Biomechanical intelligent analysis and adjustment system and method for lower limbs

Abstract

This invention proposes a biomechanical intelligent analysis and adjustment system and method for both lower limbs, belonging to the field of sports modeling and physiotherapy technology. The system includes: a data acquisition unit comprising a fully enclosed foot and ankle support structure, a 3D scanner, and an infrared thermal imager; the fully enclosed foot and ankle support structure is equipped with a matrix pressure sensor, multiple sets of vibration sensors, an airbag adjustment component, and an independent lifting module; a data processing unit that receives multidimensional data collected by the data acquisition unit, preprocesses it, and outputs optimized 3D model data and standardized biomechanical data; an analysis and decision-making unit that receives the standardized biomechanical data output by the data processing unit, analyzes it to obtain the plantar pressure imbalance area, ankle joint force line deviation, and lower limb force line offset results, and generates a personalized adjustment plan; and a human-computer interaction unit that receives user feedback information and transmits it to the analysis and decision-making unit to achieve dynamic optimization of the plan. This invention can assist in improving lower limb discomfort from a biomechanical perspective.

Description

Technical Field

[0001] This invention belongs to the field of motion modeling and physiotherapy technology, and in particular relates to a biomechanical intelligent analysis and adjustment system and method for both lower limbs. Background Technology

[0002] With increasing health awareness and the prevalence of sedentary and standing habits, foot discomfort caused by lower limb fatigue, muscle tension, and biomechanical imbalance is becoming increasingly prominent. Lower limb massagers, as commonly used devices to alleviate these problems, are attracting growing attention for their technological development. Currently, most mainstream lower limb massagers on the market focus on fixed-mode mechanical massage, using preset roller rolling and airbag compression programs to massage the soles and calves. For example, the device for preventing deep vein thrombosis in the lower limbs proposed in Chinese invention patent application number CN202011287357. While these devices can achieve basic relaxation functions, they have significant limitations: their massage modes and intensities are mostly fixed parameters set by the factory, unable to be personalized according to different users' foot sizes, foot types (such as normal feet, flat feet, high arches), and lower limb stress states. This results in some users experiencing discomfort due to excessive massage intensity or insufficient intensity to achieve a relaxing effect, failing to meet diverse comfort needs.

[0003] From a structural design perspective, existing lower limb massagers generally have deficiencies in their coverage and support for the ankle joint. Most products focus only on massaging the soles of the feet or the lower part of the calves, failing to design a dedicated wrapping structure for the ankle joint—a crucial joint connecting the foot and leg. They rely solely on simple straps or open frames for restraint, failing to provide comprehensive wrapping and stable support for the ankle joint. As a vital node for biomechanical transmission in the lower limbs, imbalances in the ankle joint can easily lead to uneven pressure distribution in the foot and abnormal gait. Existing devices lack targeted care for the ankle joint, not only reducing the overall comfort of the massage but also failing to provide biomechanical assistance in improving lower limb discomfort, thus falling short of users' needs for "comprehensive lower limb comfort care."

[0004] In terms of intelligence and feedback adjustment capabilities, the current technology of dual lower limb massagers is still relatively rudimentary. For example, the Chinese invention patent application CN2023100591617 proposes a method and system for constructing a skeletal muscle model of the entire length of the dual lower limbs. Although some mid-to-high-end products have introduced pressure sensors to collect plantar pressure data, they can only trigger simple intensity switching based on a single pressure value, without combining multi-dimensional data such as three-dimensional foot morphology and ankle joint stress state for comprehensive analysis; moreover, they lack dynamic feedback optimization mechanisms, and cannot adjust the program according to the user's real-time physical sensations (such as local tenderness, changes in comfort) or changes in biomechanical parameters during the massage, making it difficult to continuously optimize the massage effect. At the same time, existing devices have not established a knowledge base related to clinical comfort thresholds and foot biomechanical standards, making it impossible to scientifically judge the rationality of the massage program, and it is difficult to meet the dual needs of "comfortable relaxation" and "biomechanical correction". The intelligence and accuracy of the technical solution urgently need to be improved. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention proposes a biomechanical intelligent analysis and adjustment system and method for both lower limbs.

[0006] In a first aspect of the invention, a dual lower limb biomechanical intelligent analysis and adjustment system is proposed, the system comprising a data acquisition unit, a data processing unit, an analysis and decision-making unit, and a human-computer interaction unit.

[0007] The data acquisition unit includes a fully enclosed ankle support structure, a 3D scanner, and an infrared thermal imager.

[0008] The fully enclosed ankle support structure is equipped with a matrix pressure sensor, multiple sets of vibration sensors, an airbag adjustment component, and an independent lifting module.

[0009] The data processing unit receives the multidimensional data collected by the data acquisition unit, performs preprocessing, and outputs optimized three-dimensional model data and standardized biomechanical data.

[0010] The analysis and decision-making unit receives the standardized biomechanical data output by the data processing unit, analyzes it to obtain the results of plantar pressure imbalance area, ankle joint force line deviation and lower limb force line offset, and generates a personalized adjustment plan.

[0011] The human-computer interaction unit receives user feedback information and transmits the feedback information to the analysis and decision-making unit to achieve dynamic optimization of the solution.

[0012] The 3D scanner is used to collect 3D point cloud data of both lower limbs to obtain a 3D model of both lower limbs;

[0013] The infrared thermal imager is used to collect infrared thermal imaging data of both lower limbs to locate bony landmarks.

[0014] The matrix pressure sensor is used to collect plantar pressure data;

[0015] The multiple sets of vibration sensors are used to collect ankle joint vibration feedback data;

[0016] The airbag adjustment assembly is used for ankle posture correction;

[0017] The independent lifting module is used to adjust the support height of both lower limbs.

[0018] The system also includes an execution adjustment module, which is communicatively connected to the analysis and decision-making unit;

[0019] The adjustment module includes an airbag control unit, a lifting control unit, and a vibration control unit;

[0020] The airbag control unit is used to control the inflation / deflation pressure of the airbag adjustment component inside the fully enclosed ankle support structure; the lifting control unit is used to control the lifting height of the independent lifting module at the bottom of the support structure; and the vibration control unit is used to control the mode parameters of the vibration signal output by the vibration sensor.

[0021] The airbag adjustment component of the fully enclosed ankle support structure adopts a zoned control design, dividing the airbags around the ankle joint into four independent control areas: medial, lateral, anterior, and posterior. Based on the direction of ankle joint force line deviation output by the analysis and decision unit, the inflation pressure of the corresponding area airbag is adjusted individually to achieve precise posture correction.

[0022] The human-computer interaction unit includes a touch screen display.

[0023] The data processing unit also includes a 3D modeling and visualization submodule, which converts the preprocessed 3D point cloud data of the lower limbs into a rotatable and scalable 3D visualization model, and marks the location of bony landmarks, the heat map of plantar pressure distribution, and the direction of ankle joint force line deviation on the model, and displays them in real time through the touch screen.

[0024] The analysis and decision-making unit is connected to a pre-trained large-scale biomechanical analysis model of the lower limbs and a clinical knowledge base;

[0025] The clinical knowledge base includes normal plantar pressure distribution maps for different age groups and weight ranges, ankle joint neutral position force line parameters, and standard values ​​of the angle between the force lines of the two lower limbs.

[0026] The normal biomechanical parameters in the clinical knowledge base are all derived from publicly available clinical literature and independent experimental verification.

[0027] In a second aspect of the invention, based on the system described in the first aspect, a biomechanical intelligent analysis and adjustment method for both lower limbs is also proposed, the method comprising:

[0028] Data acquisition phase: Perform 3D scanning of the user's lower limbs to obtain 3D point cloud data of the lower limbs to construct an initial 3D model; acquire infrared thermal imaging data of the lower limbs to identify and locate bony landmarks of the lower limbs; acquire plantar pressure distribution data, ankle joint movement and force feedback data;

[0029] Data preprocessing stage: The data obtained in the data acquisition stage is denoised, registered and meshed to optimize the initial 3D model to obtain a high-precision 3D visualization model of both lower limbs; the infrared thermal imaging data is calibrated for temperature and segmented to extract the 3D coordinate information of bony landmarks; the plantar pressure distribution data and ankle joint vibration feedback data are filtered and normalized to eliminate environmental interference signals.

[0030] Biomechanical analysis stage: The pre-processed plantar pressure distribution data, ankle joint vibration feedback data and bony landmark coordinate information are input into the pre-trained biomechanical analysis model of the two lower limbs to analyze and obtain the user's plantar pressure imbalance area, ankle joint force line deviation angle and overall force line offset of the two lower limbs.

[0031] Personalized adjustment stage: Based on the analysis results output by the large biomechanical analysis model of the lower limbs, a personalized plan is generated, which includes plantar pressure adjustment strategy, ankle posture correction strategy and lower limb support height adjustment strategy.

[0032] Feedback and optimization phase: During the execution of the personalized plan, plantar pressure data, ankle joint motion data, and somatosensory feedback information are collected in real time; the real-time collected data are compared with the target parameters output by the large model, and the deviation value is calculated; the airbag pressure parameters, lifting height parameters, and vibration mode parameters in the personalized plan are dynamically adjusted according to the deviation value until the biomechanical parameters of both lower limbs reach the normal range in the clinical knowledge base.

[0033] The large-scale biomechanical analysis model for both lower limbs is trained using multi-source lower limb biomechanical sample data and has the ability to perform pressure-force line correlation analysis and identify abnormal parameters.

[0034] The feedback optimization phase also includes storing the user's biomechanical data, adjustment parameters, and somatosensory feedback information after each adjustment into a database, which serves as incremental training data for the large-scale biomechanical analysis model of the lower limbs, thereby improving the accuracy of subsequent analysis and adjustments.

[0035] The second aspect describes a method for intelligent biomechanical analysis and adjustment of the lower limbs, in which some or all of the steps can be automated through various forms of electronic devices and computer program instructions; the computer program instructions can be stored in different forms of storage media and loaded into computer electronic devices for execution.

[0036] Therefore, in a third aspect of the present invention, a computer-readable storage medium is also provided for storing computer instructions that, when executed on an electronic device, cause the electronic device to perform the intelligent biomechanical analysis and adjustment method for both lower limbs described in the second aspect.

[0037] The technical solution of this invention uses a fully enclosed ankle support structure to achieve zoned correction of the ankle joint. It combines 3D scanning and infrared thermal imaging to acquire multi-source data, and uses a large model to analyze and locate pressure imbalance and force line deviation, generating a personalized solution to solve the problem of poor compatibility of existing equipment. During the implementation, pressure, vibration and user feedback data are collected in real time during the massage. The airbag, lifting and vibration parameters are dynamically adjusted by comparing with the target parameters. The data is also used for incremental training of the large model to continuously improve the accuracy of adjustment, breaking through the limitations of static massage. The lower limb model and pressure heat map can be visualized through 3D modeling. Combined with a clinical knowledge base containing multi-dimensional standards, it takes into account intuitive interaction and scientific analysis, balancing the needs of comfortable massage and biomechanical correction.

[0038] Further advantages of the present invention will be further detailed in the Specific Embodiments section in conjunction with the accompanying drawings. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 This is a schematic diagram of the internal functional units of a biomechanical intelligent analysis and adjustment system for both lower limbs according to an embodiment of the present invention.

[0041] Figure 2 This is a schematic diagram of the data acquisition unit in the intelligent biomechanical analysis and adjustment system for both lower limbs of the present invention.

[0042] Figure 3 This is a flowchart illustrating an embodiment of the present invention of a biomechanical intelligent analysis and adjustment method for both lower limbs. Detailed Implementation

[0043] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments identical to those described in this application. Rather, they are merely examples of apparatuses and methods identical to some aspects of this application as detailed in the appended claims.

[0044] In specific embodiments of this application, if user-related data is involved, user permission or consent must be obtained when the embodiments of this application are applied to specific products or technologies, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0045] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0046] First see Figure 1 , Figure 1 This is a schematic diagram of the internal functional units of a biomechanical intelligent analysis and adjustment system for both lower limbs according to an embodiment of the present invention.

[0047] exist Figure 1 The diagram shows that the system includes a data acquisition unit, a data processing unit, an analysis and decision-making unit, and a human-computer interaction unit.

[0048] The following sections will describe each of the aforementioned functional units in detail. For ease of understanding, specific hardware structure examples (including hardware model, software version, etc.) will be provided for some functional units during the description. However, it is understood that these limitations are merely illustrative, and other software or hardware products capable of performing the same function can also constitute corresponding functional units. This invention does not impose any specific limitations in this regard.

[0049] The data acquisition unit includes a fully enclosed ankle support structure, a 3D scanner, and an infrared thermal imager.

[0050] The fully enclosed ankle support structure is equipped with a matrix pressure sensor, multiple vibration sensors, an airbag adjustment component, and an independent lifting module.

[0051] The fully enclosed ankle support structure is equipped with a matrix pressure sensor, multiple sets of vibration sensors, an airbag adjustment component, and an independent lifting module.

[0052] Preferably, the airbag adjustment component of the fully enclosed ankle support structure adopts a zoned control design, dividing the airbags around the ankle joint into four independent control areas: medial, lateral, anterior, and posterior. Based on the direction of ankle joint force line deviation output by the analysis and decision unit, the inflation pressure of the corresponding area airbag is adjusted individually to achieve precise posture correction.

[0053] In a more specific embodiment, the four independent airbag areas (inner / outer / front / rear) are all made of TPU material (0.1mm thick, aging resistant and with good airtightness). The thickness of a single airbag is 3mm when uninflated and 15mm when inflated. The airbags are detachably fixed to the inner side of the ring frame with medical-grade Velcro for easy cleaning and replacement. Each airbag is connected to an independent micro air nozzle (2mm in diameter), which is connected to the air pump of the airbag control unit through a PU air tube (1mm inner diameter). The air tube is 15cm long to prevent it from being pulled off.

[0054] Preferably, the ring frame of the fully enclosed ankle support structure is made of medical-grade ABS material (hardness 75 Shore D, with impact resistance and biocompatibility), with an overall height of 8cm (covering the key stress area of ​​the ankle joint "from the top of the circular standing base to the top of the frame, 2cm above the ankle bone"); the inner diameter of the frame can be continuously adjusted from 20-30cm via a side adjustment knob (suitable for the common range of 18-28cm ankle circumference for adult males / females), with an adjustment accuracy of ±0.5cm; the inner side of the frame is provided with a 0.5cm thick medical sponge cushioning layer to avoid discomfort caused by direct contact between the frame and the skin.

[0055] See the analysis. Figure 2 , Figure 2 A detailed schematic diagram of the data acquisition unit in the intelligent biomechanical analysis and adjustment system for both lower limbs of the present invention is shown.

[0056] The 3D scanner is used to collect 3D point cloud data of both lower limbs to obtain a 3D model of both lower limbs;

[0057] The infrared thermal imager is used to collect infrared thermal imaging data of both lower limbs to locate bony landmarks.

[0058] The matrix pressure sensor is used to collect plantar pressure data;

[0059] The multiple sets of vibration sensors are used to collect ankle joint vibration feedback data;

[0060] The airbag adjustment assembly is used for ankle posture correction;

[0061] The independent lifting module is used to adjust the support height of both lower limbs.

[0062] Preferably, the independent lifting module supports independent lifting adjustment in zones, meaning that both sides can be lifted and lowered independently, and the height of the left and right limb supports can be adjusted separately.

[0063] Preferably, the sole contact surface is made of medical-grade silicone material with a Shore A hardness of 40 (combining support and resilience, compression deformation rate ≤15%), with a thickness of 4cm (±0.2cm). The silicone interior is divided into three support zones along the forefoot (1-5 metatarsal region), midfoot (arch region), and hindfoot (calcaneal region). Each zone contains a 20×20 array of matrix pressure sensors (model FSR402, sampling accuracy 0.1kPa, acquisition frequency 100Hz), with a sensor spacing of 5mm to ensure no pressure acquisition blind spots.

[0064] Based on this, return to Figure 1 The data processing unit receives the multidimensional data collected by the data acquisition unit, performs preprocessing, and outputs optimized three-dimensional model data and standardized biomechanical data.

[0065] The analysis and decision-making unit receives the standardized biomechanical data output by the data processing unit, analyzes it to obtain the results of plantar pressure imbalance area, ankle joint force line deviation and lower limb force line offset, and generates a personalized adjustment plan.

[0066] The human-computer interaction unit receives user feedback information and transmits the feedback information to the analysis and decision-making unit to achieve dynamic optimization of the solution.

[0067] Preferably, the system further includes an execution adjustment module, which is communicatively connected to the analysis and decision-making unit;

[0068] The adjustment module includes an airbag control unit, a lifting control unit, and a vibration control unit;

[0069] The airbag control unit is used to control the inflation / deflation pressure of the airbag adjustment component inside the fully enclosed ankle support structure; the lifting control unit is used to control the lifting height of the independent lifting module at the bottom of the support structure; and the vibration control unit is used to control the mode parameters of the vibration signal output by the vibration sensor.

[0070] The human-computer interaction unit includes a touch screen display.

[0071] The data processing unit also includes a 3D modeling and visualization submodule, which converts the preprocessed 3D point cloud data of the lower limbs into a rotatable and scalable 3D visualization model, and marks the location of bony landmarks, the heat map of plantar pressure distribution, and the direction of ankle joint force line deviation on the model, and displays them in real time through the touch screen.

[0072] The analysis and decision-making unit is connected to a pre-trained large-scale biomechanical analysis model of the lower limbs and a clinical knowledge base;

[0073] The clinical knowledge base includes normal plantar pressure distribution maps for different age groups and weight ranges, ankle joint neutral position force line parameters, and standard values ​​of the angle between the force lines of the two lower limbs.

[0074] In a specific example, a 3D scanner (preferably a Time-of-Flight (ToF) 3D scanner, with the Intel RealSense D455 hardware model as a reference) acquires millions of 3D coordinate points in real time by emitting infrared light signals to the lower limbs and receiving reflected signals, forming 3D point cloud data of the lower limbs (from the hip joint to the sole of the foot, including the ankle joint). This data contains geometric morphological details of the lower limbs, such as arch height, ankle inversion / vaulting curvature, differences in lower limb length, and calf thickness, which can be directly used to construct an initial 3D model of the lower limbs, providing intuitive geometric evidence for subsequent analysis of "whether the force lines of the lower limbs are symmetrical" and "whether the ankle joint posture is abnormal."

[0075] After receiving the raw point cloud data from the 3D scanner, the data processing unit performs preprocessing operations such as denoising (eliminating noise points caused by ambient light interference), multi-view registration (merging point clouds scanned from different angles to avoid occlusion blind spots), and triangular mesh generation (converting discrete point clouds into continuous 3D model surfaces), ultimately generating a high-precision 3D visualization model of both lower limbs. Simultaneously, the data processing unit extracts standardized geometric parameters (such as the difference in lower limb length, arch height, and ankle joint center coordinates) from the optimized model, transforming them into structured data recognizable by the analysis and decision-making unit, providing the "geometric dimension" input for biomechanical analysis.

[0076] After receiving the standardized geometric data output by the data processing unit, the analysis and decision-making unit combines it with the "geometric standard parameters of the lower limbs corresponding to different age groups and weight ranges" in the clinical knowledge base (such as the normal threshold for the difference in lower limb length in adult males ≤2mm, and the normal range for arch height 18-25mm) to determine whether the user has a deviation in the force line of the lower limbs (such as force line asymmetry caused by leg length discrepancy) or abnormal ankle joint geometric posture (such as ankle inversion caused by arch collapse). For example, if the 3D model shows that the user's left lower limb is 3mm shorter than the right, the analysis and decision-making unit will use this parameter as the core basis for the "lower limb support height adjustment strategy" to guide the lifting control unit of the adjustment module to raise the left independent lifting module to achieve force line balance; at the same time, the 3D visualization model will also be displayed on the touch screen of the human-computer interaction unit to help the user intuitively understand the force line deviation problem.

[0077] Infrared thermal imagers (such as the preferred high-sensitivity infrared thermal imager, hardware model FLIRE54) can accurately identify and locate key bony landmarks in the lower limbs by capturing the differences in thermal radiation between different tissues (bone structures exhibit significant temperature contrasts in thermal images due to their different thermal conductivity compared to soft tissues). These landmarks include the femoral condyle (at the knee joint), tibial plateau (above the ankle joint), apex of the medial and lateral malleoli (at the ankle joint), and plantar fascia attachment points (bone prominences on the sole of the foot). These bony landmarks serve as the "anatomical benchmarks" for the biomechanical alignment of the lower limbs. For example, the "neutral ankle alignment" requires the line connecting the apex of the medial and lateral malleoli as a benchmark to determine its deviation angle from the vertical direction; the "overall alignment of the lower limbs" requires the line connecting the femoral condyle, tibial plateau, and the center of the ankle joint as a benchmark to determine whether there is bending or deviation.

[0078] After receiving thermal imaging data from the infrared thermal imager, the data processing unit performs operations such as temperature calibration (eliminating the interference of ambient temperature on the thermal image), threshold segmentation (separating bony structures from soft tissue areas based on temperature differences), and coordinate matching (converting the two-dimensional thermal coordinates of bony markers into three-dimensional spatial coordinates). Ultimately, it outputs precise three-dimensional coordinate information for each bony marker. Simultaneously, the data processing unit performs coordinate alignment with the three-dimensional model of the lower limbs generated by the 3D scanner, marking the specific locations of the bony markers on the 3D model (e.g., marking the apex of the medial and lateral malleoli with red dots). This establishes a connection between the "geometric model" and the "anatomical structure," allowing subsequent analysis to move beyond geometric morphology and be based on real anatomical benchmarks.

[0079] After receiving the three-dimensional coordinates of the bony landmarks from the data processing unit, the analysis and decision-making unit calculates key biomechanical parameters using algorithms. For example, it calculates the ankle joint alignment deviation angle (normal threshold ±3°) based on the angle between the line connecting the apex of the medial and lateral malleoli and the vertical direction; and it calculates the bilateral lower limb alignment angle (normal threshold 0°-2°) based on the line connecting the femoral condyle, tibial plateau, and ankle joint center. These quantitative parameters are compared with standard values ​​in the clinical knowledge base. If an ankle inversion deviation of 5° is detected, the analysis and decision-making unit accurately locates "abnormal medial ankle force" and guides the airbag control unit of the adjustment module to "increase the pressure of the lateral ankle airbag and decrease the pressure of the medial ankle airbag" to achieve targeted posture correction. At the same time, the positional information of the bony landmarks is also used to verify the accuracy of the three-dimensional model, avoiding analytical deviations caused by geometric model errors and ensuring the anatomical rationality of the personalized adjustment plan.

[0080] In the technical solution of this invention, a 3D scanner and an infrared thermal imager work together to provide "dual-dimensional" data support for subsequent units: the 3D scanner provides "geometric morphological data" (solving the question of "what do the lower limbs look like"); the infrared thermal imager provides "anatomical benchmark data" (solving the question of "where are the key force points"). After preprocessing by the data processing unit, both form a "precise 3D model with anatomical markings," which provides both a "geometric carrier for force line analysis" and an "anatomical basis for deviation judgment" for the analysis and decision-making unit. Ultimately, this ensures that the personalized solutions generated by the analysis and decision-making unit conform to both geometric morphological adaptability and anatomical principles, while also making the "visual modeling" of the data processing unit and the "intuitive display" of the human-computer interaction unit more clinically valuable.

[0081] Based on the system implementation, the method implementation of the present invention will be described next. See [link to relevant documentation]. Figure 3 , Figure 3 The diagram shows a flowchart of a biomechanical intelligent analysis and adjustment method for both lower limbs according to an embodiment of the present invention.

[0082] exist Figure 3 The paper illustrates a biomechanical intelligent analysis and adjustment method for both lower limbs, comprising a data acquisition stage, a data preprocessing stage, a biomechanical analysis stage, a personalized adjustment stage, and a feedback optimization stage. The specific execution principles of each stage are as follows:

[0083] Data acquisition phase: Perform 3D scanning of the user's lower limbs to obtain 3D point cloud data of the lower limbs to construct an initial 3D model; acquire infrared thermal imaging data of the lower limbs to identify and locate bony landmarks of the lower limbs; acquire plantar pressure distribution data, ankle joint movement and force feedback data;

[0084] Data preprocessing stage: The data obtained in the data acquisition stage is denoised, registered and meshed to optimize the initial 3D model to obtain a high-precision 3D visualization model of both lower limbs; the infrared thermal imaging data is calibrated for temperature and segmented to extract the 3D coordinate information of bony landmarks; the plantar pressure distribution data and ankle joint vibration feedback data are filtered and normalized to eliminate environmental interference signals.

[0085] Biomechanical analysis stage: The pre-processed plantar pressure distribution data, ankle joint vibration feedback data and bony landmark coordinate information are input into the pre-trained biomechanical analysis model of the two lower limbs to analyze and obtain the user's plantar pressure imbalance area, ankle joint force line deviation angle and overall force line offset of the two lower limbs.

[0086] Personalized adjustment stage: Based on the analysis results output by the large biomechanical analysis model of the lower limbs, a personalized plan is generated, which includes plantar pressure adjustment strategy, ankle posture correction strategy and lower limb support height adjustment strategy.

[0087] Feedback and optimization phase: During the execution of the personalized plan, plantar pressure data, ankle joint motion data, and somatosensory feedback information are collected in real time; the real-time collected data are compared with the target parameters output by the large model, and the deviation value is calculated; the airbag pressure parameters, lifting height parameters, and vibration mode parameters in the personalized plan are dynamically adjusted according to the deviation value until the biomechanical parameters of both lower limbs reach the normal range in the clinical knowledge base.

[0088] The feedback optimization phase also includes storing the user's biomechanical data, adjustment parameters, and somatosensory feedback information after each adjustment into a database, which serves as incremental training data for the large-scale biomechanical analysis model of the lower limbs, thereby improving the accuracy of subsequent analysis and adjustments.

[0089] It is understood that the specific hardware basis for implementing the method embodiments is the aforementioned system embodiments, and the two correspond to each other. To highlight the key steps of the method embodiments, specific numerical examples (states, parameter values) will be used to elaborate on each of the above execution stages when introducing the method embodiments, as follows:

[0090] Data collection phase: Taking a 42-year-old female user, weighing 65kg, complaining of soreness and swelling in both feet and occasional dull pain in the right ankle after prolonged standing as an example, the data collection phase involves multi-device collaboration to capture comprehensive data. The specific operation is as follows:

[0091] 1. 3D Scanning Data Acquisition: An Intel RealSense D455 3D scanner (ToF technology, resolution 1280×720, frame rate 15fps) was used. The user stood in the scanning area (feet shoulder-width apart, knees slightly bent). The scanner scanned the user's hip joint to the sole of the foot from three angles: front, side, and back. The scan took about 30 seconds and acquired 3D point cloud data containing the geometry of both lower limbs. Key data included: right arch height 17mm, left arch height 21mm, and the difference in length between the two lower limbs 2.5mm.

[0092] 2. Infrared Thermal Imaging Data Acquisition: Using an FLIRE54 infrared thermal imager (thermal sensitivity ≤0.05℃, detection range -20℃ to 150℃), images were captured at a distance of 50cm from the user's lower limbs to obtain thermal radiation images of both lower limbs. Due to the difference in thermal conductivity between bony structures (such as the medial and lateral malleoli and tibial plateau) and surrounding soft tissues, a significant temperature contrast was observed in the thermal images (the temperature of bony structures was 1-2℃ lower than that of soft tissues). Based on this, six key bony landmarks were located: the apex of the right medial and lateral malleoli, the apex of the left medial and lateral malleoli, and the femoral condyles of both knee joints.

[0093] 3. Foot and Ankle Joint Data Acquisition: The user's feet are placed under a fully enclosed foot and ankle support structure (the sole contact surface is made of 4cm thick flexible silicone). A matrix of 20×20 pressure sensors (sampling accuracy 0.1kPa, acquisition frequency 100Hz) inside the structure collects real-time foot pressure distribution data, displaying a pressure value of 410kPa on the medial side of the right foot and 320kPa on the medial side of the left foot. At the same time, four sets of vibration sensors (frequency response 20-200Hz) located on the medial and lateral sides of the ankle joint and the arch of the foot collect static force feedback data of the ankle joint, displaying a vibration amplitude of 0.7mm on the right ankle joint (higher than 0.4mm on the left), indicating poor stability of the right ankle joint.

[0094] Data preprocessing stage: For the raw data collected above, the data processing unit (equipped with an Intel Core i7-12700H processor and running MATLAB R2023a data processing software) performs multi-step preprocessing to output standardized data. The specific process is as follows:

[0095] 1. 3D point cloud data preprocessing: First, a Gaussian filtering algorithm (standard deviation σ=0.8) is used to remove noise points (such as speckles caused by dust reflections in the environment) from the point cloud data. Then, the ICP (Iterative Closest Point) algorithm is used to complete multi-view point cloud registration (eliminating stitching errors from scanning at different angles). Finally, the Poisson surface reconstruction algorithm is used for mesh generation (generating a triangular mesh with a density of 1000 points / cm). 2 The model surface was optimized to obtain a high-precision three-dimensional visualization model of both lower limbs, which can clearly show the geometric features of right foot arch collapse and subtle differences in the length of the two lower limbs;

[0096] 2. Infrared thermal imaging data preprocessing: First, temperature calibration is performed (based on an ambient temperature of 25℃, correcting for system errors in the thermal imager). Then, the Otsu threshold segmentation algorithm (threshold set to 30℃) is used to separate the bony structure region (temperature < 30℃) from the soft tissue region (temperature ≥ 30℃). Subsequently, a coordinate transformation algorithm is used (mapping the two-dimensional pixel coordinates of the thermal image to three-dimensional spatial coordinates) to extract the three-dimensional coordinates of 6 bony landmarks. For example, the coordinates of the right medial malleolus apex are (X: 125.3mm, Y: 482.1mm, Z: 89.6mm), and the coordinates of the left medial malleolus apex are (X: 123.1mm, Y: 484.6mm, Z: 88.9mm).

[0097] 3. Pressure and vibration data preprocessing: Plantar pressure data is filtered by mean (window size 5) to eliminate high-frequency interference (such as pressure fluctuations caused by slight shaking of the user), and then normalized by normalization algorithm (mapping pressure values ​​to the 0-1 range) to obtain standardized pressure data; ankle joint vibration feedback data is filtered by Butterworth low-pass filter (cutoff frequency 50Hz) to filter environmental noise, and is also normalized. The final output standardized data can be directly used for subsequent biomechanical analysis.

[0098] Biomechanical analysis phase: The analysis decision unit calls upon a pre-trained large-scale lower limb biomechanical analysis model (based on the Transformer architecture, with a training sample size of over 100,000, including biomechanical data for different age groups and foot types), and combines it with a clinical knowledge base (storing normal parameters for individuals aged 18-65 years and weighing 40-100kg) to analyze the preprocessed data. The specific process and results are as follows:

[0099] 1. Model input data: Standardized plantar pressure data, ankle joint vibration feedback data, and three-dimensional coordinate information of 6 bony landmarks are input into the large model;

[0100] 2. Analysis of pressure imbalance area: The large model compared the normal plantar pressure range (forefoot 200-300kPa, midfoot 100-180kPa, hindfoot 250-350kPa) of a "42-year-old, 65kg female" in the clinical knowledge base. It was found that the pressure value of the right hindfoot medial side (standardized to 0.82) exceeded the normal threshold (0.7), and was judged as "right hindfoot medial side pressure imbalance area", and the degree of imbalance was "moderate" (17% exceeding the threshold).

[0101] 3. Ankle joint alignment deviation analysis: Based on the coordinates of bony landmarks, the ankle joint alignment (the angle between the line connecting the inner and outer ankle apexes and the vertical direction) was calculated. The angle of the right ankle joint was 4.2° (normal threshold ±3°), which was determined to be "mild inversion deviation of the right ankle joint"; the angle of the left ankle joint was 1.8°, which is within the normal range.

[0102] 4. Analysis of lower limb force line deviation: The length of the two lower limbs (vertical distance from the center of the femoral head to the sole of the foot) was calculated by using the coordinates of bony landmarks. The length of the right side was 92.3cm and the length of the left side was 92.55cm, with a length difference of 2.5mm (normal threshold ≤2mm). This was determined to be "slight length deviation of the two lower limbs", and the balance force line needs to be adjusted by adjusting the support height.

[0103] 5. Output Analysis Results: The large model generates an analysis report, which identifies the core issues as "excessive pressure on the right hindfoot, right ankle inversion, and slight length difference between the two lower limbs," providing a basis for personalized adjustments.

[0104] Personalized adjustment phase: Based on the above analysis results, the adjustment module (including the airbag control unit, lifting control unit, and vibration control unit) generates and executes a personalized adjustment plan. The specific operations are as follows:

[0105] 1. Ankle posture correction strategy implementation: The airbag control unit (using a miniature air pump with pressure control accuracy of ±0.5kPa) controls the inflation of the lateral airbag of the fully enclosed support structure to 22kPa and the medial airbag to 16kPa for the right ankle inversion deviation (correcting the inversion posture through the pressure difference between the lateral and medial sides); the airbag pressure of the left ankle is set to 19kPa (normal neutral position pressure) to maintain a stable posture;

[0106] 2. Execution of the dual lower limb support height adjustment strategy: The lifting control unit (driven by a stepper motor, with a position accuracy of ±0.1mm) controls the right independent lifting module to rise by 2.5mm based on the 2.5mm difference in length between the two lower limbs, while the left side maintains its initial height (30mm), thus balancing the force line of the two lower limbs and avoiding additional force caused by the length difference;

[0107] 3. Foot pressure adjustment strategy execution: The vibration control unit (piezoelectric ceramic vibration generator) controls the vibration sensors around the medial side of the right hindfoot to output a 60Hz vibration signal with an amplitude of 0.5mm (to guide the user to subconsciously adjust the pressure on the foot through vibration), while adjusting the support stiffness of the flexible material under the medial side of the right hindfoot (from 50 Shore A to 42 Shore A) to achieve pressure diversion and reduce local pressure;

[0108] 4. Adjustment process monitoring: During the adjustment process, the plantar pressure and airbag pressure data are collected in real time to ensure that the parameters operate according to the set values. For example, the pressure of the right lateral airbag is stable at 22kPa±0.3kPa, and the height error is ≤0.1mm.

[0109] Feedback and optimization phase: During the implementation of the adjustment plan, the plan is continuously optimized through a closed-loop logic of "real-time data collection - deviation calculation - dynamic optimization". The specific process is as follows:

[0110] 1. Real-time data acquisition: After 10 minutes of adjustment, plantar pressure data (the pressure on the medial side of the right hindfoot decreased to 352 kPa) and ankle vibration data (the amplitude on the right side decreased to 0.5 mm) were collected again. At the same time, the user inputs haptic feedback through the touch screen of the human-computer interaction unit: "The soreness in the right plantar is relieved, but there is a slight pressure on the outside of the ankle joint." The comfort score is 7 points (out of 10, target ≥ 8 points).

[0111] 2. Deviation calculation: The real-time data was compared with the target parameters output by the large model (right hindfoot pressure target 320-350kPa, ankle joint amplitude target 0.3-0.5mm, comfort target ≥8 points). The right hindfoot pressure was 352kPa (exceeding the upper limit of the target by 0.6%, close to normal). The pressure sensation on the outside of the ankle joint indicated that the airbag pressure needed to be finely adjusted.

[0112] 3. Dynamic optimization and adjustment: Based on the deviation, the airbag control unit reduces the airbag pressure on the outer side of the right ankle joint from 22 kPa to 20 kPa, while maintaining it at 16 kPa on the inner side; the vibration control unit reduces the vibration frequency from 60 Hz to 55 Hz, while keeping the amplitude at 0.5 mm.

[0113] 4. Final Validation and Data Storage: After 5 minutes of optimization and adjustment, the pressure on the medial side of the right hindfoot decreased to 342 kPa (within the normal range), the ankle joint vibration amplitude was 0.4 mm, the user comfort score improved to 8.5 points, and the biomechanical parameters of both lower limbs reached the normal range of the clinical knowledge base, indicating that the adjustment was complete. Simultaneously, the system stored the biomechanical data (such as pressure change curves), adjustment parameters (airbag pressure, vibration frequency), and user feedback (comfort score) of this adjustment in the database as incremental training data for the large model, improving the accuracy of subsequent adjustments for users with "ankle inversion + excessive plantar pressure" type.

[0114] In a specific training scenario, the training sample library of the large-scale lower limb biomechanical analysis model contains 120,000 valid samples. The samples are sourced from clinical data of orthopedic / rehabilitation departments of multiple tertiary hospitals and rehabilitation institutions (the acquisition and processing of related data must comply with relevant laws and regulations). The sample coverage is as follows:

[0115] Age range: 18-65 years old, divided into 10 intervals of 5 years each, with a sample size of ≥12,000 for each interval;

[0116] Weight range: 40-100kg, with each range in 10kg increments (6 ranges in total), and each range having a sample size of ≥20,000.

[0117] Foot type: normal foot (60%), flat foot (25%), high arch foot (15%). Each foot type includes samples of healthy and pathological conditions (such as plantar fasciitis, ankle instability).

[0118] Sample dimensions: Each sample includes three-dimensional point cloud data (≥1 million coordinate points), infrared thermal imaging bone marker coordinates (6 key marker points), plantar pressure distribution matrix (400 pressure values), ankle joint vibration feedback data (10 seconds duration, 100Hz sampling rate), and clinical diagnostic conclusions.

[0119] The model architecture is based on the Transformer-Base architecture, with an input layer dimension of 512 (integrating standardized geometric data, pressure data, and vibration data), 6 hidden layers, 8 attention heads, and a 3-class output layer (pressure imbalance region, force line deviation angle, and length difference).

[0120] Training environment: NVIDIA A100 GPU, batch size of 64, initial learning rate of 1e-4, cosine annealing learning rate scheduling (decreasing by 10% every 100 epochs).

[0121] Loss function: The weighted sum of the cross-entropy loss function (for classification tasks) and the mean squared error loss function (for regression tasks such as force line deviation angle and length difference) (weight ratio 1:1) is used.

[0122] Training metrics: Training set accuracy ≥ 95%, validation set accuracy ≥ 92%, inference time ≤ 1s (per sample), model saved in ONNX format, supporting deployment on edge devices.

[0123] The model establishes the correlation between "plantar pressure distribution and ankle joint force line" through an attention mechanism:

[0124] 1. Extract the feature vector of plantar pressure data (such as the proportion of pressure on the medial side of the hindfoot and the peak pressure on the lateral side of the forefoot).

[0125] 2. Calculate the baseline vector of the ankle joint force line (the line connecting the apex of the medial and lateral malleoli) by combining the coordinates of bony landmarks;

[0126] 3. By learning the mapping relationship between pressure feature vector and force line reference vector through multilayer perceptron (MLP), when the proportion of pressure on the medial side of the hindfoot exceeds 40% of the total pressure of the hindfoot (normal proportion is 25%-35%), the model determines that there is an inversion tendency of the ankle joint, and for every 5% increase in the proportion of pressure, the predicted force line deviation angle increases by 1°.

[0127] 4. When outputting the final force line deviation angle, the error is corrected by combining the standard parameters of the same age group / weight range in the clinical knowledge base to ensure that the error is ≤0.5°.

[0128] Although not shown in the accompanying drawings, further embodiments include a computer-readable storage medium for storing computer instructions that, when executed on an electronic device, enable the implementation of a biomechanical intelligent analysis and adjustment method for both lower limbs according to an embodiment of the present invention.

[0129] Although not shown in the accompanying drawings, further embodiments also include an electronic device comprising a processor and a memory, the memory for storing instructions, and the processor for calling the instructions in the memory to cause the electronic device to execute a biomechanical intelligent analysis and adjustment method for both lower limbs according to an embodiment of the present invention.

[0130] Although not shown in the accompanying drawings, further embodiments also include a computer program product comprising a computer program that, when executed, implements an embodiment of the present invention of a biomechanical intelligent analysis and adjustment method for both lower limbs.

[0131] Based on clinical application, data comparison, and user experience surveys, the technical solution proposed in this invention has at least the following advantages compared to existing technologies:

[0132] (1) Comprehensive data support + precise ankle joint wrapping solves the problems of poor adaptability and inaccurate correction in existing technologies. Existing technologies mostly rely on single plantar pressure data and lack a fully wrapped ankle joint structure, making it impossible to accurately correct ankle joint alignment deviations. This invention uses a 3D scanner, an infrared thermal imager, and a fully wrapped ankle support structure to collect multi-source data on geometric shape, bony landmarks, and pressure / vibration. At the same time, the ankle joint airbag is divided into four zones, and the pressure can be adjusted individually according to the alignment deviation (such as pressurizing the lateral airbag during inversion), achieving comprehensive adaptation from the sole of the foot to the ankle joint, avoiding the discomfort or ineffective correction caused by the "one-size-fits-all" adjustment of existing devices.

[0133] (2) Dynamic closed-loop feedback + incremental model training overcomes the shortcomings of existing static massage technology and lack of accuracy improvement. Existing technologies mostly use preset fixed programs and lack real-time feedback and iterative optimization capabilities. This invention collects pressure, vibration and user sensation data in real time during the feedback optimization stage, compares the target parameters and dynamically fine-tunes the airbag / lifting / vibration parameters (e.g., if the user reports pressure, the airbag pressure is reduced), and stores the adjusted data in the database for incremental training of large models. The subsequent analysis and adjustment accuracy for similar users (e.g., people with ankle inversion) can be improved by 15%-20%, achieving "the more you use it, the more accurate it becomes," solving the problem of no improvement in accuracy after long-term use of existing equipment.

[0134] (3) Three-dimensional visualization interaction + clinical knowledge base support to make up for the shortcomings of poor interaction and lack of scientific basis in existing technologies. Existing technologies lack intuitive display and scientific parameter support, and users cannot perceive the adjustment logic. The data processing unit of this invention generates a rotatable and scalable three-dimensional model, annotates the pressure heat map, bony landmarks and force line deviation, and presents it intuitively through the touch screen; at the same time, the analysis and decision unit combines clinical standard parameters for different age groups / weight (such as the normal threshold of hindfoot pressure 250-350kPa) to ensure that the adjustment plan is both comfortable and biomechanically reasonable, and solves the shortcomings of existing equipment in terms of interaction and science, such as "only massaging without analysis" and "users not knowing why to adjust".

[0135] (4) Multi-strategy coordinated execution overcomes the limitations of existing technologies that are functionally fragmented and only provide basic relaxation. Existing technologies mostly only achieve foot roller massage or simple airbag compression, which cannot take into account pressure balance, force line correction and comfortable relaxation. The present invention links three major strategies in the personalized adjustment stage: airbag correction of ankle joint posture, lifting module to balance the force line of the two lower limbs, and vibration to guide the diversion of foot pressure. For example, for users with excessive foot pressure, "vibration diversion + airbag posture adjustment + height balance" are executed simultaneously, which not only relieves local fatigue, but also improves the root cause from a biomechanical perspective, breaking through the limitations of existing devices that are "single-function and can only provide basic relaxation".

[0136] Those skilled in the art will understand that the implementation principles of the system (product, medium, device) embodiments and the steps of the method embodiments can correspond to and reference each other. Therefore, given that the method embodiments have already been described in detail, the implementation principles of the system (product, medium, device) embodiments need not be repeated. Furthermore, the technical problems that the method embodiments can solve and their advantages over existing technologies will necessarily be reflected in the relevant system (product, medium, device) embodiments as well.

[0137] Nevertheless, it is particularly important to note that although the present invention provides multiple embodiments, each embodiment can constitute an independent technical solution and may contribute to the prior art and solve corresponding technical problems. That is, each embodiment can solve at least one technical problem, but it is not required that each individual embodiment solve multiple or all technical problems.

[0138] Other technologies, principles, algorithms, or models not elaborated in detail in this application can be found in the prior art.

[0139] The foregoing has shown and described the method embodiments and systems of the present invention, but it will be understood by those skilled in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A biomechanical intelligent analysis and adjustment system for both lower limbs, the system comprising a data acquisition unit, a data processing unit, an analysis and decision-making unit, and a human-computer interaction unit. Its features are: The data acquisition unit includes a fully enclosed ankle support structure, a 3D scanner, and an infrared thermal imager. The fully enclosed ankle support structure is equipped with a matrix pressure sensor, multiple sets of vibration sensors, an airbag adjustment component, and an independent lifting module. The data processing unit receives the multidimensional data collected by the data acquisition unit, performs preprocessing, and outputs optimized three-dimensional model data and standardized biomechanical data. The analysis and decision-making unit receives the standardized biomechanical data output by the data processing unit, analyzes it to obtain the results of plantar pressure imbalance area, ankle joint force line deviation and lower limb force line offset, and generates a personalized adjustment plan. The human-computer interaction unit receives user feedback information and transmits the feedback information to the analysis and decision-making unit to achieve dynamic optimization of the solution; The 3D scanner is used to collect 3D point cloud data of both lower limbs to obtain a 3D model of both lower limbs; The infrared thermal imager is used to collect infrared thermal imaging data of both lower limbs to locate bony landmarks. The matrix pressure sensor is used to collect plantar pressure data; The multiple sets of vibration sensors are used to collect ankle joint vibration feedback data; The airbag adjustment assembly is used for ankle posture correction; The independent lifting module is used to adjust the support height of both lower limbs.

2. The intelligent biomechanical analysis and adjustment system for both lower limbs as described in claim 1, characterized in that, The system also includes an execution adjustment module, which is communicatively connected to the analysis and decision-making unit; The adjustment module includes an airbag control unit, a lifting control unit, and a vibration control unit; The airbag control unit is used to control the inflation / deflation pressure of the airbag adjustment component inside the fully enclosed ankle support structure; the lifting control unit is used to control the lifting height of the independent lifting module at the bottom of the support structure; and the vibration control unit is used to control the mode parameters of the vibration signal output by the vibration sensor.

3. The intelligent biomechanical analysis and adjustment system for both lower limbs as described in claim 1, characterized in that, The airbag adjustment component of the fully enclosed ankle support structure adopts a zoned control design, dividing the airbags around the ankle joint into four independent control areas: medial, lateral, anterior, and posterior. Based on the direction of ankle joint force line deviation output by the analysis and decision unit, the inflation pressure of the corresponding area airbag is adjusted individually to achieve precise posture correction.

4. The intelligent biomechanical analysis and adjustment system for both lower limbs as described in claim 1, characterized in that, The human-computer interaction unit includes a touch screen display. The data processing unit also includes a 3D modeling and visualization submodule, which converts the preprocessed 3D point cloud data of the lower limbs into a rotatable and scalable 3D visualization model, and marks the location of bony landmarks, the heat map of plantar pressure distribution, and the direction of ankle joint force line deviation on the model, and displays them in real time through the touch screen.

5. The intelligent biomechanical analysis and adjustment system for both lower limbs as described in claim 1, characterized in that, The analysis and decision-making unit is connected to a pre-trained large-scale biomechanical analysis model of the lower limbs and a clinical knowledge base; The clinical knowledge base includes normal plantar pressure distribution maps for different age groups and weight ranges, ankle joint neutral position force line parameters, and standard values ​​of the angle between the force lines of the two lower limbs.

6. A biomechanical intelligent analysis and adjustment method for both lower limbs, applied to the biomechanical intelligent analysis and adjustment system for both lower limbs as described in any one of claims 1-5, characterized in that, The method includes: Data acquisition phase: Perform 3D scanning of the user's lower limbs to obtain 3D point cloud data of the lower limbs to construct an initial 3D model; acquire infrared thermal imaging data of the lower limbs to identify and locate bony landmarks of the lower limbs; acquire plantar pressure distribution data, ankle joint movement and force feedback data; Data preprocessing stage: The data obtained in the data acquisition stage is denoised, registered and meshed to optimize the initial 3D model to obtain a high-precision 3D visualization model of both lower limbs; the infrared thermal imaging data is calibrated for temperature and segmented to extract the 3D coordinate information of bony landmarks; the plantar pressure distribution data and ankle joint vibration feedback data are filtered and normalized to eliminate environmental interference signals. Biomechanical analysis stage: The pre-processed plantar pressure distribution data, ankle joint vibration feedback data and bony landmark coordinate information are input into the pre-trained biomechanical analysis model of the two lower limbs to analyze and obtain the user's plantar pressure imbalance area, ankle joint force line deviation angle and overall force line offset of the two lower limbs. Personalized adjustment stage: Based on the analysis results output by the large biomechanical analysis model of the lower limbs, a personalized plan is generated, which includes plantar pressure adjustment strategy, ankle posture correction strategy and lower limb support height adjustment strategy. Feedback and optimization phase: During the execution of the personalized plan, plantar pressure data, ankle joint motion data, and somatosensory feedback information are collected in real time; the real-time collected data are compared with the target parameters output by the large model, and the deviation value is calculated; the airbag pressure parameters, lifting height parameters, and vibration mode parameters in the personalized plan are dynamically adjusted according to the deviation value until the biomechanical parameters of both lower limbs reach the normal range in the clinical knowledge base.

7. The intelligent biomechanical analysis and adjustment method for both lower limbs as described in claim 6, characterized in that, The large-scale biomechanical analysis model for both lower limbs is trained using multi-source lower limb biomechanical sample data and has the ability to perform pressure-force line correlation analysis and identify abnormal parameters.

8. The intelligent biomechanical analysis and adjustment method for both lower limbs as described in claim 6, characterized in that, The feedback optimization phase also includes storing the user's biomechanical data, adjustment parameters, and somatosensory feedback information after each adjustment into a database, which serves as incremental training data for the large-scale biomechanical analysis model of the lower limbs, thereby improving the accuracy of subsequent analysis and adjustments.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements all the steps of the intelligent biomechanical analysis and adjustment method for both lower limbs as described in any one of claims 6-8, and during the execution process, it calls the three-dimensional modeling visualization interface to generate a three-dimensional model of both lower limbs, and realizes the reception and processing of user feedback information through the human-computer interaction interface.

Images