Children gait abnormity recognition feedback method and system based on plantar pressure data

Abstract

The invention discloses a child gait anomaly recognition feedback method and system based on plantar pressure data, and relates to the technical field of gait anomaly recognition, and the method comprises the steps: collecting the plantar pressure data of a target wearer through a sensor array, and carrying out the feature extraction of multi-dimensional gait feature parameters; comparing and analyzing the model with a child gait development standard model, and recognizing a gait abnormal mode; carrying out risk assessment on the target shoe wearer and dividing risk levels; performing gait analysis on the target wearer to generate an individualized gait analysis report; and performing gait correction analysis on the target wearer according to the individualized gait analysis report, and matching to generate a correction training course scheme. The technical problems that in the prior art, child gait evaluation lacks quantitative standards, detection intervention is disjointed, and personalized analysis based on a dynamic development model is lacked are solved, and the technical effects of achieving objective, accurate and intelligent recognition of child gait abnormity and improving the child gait health management level are achieved.

Description

Technical Field

[0001] This invention relates to the field of gait abnormality recognition technology, specifically to a method and system for recognizing and responding to gait abnormalities in children based on plantar pressure data. Background Technology

[0002] Normal gait patterns in children rely on the coordinated development of foot structure and lower limb function. However, during the critical developmental period of 3-12 years old, approximately 15% to 20% of children exhibit varying degrees of gait abnormalities, including in-toeing, out-toeing, flat-footed, and equinus gait patterns. Failure to identify and intervene promptly can lead to foot pain, joint wear and tear, and decreased motor skills. Traditional gait assessment in children primarily relies on clinical observation and two-dimensional video analysis, which suffers from limitations such as strong subjectivity and insufficient quantification. While the development of biomechanical sensing technology has provided objective data support for gait analysis, it largely focuses on adult rehabilitation or athlete training, lacking proprietary analytical models tailored to the characteristics of children's growth and development. Furthermore, a closed-loop feedback system encompassing detection, assessment, and intervention has not been established.

[0003] Therefore, current technologies for assessing children's gait suffer from several technical problems, including a lack of quantitative standards, a disconnect between detection and intervention, and a lack of personalized analysis based on dynamic developmental models. Summary of the Invention

[0004] This application provides a method and system for identifying and responding to gait abnormalities in children based on plantar pressure data. This solves the technical problems in the prior art, such as the lack of quantitative standards for gait assessment, the disconnect between detection and intervention, and the lack of personalized analysis based on dynamic developmental models. It achieves the technical effect of realizing objective, accurate, and intelligent identification of gait abnormalities in children and improving the level of gait health management for children.

[0005] This application provides a method for identifying and responding to gait abnormalities in children based on plantar pressure data. The method includes: collecting plantar pressure data of a target shoe wearer through a sensor array; extracting features from the plantar pressure data to obtain multidimensional gait feature parameters; synchronizing the multidimensional gait feature parameters to a standard model of children's gait development for comparative analysis to identify gait abnormal patterns; conducting a risk assessment of the target shoe wearer based on the gait abnormal patterns and setting an abnormal risk level according to the assessment results; performing gait analysis of the target shoe wearer based on the gait abnormal patterns and the abnormal risk level to generate an individualized gait analysis report; and performing gait correction analysis of the target shoe wearer based on the individualized gait analysis report to generate a corrective training program.

[0006] In a possible implementation, the child gait anomaly identification feedback method based on plantar pressure data further performs the following processing: synchronously collecting pressure values ​​of each zone of the sole and foot movement posture data through a sensor array including a pressure sensing unit and an inertial measurement unit; transmitting the pressure values ​​of each zone of the sole and the foot movement posture data to a mobile terminal to activate a motion state detection module; determining whether the target shoe wearer has entered a motion state through the motion state detection module; when the target shoe wearer enters a motion state, activating the pressure sensing unit to collect data from the target shoe wearer at a first sampling frequency to obtain an initial pressure data stream, activating the inertial measurement unit to collect data from the target shoe wearer at a second sampling frequency to obtain an initial posture data stream; determining the target movement mode based on the initial pressure data stream and the initial posture data stream; dynamically adjusting the sensor array based on the target movement mode, and collecting the plantar pressure data of the target shoe wearer based on the adjustment result.

[0007] In a possible implementation, the child gait anomaly identification feedback method based on plantar pressure data further performs the following processing: retrieving the gait record data log of the target shoe wearer, performing periodic repetitive analysis based on the gait record data log to determine the gait cycle; segmenting the plantar pressure data according to the gait cycle to obtain a single gait cycle pressure dataset; calculating the pressure center based on the single gait cycle pressure dataset to obtain pressure center trajectory data; performing time-domain analysis on the pressure center trajectory data to obtain pressure time-series data; performing frequency-domain transformation on the pressure time-series data to obtain frequency-domain energy distribution characteristics; performing pressure calculation based on the single gait cycle pressure dataset according to the sensor array to obtain pressure peak parameters and pressure mean parameters; calculating the contact duration of the target shoe wearer based on the pressure peak parameters and the pressure mean parameters to obtain foot-ground contact time data; and fusing the frequency-domain energy distribution characteristics, the foot-ground contact time data, the pressure peak parameters, and the pressure mean parameters to construct the multidimensional gait feature parameters.

[0008] In a possible implementation, the feedback method for identifying gait abnormalities in children based on plantar pressure data further performs the following processing: synchronizing the multidimensional gait feature parameters to a standard model of gait development in children for deviation calculation to obtain the deviation degree of multiple feature parameters; performing gait influence analysis on the multidimensional gait feature parameters to determine multiple influencing factors; assigning weights to the multidimensional gait feature parameters based on the multiple influencing factors to generate multiple weight parameters; performing a fusion calculation based on the deviation degree of the multiple feature parameters and the multiple weight parameters to obtain a comprehensive gait abnormality index; and identifying multiple gait abnormalities based on the comprehensive gait abnormality index to determine the gait abnormality pattern.

[0009] In a possible implementation, the gait abnormality identification and feedback method for children based on plantar pressure data further performs the following processing: collecting raw plantar pressure data from healthy children of multiple age groups and genders; performing feature analysis based on the raw plantar pressure data; setting multidimensional standard gait feature parameters according to multiple data features; performing normalization analysis based on the multidimensional standard gait feature parameters to generate normalization analysis results; grouping the normalization analysis results according to multiple age groups and genders to obtain multiple feature parameter groups; performing fluctuation analysis on the multiple feature parameter groups, extracting fluctuation extreme values, and dividing the normal fluctuation range of features according to the fluctuation extreme values; structurally storing the normal fluctuation range of features according to multiple age groups and genders to obtain a parameter data aggregation library; and using the parameter data aggregation library as a standard threshold to construct the standard model of children's gait development.

[0010] In a possible implementation, the gait anomaly identification feedback method for children based on plantar pressure data further performs the following processing: deviation calculation is performed based on the deviation degree of the gait anomaly pattern according to multiple feature parameters to obtain a deviation amplitude value; multiple risk thresholds are set, and the deviation amplitude value is matched with the multiple risk thresholds to determine a target risk threshold range; a risk assessment is performed based on the target risk threshold range to generate a risk assessment score; the risk assessment score is weighted and corrected based on the multiple weight parameters to generate an anomaly assessment score; and the anomaly assessment score is added to the assessment result.

[0011] In a possible implementation, the gait abnormality identification and feedback method for children based on plantar pressure data further performs the following processing: extracting the target abnormality pattern and target abnormality risk level based on the individualized gait analysis report; using the target abnormality pattern and target abnormality risk level as the main index, introducing the user profile information of the target shoe wearer, and using the user profile information as auxiliary constraints; mapping the main index to the correction training course library according to the auxiliary constraints for multi-condition matching and filtering to generate an initial course unit set; optimizing and sorting the initial course unit set to construct the correction training course scheme.

[0012] In a possible implementation, the gait abnormality identification and feedback method for children based on plantar pressure data further performs the following processing: mapping the master index to the correction training course library for preliminary matching to generate multiple candidate course sets; performing user preference analysis based on the user profile information to generate user preference parameters; filtering the multiple candidate course sets according to the user preference parameters to generate multiple target candidate courses; performing association analysis based on the master index and the correction training course library to obtain multiple association strengths; sorting the multiple target candidate courses in descending order of priority according to the multiple association strengths, selecting the first-ranked target candidate course as the core course, and using the remaining target candidate courses as complementary auxiliary training courses; combining the core course and the complementary auxiliary training courses to construct an initial course unit set.

[0013] In a possible implementation, the gait abnormality identification and feedback method for children based on plantar pressure data further performs the following processing: distributing the corrective training course plan to a user terminal; receiving the plan feedback parameters from the target shoe wearer through the user terminal and generating a plan modification feedback instruction; updating the corrective training course plan through the plan modification feedback instruction and generating an optimized corrective training course plan; executing the optimized corrective training course plan for course tracking and generating a closed-loop optimization of the corrective training course plan based on the corrective training effect.

[0014] This application also provides a feedback system for identifying gait abnormalities in children based on plantar pressure data. The system includes: a feature extraction module, used to collect plantar pressure data of the target shoe wearer through a sensor array, extract features from the plantar pressure data, and obtain multidimensional gait feature parameters; a comparison and analysis module, used to synchronize the multidimensional gait feature parameters to a standard model of children's gait development for comparison and analysis, and identify gait abnormal patterns; a risk assessment module, used to conduct a risk assessment of the target shoe wearer based on the gait abnormal patterns, and set an abnormal risk level according to the assessment results; a gait analysis module, used to conduct gait analysis of the target shoe wearer based on the gait abnormal patterns and the abnormal risk level, and generate an individualized gait analysis report; and a correction plan generation module, used to conduct gait correction analysis of the target shoe wearer based on the individualized gait analysis report, and generate a correction training course plan.

[0015] This application proposes a method and system for identifying and responding to gait abnormalities in children based on plantar pressure data. The system uses a sensor array to collect plantar pressure data from the target shoe wearer and extracts multi-dimensional gait feature parameters. This data is then compared and analyzed with a standard model of children's gait development to identify gait abnormality patterns. A risk assessment is conducted on the target shoe wearer, and a risk level is assigned. Gait analysis is performed to generate an individualized gait analysis report. Based on the individualized gait analysis report, gait correction analysis is conducted, and a corrective training program is generated. This addresses the technical problems in existing technologies, such as the lack of quantitative standards for children's gait assessment, the disconnect between detection and intervention, and the lack of personalized analysis based on dynamic developmental models. It achieves the technical effect of objective, accurate, and intelligent identification of gait abnormalities in children, thereby improving the level of children's gait health management. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings of the embodiments of this disclosure will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0017] Figure 1 A schematic diagram of the gait abnormality identification and feedback method for children based on plantar pressure data provided in this application embodiment.

[0018] Figure 2 A schematic diagram of the structure of a feedback system for identifying abnormal gait in children based on plantar pressure data, provided in an embodiment of this application.

[0019] Figure labeling: Feature extraction module 10, comparison analysis module 20, risk assessment module 30, gait analysis module 40, correction plan generation module 50. Detailed Implementation

[0020] To further illustrate the technical means and effects adopted by the present invention in order to achieve the intended purpose, the following detailed description is provided in conjunction with the accompanying drawings and preferred embodiments, based on the specific implementation methods, structures, features and effects of the present invention.

[0021] This application provides a method for identifying and responding to gait abnormalities in children based on plantar pressure data, such as... Figure 1 As shown, the method includes: Step S100: Collect plantar pressure data of the target shoe wearer through a sensor array, extract features from the plantar pressure data, and obtain multidimensional gait feature parameters.

[0022] Step S100 further includes step S101, synchronously collecting pressure values ​​of each zone of the sole and foot movement posture data through a sensor array including a pressure sensing unit and an inertial measurement unit; step S102, transmitting the pressure values ​​of each zone of the sole and the foot movement posture data to a mobile terminal to activate the motion state detection module; step S103, determining whether the target shoe wearer has entered a movement state through the motion state detection module; step S104, when the target shoe wearer enters a movement state, activating the pressure sensing unit to collect data from the target shoe wearer at a first sampling frequency to obtain an initial pressure data stream, and activating the inertial measurement unit to collect data from the target shoe wearer at a second sampling frequency to obtain an initial posture data stream; step S105, determining the target movement mode based on the initial pressure data stream and the initial posture data stream; step S106, dynamically adjusting the sensor array based on the target movement mode, and collecting the sole pressure data of the target shoe wearer based on the adjustment result.

[0023] Preferably, a sensor array integrated into the insole is used, including a pressure sensing unit and an inertial measurement unit, to measure the pressure distribution, movement speed, and acceleration of different areas of the sole, while simultaneously recording the pressure values ​​of different areas of the sole and the foot's motion posture parameters such as angular velocity, tilt angle, and acceleration in three-dimensional space. The pressure values ​​of each area of ​​the sole and the foot's motion posture data are transmitted to a mobile terminal, activating a pre-set motion state detection module. This module determines whether the target shoe wearer has entered a motion state, i.e., it identifies the motion state based on the received data, determining in real time whether the child is currently stationary, walking, or running. When the target shoe wearer enters a motion state, high-frequency data acquisition is performed, activating the pressure sensing unit to continuously acquire data at a first sampling frequency to form an initial pressure data stream, while simultaneously controlling the inertial measurement unit to acquire data at a second sampling frequency. Continuous sampling at different frequencies forms an initial posture data stream. The first and second sampling frequencies can be the same or different to meet the acquisition requirements of different signal types. The target movement pattern is determined based on the initial pressure and initial posture data streams. This involves real-time analysis of the acquired initial pressure and initial posture to identify the child's current specific movement pattern, such as walking slowly, walking quickly, running, or jumping. The sensor array is dynamically adjusted based on the target movement pattern. This may include adjusting the sampling frequency of the pressure sensor and inertial measurement unit, changing the filtering parameters of the data processing, and focusing on acquiring data from certain key areas or features for specific movement patterns. Finally, according to the adjusted parameter settings, the plantar pressure data of the target shoe wearer is acquired and output, thereby ensuring that the acquired data can most effectively reflect the gait characteristics in the target movement pattern.

[0024] Further, step S100 also includes step S107, retrieving the gait record data log of the target shoe wearer, performing periodic repetitive analysis based on the gait record data log to determine the gait cycle; step S108, segmenting the plantar pressure data according to the gait cycle to obtain a single gait cycle pressure dataset; step S109, calculating the pressure center based on the single gait cycle pressure dataset to obtain pressure center trajectory data; step S110, performing time-domain analysis on the pressure center trajectory data to obtain pressure time-series data; step S111, processing the pressure... Step S112: Perform frequency domain transformation on the force time series data to obtain frequency domain energy distribution characteristics; Step S113: Calculate pressure based on the single gait cycle pressure dataset according to the sensor array to obtain peak pressure parameters and average pressure parameters; Step S114: Calculate the contact time of the target shoe wearer based on the peak pressure parameters and average pressure parameters to obtain foot-ground contact time data; Step S115: Perform feature fusion on the frequency domain energy distribution characteristics, the foot-ground contact time data, the peak pressure parameters, and the average pressure parameters to construct the multidimensional gait feature parameters.

[0025] Preferably, the gait recording data log of the target shoe wearer is retrieved to obtain historical gait recording data. Periodic repetitive analysis is performed based on the gait recording data log, that is, by analyzing the repetitive patterns of pressure signals, the start and end times of a complete gait movement are determined, thereby calculating the gait cycle. The plantar pressure data is then segmented according to the gait cycle, dividing it into multiple independent and complete single-gait cycle pressure datasets. Based on the single-gait cycle pressure datasets, pressure center calculation is performed, that is, for each single-gait cycle pressure dataset, based on the pressure values ​​and position coordinates of each point on the sensor array, the pressure center of the foot is calculated throughout the entire gait cycle. The continuous movement trajectory data of the bottom pressure center point on a two-dimensional plane is used to determine the pressure center trajectory data. Then, time domain analysis is performed on the pressure center trajectory data to extract parameters such as trajectory length, average movement speed, heel touch-down and toe lift-off, etc., to obtain pressure time series data, which characterizes the law of pressure center changing with time within the gait cycle. Through fast Fourier transform and other methods, the pressure time series data is transformed into the frequency domain, that is, it is transformed from the time domain to the frequency domain. The distribution of its energy on different frequency components is analyzed to obtain the frequency domain energy distribution characteristics, which characterize the rhythmicity, smoothness and stability information of gait.

[0026] Preferably, for each single gait cycle pressure dataset, pressure calculation is performed according to the foot regions divided by the sensor array. The peak pressure and average pressure are calculated for each region throughout the entire cycle to obtain peak pressure parameters and average pressure parameters, which are used to quantify the load on anatomical regions of the foot such as the heel, arch, and forefoot. Based on a pre-set pressure threshold and combined with the peak and average pressure values ​​of each region, the time point from contact to departure from the ground in that region during the gait cycle is determined, thereby calculating the total contact time and the contact sequence of each region, and obtaining the temporal phase characteristics of the gait, such as the proportion of the support phase and swing phase and the order of load on each region. Finally, the frequency domain energy distribution characteristics, foot-ground contact time data, peak pressure parameters, and average pressure parameters are fused to determine structured multidimensional gait characteristic parameters, comprehensively characterizing the spatiotemporal characteristics, load distribution, and dynamic stability of the gait.

[0027] Step S200: Synchronize the multidimensional gait feature parameters to the standard model of children's gait development for comparative analysis and identify abnormal gait patterns.

[0028] Step S200 further includes step S201, synchronizing the multidimensional gait feature parameters to a standard model of children's gait development for deviation calculation to obtain the deviation degree of multiple feature parameters; step S202, performing gait influence analysis on the multidimensional gait feature parameters to determine multiple influencing factors, and assigning weights to the multidimensional gait feature parameters based on the multiple influencing factors to generate multiple weight parameters; step S203, performing a fusion calculation based on the deviation degree of the multiple feature parameters and the multiple weight parameters to obtain a comprehensive gait abnormality index; and step S204, performing multi-gait abnormality identification based on the comprehensive gait abnormality index to determine the gait abnormality pattern.

[0029] Preferably, the multidimensional gait feature parameters are synchronized to a standard model of gait development in children for comparative analysis. Specifically, each multidimensional gait feature parameter extracted from the target child is compared with the normal reference range of feature parameters for healthy children of the corresponding age and gender in the standard model of gait development in children. By calculating the Z-score, which is the difference from the mean divided by the standard deviation, the degree of deviation of each feature parameter from the normal range is quantified, and the deviation of multiple feature parameters is obtained. This value directly reflects the severity of the abnormality of the feature. Since the deviation of all feature parameters contributes differently to gait abnormalities, a gait influence analysis is performed on the multidimensional gait feature parameters, that is, the importance of each feature in identifying specific gait abnormality patterns, such as flat feet and pigeon toes, is assessed to determine multiple influencing factors. Then, the multidimensional gait feature parameters are weighted according to the multiple influencing factors. The more critical the feature, the greater its weight value. For example, when identifying flat feet, the average pressure in the arch area has a higher weight. This generates multiple weight parameters to ensure that the comprehensive assessment can more accurately reflect the differences in the contribution of different features to the overall gait abnormality. The deviation of multiple feature parameters is fused with multiple weight parameters for calculation, such as weighted summation, to generate a comprehensive gait abnormality index, which represents the overall degree to which a child's gait deviates from the normal developmental standard. Then, based on the comprehensive gait abnormality index, combined with the specific deviation direction and combination pattern of each feature parameter, multi-gait abnormality identification is performed. That is, different gait abnormality patterns are preset or learned based on historical data, such as feature deviation fingerprints of "pigeon toeing", "out-toeing", "flat feet", "pointed feet", etc. By comparing the individual feature deviation vector with the fingerprints of each gait abnormality pattern, the most matching gait abnormality pattern is determined, such as the presence of moderate flat feet with a mild in-toeing tendency.

[0030] Furthermore, step S201 also includes: collecting raw plantar pressure data from healthy children of multiple age groups and genders; performing feature analysis based on the raw plantar pressure data; setting multidimensional standard gait feature parameters according to multiple data features; performing normalization analysis based on the multidimensional standard gait feature parameters to generate normalization analysis results; grouping the normalization analysis results according to multiple age groups and genders to obtain multiple feature parameter groups; performing fluctuation analysis on the multiple feature parameter groups, extracting fluctuation extreme values, and dividing the normal fluctuation range of features according to the fluctuation extreme values; structurally storing the normal fluctuation range of features according to multiple age groups and genders to obtain a parameter data aggregation library; and using the parameter data aggregation library as a standard threshold to construct the standard model of children's gait development.

[0031] Preferably, raw plantar pressure data are collected from healthy children of different ages and genders under normal walking conditions. Feature analysis is performed based on the raw plantar pressure data, and multidimensional standard gait characteristic parameters are set according to multiple data features, such as pressure center trajectory, peak / average pressure in each region, and contact time. Normalization analysis is performed on the multidimensional standard gait characteristic parameters to eliminate the influence of data scaling caused by physiological differences in children's weight, foot length, etc., so that the characteristic parameters between different individuals are comparable, and normalization analysis results are generated. Based on the normalization analysis results, the children are grouped according to multiple age groups and genders, such as 3-4 years old, 5-6 years old, etc., to obtain multiple characteristic parameter groups to identify changes in children's gait characteristics with growth and development and gender differences.

[0032] Preferably, for each feature parameter group in each group, fluctuation analysis is performed. By calculating the mean and standard deviation, and according to statistical rules, such as taking the mean ± 1.96 times the standard deviation to cover a 95% confidence interval, the normal fluctuation extreme value of the feature in the healthy population is extracted. This allows the normal fluctuation range of the feature in that age group and gender to be defined, and the normal variation boundary of each feature in the healthy children population is quantitatively defined. Individual data is compared with this range to determine whether it is abnormal. The normal fluctuation range of all features in all age-gender groups is stored in a structured manner to form a standard parameter data aggregation library. Each record may contain age range, gender, feature parameter name, upper limit of normal range, lower limit of normal range, reference mean, etc. Finally, the parameter data aggregation library is used as a standard threshold and encapsulated and integrated with the corresponding comparison and calculation logic to construct a standard model of children's gait development for automatic assessment of individual gait movement patterns.

[0033] Step S300: Based on the gait abnormality pattern, a risk assessment is performed on the target shoe wearer, and an abnormal risk level is set according to the assessment results.

[0034] Step S300 further includes step S301, calculating the deviation based on the deviation of multiple feature parameters according to the gait anomaly pattern to obtain the deviation magnitude value; step S302, setting multiple risk thresholds, matching the deviation magnitude value with the multiple risk thresholds to determine a target risk threshold range; step S303, performing a risk assessment based on the target risk threshold range to generate a risk assessment score; step S304, weighting and correcting the risk assessment score based on the multiple weight parameters to generate an anomaly assessment score; and step S305, adding the anomaly assessment score to the assessment result.

[0035] Preferably, for identified gait abnormality patterns, multiple key feature parameters forming the pattern are extracted, such as the average arch pressure and the extent to which the pressure center trajectory deviates medially in the flat foot pattern. The deviation of each feature from its normal reference range is calculated using the absolute value or percentage offset of the Z-score. The deviations of multiple key features are then combined by summing, taking the maximum value, or weighted averaging to form a deviation magnitude value representing the overall deviation of the abnormal pattern, thereby quantifying the severity of the specific abnormal pattern. Multiple consecutive risk thresholds are preset to divide the risk level into different intervals. For example, a deviation magnitude value of 0-1 indicates low risk, 1-2 indicates medium risk, and >2 indicates high risk. The deviation magnitude value is then compared with the preset risk thresholds to determine the target risk threshold interval it falls into. For example, if the deviation magnitude value is 1.5, it falls within the medium risk interval.

[0036] Preferably, based on the determined target risk threshold range, a basic risk assessment score is calculated and assigned. For example, low, medium, and high risk ranges may correspond to scores of 10, 50, and 90, respectively, thus converting the target risk threshold range into a quantifiable score. The risk assessment score is then weighted and adjusted based on multiple weight parameters. This involves combining the weight of each feature with its deviation to calculate an adjustment coefficient, which is then multiplied by the basic risk assessment score. Features with higher weights that deviate more significantly will have a greater impact on the final score adjustment. Finally, an anomaly assessment score is generated, ensuring that the final score reflects not only the quantity but also the quality of the deviation—that is, whether key features are affected. This makes the assessment more accurate. For example, even with the same moderate deviation, if the deviation is mainly concentrated on key features, the corrected score may be raised to near a high-risk level. The calculated abnormal assessment score is written into the core component of the gait risk assessment result. Finally, the abnormal risk level is set according to the assessment result. For example, if the abnormal assessment score is <30, the risk level is set to low risk; if 30 ≤ abnormal assessment score <70, the risk level is set to medium risk; if the abnormal assessment score is ≥70, the risk level is set to high risk. Finally, a clear risk level label is output, such as low risk for observation, medium risk for attention, and high risk for immediate intervention.

[0037] Step S400: Based on the gait anomaly pattern and the anomaly risk level, perform gait analysis on the target shoe wearer and generate an individualized gait analysis report.

[0038] Preferably, the identified gait abnormality patterns and quantified abnormality risk levels are combined with the individual information of the target shoe wearer to conduct gait analysis. Specifically, based on the identified gait abnormality patterns, a knowledge base is invoked to generate a simplified description of the pattern, common causes such as muscle strength imbalance and skeletal development characteristics, and possible long-term effects. Then, the abnormality risk level is translated into specific clinical or health management meanings. For example, a medium risk level translates to the presence of a clear functional abnormality that, while not immediately causing pain or injury, may affect exercise efficiency and, if present in the long term, carries the risk of poor joint alignment, with intervention recommended within 3-6 months. Finally, this is linked to the individual information of the target shoe wearer to generate an individualized gait analysis report, which includes at least basic information such as the target shoe wearer's age and gender, a summary of the core content of the main gait abnormality patterns and risk levels, a pressure center trajectory comparison diagram and plantar pressure distribution heatmap of the abnormality pattern, key characteristic parameters and risk level-related data, causal analysis and impact assessment, and preliminary recommendations based on the current analysis, such as recommending further muscle strength assessment and personalized corrective training.

[0039] Step S500: Based on the individualized gait analysis report, perform gait correction analysis on the target shoe wearer and generate a correction training course plan.

[0040] Step S500 further includes step S501, extracting the target abnormality pattern and target abnormality risk level based on the individualized gait analysis report; step S502, using the target abnormality pattern and target abnormality risk level as the main index, introducing the user profile information of the target shoe wearer, and using the user profile information as auxiliary constraints; step S503, mapping the main index to the correction training course library according to the auxiliary constraints for multi-condition matching and filtering, generating an initial course unit set; step S504, optimizing and sorting the initial course unit set to construct the correction training course scheme.

[0041] Preferably, gait correction analysis is performed on the target shoe wearer based on the individualized gait analysis report. Specifically, the target abnormality pattern and target abnormality risk level are extracted from the individualized gait analysis report. The target abnormality pattern and target abnormality risk level are used as the main index, i.e., the primary basis and the first screening condition. At the same time, the user profile information of the target shoe wearer is retrieved, such as age 6 years old, weight 25 kg, preference for gamified training, available training time of 30 minutes per day, and no professional equipment at home. The user profile information is used as auxiliary constraints, i.e., secondary screening conditions and optimization adjustment parameters, thereby ensuring that the training course is targeted, practical, and personalized. Then, according to the auxiliary constraints, the main index is mapped to the correction training course library for multi-condition matching and screening. The correction training course library is a structured database in which each course or course unit is labeled with its attributes, such as the abnormality pattern it targets, suitable for... The system considers factors such as risk level, applicable age, required equipment, training duration, and training format. It then performs a search to filter all potential training courses that meet the main index and auxiliary constraints, forming an initial set of course units. This initial set of course units is then optimized and sorted. The sorting rules may be based on priority of therapeutic efficacy, user suitability, feasibility, or inter-course logic. For example, the strength of recommendations based on the current abnormal pattern and risk level, the degree of matching with user profile information, the priority considering constraints such as user time and equipment, and the order of training movements from easy to difficult or the process logic of warm-up-main training-cool-down. Based on the sorting results, according to weekly training frequency and the units included in each training session, the top-ranked course units are selected and combined to ultimately form a corrective training program. This program details the daily / weekly training content, methods, duration, and precautions to improve children's gait health management.

[0042] Furthermore, step S503 also includes: mapping the master index to the correction training course library for preliminary matching to generate multiple candidate course sets; performing user preference analysis based on the user profile information to generate user preference parameters; filtering the multiple candidate course sets according to the user preference parameters to generate multiple target candidate courses; performing association analysis based on the master index and the correction training course library to obtain multiple association strengths; sorting the multiple target candidate courses in descending order of priority according to the multiple association strengths, selecting the first-ranked target candidate course as the core course, and using the remaining target candidate courses as complementary auxiliary training courses; combining the core course and the complementary auxiliary training courses to construct an initial course unit set.

[0043] Preferably, a preliminary matching and screening process is conducted in the correction training course library based on the master index to identify all courses labeled as applicable to the abnormal pattern and level, thus determining multiple sets of alternative courses. User preference analysis is performed based on user profile information to extract inferred user preference parameters, which may include preferences for gamified training, video follow-up or text-based guidance, tolerance for training duration, availability or aversion to specific training equipment, completion rate and feedback scores in historical training, etc. These user preference parameters are then used to filter out courses that clearly do not match user preferences, such as all courses requiring large equipment or training formats with historically low user ratings. This generates multiple target alternative courses, improving compliance and user experience, and ensuring that users are willing and able to implement the plan.

[0044] Preferably, a correlation analysis is performed based on the main index and the correction training course library to evaluate the correlation strength between each target candidate course and the current main index. This may include evaluating the effectiveness score of the course's historical efficacy data for the specific pattern, the biomechanical targeting strength of the course content and the current abnormal characteristics, etc., to obtain multiple correlation strengths. The target candidate courses are then prioritized in descending order according to the multiple correlation strengths, and the first-ranked target candidate course is selected as the core course, which is the main training unit for the main abnormalities. The remaining target candidate courses are used as complementary auxiliary training courses to consolidate the effect of the core course, target secondary or related abnormal characteristics, provide training diversity to prevent boredom, and serve as advanced or regressive options for the core course. Finally, the core course and the complementary auxiliary training courses are combined to construct a structured initial course unit set, ensuring that the personalized training program has high targeting, high user acceptance, and good structure, thereby improving the level of children's gait health management.

[0045] Furthermore, step S500 also includes step S505, sending the corrective training course plan to the user terminal; step S506, receiving the plan feedback parameters from the target shoe wearer through the user terminal and generating a plan modification feedback instruction; step S507, updating the corrective training course plan through the plan modification feedback instruction and generating an optimized corrective training course plan; step S508, executing the optimized corrective training course plan to track the course and generating a closed-loop optimization of the corrective training course plan based on the corrective training effect.

[0046] Preferably, the generated corrective training program is transmitted over a network and pushed to a user terminal, such as a parent's mobile app or a child's tablet. The user terminal communicates with the child's smart insole and receives program feedback parameters from the target wearer. These parameters include at least information actively submitted by the user, such as subjective feedback like the movements being too difficult to understand, the training time being inappropriate, or feeling pain, as well as user behavior data in the app, such as the number of times a training video is repeatedly watched, the frequency of skipping a particular exercise, and the training completion rate. The user then analyzes this data to generate program modification feedback instructions. For example, for feedback that the movements are too difficult, instructions to reduce the complexity of the movements are generated; for a high skip rate, instructions to replace the training unit are generated.

[0047] Preferably, the corrective training course plan is updated through feedback instructions for plan modification. This may include replacing individual training movements, adjusting training duration and frequency, and modifying training prompts, thereby generating an optimized corrective training course plan. This allows the plan to quickly adapt to the user's actual situation, improving user compliance and training experience. Then, the user executes the optimized corrective training course plan and starts course tracking. This not only tracks course completion but also re-collects plantar pressure data of the target shoe wearer, uses the same model for evaluation, and generates quantitative indicators of corrective training effectiveness, such as the decrease in abnormal assessment scores and the rate of change in the comprehensive gait abnormality index. The corrective training effectiveness data is then used as feedback for closed-loop optimization. This involves analyzing the correlation between the optimized corrective training course plan and the corrective training effectiveness, and further personalizing the user's training course based on this correlation. Simultaneously, the correlation strength calculation rules for recommending courses to all users in the future are optimized, making corrective training course recommendations more accurate.

[0048] In the above text, refer to Figure 1 A method for identifying and responding to gait abnormalities in children based on plantar pressure data, according to an embodiment of the present invention, is described in detail. Next, reference will be made to... Figure 2 A feedback system for identifying gait abnormalities in children based on plantar pressure data, according to an embodiment of the present invention, is described.

[0049] The gait abnormality identification and feedback system for children based on plantar pressure data, according to embodiments of the present invention, addresses the technical problems in existing technologies, such as the lack of quantitative standards for gait assessment, the disconnect between detection and intervention, and the lack of personalized analysis based on dynamic developmental models. It achieves the technical effect of objective, accurate, and intelligent identification of gait abnormalities in children, thereby improving the level of children's gait health management. Figure 2 As shown, the gait abnormality identification and feedback system for children based on plantar pressure data includes: feature extraction module 10, comparison and analysis module 20, risk assessment module 30, gait analysis module 40, and correction plan generation module 50.

[0050] The feature extraction module 10 is used to collect plantar pressure data of the target shoe wearer through a sensor array, extract features from the plantar pressure data, and obtain multidimensional gait feature parameters; the comparison and analysis module 20 is used to synchronize the multidimensional gait feature parameters to a standard model of children's gait development for comparison and analysis, and identify gait abnormal patterns; the risk assessment module 30 is used to conduct a risk assessment of the target shoe wearer based on the gait abnormal patterns, and set an abnormal risk level according to the assessment results; the gait analysis module 40 is used to conduct gait analysis of the target shoe wearer based on the gait abnormal patterns and the abnormal risk level, and generate an individualized gait analysis report; the correction plan generation module 50 is used to conduct gait correction analysis of the target shoe wearer based on the individualized gait analysis report, and generate a correction training course plan.

[0051] The specific configuration of the feature extraction module 10 will be described in detail below. The feature extraction module 10 further includes: synchronously collecting pressure values ​​of each zone of the sole and foot movement posture data through a sensor array containing a pressure sensing unit and an inertial measurement unit; transmitting the pressure values ​​of each zone of the sole and the foot movement posture data to a mobile terminal to activate a motion state detection module; determining whether the target shoe wearer has entered a motion state through the motion state detection module; when the target shoe wearer enters a motion state, activating the pressure sensing unit to collect data from the target shoe wearer at a first sampling frequency to obtain an initial pressure data stream, activating the inertial measurement unit to collect data from the target shoe wearer at a second sampling frequency to obtain an initial posture data stream; determining the target motion mode based on the initial pressure data stream and the initial posture data stream; dynamically adjusting the sensor array based on the target motion mode, and collecting the sole pressure data of the target shoe wearer based on the adjustment result.

[0052] The specific configuration of the feature extraction module 10 will be described in detail below. The feature extraction module 10 further includes: retrieving gait record data logs of the target shoe wearer; performing periodic repetitive analysis based on the gait record data logs to determine the gait cycle; segmenting the plantar pressure data according to the gait cycle to obtain a single gait cycle pressure dataset; calculating the pressure center based on the single gait cycle pressure dataset to obtain pressure center trajectory data; performing time-domain analysis on the pressure center trajectory data to obtain pressure time-series data; performing frequency-domain transformation on the pressure time-series data to obtain frequency-domain energy distribution characteristics; performing pressure calculation based on the single gait cycle pressure dataset according to the sensor array to obtain pressure peak parameters and pressure mean parameters; calculating the contact duration of the target shoe wearer based on the pressure peak parameters and the pressure mean parameters to obtain foot-to-ground contact time data; and fusing the frequency-domain energy distribution characteristics, the foot-to-ground contact time data, the pressure peak parameters, and the pressure mean parameters to construct the multi-dimensional gait feature parameters.

[0053] The specific configuration of the comparative analysis module 20 will be described in detail below. The comparative analysis module 20 further includes: synchronizing the multidimensional gait feature parameters to a standard model of children's gait development for deviation calculation to obtain the deviation degree of multiple feature parameters; performing gait influence analysis on the multidimensional gait feature parameters to determine multiple influencing factors; assigning weights to the multidimensional gait feature parameters based on the multiple influencing factors to generate multiple weight parameters; performing a fusion calculation based on the deviation degree of the multiple feature parameters and the multiple weight parameters to obtain a comprehensive gait abnormality index; and performing multi-gait abnormality identification based on the comprehensive gait abnormality index to determine the gait abnormality pattern.

[0054] The specific configuration of the comparative analysis module 20 will be described in detail below. The comparative analysis module 20 further includes: collecting raw plantar pressure data from healthy children of multiple age groups and genders; performing feature analysis based on the raw plantar pressure data; setting multidimensional standard gait feature parameters according to multiple data features; performing normalization analysis based on the multidimensional standard gait feature parameters to generate normalization analysis results; grouping the normalization analysis results according to multiple age groups and genders to obtain multiple feature parameter groups; performing fluctuation analysis on the multiple feature parameter groups, extracting fluctuation extreme values, and dividing the normal fluctuation range of features according to the fluctuation extreme values; structurally storing the normal fluctuation range of features according to multiple age groups and genders to obtain a parameter data aggregation library; and using the parameter data aggregation library as a standard threshold to construct the standard model of children's gait development.

[0055] The specific configuration of the risk assessment module 30 will be described in detail below. The risk assessment module 30 further includes: calculating the deviation based on the deviation of multiple feature parameters according to the gait anomaly pattern to obtain a deviation magnitude value; setting multiple risk thresholds, matching the deviation magnitude value with the multiple risk thresholds to determine a target risk threshold range; performing a risk assessment based on the target risk threshold range to generate a risk assessment score; weighting and correcting the risk assessment score based on the multiple weight parameters to generate an anomaly assessment score; and adding the anomaly assessment score to the assessment result.

[0056] The specific configuration of the correction plan generation module 50 will be described in detail below. The correction plan generation module 50 further includes: extracting the target abnormality pattern and target abnormality risk level based on the individualized gait analysis report; using the target abnormality pattern and target abnormality risk level as the main index, introducing the user profile information of the target shoe wearer, and using the user profile information as auxiliary constraints; mapping the main index to the correction training course library according to the auxiliary constraints for multi-condition matching and filtering, generating an initial course unit set; optimizing and sorting the initial course unit set to construct the correction training course plan.

[0057] The specific configuration of the correction plan generation module 50 will be described in detail below. The correction plan generation module 50 further includes: mapping the master index to the correction training course library for preliminary matching to generate multiple candidate course sets; performing user preference analysis based on the user profile information to generate user preference parameters; filtering the multiple candidate course sets according to the user preference parameters to generate multiple target candidate courses; performing association analysis based on the master index and the correction training course library to obtain multiple association strengths; sorting the multiple target candidate courses in descending order of priority according to the multiple association strengths, selecting the first-ranked target candidate course as the core course, and using the remaining target candidate courses as complementary auxiliary training courses; combining the core course and the complementary auxiliary training courses to construct an initial course unit set.

[0058] The specific configuration of the correction plan generation module 50 will be described in detail below. The correction plan generation module 50 further includes: distributing the correction training course plan to a user terminal; receiving the plan feedback parameters from the target shoe wearer through the user terminal and generating a plan modification feedback instruction; updating the correction training course plan through the plan modification feedback instruction to generate an optimized correction training course plan; executing the optimized correction training course plan to track the course progress and generate a closed-loop optimization of the optimized correction training course plan based on the correction training effect.

[0059] The gait abnormality identification and feedback system for children based on plantar pressure data provided in this embodiment of the invention can execute the gait abnormality identification and feedback method for children based on plantar pressure data provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0060] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for identifying and responding to gait abnormalities in children based on plantar pressure data, characterized in that, The method includes: The foot pressure data of the target shoe wearer is collected by a sensor array, and features are extracted from the foot pressure data to obtain multidimensional gait feature parameters. The multidimensional gait feature parameters are synchronized to a standard model of children's gait development for comparative analysis to identify abnormal gait patterns. Based on the aforementioned gait abnormality pattern, a risk assessment is conducted on the target shoe wearer, and an abnormality risk level is set according to the assessment results; Based on the gait anomaly pattern and the anomaly risk level, gait analysis is performed on the target shoe wearer to generate an individualized gait analysis report; Based on the individualized gait analysis report, gait correction analysis is performed on the target shoe wearer, and a correction training program is generated accordingly.

2. The method for identifying and responding to gait abnormalities in children based on plantar pressure data as described in claim 1, characterized in that, Methods for acquiring plantar pressure data of a target shoe wearer using a sensor array include: The sensor array, which includes pressure sensing units and inertial measurement units, synchronously collects pressure values ​​and foot movement posture data for each zone of the sole. The pressure values ​​of each zone on the sole of the foot and the foot movement posture data are transmitted to the mobile terminal to activate the motion state detection module; The motion state detection module determines whether the target shoe wearer has entered a state of motion. When the target wearer enters a state of motion, the pressure sensing unit is activated to collect data from the target wearer at a first sampling frequency to obtain an initial pressure data stream, and the inertial measurement unit is activated to collect data from the target wearer at a second sampling frequency to obtain an initial attitude data stream. The target motion pattern is determined based on the initial pressure data stream and the initial attitude data stream; The sensor array is dynamically collected and adjusted based on the target motion pattern, and the plantar pressure data of the target shoe wearer is collected according to the adjustment results.

3. The method for identifying and responding to gait abnormalities in children based on plantar pressure data as described in claim 1, characterized in that, The method for extracting features from the plantar pressure data to obtain multidimensional gait feature parameters includes: Retrieve the gait record log of the target shoe wearer, and perform periodic repetitive analysis based on the gait record log to determine the gait cycle; The plantar pressure data is segmented according to the gait cycle to obtain a single gait cycle pressure dataset; Pressure center is calculated based on the single-step periodic pressure dataset to obtain pressure center trajectory data. Time-domain analysis is performed on the pressure center trajectory data to obtain pressure time-series data; The pressure time-series data is transformed in the frequency domain to obtain the frequency domain energy distribution characteristics. Based on the single-step periodic pressure dataset, pressure is calculated according to the sensor array to obtain the peak pressure parameter and the average pressure parameter. Based on the peak pressure parameter and the average pressure parameter, the contact time of the target shoe wearer is calculated to obtain foot-ground contact time data. The frequency domain energy distribution characteristics, foot-ground contact time data, pressure peak parameters, and pressure mean parameters are fused to construct the multidimensional gait feature parameters.

4. The method for identifying and responding to gait abnormalities in children based on plantar pressure data as described in claim 1, characterized in that, The multidimensional gait feature parameters are synchronized to a standard model of children's gait development for comparative analysis to identify abnormal gait patterns. The method includes: The multidimensional gait feature parameters are synchronized to the standard model of children's gait development for deviation calculation to obtain the deviation degree of multiple feature parameters; Gait influence analysis is performed on the multidimensional gait feature parameters to determine multiple influencing factors. Based on the multiple influencing factors, weights are assigned to the multidimensional gait feature parameters to generate multiple weight parameters. The comprehensive gait anomaly index is obtained by fusing the deviation of the multiple feature parameters with the multiple weight parameters. Based on the comprehensive gait anomaly index, multi-gait anomaly identification is performed to determine the gait anomaly pattern.

5. The method for identifying and responding to gait abnormalities in children based on plantar pressure data as described in claim 1, characterized in that, The process and methods for constructing a standard model of children's gait development include: Raw plantar pressure data of healthy children of multiple ages and genders were collected. Feature analysis was performed based on the raw plantar pressure data, and multidimensional standard gait feature parameters were set according to multiple data features. Normalization analysis is performed based on the aforementioned multidimensional standard gait feature parameters to generate normalization analysis results; Based on the normalization analysis results, multiple groups of feature parameters were obtained by grouping them according to multiple age groups and multiple genders. Fluctuation analysis is performed on the multiple feature parameter groups to extract the extreme fluctuation values, and the normal fluctuation range of the features is divided according to the extreme fluctuation values. The normal fluctuation range of the aforementioned features is structured and stored according to multiple age groups and multiple genders to obtain a parameter data aggregation library; The parameter data aggregation library was used as a standard threshold to construct the standard model of children's gait development.

6. The method for identifying and responding to gait abnormalities in children based on plantar pressure data as described in claim 4, characterized in that, Risk assessment of target shoe wearers based on the aforementioned gait abnormality patterns includes the following methods: Based on the gait anomaly pattern, deviation is calculated according to the deviation of multiple feature parameters to obtain the deviation magnitude value; Multiple risk thresholds are set, and the deviation magnitude value is matched with the multiple risk thresholds to determine the target risk threshold range; A risk assessment is performed based on the target risk threshold range, and a risk assessment score is generated. The risk assessment score is weighted and corrected based on the multiple weight parameters to generate an anomaly assessment score. Add the anomaly assessment score to the assessment result.

7. The method for identifying and responding to gait abnormalities in children based on plantar pressure data as described in claim 1, characterized in that, Based on the individualized gait analysis report, gait correction analysis is performed on the target shoe wearer, and a correction training program is generated accordingly. The method includes: Based on the individualized gait analysis report, the target's abnormal pattern and the target's abnormal risk level are extracted; The target anomaly pattern and the target anomaly risk level are used as the main indexes, and the user profile information of the target shoe wearer is introduced as the auxiliary constraint condition. According to the auxiliary constraints, the main index is mapped to the correction training course library for multi-condition matching and filtering to generate an initial set of course units; The initial set of course units is sorted and optimized to construct the corrective training course scheme.

8. The method for identifying and responding to gait abnormalities in children based on plantar pressure data as described in claim 7, characterized in that, According to the auxiliary constraints, the main index is mapped to the correction training course library for multi-condition matching and filtering to generate an initial set of course units. The method includes: The master index is mapped to the correction training course library for initial matching, generating multiple sets of alternative courses; Based on the user profile information, user preference analysis is performed to generate user preference parameters; The multiple sets of alternative courses are filtered according to the user preference parameters to generate multiple target alternative courses; Based on the association analysis between the main index and the correction training course library, multiple association strengths are obtained; The multiple target candidate courses are sorted in descending order of priority according to the multiple association strengths, and the target candidate course with the first priority is selected as the core course, and the remaining target candidate courses are used as complementary auxiliary training courses. The core courses are combined with the complementary auxiliary training courses to construct an initial set of course units.

9. The method for identifying and responding to gait abnormalities in children based on plantar pressure data as described in claim 7, characterized in that, After generating the corrective training program plan, the methods include: The corrective training course plan will be distributed to the user terminal. The user terminal receives the solution feedback parameters from the target shoe wearer and generates a solution modification feedback instruction. The corrective training course plan is updated by modifying the feedback instructions of the plan, and an optimized corrective training course plan is generated. The corrective training course optimization plan is implemented to track the course progress and generate corrective training results for closed-loop optimization of the corrective training course optimization plan.

10. A gait abnormality recognition and feedback system for children based on plantar pressure data, characterized in that, The system is used to implement the gait abnormality identification and feedback method for children based on plantar pressure data according to any one of claims 1 to 9, the system comprising: The feature extraction module is used to collect plantar pressure data of the target shoe wearer through a sensor array, extract features from the plantar pressure data, and obtain multidimensional gait feature parameters. The comparison and analysis module is used to synchronize the multidimensional gait feature parameters to a standard model of children's gait development for comparison and analysis, and to identify abnormal gait patterns. The risk assessment module is used to assess the risk of the target shoe wearer based on the gait abnormality pattern and set the abnormal risk level according to the assessment results; The gait analysis module is used to perform gait analysis on the target shoe wearer based on the gait anomaly pattern and the anomaly risk level, and generate an individualized gait analysis report. The correction plan generation module is used to perform gait correction analysis on the target shoe wearer based on the individualized gait analysis report and generate a correction training course plan.

Images