Insole design method and system based on foot physiology

Abstract

The invention discloses an insole design method and system based on foot physiology, and relates to the technical field of insoles, the method comprises the following steps: dividing an insole into a heel area, an arch area and a half sole area to obtain a partition model; an upper-layer pad and a lower-layer pad are formed at the corresponding positions of the partition model in an injection molding mode, a pressure sensor module is arranged between the upper-layer pad and the lower-layer pad, a plurality of air bag cavities are preset in the heel area, the arch area and the half sole area respectively, and an embedding groove is formed in the arch area; a processing module is embedded in the lower-layer cushion embedding groove, is used for being connected with the pressure sensor module and is further used for controlling the air pressure of the air bag cavity; superposing the upper layer pad, the functional layer containing the pressure sensor module and the lower layer pad, and fusing the edges of the partitions through hot pressing; the joint motion capturing module and the myoelectricity sensing module are configured with the processing module, and intelligent design of the insole is completed. The shoe pad designed by the invention can comprehensively reflect the gait of the user.

Description

Technical Field

[0001] This invention relates to the technical field of insoles, and more particularly to an insole design method and system based on foot physiology. Background Technology

[0002] Intelligent insoles, as wearable biomechanical monitoring devices, collect real-time data on foot pressure distribution through a piezoresistive / capacitive sensor array.

[0003] Existing technologies, such as Chinese Patent Publication No. CN117731272A, disclose a gait analysis method and system that integrates a plantar pressure insole and multiple IMUs. The integrated inertial measurement unit (IMU) monitors the ankle joint angle to diagnose and analyze gait disorders. However, it cannot fully reflect the user's gait.

[0004] For example, when abnormal plantar pressure distribution is detected (such as decreased medial pressure and increased lateral pressure), data on ankle angle and plantar pressure distribution alone cannot accurately distinguish whether this abnormality is caused by lower limb structural lesions (such as fixed arch collapse or bony deformities) or by compensatory gait resulting from pain avoidance responses (such as plantar fasciitis or acute arthritis). If the pain-compensating gait is mistakenly identified as a structural abnormality, the insole may guide the user to forcibly correct themselves to a so-called normal posture, such as increasing arch support or forcibly inverting the foot, which may actually aggravate the pain source and cause secondary injury. If the structural abnormality is misjudged as a temporary compensation, the early corrective intervention window will be missed, leading to worsening of the deformity or secondary damage, such as knee osteoarthritis.

[0005] Therefore, how to design insoles that comprehensively reflect a user's gait has become a pressing technical problem that needs to be solved. Summary of the Invention

[0006] The technical problem solved by this invention is that existing technologies cannot comprehensively reflect the user's gait during the design process.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a method for designing insoles based on foot physiology, the method comprising:

[0008] The insole is divided into the heel area, arch area, and forefoot area to obtain a partitioned model;

[0009] An upper pad and a lower pad are injection molded at the corresponding positions of the partition model. A pressure sensor module is provided between the upper pad and the lower pad. Multiple air bladder cavities are pre-set in the heel area, arch area and forefoot area respectively. An interlocking groove is provided in the arch area.

[0010] A processing module is embedded in the lower pad fitting groove. The processing module is used to connect to the pressure sensor module and is also used to control the air pressure in the airbag cavity.

[0011] The upper pad, the functional layer containing the pressure sensor module, and the lower pad are stacked together, and the partition edges are fused together by hot pressing.

[0012] Configure the joint motion capture module and the electromyography sensing module with the processing module respectively.

[0013] Preferably, the processing module is further configured to wirelessly connect to the joint motion capture module and the electromyography (EMG) sensing module, respectively. The joint motion capture module is used to be attached to the knee joint to monitor the dynamic changes in the flexion and extension angles of the knee joint, obtain joint angle change data, and send the joint angle change data to the processing module. The EMG sensing module is used to be attached to the belly of the gastrocnemius muscle to monitor EMG signals, determine the EMG signals within a preset time window after heel strike, and send the determined EMG signals to the processing module, so that the processing module can monitor the foot status of the target user based on the joint angle change data and the EMG signals within the preset time window after heel strike.

[0014] Preferably, the foot condition monitoring method executed by the processing module includes:

[0015] Acquire temporal data of plantar pressure, joint angle change data, and electromyographic signals within a preset time window after heel strike;

[0016] Based on the plantar pressure time-series data, joint angle change data, and electromyography signals, the data corresponding to the abnormal pressure segment, abnormal fluctuation segment, and abnormal electromyography segment are identified respectively.

[0017] Gait coupling index is determined based on the spatiotemporal correlation characteristics of data corresponding to abnormal pressure segments, abnormal fluctuation segments, and abnormal electromyography segments.

[0018] If the gait coupling index exceeds the preset coupling threshold, a diagnosis of pain compensatory abnormality is output.

[0019] If an abnormal pressure segment exists and the gait coupling index does not exceed the preset coupling threshold, a structural anomaly diagnosis is output.

[0020] Preferably, the foot condition monitoring method further includes:

[0021] If the diagnosis is the aforementioned pain compensation abnormality, then the mechanical output of the arch area of ​​the insole is suppressed, and the cushioning of the heel area of ​​the insole is enhanced.

[0022] If the diagnosis result is the structural abnormality diagnosis, then the mechanical output of the arch area is increased to the preset correction requirement level.

[0023] Preferably, the arch region includes the medial longitudinal arch region and the lateral metatarsal head region; the step of identifying the data corresponding to the abnormal pressure segment, abnormal fluctuation segment, and abnormal electromyographic segment based on plantar pressure time-series data, joint angle change data, and electromyographic signals respectively includes:

[0024] The pressure center trajectory is determined based on the plantar pressure time-series data.

[0025] The time-series data of plantar pressure are used to determine whether they represent the first and second abnormal phenomena.

[0026] Detect whether the pressure center trajectory represents the third and fourth anomalies;

[0027] If the plantar pressure time series data simultaneously represents the first abnormal phenomenon and the second abnormal phenomenon, and the pressure center trajectory represents either the third abnormal phenomenon or the fourth abnormal phenomenon, then the data in the plantar pressure time series data that simultaneously represents the first abnormal phenomenon and the second abnormal phenomenon, and the pressure center trajectory represents either the third abnormal phenomenon or the fourth abnormal phenomenon, is determined as the data corresponding to the abnormal pressure segment.

[0028] Based on joint angle change data and electromyography signals, the data corresponding to abnormal fluctuation segments and abnormal electromyography segments are identified respectively.

[0029] The time period during which the pressure center trajectory represents either the third or the fourth abnormal phenomenon is defined as the trajectory abnormal period.

[0030] The second derivative of the joint angle change data is extracted, and the second derivative of the joint angle change data is used to characterize angular acceleration;

[0031] The joint angle change data during the period when the absolute value of the angular acceleration is greater than the preset acceleration threshold during the abnormal trajectory period are determined as the data corresponding to the abnormal fluctuation segment;

[0032] Within the time window corresponding to the abnormal fluctuation period, the period when the electromyographic signal corresponding to the tibialis posterior muscle is lower than the preset benchmark electromyographic value, or the period when the electromyographic value corresponding to the peroneus longus muscle is higher than the benchmark electromyographic value, is determined as the abnormal electromyographic segment.

[0033] The electromyographic signals within the abnormal electromyographic segment are identified as the data corresponding to the abnormal electromyographic segment; wherein...

[0034] The first abnormal phenomenon includes the fact that the pressure corresponding to the medial longitudinal arch area of ​​the insole in the foot pressure time series data is lower than the first proportion value of the dynamic reference pressure corresponding to the first preset time period within a first preset time period. The dynamic reference pressure corresponding to the first preset time period is the average value of multiple pressure values ​​of the insole within the first preset time period.

[0035] The second abnormal phenomenon includes the pressure corresponding to the lateral metatarsal head region of the insole in the plantar pressure time series data being higher than the second proportional value of the dynamic reference pressure in the second preset time period, where the first proportional value is less than the second proportional value.

[0036] The third abnormal phenomenon includes the instantaneous outward movement speed of the pressure center trajectory corresponding to the middle and late support phases of the gait cycle exceeding a preset outward movement speed threshold.

[0037] The fourth abnormal phenomenon includes the outward displacement of several consecutive trajectory points of the preset number of pressure center trajectory points corresponding to the middle and late support phases of the gait cycle, which exceeds the trajectory range corresponding to the preset maximum eversion angle of the foot.

[0038] Preferably, the determination of the gait coupling index based on the spatiotemporal correlation features of data corresponding to abnormal pressure segments, abnormal fluctuation segments, and abnormal electromyography segments includes:

[0039] The abnormal pressure segment, abnormal fluctuation segment, and abnormal electromyography segment identified in the gait cycle are time-aligned using the heel strike event in the gait cycle as the synchronization point.

[0040] Determine the coexistence rate of abnormal pressure segment, abnormal fluctuation segment, and abnormal electromyographic segment in a single gait cycle;

[0041] The time difference between the abnormal fluctuation segment and the abnormal electromyography segment is determined based on the starting point of the abnormal fluctuation segment and the starting point of the abnormal electromyography segment;

[0042] When the coexistence rate is higher than a preset association threshold and the interval delay time difference is lower than a preset delay threshold, the spatiotemporal association feature is determined based on the ratio of the coexistence rate to the interval delay time difference.

[0043] The gait coupling index is determined based on the spatiotemporal correlation characteristics.

[0044] Preferably, the gait coupling index is positively correlated with the ratio of the coexistence rate to the time difference of the interval delay.

[0045] Preferably, the step of outputting a diagnosis of pain compensation abnormality if the gait coupling index exceeds a preset coupling threshold includes:

[0046] If the abnormal electromyographic segment is present and the gait coupling index exceeds a preset coupling threshold, a diagnosis of pain compensation abnormality is output.

[0047] Preferably, if an abnormal pressure segment exists and the gait coupling index does not exceed a preset coupling threshold, then a structural anomaly diagnosis is output, including:

[0048] The inter-step variation data corresponding to the abnormal pressure segment are determined based on the outward velocity sequence in the pressure center trajectory corresponding to the abnormal pressure segment.

[0049] If the gait coupling index does not exceed the preset coupling threshold, the abnormal pressure segment appears in a preset number of consecutive gait cycles, and the inter-step variation coefficient corresponding to the abnormal pressure segment is lower than the preset variation upper limit, then a structural abnormality diagnosis is output.

[0050] On the other hand, the present invention also provides an insole design system based on foot physiology, the system comprising:

[0051] The partitioning module is used to divide the insole into the heel area, arch area, and forefoot area to obtain a partitioning model.

[0052] The injection molding module is used to inject and form an upper pad and a lower pad at the corresponding positions of the partition model. A pressure sensor module is provided between the upper pad and the lower pad. Multiple air bladder cavities are pre-set in the heel area, arch area and forefoot area respectively. An interlocking groove is provided in the arch area.

[0053] An embedding module is used to embed a processing module in the fitting groove of the lower pad. The processing module is used to connect to the pressure sensor module and is also used to control the air pressure in the airbag cavity.

[0054] The stacking module is used to stack an upper pad, a functional layer containing the pressure sensor module, and a lower pad, and fuses the partition edges together by hot pressing.

[0055] A configuration module is used to configure the joint motion capture module and the electromyography sensing module with the processing module, respectively.

[0056] The beneficial effects of this invention are as follows: By configuring the joint motion capture module and the electromyography (EMG) sensing module with the processing module respectively, the processing module can acquire plantar pressure time-series data, joint angle change data, and EMG signals within a preset time window after heel strike. This allows the insole to comprehensively reflect the user's gait based on the plantar pressure time-series data, joint angle change data, and EMG signals within a preset time window after heel strike. Attached Figure Description

[0057] Figure 1A schematic diagram illustrating the basic process of a foot physiology-based insole design method according to an embodiment of the present invention;

[0058] Figure 2 This is a basic flowchart illustrating a foot condition monitoring method according to an embodiment of the present invention. Detailed Implementation

[0059] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0060] like Figure 1 As shown, the method includes S1 to S5.

[0061] S1 divides the insole into the heel area, arch area, and forefoot area to obtain a partitioned model.

[0062] First, the entire insole design area is clearly divided into three key functional areas: the heel area, closest to the heel; the arch area, located below the center of the foot and supporting the arch; and the forefoot area, corresponding to the toes and the ball of the foot. This partitioned design creates a basic partition model, providing a blueprint for the subsequent implementation of differentiated functions in different areas.

[0063] S2, injection molding of upper and lower pads at corresponding positions in the partition model, wherein a pressure sensor module is provided between the upper and lower pads, and multiple air bladder cavities are pre-set in the heel area, arch area and forefoot area, and a fitting groove is provided in the arch area.

[0064] Next, based on the aforementioned partitioned model, the core structural layers of the insole are manufactured using injection molding: the upper insole (the side that directly contacts the sole of the foot) and the lower insole (the side closest to the sole). During the injection molding process, it is particularly important to pre-create cavities within the heel, arch, and forefoot areas; these are the air bladder cavities (functionally similar to rudimentary air bladders). Simultaneously, a specific shape and size recessed structure, called a fitting groove, is precisely injection molded into the lower insole in the arch area to accurately secure the subsequent electronic modules. The most crucial layer is a pressure sensor module placed between the upper and lower insole layers after injection molding. This module is carefully positioned to cover or correspond to key load-bearing areas in the partitioned model (especially the heel and forefoot) to detect pressure distribution during walking.

[0065] S3, a processing module is embedded in the lower pad fitting groove. The processing module is used to connect the pressure sensor module and also to control the air pressure in the airbag cavity.

[0066] Then, the processing module (the core calculation and control unit of the insole) is securely embedded and fixed in the fitting groove of the lower arch area. It is connected to the pressure sensor module set in the previous step through fine wiring to receive pressure signals.

[0067] The processing module achieves precise control of the air pressure in the airbag cavity through an integrated airway control system. This airway control system comprises three core units: a micro air pump unit, an air pressure sensor unit, and an intelligent valve array, all of which work collaboratively through the central processing unit of the processing module.

[0068] Each air bladder cavity is embedded with a miniature pressure sensor to monitor the internal pressure (unit: kPa) in real time. These sensors transmit data to the core chip of the processing module via a network of microwires at millisecond-level frequencies. The algorithm built into the processing module dynamically analyzes the pressure requirements of the three zones (heel, arch, and forefoot): for example, during walking, the forefoot requires high pressure to support the push-off action, while the arch needs to maintain medium pressure to stabilize the arch structure; when standing statically, the pressure needs to be evenly distributed.

[0069] During the inflation phase, the miniature brushless air pump (size <2cm) is activated. 3 Integrated within the module, the air pump draws in external air and delivers it to the target airbag cavity through silicone microtubes (0.8-1.2mm in diameter) pre-embedded in the padding layer. The inflation rate is precisely controlled by the processing module's PWM pulse width modulation technology, ensuring a smooth pressure increase in a gradient of 5-10kPa / s. Once the target pressure is reached, the processing module closes the electromagnetic one-way valve in the corresponding area, cutting off the airflow channel, and dynamically compensates through continuous data feedback from the pressure sensor (e.g., initiating micro-inflation for 0.3 seconds upon detecting a pressure drop of 0.5kPa). If excessive pressure is detected (e.g., a sudden pressure increase due to impact on the heel area), the processing module opens the differential pressure relief valve at the top of the airbag cavity and simultaneously initiates the suction mode. At this time, the air pump reverses its rotation, expelling excess gas through a nano-level filter membrane, with a pressure relief rate of up to 15kPa / s.

[0070] Furthermore, a buffer solution can be placed in the airbag cavity, and the processing module is also used to control the flow of the buffer solution between the airbag cavities.

[0071] S4, the upper pad, the functional layer containing the pressure sensor module and the lower pad are stacked, and the partition edges are fused together by hot pressing.

[0072] The next step is to precisely align and stack the assembled layers. This includes: the upper pad, the middle functional layer containing the pressure sensor module, and the lower pad with the processing module already embedded. After stacking, a precisely controlled hot-pressing process is used to heat and pressurize these three layers at the partition edges (i.e., the boundaries between the heel, arch, and forefoot areas defined by the partition model). Hot pressing melts and ultimately fuses the material in this edge area, firmly bonding the layers together. More importantly, this hot-pressing seal also ensures that the connecting ducts between the air bladder cavity (in the heel and forefoot) and the processing module (in the arch) are completely sealed, thus forming a closed, internally connected, complete airflow circuit system inside the insole.

[0073] S5 configures the joint motion capture module and electromyography sensing module with the processing module respectively to complete the intelligent design of the insole.

[0074] To achieve more comprehensive gait and biomechanical monitoring, joint motion capture and electromyography (EMG) sensing modules are connected to a processing module embedded in the arch area of ​​the insole. This connection can be wired (with tiny wires guided within the shoe) or a low-power wireless connection (such as Bluetooth). Through configuration and communication with the processing module, the kinematic and physiological data collected by these external sensors can be transmitted to the core processing unit.

[0075] In this embodiment, the joint motion capture module and the electromyography (EMG) sensing module are configured with the processing module respectively, so that the processing module can acquire plantar pressure time-series data, joint angle change data, and EMG signals within a preset time window after heel strike. This allows the insole to comprehensively reflect the user's gait based on the plantar pressure time-series data, joint angle change data, and EMG signals within a preset time window after heel strike.

[0076] When existing insole systems detect abnormal plantar pressure distribution (such as decreased medial pressure and increased lateral pressure), they rely solely on plantar pressure (single modality) data and cannot accurately distinguish whether such abnormalities are caused by lower limb structural lesions (such as fixed arch collapse and bony deformities) or by compensatory gait resulting from pain avoidance responses (such as plantar fasciitis and acute arthritis).

[0077] For example, conventional methods for analyzing pressure distribution and COP trajectories are completely unable to distinguish between structural lesions and pain avoidance responses.

[0078] For example, simple methods that integrate IMU (such as foot / lower leg angles) can detect foot inversion / outversion angles, lower leg tilt, etc., and make a comprehensive judgment based on pressure abnormalities. However, this can only confirm whether the foot posture is normal, and cannot answer whether the cause of the posture abnormality is structural or pain. For example, foot eversion + knee varus can be seen in both structural flat feet and pain avoidance responses caused by medial arch pain.

[0079] On the other hand, systems based on subjective user input (app questionnaires) require users to report whether they experience pain. This approach is subjective, delayed, and unreliable (users may not report or may give false reports), resulting in poor applicability.

[0080] Furthermore, machine learning classification (black box model): a classification model (normal / abnormal) is trained using stress + IMU data. However, the model has poor interpretability; weak generalization ability (easily fails in the face of individual differences, new shoe styles / environments); it does not explicitly model the biomechanical differences in structure / pain, and relies on a large amount of labeled data (rare and costly).

[0081] Therefore, existing technologies lack the ability to deeply analyze and infer the underlying human system state (structural stability and pain transient stability) behind the characterization signals (pressure / angle), leading to misjudgment and incorrect intervention.

[0082] If the pain-compensating gait is mistakenly identified as a structural abnormality, the insole system may guide the user to forcibly "correct" to a so-called "normal" posture (such as increasing arch support or forcibly inverting), which may aggravate the pain and cause secondary injury. If the structural abnormality is misjudged as a temporary compensation, the early corrective intervention window will be missed, leading to the aggravation of deformity or secondary damage (such as knee arthritis).

[0083] Based on this, the inventors of this application conducted in-depth research and discovered that, at the phenomenological level, both structural abnormalities and pain compensation can manifest as follows: during the mid / late support phase, the center of pressure trajectory (COP) shifts excessively outward (insufficient varus or increased valgus); in pressure distribution, pressure is significantly reduced in the medial longitudinal arch region and significantly increased in the lateral metatarsal head region. Therefore, traditional threshold judgments (such as medial pressure < threshold X) are completely ineffective.

[0084] However, in terms of biomechanical mechanisms, the root cause of structural abnormalities lies in persistent anatomical changes in bone and joint structures, ligaments, or muscles (such as posterior tibial tendon dysfunction leading to arch collapse); the gait characteristics of structural abnormalities are stable and highly repetitive abnormal patterns. Changes in joint kinematics and abnormal plantar pressure show a high degree of linear or constant correlation (e.g., the degree of arch collapse is positively correlated with the genu varum angle). The body attempts to adapt to these structural changes with relatively fixed compensatory patterns.

[0085] In terms of biomechanical mechanisms, the root cause of pain-compensating gait is a neuroreflexive pain avoidance response triggered by local tissue pain (such as the heel or medial arch). The gait is characterized by fluctuating abnormal patterns and pain avoidance orientation. To reduce the load on the painful area, the body will momentarily and dynamically adjust its gait (e.g., shortening the support time on the affected side and quickly shifting the center of gravity to the lateral side). This may be accompanied by a chain reaction caused by pain avoidance (e.g., excessive load on the healthy side, sudden inward valgus of the knee).

[0086] Based on the differences in biomechanical mechanisms between structural abnormalities and pain compensation, the key differences between the two modes can be found to be: the dynamic changes of structural abnormalities are smooth, predictable, and coordinated as a whole; while the changes of pain compensation are sudden, time-varying, and localized, and may disrupt the original coordination between joints.

[0087] Based on this, the inventors of this application have discovered that abnormal pressure segments in plantar pressure time-series data can reflect the persistence or fluctuation of abnormal plantar pressure distribution, such as a significant decrease in medial pressure and a significant increase in lateral pressure during the mid / late support phase. Abnormal fluctuation segments in joint angle change data can reflect transient and discontinuous fluctuations in joint dynamic parameters, such as sudden and drastic changes in joint angle or angular acceleration at the moment of heel strike or at the point of maximum weight-bearing during the mid-support phase. Abnormal electromyographic segments in electromyographic signals within a preset time window after heel strike can reflect abnormal muscle activity caused by nerve reflex pain avoidance responses, such as abrupt changes in the intensity or timing pattern of electromyographic signals caused by medial arch or heel pain. By combining these three factors, the inherent coordination differences in lower limb gait patterns can be reflected, namely, the stable linear correlation of joint movement trajectories in structural abnormalities and the local sudden decoupling of dynamic disturbances in pain compensation. This allows for the characterization of the overall stability of the gait system, thereby enabling accurate differentiation between structural abnormalities and pain compensation when plantar pressure time-series data is abnormal.

[0088] Specifically, the processing module built into the insole described in this embodiment aims to accurately distinguish the root cause of abnormal gait—structural abnormalities (such as fixed arch collapse or bony deformities) or pain compensation responses (such as pain-avoidance gait triggered by plantar fasciitis or acute arthritis)—through multimodal sensor fusion and spatiotemporal correlation analysis. The multimodal sensor mainly consists of the insole, a joint motion capture module (usually an IMU inertial measurement unit), an electromyography (EMG) module, and the processing module. The insole includes a pressure sensing module and a processing module. The pressure sensing module monitors plantar pressure data to obtain temporal plantar pressure data. A joint motion capture module, wirelessly connected to the processing module, is attached to the knee joint to monitor dynamic changes in the knee's flexion and extension angles, obtaining joint angle change data, and sending this data to the processing module. An electromyography (EMG) sensing module, also wirelessly connected to the processing module, is attached to the gastrocnemius muscle belly to monitor EMG signals and determine the EMG signals within a preset time window after heel strike, sending the determined EMG signals to the processing module. The processing module performs actions such as... Figure 2 The foot condition monitoring method shown is to design the insole to have foot condition monitoring function, and when the monitoring function is implemented, it can accurately distinguish the pressure source when abnormal foot pressure data occurs.

[0089] Pressure sensing module: High-precision piezoresistive / capacitive sensor array (e.g., 8x8 grid, 100Hz sampling) to generate real-time plantar pressure time-series data (P_t(x, y)).

[0090] Joint motion capture module (IMU Unit): Sensor, a miniature IMU integrating a three-axis accelerometer and a three-axis gyroscope, tightly strapped to the lateral or anterior side of the lower leg near the knee joint (optimal position for capturing knee flexion-extension angle changes (θ_knee)). Data stream: raw acceleration (a), angular velocity (ω) → onboard / processing module fusion filtering algorithm → output joint flexion / extension angle time-series data (θ_knee(t)) → processing module second derivative to obtain angular acceleration (α_knee(t) = d). 2 θ_knee / dt 2 Data transmission: Bluetooth LE / BLE Mesh (Low Power).

[0091] Electromyography (EMG) Unit: Sensors, paired surface electrodes (Ag / AgCl), precisely attached to the belly of the Tibialis posterior muscle (posterosuperior to the medial malleolus, parallel to the muscle fibers) and the belly of the Peroneus Longus muscle (outer middle and lower 1 / 3 of the lower leg), respectively. Includes a reference electrode. Signal chain: Raw EMG (mV) → preamplifier → bandpass filter (20-500Hz) → power frequency notch filter (50 / 60Hz) → rectification (|EMG|) → (optional: RMS smoothing). Event triggering: Receives HS event synchronization signal from the insole processing module → extracts and sends complete EMG timing data (EMG_TibPost(t), EMG_PerLong(t)) within the 0ms-200ms time window after HS.

[0092] The processing module (Embedded Processor) is hardware integrated into the insole or a separate controller, containing a main control MCU (such as a Cortex-M7) and a real-time signal processing unit (such as a DSP core). It coordinates the data flow of all modules (S110) and executes all algorithms (S120-S170) (recognition, association, calculation, diagnosis, and control). Furthermore, it has a storage unit to store personalized thresholds / benchmark values, historical gait data, and diagnostic results.

[0093] like Figure 1 The foot condition monitoring method shown includes S110 to S150:

[0094] S110 acquires plantar pressure time-series data, joint angle change data, and electromyographic signals within a preset time window after heel strike.

[0095] This step is the starting point of the data processing flow and is responsible for collecting raw multimodal sensor data.

[0096] Plantar pressure time-series data originates from pressure sensing modules integrated within the insole (such as piezoresistive sensor arrays or capacitive sensors). These sensors are densely distributed across different areas of the insole (such as the medial longitudinal arch, lateral metatarsal heads, and heel), continuously measuring and outputting the pressure value at each sensor location at a preset high frequency (e.g., 100Hz) during the user's walking process. Therefore, the output "plantar pressure time-series data" is a complex time-series dataset containing timestamp information and pressure values ​​for different plantar areas (possibly organized in matrix or vector form). For example, when a user walks, the sensors record pressure values ​​at 100Hz for the medial longitudinal arch region as 30kPa, 32kPa, ..., and for the lateral metatarsal head region as 80kPa, 82kPa, ..., forming a curve that changes over time.

[0097] Joint angle change data originates from a joint motion capture module (typically an IMU containing a gyroscope and accelerometer) attached to the user's knee joint (or other critical joints such as the ankle). The core function of this module is to capture the dynamic changes in joint angles in the sagittal (flexion / extension) and coronal (adduction / abduction or varus / valgus) planes. The IMU unit continuously measures angular velocity (via the gyroscope) and linear acceleration (via the accelerometer), and fuses this raw data using algorithms built into the module or processing it (e.g., complementary filtering or Kalman filtering) to calculate and output the changes in joint flexion / extension angles and their time-series data in real time (typically at a sampling rate of 100Hz or higher). For example, the IMU calculates the current knee flexion angle as 15 degrees, then shows it changing over time to 14 degrees, 13 degrees... or extension to 5 degrees, 0 degrees, etc., forming a time-series curve of the knee flexion angle.

[0098] The electromyographic (EMG) signals within a preset time window after heel strike originate from an EMG sensor module. This module needs to be attached to the belly of a specific tibialis posterior muscle (whose main function is to support the arch of the foot and contribute to foot inversion) and / or the belly of a peroneus longus muscle (whose main function is to plantarflex and evert the foot). EMG signals reflect the electrophysiological activity generated during muscle activity. EMG sensors (usually surface electrodes) capture the electrical signals from muscle fibers, which are then pre-amplified and filtered (typically removing 50 / 60Hz power frequency interference and noise outside the EMG signal range, retaining typical frequency bands such as 20-500Hz) to obtain a time-series voltage signal representing the degree of muscle activation.

[0099] Within a preset time window following heel strike, the processing module (or the EMG module itself) uses impact signals from pressure sensors or accelerometers to precisely identify the timing of the "heel strike" (HS) event in the gait cycle. After identification, the system extracts all electromyographic signal data within a crucial preset time window immediately following the HS (e.g., HS+0ms to HS+200ms). The selection of this time window is physiologically significant because it captures the moment of heel strike impact and the subsequent proprioceptive reflexes and / or potential pain avoidance reflexes of the supporting muscles. Pain avoidance responses are often most pronounced during this phase. For example, when the pressure sensor detects a sudden increase in pressure in the heel area (i.e., the HS event), the system immediately marks this moment (t0). Subsequently, it extracts the EMG signal amplitude (e.g., in microvolts μV) and its temporal changes of the tibialis posterior and peroneus longus muscles during the time interval from t0 to t0+200ms.

[0100] Existing technologies are limited to a single modality (pressure only) or a dual modality (pressure + IMU joint angle), failing to uncover the neuromuscular mechanisms underlying gait abnormalities. The core of pain compensation lies in the rapid feedback regulation of the nervous system (avoidance reflex), which manifests instantly in the electromyographic activity of muscles. Structural abnormalities typically present as fixed, stable biomechanical and muscle activity patterns (even if anomalous), while pain compensation manifests as transient, fluctuating kinematic and muscle activity patterns (especially in the initial stages of impact). Therefore, introducing electromyographic signals during the critical period of the avoidance reflex (after heel strike impact) is a key aspect of this approach, aiming to capture the unique neuromuscular response characteristics specific to pain—a crucial dimension that existing technologies (pressure only or pressure + angle) completely ignore or cannot effectively acquire.

[0101] In S110, by acquiring temporal data of plantar pressure, the system captured the final biomechanical manifestations of gait abnormalities (decreased medial pressure / increased lateral pressure, etc.); by acquiring joint angle change data (especially the second derivative information provided by the IMU, which can be used to derive angular acceleration laterally), the system captured the kinematic characteristics and dynamic changes of joint movement; by accurately capturing electromyographic signals within a preset time window after heel strike, the system introduced for the first time key evidence reflecting neuromuscular control and pain avoidance reflex activity, providing unique bioelectrical signal evidence for distinguishing between structural adaptation and pain-induced transient coordinated changes. The data from these three modalities together form the cornerstone of multidimensional, spatiotemporally correlated analysis.

[0102] S120 identifies the data corresponding to abnormal pressure segments, abnormal fluctuation segments, and abnormal electromyographic segments based on plantar pressure time-series data, joint angle change data, and electromyographic signals, respectively.

[0103] This step is responsible for the initial processing and analysis of the raw data, identifying data segments that conform to specific abnormal patterns (abnormal pressure segments, abnormal fluctuation segments, abnormal electromyography segments).

[0104] Among them, the abnormal pressure segment, abnormal fluctuation segment, and abnormal electromyography segment are all time-period data.

[0105] Preferably, S120 includes sub-steps S121 to S129:

[0106] S121, determine the pressure center trajectory based on plantar pressure time series data.

[0107] The readings from all pressure sensors on the insole are processed. At any given time point t, the position of COP (CoPx, CoPy) is typically calculated using the following formulas: CoPx = ∑(P_i*x_i) / ∑P_i; CoPy = ∑(P_i*y_i) / ∑P_i, where P_i is the pressure value of the i-th sensor, and (x_i, y_i) are the coordinates of that sensor in the insole coordinate system. These points are connected over time to form the COP trajectory. Example: If the insole has 100 sensors, each with its own (x, y) coordinates and real-time pressure P, the CoP position is calculated every 100 ms, and the points are connected to form a trajectory line.

[0108] The COP trajectory visually displays the movement path of the center of gravity (strictly speaking, the center of force on the sole of the foot) in the sole area during walking. Abnormal COP lateral shift is a common manifestation of both underpronation and overpronation.

[0109] S122, detect whether the plantar pressure time series data represents a first abnormality and a second abnormality. The first abnormality includes that the pressure corresponding to the medial longitudinal arch area of ​​the insole in the plantar pressure time series data is lower than a first proportion value of the dynamic reference pressure corresponding to the first preset time period within a first preset time period. The dynamic reference pressure corresponding to the first preset time period is the average value of multiple pressure values ​​corresponding to the insole within the first preset time period. The second abnormality includes that the pressure corresponding to the lateral metatarsal head area of ​​the insole in the plantar pressure time series data is higher than a second proportion value of the dynamic reference pressure within a second preset time period, and the first proportion value is less than the second proportion value.

[0110] The first anomaly needs to be verified to determine whether the average pressure value of the sensor corresponding to the "medial longitudinal arch region" in the "first preset time period" of the gait cycle (e.g., from the middle to the late support phase, i.e. from the foot flattening to the heel lifting off the ground) is lower than the first proportional value of the "dynamic reference pressure" in that region during the same time period.

[0111] First, calculate the average pressure value A_avg_inner of the selected sensors in the inner longitudinal arch region during the "first preset time period". Second, calculate the average value B_ref_dynamic of the set of pressure values ​​of all selected sensors in this region at all times during the "first preset time period" (this is the dynamic reference pressure for this period). Finally, check if A_avg_inner < (C1 * B_ref_dynamic) is true? Where C1 is the first proportional value (e.g., 0.7, i.e., the inner pressure is 70% lower than the reference). For example, from the middle to the later stage of the support (0.2s), the average pressure in the inner longitudinal arch region is 30kPa, and the dynamic reference pressure of this region during this period is 50kPa (C1 = 0.7). 30 < 0.7 * 50 (35)? If 30 < 35 is true, then the first anomaly is true for this period.

[0112] The second abnormal phenomenon needs to verify whether the average pressure of the sensor corresponding to the "outer metatarsal head area" in the "second preset period" (which may be the same as the first period, also from mid-stance to late-stance) during the gait cycle is higher than the second proportional value of the "dynamic reference pressure" in the same period in this area.

[0113] Similar to the first phenomenon, verifying the second abnormal phenomenon requires calculating the average pressure A_avg_outer of the outer metatarsal head area during this period. Calculate the dynamic reference pressure B_ref_dynamic of this area (the calculation method is the same as that of the medial side). Check whether A_avg_outer > (C2 * B_ref_dynamic)? Where C2 is the second proportional value (for example, 1.2, that is, the outer pressure is 120% higher than the reference), and the key requirement is that the first proportional value < the second proportional value (that is, C1 < C2, for example, C1 = 0.7, C2 = 1.2).

[0114] S123, detect whether the center of pressure trajectory characterizes the third abnormal phenomenon and the fourth abnormal phenomenon. The third abnormal phenomenon includes that the instantaneous outward movement speed of the center of pressure trajectory corresponding to the mid-late stance in the gait cycle exceeds the preset outward movement speed threshold; the fourth abnormal phenomenon includes that the outward movement positions of a preset number of consecutive trajectory points of the center of pressure trajectory corresponding to the mid-late stance in the gait cycle exceed the trajectory range corresponding to the preset maximum foot eversion angle.

[0115] The third abnormal phenomenon is to detect whether the instantaneous outward movement speed of the COP trajectory exceeds the preset outward movement speed threshold during the mid-late stance. Calculate the COP time-sequence trajectory points (x_t, y_t) according to S121. Calculate the instantaneous speed (the component of the vector speed in the coronal plane, that is, the left-right direction) at a certain point t during the mid-late stance. The speed vx_t = (x_t - x_{t - Δt}) / Δt. If the speed value VX_t (assuming the positive x direction is the outward movement direction) is greater than the preset threshold V_threshold (for example, 0.3 m / s), then this point t triggers the third abnormality. Example: The COP coordinate x of point t is 0.05 m, and the COP coordinate x before Δt = 0.01 s is 0.03 m, then vx_t = (0.05 - 0.03) / 0.01 = 2 m / s > 0.3 m / s, then the third abnormality holds at point t.

[0116] The fourth abnormal phenomenon is to detect whether there are consecutive preset number of trajectory points (e.g., 5 consecutive points, corresponding to 50 ms) in the COP trajectory points in the middle and late stages of support, and their positions (x coordinates) exceed the trajectory range corresponding to the preset maximum hallux valgus angle of the foot. By setting a "safety boundary line" based on the normal / maximum physiological valgus angle of the foot (usually a critical x coordinate line on the COP path, exceeding this line indicates that the hallux valgus may be too large). Scan the x coordinate sequence of COP in the middle and late stages of support, and count whether there are N consecutive points (N is a preset number, such as 5) whose x coordinates are all greater than the boundary value X_boundary. If so, trigger the fourth abnormality. Example: The x coordinate sequence of COP points in the middle and late stages of support is: 0.08, 0.085, 0.09, 0.095, 0.10,... (unit: m). For example, X_boundary = 0.09 m, and the preset consecutive number of points N = 5. If the consecutive 5 points >= 0.09? And the last 5 points are all greater than 0.09 m, then the fourth abnormality is established.

[0117] S124, if the plantar pressure time series data simultaneously represents the first abnormal phenomenon and the second abnormal phenomenon, and the center of pressure trajectory represents any one of the third abnormal phenomenon and the fourth abnormal phenomenon, then determine the data in the plantar pressure time series data that simultaneously represents the first abnormal phenomenon and the second abnormal phenomenon, and when the center of pressure trajectory represents any one of the third abnormal phenomenon and the fourth abnormal phenomenon as the data corresponding to the abnormal pressure segment.

[0118] The plantar pressure data represents the first abnormal phenomenon (A_avg_inner < C1 * B_ref_dynamic) in the time period T, the plantar pressure data represents the second abnormal phenomenon (A_avg_outer > C2 * B_ref_dynamic) in the time period T, and within the time period T, the center of pressure trajectory (COP) represents the third abnormal phenomenon or the fourth abnormal phenomenon in this period, then mark the plantar pressure time series data segment corresponding to the time period T as "the data corresponding to the abnormal pressure segment".

[0119] A single low medial pressure or high lateral pressure may not be sufficient to represent an abnormal situation, but when both occur simultaneously (a significant decrease in the medial and a significant increase in the lateral), and combined with the abnormal outward movement of COP (too fast speed or out of range), it highly indicates an abnormal pressure pattern that needs attention.

[0120] S125, determine the time period when the center of pressure trajectory represents any one of the third abnormal phenomenon and the fourth abnormal phenomenon as the trajectory abnormal time period.

[0121] The time period when the center of pressure trajectory represents the third abnormal phenomenon or the fourth abnormal phenomenon is marked as the trajectory abnormal time period.

[0122] S126, extract the second derivative of the joint angle change data. The second derivative of the joint angle change data is used to characterize angular acceleration.

[0123] Numerical differentiation is performed on the time-series data of joint angles (such as the knee flexion-extension angle θ_k) obtained from S110. First, the first derivative is calculated to obtain the angular velocity ω_k = dθ_k / dt. Then, the first derivative with respect to ω_k (or the second derivative directly with respect to θ_k) is calculated to obtain the angular acceleration α_k = dθ_k / dt. 2 θ_k / dt 2 Angular acceleration α can more sensitively capture abrupt changes and fluctuations in joint motion.

[0124] S127, the joint angle change data during the period when the absolute value of the angular acceleration is greater than the preset acceleration threshold during the trajectory abnormal period are identified as the data corresponding to the abnormal fluctuation segment.

[0125] Identify the data corresponding to the abnormal fluctuation segment. Within the "trajectory abnormal period" determined in S125, check the absolute value of the second derivative (angular acceleration α_k) of the joint angle change data. If the absolute value |α_k| is greater than a preset acceleration threshold α_threshold (e.g., 1000 deg / s),... 2 If a value is detected at time t, then a drastic fluctuation is considered to exist at that time. All joint angle change data (raw angles or calculated angular velocities and angular accelerations) corresponding to the time periods (usually several consecutive time points) detected within the "trajectory anomaly period" are marked as data corresponding to the abnormal fluctuation segment. For example, if |α_k| = 1200 deg / s is detected during the trajectory anomaly period (e.g., from 0.4s to 0.6s), then... 2 >1000deg / s 2 For time points (e.g., t = 0.45s, 0.46s), the joint angle data related to 0.45s-0.46s are extracted as abnormal fluctuation segment data.

[0126] S128, within the time window corresponding to the abnormal fluctuation period, the period when the electromyographic signal corresponding to the tibialis posterior muscle is lower than the preset benchmark electromyographic value, or the period when the electromyographic value corresponding to the peroneus longus muscle is higher than the benchmark electromyographic value, is determined as the abnormal electromyographic segment.

[0127] S129, the electromyographic signal within the abnormal electromyographic segment is identified as the data corresponding to the abnormal electromyographic segment.

[0128] Identify the abnormal electromyographic segments and their corresponding data. Within the time window corresponding to the detected abnormal fluctuation segment (usually the time period of the abnormal fluctuation segment itself), examine the electromyographic signals from the tibialis posterior and / or peroneus longus muscles:

[0129] Tibialis posterior muscle abnormality: If the average amplitude of the electromyographic signal is lower than the user-personalized or preset baseline electromyographic value EMG_ref_TibPost (which can be calibrated when the user is standing or walking in a pain-free, relaxed state), then inhibition is considered to exist (due to pain inhibiting its activity).

[0130] Peroneus longus abnormality: If the average amplitude of the electromyographic signal is higher than the user-personalized or preset baseline electromyographic value EMG_ref_PerLong, then enhanced activation is considered to be present (due to eversion compensation or pain avoidance response activation).

[0131] The time periods that meet any of the above conditions (tibialis posterior muscle activity is significantly lower than the baseline or peroneus longus muscle activity is significantly higher than the baseline) are marked as "abnormal electromyographic segments", and the original or processed electromyographic signal data corresponding to these time periods are marked as "data corresponding to abnormal electromyographic segments".

[0132] The core logic of S124 and S127 is to temporally correlate abnormal pressure with abnormal joint movement. Simply exhibiting an abnormal pressure pattern is insufficient; this approach requires the simultaneous occurrence of abnormal fluctuations in joint movement (drastic changes in angular acceleration during periods of abnormal COP). The joint motion capture module (IMU) explicitly captures these sudden, discontinuous fluctuations in joint dynamic parameters by providing angular acceleration information. Such dramatic, transient movement pattern disturbances are rare in structural abnormalities but are highly likely to occur when pain compensation forces a change in gait to reduce local load (e.g., avoiding pain points on the heel or medial arch).

[0133] S128-129 further extract and analyze electromyographic signals within the identified periods of abnormal joint dynamic fluctuations. The tibialis posterior muscle is a key muscle group supporting the arch of the foot and preventing excessive eversion. Pain (especially near the medial arch) reflexively inhibits the activity of this muscle group (“abnormal electromyographic segment” is characterized by electromyographic signals below baseline). The peroneus longus muscle is a key muscle group for foot eversion; in order to quickly transfer pressure to the lateral side when avoiding pain, it may undergo compensatory activation enhancement (“abnormal electromyographic segment” is characterized by electromyographic signals above baseline). This analysis step directly detects sudden deviations from neuromuscular control patterns caused by pain.

[0134] The abnormal pressure segments identified by the S120 typically exist stably in continuous gait cycles (structural abnormalities) or appear unstablely (pain compensation). The S120 identifies abnormal pressure segments by combining specific pressure threshold judgments (significantly reduced medial pressure and significantly increased lateral pressure) and COP trajectory abnormalities (excessive lateral movement speed or range), effectively locating key biomechanical manifestation periods of gait abnormalities. By detecting high-amplitude angular acceleration (|α_k|>α_threshold) during trajectory abnormality periods, abnormal fluctuation segments are identified, successfully capturing transient and discontinuous violent fluctuations in joint dynamics that may indicate pain avoidance responses. By analyzing the abnormal electromyographic activity of key muscles (tibialis posterior inhibition, peroneus longus enhancement) within the time window corresponding to the abnormal fluctuation segments, the changes in neuroreflexive muscle activation patterns that may be triggered by pain are successfully captured, providing direct physiological signal evidence for determining whether it is pain compensation. The entire S120 (S121~S129) outputs precisely located multimodal data segments (abnormal pressure segment, abnormal fluctuation segment, abnormal electromyography segment) that reflect anomalies at different levels for the subsequent spatiotemporal correlation analysis (S130).

[0135] Unlike black-box machine learning models, S120 defines clear anomaly judgment criteria based on biomechanical principles (internal low-external high pressure pattern, excessively rapid / excessive COP outward shift, sudden increase in angular acceleration, and changes in specific muscle electromyography patterns). Furthermore, in S120, pressure anomalies must occur simultaneously with COP trajectory anomalies, abnormal fluctuations in joint movement must be detected within the COP anomaly period, and electromyography anomalies must be observed within the joint movement fluctuation period. This establishes a strong temporal correlation constraint between modalities, improving the reliability of anomaly detection. S120 focuses on the mid-to-late support phase, a critical period for the manifestation of pressure distribution and COP outward shift issues. It also focuses on the time window after heel strike (for electromyography), a high-incidence period for proprioception and pain avoidance reflexes. Finally, S120 uses the average pressure during the user's current gait period as a benchmark (dynamic benchmark pressure), which is more adaptable to changes in walking speed, terrain slope, etc., compared to static thresholds.

[0136] S130, the gait coupling index is determined based on the spatiotemporal correlation characteristics of the data corresponding to the abnormal pressure segment, abnormal fluctuation segment and abnormal electromyography segment.

[0137] This step is the core of the solution, responsible for calculating the gait coupling index, a key feature that integrates multimodal anomaly information.

[0138] Preferably, S130 includes sub-steps S131 to S135:

[0139] S131 uses the heel strike event in the gait cycle as the synchronization point to time-align the identified abnormal pressure segment, abnormal fluctuation segment, and abnormal electromyography segment.

[0140] Time alignment involves normalizing or directly extracting a fixed-length time window by using the heel strike (HS) event moment of that gait cycle as the reference zero point (t=0) for all data segments corresponding to "abnormal pressure segments," "abnormal fluctuation segments," and "abnormal electromyographic segments" identified within the same gait cycle. This allows events from different modalities (such as the occurrence of abnormal fluctuations or the moment of electromyographic inhibition) to be compared on the same gait cycle time axis.

[0141] S132, determine the coexistence rate of abnormal pressure segment, abnormal fluctuation segment and abnormal electromyography segment in a single gait cycle.

[0142] Calculate the co-occurrence rate (CR). For a complete gait cycle, calculate the proportion of the total gait cycle duration in which the abnormal pressure segment, abnormal fluctuation segment, and abnormal electromyographic segment coexist. Simply put, it's the length of time these three segments appear together.

[0143] CR = (T_press∩T_fluct∩T_emg) / T_gait, where T_press is the total duration of abnormal pressure segments within the gait cycle, T_fluct is the total duration of abnormal fluctuation segments, T_emg is the total duration of abnormal electromyography segments, and T_gait is the total duration of the entire gait cycle.

[0144] Structural abnormalities exhibit stable and coordinated patterns; even when abnormal, their aberrant features (pressure, fluctuation, electromyography) typically remain synchronized and coexist persistently (response rate (CR) may be high). Pain compensation patterns are unstable and highly volatile; the aberrant features of the three modalities may not appear simultaneously, and their duration may be shorter (CR may be low). The coexistence rate is not simply a statistical measure of the presence or absence of a signal, but rather a measurement of the degree of temporal overlap among the three.

[0145] S133, determine the time difference of interval delay between the abnormal fluctuation segment and the abnormal electromyography segment based on the starting point of the abnormal fluctuation segment and the starting point of the abnormal electromyography segment.

[0146] Calculating the Inter-Segment Delay (ISD) is another key indicator for measuring pain compensation, representing the physiological delay caused by the reaction speed of nerve conduction and the avoidance reflex. This step calculates the start time t_fluct_start of the detected "abnormal fluctuation segment" and the start time t_emg_start of the detected "abnormal electromyographic segment" within each gait cycle. The time difference between the two is calculated as Δt_delay = |t_emg_start - t_fluct_start|. In healthy gait or structural abnormalities, joint movements (including fluctuations) and muscle activity are coordinated (co-activated or temporally fixed), and Δt_delay is usually small and stable. The pain-induced avoidance reflex is a rapid neural feedback mechanism aimed at minimizing movement-induced pain stimuli. This mechanism may lead to immediate changes (adjustments) in muscle activity patterns upon sensing pressure / impact or movement-induced pain, but these adjustments are either "corrections" of already occurring movements or "inhibitions" of movements that are about to exacerbate pain. Therefore, t_emg_start often occurs after or almost simultaneously with t_fluct_start, but this time difference Δt_delay can theoretically be physiologically shortened when the pain reflex is active (the nerve reflex intervenes as quickly as possible to protect). However, in actual measurements, if Δt_delay is very small (e.g., <20ms), it may also indicate a strong correlation. This scheme assumes and seeks whether Δt_delay is below a preset threshold as one of the indicators of a faster neuromuscular response during pain (S134 judgment condition).

[0147] S134, when the coexistence rate is higher than the preset association threshold and the interval delay time difference is lower than the preset delay threshold, the spatiotemporal association characteristics are determined according to the ratio of the coexistence rate and the interval delay time difference.

[0148] The coexistence rate (CR) must be higher than a preset association threshold (CR_threshold, e.g., 0.5, indicating that the three elements coexist for more than half the gait cycle time), and the interval delay time difference (Δt_delay) must be lower than a preset delay threshold (Δt_threshold, e.g., 30ms). If both conditions are met, it indicates that there is a strong spatiotemporal correlation characteristic within the current gait cycle. Then, the ratio (Ratio = CR / Δt_delay) is calculated based on the CR and Δt_delay that meet the conditions. This ratio represents the spatiotemporal correlation characteristic value of that gait cycle.

[0149] A combination of highly overlapping abnormal patterns (high CR) and extremely rapid neuromuscular responses (very low Δt_delay) is more likely to occur when pain reflexes are strongly involved (although the overall coexisting CR may not be as high as that of stable structural abnormalities, once it occurs, the neural response is extremely rapid, and the CR / Δt_delay ratio will be large). While the coexistence of all three structural abnormalities may result in higher CR, their Δt_delay (equivalent to the inherent neuromuscular coordination delay) is not necessarily abnormally short (the ratio may not be significantly large).

[0150] S135, determine the gait coupling index based on spatiotemporal correlation characteristics.

[0151] Preferably, the gait coupling index is positively correlated with the ratio of coexistence rate to interval delay time difference.

[0152] The Gait Coupling Index (GCI) is determined by calculating the average (or other aggregated statistics such as maximum / median) of the Ratio values ​​of all gait cycles satisfying the S134 condition over the entire analysis period (e.g., 30 seconds or 10 steps). This aggregated value is the final GCI (e.g., GCI = mean(Ratio)). Furthermore, the SCI is designed to be positively correlated with the co-occurrence rate (CR) (the larger the CR, the larger the numerator) and negatively correlated with the interval delay time difference (Δt_delay) (the smaller the Δt_delay, the smaller the denominator, and the larger the Ratio), ultimately resulting in a larger GCI. Therefore, a high GCI reflects a specific combination of high co-occurrence rate and extremely short neuromuscular delay, a characteristic highly suggestive of pain compensation.

[0153] The core of step S130 is to construct the GCI. Although this scheme does not explicitly calculate the traditional interstep variation coefficient (such as the standard deviation of the DTW distance of the joint trajectory), the CR indirectly reflects the stability of the coordination of the three modal abnormal features within a single gait cycle in a continuous gait cycle (a high CR indicates high synchronization and stability of the three in the current step). The variation of CR between consecutive gaits can be associated with the characteristics of small interstep variation when the structure is stable and large interstep variation when the pain fluctuates. The GCI is based on single-step correlation, and the fluctuation of the continuous Ratio and its aggregated GCI can also serve as a reference for system stability. The design of Δt_delay is to explicitly model the temporal relationship between transient fluctuations in joint dynamics (captured by t_fluct_start) and neuromuscular response (captured by t_emg_start). Very short Δt_delay is considered to be a feature of highly coordinated but biomechanically destructive pain avoidance reflexes (breaking the original smooth coordination pattern), which is different from the stable coordination under structural abnormalities.

[0154] GCI is not a simple classification, but rather an explicit model of the "human system state" by comprehensively quantifying the degree of temporal overlap (CR) of abnormal segments in the three modes of stress, movement, and electromyography, as well as the delay in the rapid response of muscles to sudden changes in joints (Δt_delay): structural abnormalities tend to be stable and coordinated (high CR, but normal Δt_delay leads to a moderate ratio); pain compensation tends to show high coordination + extremely fast response at specific points (high CR + extremely low Δt_delay leads to a large ratio), and fluctuations occur (interstep GCI fluctuations may also be large).

[0155] CR and Δt_delay have clear biomechanical interpretations, and Ratio and the final GCI also have a logical relationship, which significantly improves the interpretability of the model, allowing doctors to understand the basis of the diagnosis (e.g., "because in these steps, the three factors of abnormal pressure, joint tremor and medial muscle inhibition are highly overlapping in time, and the muscle response is 20ms faster than expected").

[0156] In S130, the identified anomalous data segments (pressure, fluctuation, electromyography) are aligned temporally (S131) ​​to ensure the comparability of multimodal signals. In S32, the coexistence rate (CR) is calculated to quantify the temporal synchronicity (stability index) of anomalous pressure, sudden joint fluctuations, and pain-related muscle activity patterns within a single gait cycle. In S133, the time difference Δt_delay is calculated (S133) to explicitly model the time difference between the onset of sudden joint dynamic fluctuations and the onset of pain-related electromyographic activity patterns (neural response speed index). The design of Δt_delay directly targets the core of pain compensation—rapid neural reflex adjustment. When the dual high correlation condition (high CR + low Δt_delay) is met (S134), the spatiotemporal correlation feature value Ratio is calculated, which comprehensively characterizes the stability of the anomalous condition and the agility of the neural response. By aggregating the Ratio values ​​of multiple gaits, the gait coupling index GCI is obtained (S135), which is designed to be positively correlated with CR and negatively correlated with Δt_delay. A high GCI value strongly indicates that the pain-related neuromuscular response is extremely rapid and powerfully synchronized after the occurrence of sudden joint fluctuations (high coexistence + fast response), which is the key biomechanical characteristic of pain-compensating gait. Moreover, GCI is a statistical aggregation of multiple gait cycles that meet the strong correlation condition, which can tolerate noise interference from individual gait to a certain extent, making GCI robust.

[0157] S140, if the gait coupling index exceeds the preset coupling threshold, output a diagnosis of pain compensatory abnormality.

[0158] The gait coupling index GCI is greater than the preset coupling threshold GCI_threshold.

[0159] Preferably, S140 includes: S141, if an abnormal electromyographic segment exists and the gait coupling index exceeds a preset coupling threshold, then output a diagnosis of pain compensation abnormality.

[0160] The meaning of GCI>GCI_threshold: This high index value indicates that within the analysis window (e.g., 10 consecutive steps), multiple strongly spatiotemporally correlated events occurred (i.e., CR>CR_threshold and Δt_delay<Δt_threshold). This combination pattern—abnormal plantar pressure, sudden violent joint fluctuations, and abnormal activation of specific muscles—highly overlaps in time, and the electromyographic response after joint fluctuation is extremely rapid is a typical characteristic of neurogenic pain avoidance. The body attempts to instantly escape or reduce the load of the pain stimulus by suddenly and unnaturally adjusting the joint (TKF) and rapidly inhibiting / activating muscles (e.g., inhibiting the tibialis posterior and activating the peroneus longus) at critical time points (e.g., the moment of impact).

[0161] S141 provides a key enhancement condition: if the identified abnormal electromyographic segment data are also present, the diagnostic confidence is improved.

[0162] Electromyography (EMG) signals are the most direct electrophysiological evidence of nerve reflexes induced by pain. Inhibition of the tibialis posterior (EMG↓) reflects a neurological "shutdown" to avoid stretching the inflamed plantar fascia or medial joint capsule; abnormal activation of the peroneus longus (EMG↑) is a compensatory "activation" to forcefully evert the foot and quickly shift the center of gravity away from the painful area. Without this specific EMG pattern, even with joint pulsation (TKF) and pressure abnormalities, it may simply be other disturbances (such as uneven road surfaces or transient imbalances) rather than specific pain.

[0163] Diagnostic conclusion generation: When both GCI > GCI_threshold and abnormal electromyographic segments are present (preferably), the processing module is certain that the user is in a pain-driven compensatory state. The diagnostic result includes: "Etiological type: neuroreflexive pain avoidance compensation", and may be associated with inferred potential pain sites (e.g., proximal origin of plantar fascia, medial arch, heel fat pad).

[0164] Traditional methods cannot distinguish whether changes in joint kinematics are due to structural stability or dynamic fluctuations in pain. The GCI threshold mechanism, combined with electromyographic verification (S141), explicitly captures the unique "time-varying, discontinuous, neurally driven" dynamic characteristics of pain, avoiding misdiagnosis as structural abnormalities.

[0165] S141 provides a higher level of confidence for diagnosis. In addition to meeting the core condition of S140 (GCI > GCI_threshold), it also requires the presence of abnormal electromyographic (EMG) segment data. This is because EMG signals (inhibition of the tibialis posterior muscle or overactivation of the peroneus longus muscle) are among the most direct evidence of pain reflexes.

[0166] Diagnostic Conclusion: When the above conditions are met, the processing module outputs the diagnostic result as "abnormal pain compensation." This diagnosis reflects that the user's gait abnormality is likely due to neuroreflexive pain avoidance compensation behavior (transient kinematic fluctuation, TKF) caused by pain in the foot (especially the medial region). The sudden fluctuations in joint movement are highly correlated in time and space with rapid electromyographic inhibition / abnormal activation, which is the biomechanical basis for this diagnosis.

[0167] S150, if there is an abnormal pressure segment and the gait coupling index does not exceed the preset coupling threshold, then output a structural anomaly diagnosis.

[0168] Preferably, step S150 includes sub-steps S151 to S152:

[0169] S151, Determine the inter-step variation data corresponding to the abnormal pressure segment based on the outward velocity sequence in the pressure center trajectory corresponding to the abnormal pressure segment;

[0170] S152, if the gait coupling index does not exceed the preset coupling threshold, the abnormal pressure segment appears in a preset number of consecutive gait cycles, and the inter-step variation coefficient corresponding to the abnormal pressure segment is lower than the preset variation upper limit, then output a structural abnormality diagnosis.

[0171] An abnormal pressure segment was identified (the plantar pressure characteristics meet the requirements), but the gait coupling index (GCI) was less than the preset coupling threshold (GCI_threshold) (the high GCI characteristic value for pain compensation was not achieved).

[0172] The specific implementation of the logical judgment in S150 is as follows:

[0173] If there is identified abnormal pressure segment data, and the gait coupling index (GCI) ≤ preset coupling threshold (GCI_threshold), and the inter-step variation coefficient (CoV) calculated based on the COP outward movement velocity sequence corresponding to the abnormal pressure segment < preset variation upper limit (CoV_upper), and the above-mentioned abnormal pressure segment continues to exist in the preset number of consecutive gait cycles, then a structural abnormality diagnosis is output.

[0174] Among them, the implementation of S151 includes: extracting the sequence of the COP outward movement speed (v_x) in the mid / late stance phase of each gait cycle in a continuous preset number of gait cycles (such as the last 5 steps) (for example, the speed values of 10 sampling points in each gait cycle); calculating the standard deviation (σ_v) and the mean value (μ_v) of the v_x sequence within each gait cycle; calculating the coefficient of variation CoV_i = (σ_v / μ_v) * 100% for this gait cycle; calculating the average value (CoV_mean) of CoV_i for these consecutive preset number of gaits; if CoV_mean < CoV_upper (for example, 15%), the condition is satisfied.

[0175] By confirming that the plantar pressure pattern indeed shows an abnormal decrease on the medial side and an increase on the lateral side, an abnormal pressure segment is determined to exist; by GCI <= GCI_threshold, it indicates that the "high coexistence rate + extremely short nerve delay" combination pattern unique to pain compensation has not been detected (the Ratio value is not high), excluding pain as the main cause; by CoV < CoV_upper, it is proved that the abnormal COP outward movement speed (v_x) shows high repeatability and stability in consecutive walking cycles. This low inter-step variation (CoV) is a key sign of an adaptive compensatory gait caused by structural anatomical changes (such as fixed arch collapse, continuous tendon failure, and abnormal bony alignment); the abnormal pressure appears stably in multiple consecutive steps rather than fluctuating occasionally, verifying the fixity of the problem.

[0176] Diagnosis conclusion generation: When the above multiple conditions are simultaneously met, the processing module diagnoses it as a structural anomaly. The diagnosis result includes: "Etiology type: driven by persistent anatomical / functional changes", and may be associated with potential structural problem sites (such as: posterior tibial tendon insufficiency, secondary genu varum, excessive subtalar joint pronation). This diagnosis reflects that the user has a fixed compensatory gait (such as continuous COP outward movement and specific pressure distribution abnormalities) caused by stable and persistent anatomical structure or functional changes (such as arch collapse, ligament laxity, muscle weakness), and lacks the high GCI (rapid nerve response) feature unique to pain compensation.

[0177] S150 overcomes the fundamental defect of the prior art that it is unable to distinguish between fixed structural changes and transient pain responses. Through the double verification of combining a low GCI value + low inter-step variation (CoV), the stable and repetitive pressure / motion pattern is locked as a clear signal of a structural anatomical / functional disorder, avoiding misidentifying it as a transient pain avoidance behavior and delaying corrective intervention.

[0178] Preferably, the method further includes S160 - S170:

[0179] S160, if the diagnosis is a compensatory abnormality of pain, then the mechanical output of the arch area of ​​the insole is suppressed and the cushioning of the heel area of ​​the insole is enhanced.

[0180] When a pain compensatory abnormality is diagnosed, the processing module sends control commands to the actuators of the insole (such as airbags, shape memory alloys, and solenoid valves).

[0181] Inhibit biomechanical output in the arch area: Reduce or completely shut down the support / lifting force provided by arch support components (such as the medial wedge, air bladder, etc.). Abnormal postures at this time (such as excessive eversion) are protective actions taken by the body to avoid painful areas (often in the heel or medial arch). If the system incorrectly imposes arch support in an attempt to "correct" the posture, it may directly compress or stimulate pain sources (such as the origin of the plantar fascia, the inflamed area of ​​the heel fat pad, and the medial joint capsule), thus worsening the pain and causing secondary injury.

[0182] Enhance the cushioning of the heel area of ​​the insole: Simultaneously (or separately) increase the cushioning capacity of the heel area of ​​the insole (e.g., with air cushions, softened material areas, etc.). This reduces the impact on the painful heel area at the moment of heel strike (HS), helping to reduce painful stimuli and lessen the intensity and frequency of the avoidance reflex. This supports a more natural adaptation of the body (rather than forced correction), aiming to allow the body to return to a healthier gait naturally after pain relief.

[0183] S170, if the diagnosis result is a structural abnormality, then increase the mechanical output of the arch area to the preset correction requirement level.

[0184] Enhance the mechanical output of the arch area to the preset correction requirement level: Gradually and controllably increase the support force or lift height of the arch support parts (medial wedge, air bladder, etc.) to the preset correction force for the user's structural problems (such as the arch support force required to correct flat feet, which is usually personalized based on the user's weight and degree of flatness).

[0185] Structural problems require continuous biomechanical support to correct biomechanical misalignment, mitigate deformation trends, and reduce pressure on compensatory structures (such as the medial aspect of the knee). Appropriate biomechanical support (such as support for collapsed arches) can improve arch function, correct COP lateral displacement, and thus alleviate secondary problems such as varus stress. Early, moderate intervention helps prevent further deformity and secondary injuries (such as knee osteoarthritis).

[0186] This application embodiment acquires plantar pressure time-series data, joint angle change data, and electromyography (EMG) signals within a preset time window after heel strike. Based on the plantar pressure time-series data, joint angle change data, and EMG signals, it identifies data corresponding to abnormal pressure segments, abnormal fluctuation segments, and abnormal EMG segments. Abnormal pressure segments reflect persistent or fluctuating abnormalities in plantar pressure distribution, such as a significant decrease in medial pressure and a significant increase in lateral pressure during the mid / late support phase. Abnormal fluctuation segments reflect transient and discontinuous fluctuations in joint dynamic parameters, such as sudden and drastic changes in joint angle or angular acceleration at the moment of heel strike or at the point of maximum weight-bearing during the mid-support phase. Abnormal EMG segments reflect abnormal muscle activity triggered by neuroreflexive pain avoidance responses, such as abrupt changes in EMG signal intensity or temporal patterns caused by medial arch or heel pain. Therefore, the spatiotemporal correlation characteristics of the data corresponding to abnormal pressure segments, abnormal fluctuation segments, and abnormal EMG segments can reflect the lower limb gait pattern. The inherent coordination differences between the gait system and the gait coupling index, determined based on spatiotemporal correlation characteristics, reflect the overall stability of the gait system. A high index value indicates time-varying fluctuations and coordination breakdowns, while a low index value indicates smooth continuity and overall synergy. In diagnostic analysis, the gait coupling index is compared with a preset coupling threshold to distinguish between structural abnormalities and pain compensation abnormalities. When the gait coupling index exceeds the preset coupling threshold, it indicates that the user is experiencing pain compensation, i.e., gait instability caused by dynamic pain avoidance response. When there is an abnormal pressure segment and the gait coupling index does not exceed the preset coupling threshold, it indicates that the user is experiencing structural abnormalities, i.e., the user is anatomically changing the fixed compensation pattern driven by the gait system. This allows for a more accurate differentiation of lower limb gait abnormality types (structural lesions and pain compensation responses).

[0187] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0188] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for designing insoles based on foot physiology, characterized in that, The method includes: The insole is divided into the heel area, arch area, and forefoot area to obtain a partitioned model; An upper pad and a lower pad are injection molded at the corresponding positions of the partition model. A pressure sensor module is provided between the upper pad and the lower pad. Multiple air bladder cavities are pre-set in the heel area, arch area and forefoot area respectively. An interlocking groove is provided in the arch area. A processing module is embedded in the lower pad fitting groove. The processing module is used to connect to the pressure sensor module and is also used to control the air pressure in the airbag cavity. The upper pad, the functional layer containing the pressure sensor module, and the lower pad are stacked together, and the partition edges are fused together by hot pressing. Configure the joint motion capture module and the electromyography sensing module with the processing module respectively.

2. The method as described in claim 1, characterized in that, The processing module is also wirelessly connected to the joint motion capture module and the electromyography (EMG) sensing module, respectively. The joint motion capture module is attached to the knee joint to monitor the dynamic changes in the flexion and extension angles of the knee joint, obtain joint angle change data, and send the joint angle change data to the processing module. The EMG sensing module is attached to the belly of the gastrocnemius muscle to monitor EMG signals, determine the EMG signals within a preset time window after heel strike, and send the determined EMG signals to the processing module, so that the processing module can monitor the foot status of the target user based on the joint angle change data and the EMG signals within the preset time window after heel strike.

3. The method as described in claim 2, characterized in that, The foot condition monitoring method executed by the processing module includes: Acquire temporal data of plantar pressure, joint angle change data, and electromyographic signals within a preset time window after heel strike; Based on the plantar pressure time-series data, joint angle change data, and electromyography signals, the data corresponding to the abnormal pressure segment, abnormal fluctuation segment, and abnormal electromyography segment are identified respectively. Gait coupling index is determined based on the spatiotemporal correlation characteristics of data corresponding to abnormal pressure segments, abnormal fluctuation segments, and abnormal electromyography segments. If the gait coupling index exceeds the preset coupling threshold, a diagnosis of pain compensatory abnormality is output. If an abnormal pressure segment exists and the gait coupling index does not exceed the preset coupling threshold, a structural anomaly diagnosis is output.

4. The method as described in claim 3, characterized in that, The foot condition monitoring method also includes: If the diagnosis is the aforementioned pain compensation abnormality, then the mechanical output of the arch area of ​​the insole is suppressed, and the cushioning of the heel area of ​​the insole is enhanced. If the diagnosis result is the structural abnormality diagnosis, then the mechanical output of the arch area is increased to the preset correction requirement level.

5. The method as described in claim 4, characterized in that, The arch region includes the medial longitudinal arch region and the lateral metatarsal head region; the data for identifying abnormal pressure segments, abnormal fluctuation segments, and abnormal electromyographic segments based on plantar pressure time-series data, joint angle change data, and electromyographic signals respectively includes: The pressure center trajectory is determined based on the plantar pressure time-series data. The time-series data of plantar pressure are used to determine whether they represent the first and second abnormal phenomena. Detect whether the pressure center trajectory represents the third and fourth anomalies; If the plantar pressure time series data simultaneously represents the first abnormal phenomenon and the second abnormal phenomenon, and the pressure center trajectory represents either the third abnormal phenomenon or the fourth abnormal phenomenon, then the data in the plantar pressure time series data that simultaneously represents the first abnormal phenomenon and the second abnormal phenomenon, and the pressure center trajectory represents either the third abnormal phenomenon or the fourth abnormal phenomenon, is determined as the data corresponding to the abnormal pressure segment. Based on joint angle change data and electromyography signals, the data corresponding to abnormal fluctuation segments and abnormal electromyography segments are identified respectively. The time period during which the pressure center trajectory represents either the third or the fourth abnormal phenomenon is defined as the trajectory abnormal period. The second derivative of the joint angle change data is extracted, and the second derivative of the joint angle change data is used to characterize angular acceleration; The joint angle change data during the period when the absolute value of the angular acceleration is greater than the preset acceleration threshold during the abnormal trajectory period are determined as the data corresponding to the abnormal fluctuation segment; Within the time window corresponding to the abnormal fluctuation period, the period when the electromyographic signal corresponding to the tibialis posterior muscle is lower than the preset benchmark electromyographic value, or the period when the electromyographic value corresponding to the peroneus longus muscle is higher than the benchmark electromyographic value, is determined as the abnormal electromyographic segment. The electromyographic signals within the abnormal electromyographic segment are identified as the data corresponding to the abnormal electromyographic segment; wherein... The first abnormal phenomenon includes the fact that the pressure corresponding to the medial longitudinal arch area of ​​the insole in the foot pressure time series data is lower than the first proportion value of the dynamic reference pressure corresponding to the first preset time period within a first preset time period. The dynamic reference pressure corresponding to the first preset time period is the average value of multiple pressure values ​​of the insole within the first preset time period. The second abnormal phenomenon includes the pressure corresponding to the lateral metatarsal head region of the insole in the plantar pressure time series data being higher than the second proportional value of the dynamic reference pressure in the second preset time period, where the first proportional value is less than the second proportional value. The third abnormal phenomenon includes the instantaneous outward movement speed of the pressure center trajectory corresponding to the middle and late support phases of the gait cycle exceeding a preset outward movement speed threshold. The fourth abnormal phenomenon includes the outward displacement of several consecutive trajectory points of the preset number of pressure center trajectory points corresponding to the middle and late support phases of the gait cycle, which exceeds the trajectory range corresponding to the preset maximum eversion angle of the foot.

6. The method as described in claim 5, characterized in that, The determination of the gait coupling index based on the spatiotemporal correlation features of data corresponding to abnormal pressure segments, abnormal fluctuation segments, and abnormal electromyography segments includes: The abnormal pressure segment, abnormal fluctuation segment, and abnormal electromyography segment identified in the gait cycle are time-aligned using the heel strike event in the gait cycle as the synchronization point. Determine the coexistence rate of abnormal pressure segment, abnormal fluctuation segment, and abnormal electromyographic segment in a single gait cycle; The time difference between the abnormal fluctuation segment and the abnormal electromyography segment is determined based on the starting point of the abnormal fluctuation segment and the starting point of the abnormal electromyography segment; When the coexistence rate is higher than a preset association threshold and the interval delay time difference is lower than a preset delay threshold, the spatiotemporal association feature is determined based on the ratio of the coexistence rate to the interval delay time difference. The gait coupling index is determined based on the spatiotemporal correlation characteristics.

7. The method as described in claim 6, characterized in that, The gait coupling index is positively correlated with the ratio of the coexistence rate to the time difference of the interval delay.

8. The method as described in claim 7, characterized in that, If the gait coupling index exceeds a preset coupling threshold, a diagnosis of pain compensatory abnormality is output, including: If the abnormal electromyographic segment is present and the gait coupling index exceeds a preset coupling threshold, a diagnosis of pain compensation abnormality is output.

9. The method as described in claim 8, characterized in that, If an abnormal pressure segment exists and the gait coupling index does not exceed a preset coupling threshold, a structural anomaly diagnosis is output, including: The inter-step variation data corresponding to the abnormal pressure segment are determined based on the outward velocity sequence in the pressure center trajectory corresponding to the abnormal pressure segment. If the gait coupling index does not exceed the preset coupling threshold, the abnormal pressure segment appears in a preset number of consecutive gait cycles, and the inter-step variation coefficient corresponding to the abnormal pressure segment is lower than the preset variation upper limit, then a structural abnormality diagnosis is output.

10. A foot insole design system based on foot physiology, characterized in that, The system includes: The partitioning module is used to divide the insole into the heel area, arch area, and forefoot area to obtain a partitioning model. The injection molding module is used to inject and form an upper pad and a lower pad at the corresponding positions of the partition model. A pressure sensor module is provided between the upper pad and the lower pad. Multiple air bladder cavities are pre-set in the heel area, arch area and forefoot area, and a fitting groove is provided in the arch area. An embedding module is used to embed a processing module in the fitting groove of the lower pad. The processing module is used to connect to the pressure sensor module and is also used to control the air pressure in the airbag cavity. The stacking module is used to stack an upper pad, a functional layer containing the pressure sensor module, and a lower pad, and fuses the partition edges together by hot pressing. A configuration module is used to configure the joint motion capture module and the electromyography sensing module with the processing module, respectively.

Images