optimization method for coupling of prosthesis and residual limb based on digital twinning

Abstract

The present application relates to an optimization method for coupling of a prosthesis and a residual limb based on digital twinning. The anatomical structure characteristics and functional attributes of the residual limb skin, soft tissue and bone are fused with stress distribution data in multi-modal data fusion; a multi-layer digital residual limb relationship modeling is constructed; a multi-scale contact space composed of a structure adaptation area, a force transmission area and an energy buffer area of the residual limb and the prosthesis in dynamic behavior is defined; in the simulation of dynamic interaction, the shear force migration trajectory, stress concentration mode and soft tissue slip response between each micro unit of the coupling surface are simulated; the physical partition relationship of the coordination domain is constructed to assist in optimizing the structure and material function gradient of the prosthesis interface; the micro-injury area caused by the prosthesis in wearing and use is described, and the minimum tissue disturbance path is calculated; through deformation vector field analysis and thermal force coupling relationship modeling, the risk distribution of tissue fatigue induced by the prosthesis interface under long-time use is evaluated; dynamic multi-scale coupling optimization and multi-objective optimization control are realized to achieve fatigue risk visualization prediction and early warning.

Description

TECHNICAL FIELD

[0001] The present application relates to an optimization method for coupling a prosthetic limb with a residual limb, in particular an optimization method for coupling a prosthetic limb with a residual limb based on digital twinning. BACKGROUND

[0002] The coupling relationship between the prosthetic limb and the residual limb has always been a key problem in the field of biomechanical engineering, rehabilitation auxiliary technology and intelligent manufacturing. The core of the problem is how to achieve the coordinated matching of the prosthetic limb interface structure and the individual residual limb in terms of geometric shape, biomechanical performance and neural-muscular function. Most of the optimization methods currently widely used still remain in the stage of static modeling and empirical debugging, which is difficult to meet the real wearing needs of users in dynamic behavior. For example, the technical solution disclosed in Chinese patent CN118986601A, a method and device for optimizing the shape of a prosthetic limb socket, still has obvious deficiencies and structural limitations, especially in terms of adaptability control, dynamic prediction ability and user feedback fusion, which cannot meet the development needs of modern high-performance prosthetic systems.

[0003] The existing technology mainly divides the surface of the residual limb into multiple regions, sets a compression center, and applies a simulated compression test in multiple regions to obtain a series of static test indicators such as compression rate, suspension force gap, support force gap, and bone movement restriction, and uses these indicators as target functions for multi-objective optimization to ultimately obtain the shape parameters of the prosthetic limb socket. Although this method does indeed achieve reasonable modeling of the local pressure bearing capacity of the residual limb and introduces an optimization algorithm to improve the comfort and support of the socket design, there are still several technical drawbacks and deficiencies. First, the model construction of this scheme is completely based on static residual limb geometric data and mechanical response results, without considering the tissue response characteristics of the wearer in dynamic behavior, nor introducing a multi-cycle motion simulation or behavior scenario change modeling mechanism, resulting in the final optimization result only applicable to standing or initial wearing state, unable to truly reflect the shear stress, slip and tissue compression response changes in continuous motion such as walking, squatting and uphill walking. This static deduction approach severely limits its generalization ability in actual multi-scenario applications.

[0004] Secondly, the optimization objective function only depends on mechanical indicators such as compression rate and support gap, and ignores important physiological factors such as heat accumulation, micro-slip trajectory, and local muscle function preservation area of the residual limb, which are often the core reasons for user pain, pressure injury, inflammation, and unstable wearing. Thirdly, the prior art does not construct a multi-scale simulation space, all area analysis is based on surface compression test, lacks the ability to divide the contact interface into biological function partitions such as structure adaptation area, force transmission area and energy buffer area, and cannot support setting different structure strategies (such as elastic gradient adjustment or multi-material composite layout) for different areas, which makes the optimization result tend to be homogeneous, and lacks personalized response ability to complex mechanical function areas of the human body. SUMMARY

[0005] The purpose of the present application is to provide an optimization method for coupling of prostheses and residual limbs based on digital twinning, so as to solve some of the problems and deficiencies pointed out in the background art.

[0006] The present application solves the above-mentioned technical problems by adopting the following technical solution: an optimization method for coupling of prostheses and residual limbs based on digital twinning, comprising: multi-modal data fusion of anatomical structure characteristics of the residual limb including skin, soft tissue and bone, and functional attributes including muscle group remaining control ability, muscle electrical response and stress distribution data; constructing a multi-layer digital residual limb relationship modeling; introducing a functional residual mapping mechanism to calibrate the force conduction of the residual limb area and the area suitable for nerve interface implantation;

[0007] defining a multi-scale contact space composed of structure adaptation area, force transmission area and energy buffer area of the residual limb and the prosthesis in dynamic behavior; simulating the shear force migration trajectory, stress concentration mode and soft tissue slip response between each micro-unit of the coupling surface in dynamic interaction; constructing the physical partition relationship of the coordination domain to assist in optimizing the interface structure and material function gradient of the prosthesis;

[0008] describing the micro-injury area caused by the prosthesis in wearing and use, calculating the minimum tissue disturbance path, i.e. the path that minimizes the strain concentration or heat accumulation caused by the prosthesis to the tissue structure; through deformation vector field analysis and thermal force coupling relationship modeling, evaluating the risk distribution of tissue fatigue induced by the prosthesis interface under long-term use.

[0009] Further, the individualized residual limb digital twin model is constructed in the simulation of dynamic interaction, which includes anatomical structure, functional attributes and stress response;

[0010] A multi-scale contact space is established between the residual limb twin model and the prosthesis digital model, which is composed of structure adaptation area, force transmission area and 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 stress change of the micro-unit, 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 mapping is output; according to the shear migration trajectory, the stress concentration trend and the slip tolerance, the structure parameters and material layout of the prosthesis interface are optimized.

[0012] Further, the individualized residual limb twin model fuses the skin, muscle, and skeletal structure information of the residual limb, combines the biomechanical characteristics and electromyographic response characteristics, and establishes a multi-layer simulation model of anatomy, function, and stress.

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

[0014] The structure adaptation area: makes the prosthesis and the residual limb geometrically fit;

[0015] The force transmission area: used to transmit the action intention of the residual limb to the prosthesis through force;

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

[0017] Further, the micro-unit grid is constructed in the contact area between the residual limb and the prosthesis, the entire contact surface is divided into multiple discrete units, and the action driving time axis is introduced to simulate the shear force dynamic flow path of the human body during movement; the shear force state of each micro-unit at any time is defined as a vector; in order to describe the migration trend and directionality of the shear force, the following shear force migration evolution function is constructed:

[0018]

[0019] Wherein:

[0020] represents the shear force migration sensitivity field intensity function at the spatial position at time t, that is, the path aggregation trend of the shear force at the point; represents the gradient under the spatial coordinate, reflecting the migration flow direction of the shear force along the position direction; represents the instantaneous shear stress change rate at position at time τ; represents the slip tolerance tensor factor of the tissue at the position, which is related to the type of soft tissue and physiological morphology; represents the action impact conversion factor of the micro-unit under the behavior driving, reflecting whether the point is in the high-frequency response path induced by the behavior; represents the integral superposition of the above factors in the history time, capturing the cumulative migration trend of the shear force on the path;

[0021] The output result of the shear force migration function is combined with the nonlinear response properties of the organization in the digital twin model to define a dynamic stagnation point, i.e., a position that cannot effectively release stress and repeatedly bear pressure within a time sequence; these regions are marked as potential fatigue-inducing points; the stress stagnation mechanism is based on the following logic: if a micro-unit point Local maximum values appear in multiple time periods without being accompanied by shear stress relief processes;

[0022] At the same time, the organization thermal decoupling parameter shows a continuous upward trend;

[0023] If so, the system marks this point as a stress-thermal coupling stagnation point, constituting a fatigue risk area;

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

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

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

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

[0028] Outputting a slip amplitude map for evaluating the matching degree of the tissue friction area and the interface structure.

[0029] Further, 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] Further, the method for evaluating the risk distribution of the prosthetic interface in inducing tissue fatigue under long-term use includes:

[0031] S1. Establishing a three-dimensional coupling model of the residual limb soft tissue and the prosthetic interface in the digital twin model; performing deformation vector field analysis on the residual limb interface area during prosthetic use, identifying deformation concentration trajectory areas based on user repetitive motion behaviors, and marking potential fatigue-inducing points;

[0032] S2. Establishing a thermal-mechanical coupling relationship modeling to simulate the accumulation and release of heat in the tissue during prosthetic wearing; combining the deformation trajectory and the thermal field change trend to generate a tissue fatigue risk spatial distribution map;

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

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

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

[0036] The beneficial effects of the present application: by fusing the anatomical structure, electromyographic function and biomechanical response of the residual limb, a multi-level digital twin model is established, realizing the transformation of prosthesis design from standard template to individual precise fitting, and making the prosthesis interface and residual limb optimally matched in morphology, mechanics and function. The contact surface is divided into structure adaptation area, force transmission area and energy buffer area, and dynamic simulation mechanisms such as shear force migration trajectory analysis, stress concentration identification and slip response modeling are introduced, realizing multi-scale dynamic optimization of coupled structure in real use scenario, and improving the stability and physiological fit of the prosthesis.

[0037] By deformation vector field and thermal force coupling modeling, tissue fatigue risk distribution map is formed, potential pressure injury or microtrauma high-risk area is identified in advance, data support is provided for interface material partition design and structure buffer optimization, and wearing safety and long-term tissue health are significantly improved. Evolutionary algorithm based on shear, thermal stress and slip behavior feedback is adopted to realize joint optimization of prosthesis interface geometry, material stiffness and buffer structure, form an automatic iterative updating mechanism, greatly reduce the artificial trial and error design time, and improve the fitting efficiency. Through dynamic response simulation of soft tissue and design of heat diffusion channel, the local stress peak and heat accumulation are effectively reduced, the continuous wearing time is prolonged, the probability of skin discomfort and pain is reduced, and the user's daily activity ability and life quality are significantly improved. BRIEF DESCRIPTION OF DRAWINGS

[0038] Figure 1 The prosthesis and residual limb coupling optimization flowchart based on digital twin of the present application.

[0039] Figure 2 The three functional areas of the prosthesis-residual limb coupling interface and the optimization design function relationship diagram of the present application.

[0040] Figure 3 The prosthesis-residual limb interface minimum tissue disturbance path modeling and closed-loop optimization relationship diagram of the present application.

[0041] Figure 4An implementation flowchart of personalized prosthesis digital twin partitioning and structure optimization of the application.

[0042] Figure 5 A system structure diagram for prosthesis interface fatigue risk distribution evaluation and multi-dimensional structure optimization of the application. DETAILED DESCRIPTION

[0043] The specific embodiments of the application will be described in detail below with reference to the accompanying drawings.

[0044] In conjunction with the accompanying drawings Figure 1 , the multi-modal raw data of the user's individualized residual limb is obtained, including the skin layer surface morphology data, soft tissue thickness and elasticity information, and bone geometry and positioning data of the residual limb, which can be collected through structured light scanning, CT image reconstruction, and ultrasonic elastography, etc. Then, through registration algorithm, the three types of anatomical structures are aligned in three-dimensional space, and a multi-level residual limb digital model with clear anatomical levels and coherent topological structure is constructed. At the same time, the response data of the user's residual muscle groups under different postures and action intentions are obtained through the electromyographic signal acquisition device, and combined with the action information recorded by the wearable inertial sensor, the residual control ability indicators of each muscle group are calculated, such as response intensity, delay time and signal-to-noise ratio, etc. In addition, in order to further realize physical and mechanical simulation, the 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 sensing array is used to record the evolution process of the stress of each region over time, and obtain the stress distribution and peak value migration trajectory data. Based on the above obtained anatomical structure, functional attribute and stress evolution information, the system unifies the spatial, temporal and physiological dimensions of the multi-source heterogeneous data, and constructs a five-layer nested digital residual limb twin model including skin layer, soft tissue layer, bone structure layer, muscle activity layer and stress response layer. In this model, a functional residual mapping mechanism is further introduced, which is based on the electromyographic intensity spatial distribution and force conduction path inversion algorithm, evaluates the actual participation of each anatomical unit under different action scenarios, and compares it with the mechanical contribution under ideal state, so as to determine the functional residual value of each residual limb region, which is used to quantify the degree of relative functional degradation or residual ability reservation of a region in actual use. According to the coupling results of the functional residual value and the stress density distribution, the system generates a region label map, which respectively marks the key structure belt mainly responsible for force conduction task (such as tibial distal load area, residual muscle tendon anchor point, etc.), and the low shear stress, high biological stability area suitable for nerve interface implantation or electrode arrangement. The partitioning results can be used as the basis for subsequent prosthesis structure interface matching design and nerve-prosthesis system docking, so as to realize the personalized collaborative optimization of human-machine interaction structure and neural response system in prosthesis design.

[0045] In conjunction with the accompanying drawings Figure 2By dynamic behavior simulation, the contact interface between the residual limb and the prosthesis that generates coupling interaction in actual use is defined as a combination of three types of functional zones, namely, the structure adaptation zone, the force transmission zone, and the energy buffer zone. The structure adaptation zone is mainly used to realize the fitting of the prosthesis geometry and the skin surface of the residual limb. This area needs to ensure high surface fitting degree and contact stability. The force transmission zone is the core path through which the residual limb internal bones or muscles activity transmits the movement intention and support force. Its position is usually located at the residual limb tendon attachment point, bone surface protruding area, and other parts with strong force feedback capability. The energy buffer zone is used to absorb excess energy and relieve soft tissue shear injury during repeated movement or impact. This area is usually distributed at the contact edge, where the soft tissue is thick, or near the area rich in blood vessels and nerves, with the design requirements of strong buffering and high wear resistance. In order to simulate the response state of the three types of functional zones in actual dynamic behavior, the contact interface is further meshed with micro-units, and the residual limb-prosthesis coupling surface is divided into high-density discrete units. Time control parameters are introduced, and the system is driven by user behavior data recorded in the twin model for dynamic simulation. The shear stress, normal pressure, micro-slip vector, and heat energy transfer of each unit in different postures are calculated in real time, and the shear force migration trajectory on the coupling surface is tracked. The flow trend of shear stress in time and space dimensions is analyzed, and the micro-area with stress retention after multiple behavior cycles is identified to form the stress concentration mode recognition result. In addition, based on the tissue hierarchical structure and the response of the residual limb shape, 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 the condition of repeated motion, thereby warning potential wear risk. After the simulation data is fused and clustered in the digital twin system, the spatial physical boundaries and behavior response boundaries of the three types of functional zones of the coupling surface can be output, and a physical partition relationship diagram of the coupling coordination domain is constructed accordingly. This diagram clearly defines the main functional role of each micro-zone in actual use, serving as an important reference for subsequent prosthesis interface design. The system maps this partition to the CAD modeling environment and assigns appropriate material types and structure parameters based on the functional attributes of different regions, such as using high-strength carbon fiber composite materials in the force transmission zone to enhance response speed, using multi-layer variable modulus silicone in the buffer zone to improve shock absorption capacity, and optimizing the elastic fitting layer and surface texture in the structure adaptation zone. Thus, a prosthesis interface structure with a gradient continuous transition of material function is formed, and the overall optimization of coupling comfort, stability, and physiological safety is finally achieved.

[0046] Combining the attached Figure 3,In view of the chronic micro-injury problem caused by the prosthesis to the stump tissue during wearing and use, a modeling and evaluation mechanism of the minimum tissue disturbance path is proposed to identify and minimize the local high strain or thermal accumulation effect of the prosthesis interface structure on the biological tissue. In the digital twin integrated model of the residual limb-prosthesis, the soft tissue in the coupling region is discretely processed by high-precision grid division, and the dynamic deformation data of each micro-unit is tracked and recorded during multiple simulation of typical use actions of the residual limb (such as walking, standing, crouching, turning, etc.), forming a continuous deformation vector field, wherein each vector represents the cumulative displacement and stress direction of a specific micro-unit in time sequence. The 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 truly reflect the physiological response under long-term wearing conditions, the system further introduces a thermal force coupling modeling mechanism, which takes into account parameters such as friction heat generation, tissue thermal conductivity, skin heat conduction hysteresis, and thermal resistance characteristics of the prosthesis material, to simulate the heat accumulation in the contact area of the residual limb during repeated wearing and movement, especially focusing on the areas where heat cannot be diffused in time and the corresponding stress state of the soft tissue. The system fuses the above mechanical deformation vector field and thermal energy conduction path data to construct a multi-dimensional energy consumption path atlas, and on this basis sets an optimization objective function, i.e. to find the path with the minimum cumulative strain intensity and thermal load on the soft tissue as the minimum tissue disturbance path, which usually avoids bone protrusion sites, tissue junctions and blood vessel dense areas, and can be used to guide the design of prosthesis interface area shape, material stiffness distribution, contact structure fine tuning and other optimization measures. In addition, through the long-term dynamic evolution of the residual limb model, the system simulates the shear force migration, deformation residual and thermal stress hysteresis effect under the multi-day wearing scenario, and according to the deformation history, temperature superposition effect and tissue stress recovery ability of each micro-unit, outputs a potential tissue fatigue risk distribution map, which visually marks the high-risk areas, provides early warning basis for chronic pressure injury, skin damage, soft tissue atrophy and other problems that occur after wearing the prosthesis, and supports updating the twin model and structure optimization suggestions according to the feedback data, realizing preventive safety control and continuous improvement of wearing comfort of the prosthesis design.

[0047] Embodiment 1

[0048] In combination with the accompanying Figure 4In this embodiment, Mr. W, a 35-year-old male, lost his right lower leg in a traffic accident four years ago. He has been wearing a conventional prosthetic limb since then, but he has repeatedly experienced problems such as local pain, skin redness and swelling, and prosthetic limb loosening. Doctors diagnosed that the long-term wearing of the prosthetic limb caused chronic fatigue and shear force accumulation in the soft tissue. Therefore, the team developed a digital twin-driven prosthetic limb interface adaptation solution for Mr. W. First, high-resolution CT scans were used to obtain the skeletal structure of Mr. W's right lower leg stump, and 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. Subsequently, these data were fused into a three-layer residual limb structure model. To supplement the functional information, an eight-channel surface electromyography sensor was used to collect the electrical signals of the residual femoral biceps, quadriceps, and hamstring muscle groups during walking and squatting. The results showed that the maximum muscle peak value was about 0.75 mV, the signal stability index was 0.82, and the reaction time delay was 70 ms, indicating that the muscle group still had certain motor control potential. After the construction of the anatomic-functional fusion residual limb digital twin model, historical wearing data of the residual limb and the prosthetic limb were introduced, and it was found that the left posterior region of the contact interface repeatedly appeared local pressure peak exceeding 120 kPa, with slip exceeding 2.3 mm, which was significantly higher than the tissue slip tolerance threshold (1.5 mm), and was marked as a high-risk stress accumulation area by the system. Second, the team coupled the twin residual limb model with the prosthetic limb structure model to be optimized, establishing a multi-scale coupled contact space including a structure adaptation zone, a force transmission zone, and an energy buffer zone. The adaptation zone covers the fitting surface corresponding to the anterior tibia and the posterior fibula, the force transmission zone is concentrated on the central axis of the tibial section, and the energy buffer zone is located in the area with thick soft tissue coverage on the medial side of the residual limb. The entire contact interface is divided into 2146 micro-unit grids, each with a side length of 3.5 mm. By driving the twin model to perform a 10-minute walking simulation, the system records the shear stress vector field data and generates a shear force migration path atlas, identifying a main shear force channel that migrates from the lower posterior to the upper anterior during gait. The average shear stress in the first 3 minutes is maintained at 37-54 kPa, and then gradually accumulates to more than 82 kPa, far exceeding the residual limb tissue threshold. Third, based on these simulation data, the system analyzes that the left-central region of the coupling surface presents a clear stress concentration evolution pattern, with a shear stress retention time of 6.8 seconds per cycle, which is much higher than the normal region's average of 1.2 seconds, and is accompanied by multiple slip response amplification. Combined with the predicted viscoelastic delay response in the soft tissue model, this region has a displacement rebound delay during each gait, forming a tissue strain trap and a potential focus for future tissue fatigue and pressure injury. The system outputs a slip amplitude mapping, marking this region as a red high-slip area.Finally, in the structure optimization link, the system adjusts the prosthetic interface buffer layer structure according to the shear force migration main path and the stress concentration heat map, increases the original silicone layer thickness in the risk area from 2 mm to 4 mm, adds a directional microporous structure to disperse shear stress, replaces the material used in this area from a single thermoplastic elastomer to a composite double-layer buffer material (soft inner layer and moderate outer layer), and strengthens the carbon fiber support skeleton at the main force transmission path to optimize the overall stiffness continuous gradient. After the modified structure is simulated again, the maximum shear stress in the risk area is reduced by 41.6%, the slip response is controlled within 1.1 mm, the stress retention time is reduced to 1.9 seconds, the fatigue risk assessment score is increased from the original 0.28 to 0.81, and Mr. W's feedback on wearing comfort is significantly improved after wearing, and there is no skin stress reaction after continuous walking for 60 minutes.

[0049] By fusing the three types of anatomical structure information of residual limb skin, muscle and bone, and combining the individual biomechanical characteristics and electromyographic response behavior, a digital residual limb simulation system integrating anatomy, function and stress is established. The three-dimensional structure of the right lower leg stump bone of Mr. W is reconstructed by CT scanning, and the residual end surface of the tibia and the lateral shape of the fibula are accurately extracted as the internal bone support layer. At the same time, the MRI data is used to obtain the muscle layer thickness, arrangement direction and wrapping relationship, and the deep soft tissue layer around the skeleton is constructed. Then, the biomechanical parameters of different regions of soft tissue are measured by near-infrared optical elastography, including Young's modulus (range 18-56 kPa), viscoelasticity coefficient and shear delay response, etc. These parameters are input into the model as muscle layer attributes. The skin layer data is obtained by high-precision 3D structured light scanning, which is used to restore the skin surface profile, capillary distribution density and tissue thickness variation. The three types of structure data are reconstructed and topologically bonded by voxel-based superposition and B-spline body registration algorithm to construct a complete three-layer structure basic model. On the basis of this structure model, electromyographic data is further introduced to establish a functional response layer. The electromyographic data of the residual biceps femoris, semitendinosus and gastrocnemius muscle groups of Mr. W during walking and stage actions are recorded by an 8-channel surface electromyography instrument. The activation threshold, peak time delay and stability characteristics in the electromyographic waveform are extracted by the system, and these characteristics are converted into muscle group control ability map by the neural network analysis module, and then mapped with the muscle anatomy layer to construct the functional contribution distribution map of each muscle region under different actions. This distribution map is used as the basis for weight calibration of the subsequent force transmission area. In addition, the stress response behavior of each tissue layer micro unit is recorded synchronously during dynamic gait simulation. Through finite element calculation, the shear stress, compression strain and residual deformation values of each grid element in the periodic motion are obtained. Combined with the fatigue accumulation model of the tissue, a stress response distribution map is formed and superimposed on the structure-function model to construct a three-in-one digital twin model of the residual limb with anatomical geometric structure, neuromuscular control ability and mechanical response ability. This model has time domain evolution ability and can be periodically updated according to the user's usage habits and tissue changes. In the case of Mr. W, the model evaluates that the gastrocnemius muscle region of his residual limb has good force transmission ability and good electromyographic response, and suggests that this region be used as the main force transmission path in the optimization design. The medial tibia is not suitable for high contact pressure area layout due to obvious stress retention. Finally, the interface structure of the prosthesis evolves towards a coupling form consistent with force, nerve and anatomy, effectively improving the structure matching degree and physiological compatibility, ensuring the stability and safety of the prosthesis in long-term wear.

[0050] The coupling interface between the prosthetic and the residual limb is divided into multiple scales of contact space and functional sub-area modeling. The sub-area strategy is based on simulation data, which divides the entire contact interface into three sub-areas: structural adaptation area, force transmission area and energy buffer area. On this basis, structure and material collaborative optimization is carried out.The system constructs a complete three-dimensional coupling space between the personalized residual limb digital twin model and the prosthetic digital interface model established by Mr. W. In the preliminary dynamic simulation, by executing five typical gaits (including walking on flat ground, uphill, 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 degree, normal pressure stability, and geometric matching error of each micro-unit in the simulation cycle, the areas with good preliminary fit are marked as structure adaptation zones. For example, between the front edge of the prosthetic bottom and the main pressure-bearing surface of the tibial stump of Mr. W, the fit error is controlled within ±1.4 mm, the contact pressure fluctuation is less than 5%, and this area is included in the structure adaptation sub-zone, which mainly functions 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 direction of the residual limb myoelectric control, and identifies a main transmission channel 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 72 kPa, and the direction deviation angle is less than 12°, which is defined as the force transmission zone by the system. The design optimization of this zone focuses on improving response accuracy and reducing power dissipation. In the above simulation, several impact dissipation zones are also identified, mainly concentrated in the areas where the thickness of the soft tissue covering the residual limb is more than 8 mm. The shear stress in these areas changes frequently but does not form a clear concentration trend, and the thermal stress superposition is about 21% higher than that in other areas, indicating that these areas participate in a large amount of kinetic energy buffering and heat accumulation process. The system marks these areas as energy buffer zones. For example, for Mr. W, the energy buffer zone is mainly concentrated at the junction between the medial upper edge of the prosthesis and the outside of the popliteal fossa of the residual limb. It is recommended to use locally embedded composite porous energy-absorbing materials, and to design a separate buffer layer from the prosthesis structure. After adding the buffer material in the simulation, the shear force peak value of the same gait simulation decreases by 32.8%, and the regional temperature rise decreases by 1.6°C, proving the feasibility and effectiveness of the buffer partition. The system outputs a complete three-zone spatial distribution map and imports it into the interface structure CAD. The local gradient hardness material strategy is adopted, the structure adaptation zone uses medium-hard thermoplastic elastomer (Shore hardness 75A), the force transmission zone uses carbon fiber reinforced brackets integrated with conductive myoelectric interfaces, and the energy buffer zone introduces three-dimensional elastic mesh silicone and heat-insulating composite foam materials for protection. The overall interface structure is verified by re-simulation, and the shear force uniformity index increases from 0.67 to 0.89, the contact stability score improves by 17.2%, and the thermal stress distribution improves significantly. Subsequent walking distance of Mr. W wearing the new structure of the prosthesis is extended to about 1.8 times, without redness, pressure marks, and fatigue pain, fully verifying the engineering practicability, physiological safety, and user experience improvement ability of the multi-scale contact space division strategy, and providing a standardized design template for digital personalized customization of prosthetic interfaces.

[0051] To further quantify and optimize the stress influence and fatigue risk control of the prosthetic interface on the long-term use of the tissue, a shear force migration evolution function modeling method is used to perform micro-scale dynamic analysis on the coupling area between the prosthesis and the residual limb. A two-dimensional micro-unit grid is constructed on the contact interface between the residual limb and the prosthesis, the entire coupling surface is divided into 2146 micro-units (with an average unit length of 3.5 mm), and a shear force vector recording unit is established on each unit. In the simulation system, Mr. W completes a continuous 60-minute gait cycle, the shear stress change of each unit at each simulation time is collected, and the following newly constructed shear force migration evolution function is used to model and quantify the path trend:

[0052]

[0053] In the simulation platform, the following parameter intervals are specifically set to support instance calculation:

[0054] The shear stress change rate of this unit is taken from the simulation data, with a unit of kPa / s, and the observed change range in Mr. W's gait is 5.6-27.3 kPa / s;

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

[0056] The action impact conversion factor represents the participation intensity of this region in the action transmission chain, and is set to 0.3-1.2 based on the peak value of electromyogram and acceleration modulus length. Mr. W can reach more than 1.1 in the knee joint force peak interval.

[0057] Taking a micro-unit point A on one of the shear main paths (located on the inner lower edge of the tibia) as an example, its average δ is 18.7 kPa / s, Γ is 0.67, and Ξ is 0.94 within the first 20 minutes of simulation. After substituting the formula, the integral output is:

[0058]

[0059] The gradient of this value in the spatial direction is calculated, combined with the evolution function values of adjacent units, and the shear migration sensitivity field intensity of point A Ψ(A, t=1200s) ≈3650 (after normalization, the shear response intensity index is 3650) is obtained. It belongs to the high sensitivity shear force convergence zone (threshold value is 3000) set by the system; further, the system will calculate the shear force migration evolution function of each time period The trend record is recorded, if the local extreme value exists continuously and the area cannot release stress in the subsequent period (i.e. no more than 40% of the downward trend), mark the point as a stress retention point; and the corresponding pyrolysis decoupling parameter (calculated by the local temperature accumulation rate + tissue thermal conductivity ratio) is 0.78 in the simulation (the system threshold is 0.65), which means that the local area is in a state of stress and heat accumulation for a long time, and the system marks this point as a stress-heat coupling retention point; In the full contact surface microelement scanning, 19 such high-risk microareas are finally identified, most of which are concentrated in the interface fitting edge of the slope area of the lower edge of the residual limb and the front end of the prosthesis. Combined with the fatigue risk map, the system assigns a red label to this area and feeds back to the interface structure CAD model, suggesting the following design optimization measures: 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 three-layer foam (soft surface, energy-absorbing core, and heat-conducting middle), secondly, introduce micro-wavy texture to the contact surface in this area to diffuse shear force concentration, after optimization, the system re-simulates the same 60-minute action period, the shear force rate drops to 12.4kPa / s, the slip response drops to 1.1mm, The peak value drops to 2670, the risk level drops from high to medium, the identification color changes from red to orange, and the thermal decoupling parameter drops to 0.62, confirming that the tissue fatigue risk in this area is effectively controlled; After Mr. W wears the new interface, the clinical observation and user feedback confirm that the walking time can be increased from 30 minutes to more than 75 minutes, the residual limb has no fever, no swelling and pressure marks, and the muscle recovery state is stable.

[0060] The soft tissue slip response modeling module is further implemented to address skin abrasion, fatigue tear, and inflammation caused by relative displacement between tissue layers during long-term prosthetic wearing. In the optimization process, the tissue adhesion state and microstructure characteristics at different positions between the skin and muscle are obtained through magnetic resonance elastography (MRE) and photoacoustic tomography of Mr. W's right lower leg stump area. After data processing, the system establishes a biological slip tolerance curve for each microzone, constructs a skin-muscle interlayer slip threshold map, and sets the main slip threshold range of Mr. W's stump between 0.9 mm and 1.6 mm based on tissue adhesion density and friction factor. The gastrocnemius muscle outer edge tissue slip threshold is 0.95 mm, and the anterior tibial side slip threshold is increased to 1.45 mm due to the thin skin and tight fascia attachment. Then, using the multi-scale simulation function of the stump twin model and the prosthetic interface model, Mr. W performs multiple high shear stress action simulations, including walking on a gentle slope, climbing stairs, and deep squatting. Each action lasts for 10 minutes. During 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 microzone at the full contact interface, called the slip displacement distribution map. For example, in continuous uphill gait, the stump tissue at the contact point C corresponding to the prosthetic front edge area has a periodic slip peak of 2.1 mm, which is about 75% higher than the slip threshold (1.2 mm) of this area and has not been alleviated after 30 cycles. The system marks this point as a slip overload risk point. Further analysis shows that this point is a sub-channel intersection area on the shear stress migration path, where multiple weak shear forces converge. Although the pressure is low, the direction changes frequently and is prone to micro-tears. The system generates a slip amplitude map based on historical data, marks high slip areas as red, and reminds designers that the structure stiffness and friction characteristics at this point do not match and need to be adjusted and replaced. For this point, the system recommends optimization schemes including replacing the interface material in this area from thermoplastic polyurethane (TPU) to low-friction coefficient coated silicone, and introducing a microspherical elastic adsorption layer to absorb and buffer tissue slip, while increasing the variable microtexture on the surface of the prosthetic interface to disrupt the consistency of the slip path and inhibit repeated friction paths. The newly designed structure is reimported into the simulation platform to perform the same behavior simulation, and the slip peak at contact point C is reduced to 0.88 mm, which is below the slip threshold (1.2 mm) of this area, and the displacement fluctuation amplitude is reduced by 47%, and the periodic rebound phenomenon is significantly weakened. At the same time, the subjective feedback of the stump user indicates that the skin tingling sensation during fast walking after wearing disappears, and the pressure marks are significantly reduced, indicating that the soft tissue slip response modeling mechanism has high accuracy and practicality in microzone resolution, dynamic behavior mapping, and structure adjustment guidance, and can significantly improve the prosthetic wearing time and tissue health life.

[0061] Further, a multi-objective evolutionary algorithm-based joint optimization process is implemented to take the shear force trajectory, stress concentration distribution, and soft tissue slip response results as input parameters, comprehensively consider the coupling efficiency, physiological safety, and material adaptability of the prosthetic interface under complex dynamic behavior, and perform multi-dimensional collaborative optimization of the interface structure in terms of geometric shape, buffer layer thickness, and material stiffness distribution. The system models the shear force migration path obtained from the previous analysis as a path energy field, identifies the main channel and high shear stress concentration areas, and outputs a stress risk map in combination with a stress retention point identification mechanism, and generates a three-dimensional risk composite matrix R(x, y, z) by superimposing the slip amplitude map, which serves as the environmental mapping basis for the optimization function. Secondly, the system sets three main optimization objectives: ① minimizing the shear stress peak value per unit area of the contact area (objective function F1), ② controlling the thermal-mechanical coupling energy in all potential retention point areas to be below the upper limit of the tissue threshold (F2), and ③ ensuring that the maximum displacement in each high slip risk area is below its corresponding slip tolerance threshold (F3). Based on this, the joint objective function F = w1·F1 + w2·F2 + w3·F3 is constructed, where the weight factors w1 = 0.4, w2 = 0.35, and w3 = 0.25 are set according to the sensitivity of the residual limb tissue of Mr. Shi, and the NSGA-III multi-objective optimization algorithm is used to perform population evolution iteration. The initial population consists of 128 sets of parameter sets, covering three types of variables: interface geometric edge fitting curvature (range: ±10° adjustment), buffer layer thickness (range: 2mm-7mm), and material elastic modulus (range: 15-85kPa). The system performs action simulation and quickly evaluates the response of the above objective functions through the digital twin model at each generation, calculates the fitness function, and performs selection, crossover, and mutation. After 42 generations of evolution, the population converges, and the final output is a set of Pareto optimal solutions. In the optimization results of Mr. Shi, the optimal solution increases the buffer layer thickness in the middle area of the prosthetic front edge from 3mm to 5.5mm, changes the original homogeneous structure to a central high-elasticity area (modulus 45kPa) and a soft transition layer (modulus 28kPa) distributed on the edge, and slightly adjusts the geometric curvature of the front edge from 5.3° to 2.1° to reduce shear concentration. Simulation verification shows that the overall shear peak value decreases by 38.6%, the energy of high-risk retention points decreases by 31%, and the slip peak value is controlled within 0.92mm, meeting the safety threshold of all objective functions. The structure sample is manufactured through 3D printing and multi-material fusion process, and Mr. Shi completes a 72-hour continuous use test in the wearing experiment, with no skin redness, indentation, or stinging feedback, and the wearing stability score increases from 72 to 91 (full score 100).

[0062] Example 2:

[0063] In combination with the Figure 5, a method for assessing the risk distribution of tissue fatigue induced by prosthetic interface under long-term use is implemented, which considers the history of tissue deformation and the effect of heat energy accumulation, combines simulation analysis and structural optimization feedback mechanism, and dynamically models the risk of prosthetic interface and adjusts the material. In step S1, the team first reconstructs the three-dimensional model of the soft tissue of Mr. W's right lower leg based on the high-resolution MRI image data, and integrates the previously established geometric model of the prosthetic interface to complete the high-precision registration of the residual limb and the prosthetic interface in the digital twin platform, forming an interactive three-dimensional coupled model; then, dynamic gait simulation is performed on the model, including six types of common daily activities such as walking on flat ground, slow running uphill, and squatting up, each type of action is repeated 20 times, the total simulation time is 60 minutes, and the system records the three-dimensional deformation vector of each micro-unit point at each time during the simulation process, and generates deformation vector field data through algorithm; among them, in the uphill walking action of Mr. W, the residual limb tissue in the anterior tibial region shows a continuous displacement focusing trend, and the cumulative displacement of the deformation vector reaches 3.2mm, which is much higher than the upper limit of the average deformation of the soft tissue in this region, 1.9mm, and the system marks the trajectory aggregation area of this region as a potential fatigue induction point P1; entering step S2, the system further starts the thermal coupling modeling module, simulates the local temperature rise caused by friction and compression between the tissue and the interface during wearing, and combines the thermal conductivity of the tissue, the skin heat dissipation ability and the residual limb area blood flow data to establish a heat diffusion and retention model, the results show that after 42 minutes of simulation, the temperature of P1 region rises from the initial 36.5℃ to 39.2℃, with a temperature rise of 2.7℃, and remains above 38.7℃ in the next 18 minutes, the heat release efficiency is significantly lower than that of the surrounding area, and the system generates a fatigue risk spatial distribution map by superimposing the deformation trajectory graph and the thermal field response curve of the P1 region, where the point is marked as a red high-risk area; entering step S3, according to the system analysis results of the risk map, the team optimizes the structure and material configuration of the interface region where P1 is located, adjusts the thickness of the interface buffer structure, adjusts the original 3.5mm silicone layer to a composite structure: 5mm microporous thermal conductive silicone in the inner layer, high thermal conductive graphite sheet as a heat diffusion element in the middle layer, and an anti-shear polyurethane coating on the outer layer, and the corner curvature of the edge transition area of this region is optimized from 45° to 30° to reduce the local compression gradient, and the system re-simulates the data after the system re-simulates the data: the maximum displacement of the deformation focusing path in this region decreases to 1.8mm(reduced by 43.8%), the heat accumulation peak temperature decreases to 37.9℃(decreased by 1.3℃), and the heat retention time is shortened to less than 8 minutes; the risk distribution map changes from the original red risk to the orange controllable risk level, and the system risk score decreases 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 after wearing the optimized structure in the subsequent actual test, and the residual limb was not red, there was no stinging pain, and the local temperature rise was controlled within the normal temperature range, and the structure adaptation was significantly improved.

[0064] A residual limb micro-zone fatigue prediction mechanism based on deformation vector field analysis and thermal coupling relationship modeling is introduced. By tracking the deformation path of the tissue and the heat retention behavior in detail, the risk points of chronic pressure injury are found in advance, and the iterative optimization of the prosthesis structure is assisted. A micro-unit discrete grid is established in the contact area between the residual limb and the prosthesis of Mr. W, with a total area of about 148 cm 2, a total of 2124 quadrilateral micro-units are divided, with each unit having a side length controlled between 3-4 mm. The system performs five types of high-frequency lower limb motion simulation on Mr. W (respectively, uphill fast walking, standing up and down, deep squatting, stepping, and static turning) through digital twin driving platform. Each motion is repeated for 30 cycles, and the total simulation time is 96 minutes. During this process, the system continuously records the three-dimensional displacement vector of each micro-unit in each motion cycle and forms a time-continuous deformation vector field. For example, unit #0789 located at the lower edge of the tibia shows a consistent deviation in the backward and upward direction during uphill fast walking, with an average displacement amplitude of 2.3 mm and a direction angle change of less than ±8°, forming a clear linear deformation trajectory. The system performs trajectory path stability and repetition amplitude clustering on all grid units, and the results show that 16 areas form high-frequency repetitive deformation trajectories, with P3 area (the front edge of the prosthesis inside) as the representative. The average repetitive deviation trajectory of this area is 2.1 mm, with a fluctuation amplitude controlled within 0.4 mm and a direction consistency score of 0.92 (full score 1.0), marking it as a high-risk fatigue-inducing area. To evaluate the thermal stress coupling effect caused by mechanical deformation, a thermal-mechanical coupling model is further established in the twin model. This model considers parameters such as the contact friction coefficient between the skin and the prosthesis interface (0.43 for Mr. W's area), tissue thermal conductivity (0.37 W / m·K for skin and 0.21 W / m·K for fat layer), and the thermal conductivity of the prosthesis cushion material (0.15 W / m·K for EVA foam). Based on the unit time heat accumulation curve, the temperature response trajectory of the P3 area under continuous motion is analyzed through multi-cycle simulation. The simulation data shows that the temperature of the P3 area gradually rises from the initial 36.7°C to 39.1°C at the 45th minute, and remains above 38.6°C for more than 20 minutes in the subsequent cycles, indicating that this area has typical heat retention characteristics, which is highly consistent with the continuous deformation trajectory. The system fuses the deformation focus path map of the P3 area with the thermal accumulation and thermal force map to generate a fatigue-inducing risk superimposed map, with a risk index of 0.87 (threshold value of 0.65), marking it as a first-level warning area. For this area, optimization suggestions include replacing the original single-layer EVA material with a three-layer structure, i.e., a high-thermal-conductivity coating on the surface, a thermal-conductivity elastic material containing graphite particles in the middle layer (thermal conductivity 0.75 W / m·K), and a low-modulus silicone energy-absorbing layer at the bottom, and introducing a 0.5 mm concave buffer groove in the geometric structure to reduce strain concentration. The re-simulation after optimization shows that the maximum deformation amplitude of the P3 area decreases to 1.3 mm, the trajectory consistency fluctuation increases (the direction change range expands to ±14°), the deformation trajectory tends to be discrete, the peak temperature decreases to 37.8°C, the heat retention time is shortened to less than 7 minutes, and the risk index decreases to 0.41, effectively removing the fatigue high-risk level. Finally, Mr. W continuously uses the optimized structure for three days, with daily walking time exceeding 90 minutes, without any symptoms such as skin discoloration, blistering, or stinging, and the gait posture remains stable.

[0065] When the system completes the fatigue risk space distribution map generation and identifies the high-risk stress-thermal superposition area, a structure parameter adjustment mechanism based on risk map feedback is further implemented. The four key optimization dimensions, namely, buffer material distribution, regional elasticity control, contact texture design, and heat conduction performance optimization, are intervened one by one to form a multi-level and physiological response matching prosthetic interface optimization structure. According to the fatigue induction point P3 identified in the digital twin model simulation, its spatial positioning is located at the junction area of the prosthetic interface and the lower edge of Mr. W's tibia stump. The cumulative shear stress in this area frequently exceeds 80 kPa in continuous gait, and the heat retention time exceeds 25 minutes. The system marks this area as a high-risk first level. Therefore, in terms of buffer material distribution, the system replaces the original 3mm single-layer EVA foam structure with a multi-layer partition structure. The central area uses 5mm high-elasticity silicone material (modulus about 25kPa) as the main buffer and energy-absorbing layer, and the peripheral area gradually transitions to 3mm porous foam to achieve pressure relief. A dynamic thickness transition algorithm is used to ensure that the interface hardness gradient change is less than ±20%, so as to ensure the faultless 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, buffer, and soft-elasticity zones respectively. The P3 high-strain area is included in the soft-elasticity response zone, and the contact structure local deformation rate is set to 0.37-0.45 (adjusted dynamically according to the simulation feedback curve) to avoid stress retention due to insufficient local tissue rebounding ability. Thirdly, in terms of contact texture design, to enhance the shear stress diffusion ability and reduce the skin slip path repetition, the system introduces a random micro-texture array with a radius of 1.5mm in the fatigue risk area surface, with a distribution density controlled at 25 textures / cm 2, the texture depth is 0.4mm, the design can effectively break the continuous path of the shear main channel in the simulation, make the stress distribution more uniform, and the slip response drops by 32%; finally, in terms of thermal conductivity optimization, the system introduces a 0.3mm thick graphite thermal conductive sheet in the middle of the risk area, with a thermal conductivity of up to 8.5W / m·K, which can significantly improve the lateral diffusion speed of thermal energy, while applying a thermal conductive coating (such as boron nitride composite resin coating) to the surface of the original buffer structure, with an overall thermal conductivity of 0.78W / m·K, ensuring that heat is quickly dissipated to the prosthesis shell and avoiding local temperature rise of the residual limb; after optimization, the system re-executes a 72-minute gait simulation test on this area in the digital twin platform, and the data shows that the maximum shear stress in P3 area decreases to 54kPa, the peak temperature of heat accumulation decreases to 37.6℃, the tissue rebound rate improves by 16.4%, the slip control is within 0.91mm, the system fatigue risk index decreases from 0.87 to 0.36, and the risk level changes from red to green; in application, Mr. W wears the optimized interface for continuous three-day, 90-minute daily exercise tests, and the skin performance is stable, with no redness, blisters or pain phenomena, the symmetry of exercise posture is improved by about 14%, and the overall user satisfaction evaluation is improved to 94 points (full score 100).

[0066] The above shows and describes the basic principles, main features and advantages of the present application. Those skilled in the art should understand that the present application is not limited to the above embodiments, and the above embodiments and descriptions in the specification are only to illustrate the principles of the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the claimed present application. The scope of protection of the present application is defined by the appended claims and their equivalents.

Claims

1. A method for optimization of prosthetic and residual limb coupling based on digital twinning, characterized in that The method comprises the following steps: The anatomical structure characteristics of the residual limb including skin, soft tissue and bone are fused with the functional attributes including muscle group remaining control ability, electromyographic response and stress distribution data; a multi-layer digital residual limb relationship model is constructed; a functional residual mapping mechanism is introduced, which is based on electromyographic intensity spatial distribution and force transmission path inversion algorithm, evaluates the actual participation of each anatomical unit under different action scenarios and compares it with the mechanical contribution under the ideal state to obtain the functional residual value of each residual limb region; and a region label map is generated according to the coupling results of the functional residual value and the stress density distribution, marking the key structure zone mainly responsible for force transmission and the region suitable for nerve interface implantation or electrode arrangement; The multi-scale contact space of the residual limb and the prosthesis in dynamic behavior is defined as three parts: a structure adaptation zone, a force transmission zone and an energy buffer zone; In the simulation of dynamic interaction, the shear force migration trajectory, stress concentration evolution mode and soft tissue slip response between the units of the coupling surface are coupled; the physical partition relationship of the coordination domain is constructed to assist in optimizing the structure and material function gradient of the prosthesis interface; The wound area caused by the prosthesis in wearing and using is described, and the minimum tissue disturbance path is calculated to minimize the path of strain concentration or heat accumulation caused by the prosthesis to the tissue structure; Through deformation vector field analysis and thermal force coupling relationship modeling, the risk distribution of tissue fatigue induced by the prosthesis interface under long-term use is evaluated; The multi-scale contact space is divided into: The structure adaptation zone: makes the prosthesis geometrically fit the residual limb; The force transmission zone: used for transmitting the action intention of the residual limb to the prosthesis through force; The energy buffer zone: used for absorbing shear force and impact force and relieving tissue stress accumulation.

2. The digital twin-based optimization method of prosthetic and residual limb coupling according to claim 1, characterized in that An individualized residual limb digital twin model including the anatomical structure, functional attributes and stress response of skin, soft tissue and bone is constructed in the simulation of dynamic interaction; A multi-scale contact space composed of a structure adaptation zone, a force transmission zone and an energy buffer zone is established between the individualized residual limb digital twin model and the digital model of the prosthesis; based on dynamic behavior simulation, the contact space is divided into unit grids, and the shear force migration trajectory between the units is simulated; Based on the force change of the unit, the stress concentration evolution mode in the residual limb-prosthesis coupling region is identified; the soft tissue slip response of the residual limb-prosthesis contact interface is modeled, and a slip amplitude mapping graph is output; according to the shear force migration trajectory, the stress concentration evolution mode and the slip tolerance, the structure parameters and material layout of the prosthesis interface are optimized.

3. The digital twin-based optimization method of prosthetic and residual limb coupling according to claim 2, characterized in that The individualized residual limb digital twin model fuses the structural information of the skin, soft tissue and bone of the residual limb, combines the biomechanical properties and electromyographic response characteristics, and establishes a multi-layer simulation model of anatomy, function and stress.

4. The digital twin based optimization method of prosthetic and residuum coupling according to claim 2, characterized in that The simulation of the shear force migration trajectory is based on unit grid division, and the twin model is driven by time-series residual limb action data to continuously track the shear force migration trajectory in the contact space; the identification of the stress concentration evolution mode is based on a nonlinear tissue response model to monitor the stress retention and accumulation evolution of the unit region under dynamic behavior.

5. The digital twin based optimization method of prosthetic and residuum coupling according to claim 2, characterized in that The soft tissue slip response modeling includes: Establishing a slip threshold between the skin and the muscle layer; The relative displacement distribution of the contact interface under different motion states is analyzed; The output slip amplitude map is used to evaluate the matching degree of the tissue friction area and the interface structure.

6. The digital twin-based optimization method of prosthetic coupling with residual limb according to claim 1, characterized in that The auxiliary optimization process of the functionally graded structure and material of the prosthesis interface is based on the shear force migration trajectory, stress concentration evolution mode and soft tissue slip response result, and a multi-objective evolutionary algorithm is used to jointly optimize the geometric structure, buffer material thickness and stiffness distribution of the prosthesis interface.

7. The digital twin-based optimization method of prosthetic coupling with residual limb according to claim 2, characterized in that The evaluation of the risk distribution of tissue fatigue induced by the prosthesis interface under long-term use includes: S1, a three-dimensional coupling model of the soft tissue of the residual limb and the prosthesis interface is established in the digital twin model; the deformation vector field analysis of the residual limb interface area during the use of the prosthesis is carried out, the deformation concentrated track area is identified based on the repeated motion behavior of the user, and the potential fatigue inducing point is marked; S2, thermal coupling relationship modeling is carried out to simulate the accumulation and release process of heat in the tissue during the wearing of the prosthesis; the deformation track and the heat field change trend are fused to generate a tissue fatigue risk space distribution map; S3, according to the tissue fatigue risk space distribution map, the structure or material configuration of the prosthesis interface is optimized.

8. The digital twin-based optimization method of prosthesis-residual limb coupling as claimed in claim 7, characterized in that The deformation vector field analysis records the displacement direction and deformation amplitude of each grid element in the multi-cycle motion by establishing an element discrete grid in the contact area, and forms a continuous deformation track path; The deformation track path is used to identify the high-frequency repeated deformation area as the basis for local fatigue risk evaluation; the thermal coupling relationship modeling integrates the thermal friction of skin and prosthesis interface, the thermal conductivity performance of biological tissue and the heat conduction characteristics of prosthesis material, and obtains the heat accumulation area through multi-cycle simulation.

9. The digital twin-based optimization method of prosthetic and residuum coupling according to claim 8, characterized in that According to the tissue fatigue risk space distribution map, the structure or material configuration of the prosthesis interface is optimized, including buffer material distribution, regional elasticity control, contact texture design or heat conduction performance optimization.

Images