Health data entry method and system and storage medium

Abstract

The invention relates to the technical field of biomechanical analysis, in particular to a health data input method and system and a storage medium. The method comprises the following steps: collecting multi-modal data of a subject, and carrying out gait event recognition to obtain a gait event mark; bilateral gait synchronous alignment is carried out according to the gait event marks, bilateral gait cycle data are obtained, and the bilateral gait cycle data comprise plantar pressure data, ground force data, joint angle data and motion state data; plantar pressure power processing is carried out according to the plantar pressure data and the ground force data, and plantar power parameters are obtained; estimating body segment inertial parameters of the subject; joint torque energy processing is carried out according to the body segment inertial parameters and the joint angle data, and joint power parameters are obtained; and performing gait acceleration calculation according to the motion state data to obtain a gait stability index. The accuracy, comprehensiveness and convenience of gait analysis are remarkably improved through a biomechanical analysis technology.

Description

Technical Field

[0001] The present invention relates to the field of biomechanical analysis technology, and in particular to a health data entry method, system and storage medium. Background Art

[0002] Existing health data entry methods generally have data fragmentation problems. Data such as plantar pressure, joint angles, and acceleration are usually collected independently by different devices, lacking a unified time reference frame; each biomechanical parameter is analyzed in isolation, and the intrinsic connection between parameters at different levels cannot be revealed; incompatible data formats between analysis systems lead to information islands; there is a conversion gap from raw data to clinical decision support, and high-value biomechanical data is difficult to convert into practical health indicators. This fragmentation severely limits the comprehensive value of multimodal data and hinders the scientific evaluation of the overall performance of the human movement system.

[0003] Existing methods are seriously deficient in individualized analysis. They mainly rely on the group average as the evaluation standard, treating individual differences as "noise" rather than meaningful personalized characteristics; they only standardize through simple height and weight normalization, failing to fully consider individual factors such as body segment proportions and muscle strength levels; they lack scientific methods to determine the optimal gait parameters for specific individuals, resulting in a lack of personalized basis for setting rehabilitation training goals; they fail to identify and quantify individual-specific functional adaptation strategies, ignoring the essential differences in biomechanical properties between different individuals. This standardization-oriented analysis method cannot meet the needs of precision medicine and personalized health management, limiting the practical value of health data at the individual level.

[0004] In summary, the problems of health data fragmentation and insufficient individualized analysis in existing technologies need to be addressed urgently. Summary of the invention

[0005] Based on this, it is necessary to provide a method, system and storage medium for entering health data to solve at least one of the above technical problems.

[0006] To achieve the above purpose, a method for entering health data includes the following steps: Step S1: collecting multimodal data of the subject, and performing gait event recognition to obtain gait event markers; performing bilateral gait synchronization alignment according to the gait event markers to obtain bilateral gait cycle data, wherein the bilateral gait cycle data includes plantar pressure data, ground force data, joint angle data and motion state data; Step S2: performing plantar pressure dynamic processing according to the plantar pressure data and the ground force data to obtain plantar dynamic parameters; estimating the subject's body segment inertia parameters; performing joint torque energy processing according to the body segment inertia parameters and the joint angle data to obtain joint dynamic parameters; calculating gait acceleration according to the motion state data to obtain gait stability index; extracting phase coordination features according to the joint angle data and the motion state data to obtain a comprehensive coordination index; Step S3: Calculate the energy conversion efficiency based on the bilateral gait cycle data, plantar power parameters and joint power parameters to obtain efficiency score data; optimize the gait parameters based on the gait stability index and comprehensive coordination index of the efficiency score data to obtain individual target parameters; perform gait health assessment based on the individual target parameters, efficiency score data and comprehensive coordination index to obtain gait health data.

[0007] The present invention synchronously collects multimodal gait data such as plantar pressure, ground reaction force, lower limb joint angle and body movement state, and performs accurate gait event recognition and bilateral gait cycle synchronization alignment, providing a comprehensive, reliable and time-referenced unified data basis for subsequent gait analysis. This not only overcomes the problems of data fragmentation and time asynchrony in traditional gait analysis, but also lays a solid foundation for in-depth analysis of various aspects of gait (such as dynamics, kinematics, energetics, coordination, etc.), making subsequent analysis results more accurate, more reliable and more comparable, and creating conditions for achieving comprehensive evaluation and personalized analysis of gait. Through in-depth dynamic analysis of multimodal data, including plantar pressure dynamic processing, joint torque energy processing, gait acceleration calculation and phase coordination feature extraction, detailed information on multiple levels such as plantar pressure distribution, joint load, muscle work, energy transfer, body stability and coordination between limbs and joints during gait is revealed. These dynamic characteristic parameters can not only quantitatively describe the gait characteristics of individuals, but also deeply reveal the internal mechanism of gait, providing an important scientific basis for identifying abnormal gait, evaluating motor function, predicting injury risk, and formulating personalized rehabilitation training programs, thus achieving a deep and multi-dimensional understanding of gait. Through comprehensive analysis of data, including energy conversion efficiency calculation, gait parameter optimization processing, and gait health assessment, gait analysis is elevated from the basic data and characteristic parameter level to the function and application level. Energy efficiency analysis reveals the utilization and loss of energy during gait, gait parameter optimization processing provides individuals with personalized gait improvement suggestions, and gait health assessment evaluates the health of gait from aspects such as joint load, gait symmetry, and fall risk. The gait analysis results are converted into information with practical application value, providing scientific guidance for improving exercise efficiency, preventing sports injuries, improving gait function, and promoting healthy aging, thus realizing the clinical transformation and application value of gait analysis results. Therefore, the present invention provides a method for entering health data, which effectively solves the problems of data fragmentation and insufficient individualized analysis in existing health data entry methods through integrated data collection, standardized processing flow, individualized parameter estimation, multi-dimensional comprehensive analysis, and personalized model establishment, and significantly improves the accuracy, comprehensiveness, and convenience of gait analysis.

[0008] Preferably, step S1 comprises the following steps: Step S11: using a pre-calibrated pressure sensor array to perform multimodal data on the subject to obtain a multimodal signal set; Step S12: performing gait event recognition on the multimodal signal set to obtain gait event labels; Step S13: dividing the multimodal signal set into gait cycles according to the gait event markers to obtain cycle standardized data; Step S14: performing bilateral gait synchronization alignment according to the cycle standardization data to obtain bilateral gait cycle data.

[0009] The present invention uses a pre-calibrated high-density pressure sensor array (for example, a 960-sensor insole system) and multiple inertial measurement units and electronic goniometers to achieve synchronous and high-precision acquisition of multiple key biomechanical parameters during the subject's gait. This not only provides detailed plantar pressure distribution information, but also simultaneously obtains data on lower limb joint kinematics (angle, angular velocity) and dynamics (ground reaction force) and overall body motion state (acceleration). This multimodal data method overcomes the problem of time asynchrony of data from different devices in traditional methods, provides a comprehensive and reliable data basis for subsequent gait analysis, and avoids deviations in analysis results due to incomplete or asynchronous data. By analyzing the collected multimodal signal set, especially combining the changing characteristics of the vertical ground reaction force (VGRF) and the center of plantar pressure (COP) trajectory, key events in the gait cycle (such as heel strike, plantar flat contact, heel off, toe off) are accurately identified. These gait event markers provide an accurate time reference for subsequent gait cycle division, parameter calculation and analysis. Compared with gait event recognition based on a single signal source (such as using only an accelerometer), this recognition method based on multimodal signals is more reliable and can effectively avoid misjudgment caused by noise interference or individual differences, thereby improving the accuracy and robustness of gait event recognition. The continuous gait data stream is divided into independent gait cycles, and each cycle is time-normalized (for example, each cycle is equally divided into 100 time points). This cycle standardization eliminates the influence of the duration differences between different individuals and between different gait cycles of the same individual, so that the data of different gait cycles can be directly compared and averaged. This provides a unified time scale for subsequent statistical analysis, pattern recognition, and feature extraction, making the data of different gait cycles comparable and convenient for subsequent in-depth analysis. By accurately aligning the gait cycle data on the left and right sides in time (for example, aligning the heel strike events on both sides and linearly stretching the shorter cycle), the synchronization relationship between the movements of the lower limbs on both sides is established. This allows for direct comparison of the kinematic and kinetic parameters of the bilateral lower limbs and analysis of the symmetry or asymmetry of bilateral gait. This bilaterally synchronized gait data provides key information for assessing gait coordination, identifying potential gait abnormalities (such as limping, compensatory gait), and evaluating rehabilitation effectiveness, highlighting the coordination and differences in bilateral limb movements, and providing an important basis for a comprehensive assessment of gait.

[0010] Preferably, the foot pressure dynamic processing in step S2 includes: Extract the plantar pressure center trajectory according to the plantar pressure data to obtain the pressure center trajectory data; Perform pressure center velocity analysis on pressure center trajectory data to obtain pressure center dynamic parameters; The regional pressure distribution time series analysis is performed based on the ground force data to obtain the regional pressure pattern; Perform pressure integral analysis based on regional pressure patterns to obtain cumulative pressure maps; The pressure peak time series analysis is performed based on the dynamic parameters of the pressure center, the regional pressure pattern and the cumulative pressure spectrum to obtain the pressure time series characteristics.

[0011] The present invention accurately extracts the trajectory of the plantar pressure center (COP) changing over time by calculating the weighted average position of all pressure sensors on the sole of the foot at each moment. The COP trajectory reflects the movement path of the plantar pressure action point during gait, and is an important indicator for evaluating foot function, balance control and gait stability. Compared with focusing only on the overall pressure size, the COP trajectory provides more detailed plantar pressure distribution information, can reveal the pattern of plantar pressure transfer during gait, and provides a key basis for identifying abnormal gaits (such as flat feet and high arch foot gait) and evaluating rehabilitation effects. By deriving the COP trajectory data, the movement speed of the COP in the front-back and inside-out directions is calculated, and the speed value at the key moment (such as the average speed in the early, middle and late stages of the support phase) is extracted. The COP speed reflects the speed and smoothness of the movement of the center of pressure during gait, and is an important indicator for evaluating gait efficiency and stability. Compared with focusing only on the COP position, the COP speed provides richer dynamic information, can reveal the differences in balance control strategies during gait, and provides an important basis for identifying gait abnormalities (such as unstable gait, small steps) and evaluating motor function. The sole of the foot is divided into multiple anatomical regions (such as the medial heel, lateral heel, midfoot, and each metatarsal head region), and the pressure share of each region at different times in the gait cycle is calculated, and a curve of the pressure share of each region over time is drawn. This regional pressure pattern can clearly show the force conditions and pressure transfer order of different regions of the sole during gait, and provides an intuitive basis for evaluating the load distribution of the foot, identifying abnormal pressure patterns (such as excessive forefoot pressure and low heel pressure), and evaluating the effect of orthopedic insoles. Compared with focusing only on the overall pressure, the regional pressure pattern provides more detailed information on the distribution of plantar pressure, which helps to deeply understand the function of the foot and gait characteristics. By integrating the pressure share curve of each region over time, the cumulative pressure load of each region in the entire support phase is calculated. The cumulative pressure map reflects the total pressure borne by different regions of the sole over a period of time, and is an important indicator for assessing the risk of foot tissue injury. Compared with focusing only on the instantaneous pressure peak, the cumulative pressure map takes into account the time factor of pressure action, which can more comprehensively evaluate the fatigue and injury risk of foot tissues, and provide an important basis for preventing foot diseases (such as plantar fasciitis and metatarsalgia) and guiding shoe design. The pressure peak size, occurrence time (relative to the percentage of the gait cycle), and the order of peak occurrence in each anatomical area are extracted, and the pressure impact index (the product of peak pressure and COP velocity) is calculated. These pressure time series characteristics can reflect the moment and intensity of the maximum pressure on different areas of the sole, as well as the transmission process of the pressure peak from the heel to the forefoot. Compared with focusing only on the pressure size, the pressure time series characteristics provide richer dynamic information, which can reveal the impact load and pressure transmission mode of the sole during gait, and provide an important basis for identifying abnormal gaits (such as high-impact gait, delayed pressure transmission) and assessing the risk of sports injuries.

[0012] Preferably, the joint torque energy processing in step S2 includes: Perform inverse dynamics calculation based on body segment inertia parameters, joint angle data and ground force data to obtain joint torque data; Perform joint power analysis based on joint torque data and joint angle data to obtain energy transfer data; Calculate joint stiffness based on joint torque data to obtain stiffness characteristic data; The torque feature is extracted according to the joint torque data and gait event markers to obtain the torque feature data.

[0013] This aspect uses inverse dynamics calculation from bottom to top (from foot to hip) to combine body segment inertia parameters (mass, center of mass position, moment of inertia), joint angle data and ground force data to accurately calculate the net torque of the ankle, knee and hip joints in the sagittal plane, frontal plane and transverse plane. Joint torque reflects the internal force generated by soft tissues such as muscles and ligaments around the joints, and is an important indicator for evaluating joint load, muscle function and movement control. Compared with focusing only on kinematic parameters such as joint angles, joint torque provides deeper dynamic information, which can reveal the work done by muscles and the stress state of joints during gait, and provides a key basis for identifying abnormal gait (such as insufficient muscle strength, joint instability) and evaluating rehabilitation effects. By multiplying the joint torque with the joint angular velocity, the instantaneous power of each joint is calculated, and the positive and negative work stages of the power curve are analyzed. Joint power reflects the generation (positive work) and absorption (negative work) of energy around the joints, and is an important indicator for evaluating gait efficiency and energy transfer. By analyzing the transfer of energy between joints (for example, from the ankle to the hip, or from the hip to the knee), we can understand the synergistic effects of different muscle groups at different stages of the gait cycle. Compared with focusing only on joint torque, joint power provides richer energy information, which can reveal the flow and utilization of energy during gait, and provides an important basis for optimizing gait, improving movement efficiency, and evaluating rehabilitation effects. By drawing the joint torque-angle curve and calculating the slope of the curve in the middle of the stance phase, the dynamic stiffness of the joint is obtained. Joint stiffness reflects the ability of the joint to resist deformation and is an important indicator for evaluating joint stability, buffering capacity, and motion control. Compared with focusing only on joint angles or torques, joint stiffness combines information on force and deformation and can more comprehensively evaluate the functional status of the joint. For example, higher ankle joint stiffness may help improve the propulsion efficiency of gait, but it may also increase the impact load. By analyzing the stiffness characteristics of different joints, potential joint instability or excessive stiffness can be identified, providing a reference for sports injury prevention and rehabilitation training. By extracting the torque peak value, peak occurrence time, and torque value at key moments (such as heel strike, mid-stance phase, toe-off) of each joint in different planes (sagittal plane, frontal plane, and transverse plane), and analyzing the morphological characteristics of the torque curve (such as whether there is a double-peak pattern and whether there is an abnormal mutation), the torque characteristic data is obtained. These characteristics can reflect the key information of joint torque during gait and are important indicators for evaluating muscle function, joint load, and motion control. Compared with focusing only on the average value of torque, torque characteristics provide more detailed dynamic information, which can reveal the activation pattern of muscles and the force characteristics of joints during gait, and provide an important basis for identifying abnormal gait (such as reduced torque peak due to insufficient muscle strength and torque mutation due to joint instability) and evaluating rehabilitation effects.

[0014] Preferably, the gait acceleration calculation in step S2 includes: Perform acceleration feature statistics according to the motion state data to obtain acceleration features; Spectral characteristics are analyzed based on motion state data and gait event markers to obtain acceleration spectrum characteristics; Perform shock transfer analysis based on acceleration characteristics and acceleration spectrum characteristics to obtain shock transfer characteristics; The stability index is calculated according to the impact transmission characteristics and acceleration characteristics to obtain the gait stability index.

[0015] The present invention obtains acceleration characteristics by calculating the statistics (such as mean, standard deviation, root mean square value, kurtosis, skewness) of the triaxial accelerometers installed on different parts of the body (such as the calf, thigh, and trunk) during the gait cycle. These statistics can reflect the vibration intensity, fluctuation degree and distribution characteristics of various parts of the body during the gait process. Compared with only observing the original acceleration curve, the acceleration characteristics provide a more concise and easier to compare quantitative index. For example, a higher root mean square value indicates a higher vibration intensity, which may be related to a higher impact load; a higher kurtosis may indicate the presence of more spike pulses, which may be related to an unstable gait. These characteristics provide an important basis for evaluating the stability of gait, impact load and potential injury risks. By performing a fast Fourier transform (FFT) on the acceleration signal, the time domain signal is converted into a frequency domain signal, and the power spectral density (PSD) is calculated, the main frequency, the main frequency power ratio and the energy ratio of different frequency bands (such as 0-3Hz, 3-8Hz, and above 8Hz) are extracted to obtain the acceleration spectrum characteristics. These features can reflect the contribution of different frequency components during gait and reveal the rhythmicity, impact characteristics and vibration patterns of gait. For example, the main frequency is usually related to the step frequency, reflecting the basic rhythm of gait; the energy proportion of the high frequency band (above 8Hz) is usually related to the impact load. Compared with focusing only on time domain features, spectral features provide richer frequency information, which helps to deeply understand the dynamic characteristics of gait and provide an important basis for identifying abnormal gaits (such as high-impact gait and uncoordinated gait) and assessing the risk of sports injuries. The impact transfer characteristics are obtained by calculating the peak attenuation rate of vertical acceleration of different body parts (such as calf, thigh, and trunk) in a short period of time after heel strike (IC) and the change in the energy proportion of the high frequency band (above 8Hz). These features can reflect the transmission and attenuation of impact loads in the human body during gait, and are important indicators for assessing the buffering capacity and potential injury risk of the lower limbs. For example, a lower impact attenuation rate indicates that the impact load is less attenuated from the foot to the trunk, which may increase the risk of injury to the upper joints and tissues. Compared with focusing only on the acceleration of a single position, impact transfer analysis provides a more comprehensive assessment of impact loads, helps to understand the body's ability to absorb and disperse impacts, and provides a reference for optimizing gait and reducing sports injuries. Gait stability indices are obtained by calculating the root mean square value (RMS) of trunk acceleration, the coefficient of variation (CV) of the gait cycle, and the symmetry index (SI) of the acceleration of the left and right sides. These indices can reflect the overall stability of the body during gait, the regularity of the gait rhythm, and the symmetry of the left and right movements. For example, a higher trunk RMS value indicates greater body sway and poor stability; a higher CV value indicates an irregular gait rhythm, which may be related to decreased balance ability; a lower SI value indicates asymmetric movement of the left and right sides, which may be related to unilateral injury or dysfunction. These stability indices provide an important quantitative basis for assessing the risk of falling, identifying potential gait abnormalities, and evaluating the effectiveness of rehabilitation training.

[0016] Preferably, the phase coordination feature extraction in step S2 includes: calculating joint coordination angles of joint angle data according to gait event markers; Perform segmented relative phase analysis based on joint coordination angle and motion state data to obtain phase difference data; Perform force-action coordination analysis based on joint angle data and ground force data to obtain force-action coordination data; Pressure-posture coupling analysis is performed based on pressure center trajectory data, regional pressure patterns, and motion state data to obtain pressure-posture coupling features; The joint coordination angle, phase difference data, force-action coordination data and pressure-posture coupling characteristics are integrated to obtain a comprehensive coordination index.

[0017] The present invention obtains the joint coordination angle by constructing an angle-angle phase diagram of adjacent joints (such as hip-knee, knee-ankle) and calculating the angle between each point on the phase diagram and the origin. The joint coordination angle reflects the relative relationship between the adjacent joint movements during gait and is an important indicator for evaluating the joint coordination movement pattern. Compared with analyzing the angle changes of each joint separately, the joint coordination angle can more clearly reveal the coordination relationship between joints. For example, in the early stage of the support phase, the hip joint and the knee joint may present a "flexion-flexion" coordination pattern, while in the swing phase, they may present a "flexion-extension" coordination pattern. The analysis of this coordination pattern helps to identify abnormal gait (such as joint coordination pattern disorder) and evaluate motor function. By calculating the phase angle of each limb segment (such as foot, calf, thigh) relative to the vertical direction and calculating the phase difference between adjacent segments, the phase difference data is obtained. The phase difference reflects the leading or lagging relationship of the movement of different limb segments in the gait cycle and is an important indicator for evaluating the coordination between limb segments. For example, in normal gait, during the swing phase, the phase of the thigh usually leads the phase of the calf, and the phase of the calf leads the phase of the foot, forming a "whiplash" effect. The analysis of phase difference helps to identify abnormal gait (such as disordered phase relationship between segments) and evaluate movement control ability. By calculating the cross-correlation coefficient between the vertical ground reaction force (VGRF) and each joint angle, and finding the maximum correlation coefficient and its corresponding time delay, the force-action coordination data is obtained. These data reflect the synergistic relationship between the ground reaction force and joint movement during gait, and are important indicators for evaluating force and action coordination. For example, in normal gait, the peak of VGRF is usually synchronized with or slightly ahead of the peak of the plantar flexion torque of the ankle joint. The analysis of force-action coordination helps to identify abnormal gait (such as asynchrony between force and action) and evaluate movement control strategies. By calculating the cross-correlation coefficient between the position of the center of plantar pressure (COP) and the trunk acceleration, and finding the maximum correlation coefficient and its corresponding time delay, the pressure-posture coupling characteristics are obtained. These characteristics reflect the relationship between the change in plantar pressure distribution and the adjustment of trunk posture during gait, and are important indicators for evaluating balance control and postural stability. For example, in normal gait, when the COP moves forward quickly, the trunk may experience a backward acceleration to maintain balance. Pressure-posture coupling analysis helps to identify abnormalities in balance control strategies (such as delayed reaction, overreaction) and assess the risk of falling. By integrating the coordination indicators (joint coordination angle, phase difference, force-action coordination data, pressure-posture coupling characteristics) obtained from the above four aspects, such as using principal component analysis (PCA) and other methods, the main coordination patterns are extracted, and the scores of each individual on these patterns are calculated to obtain a comprehensive coordination index. This comprehensive index can comprehensively reflect the overall coordination level of an individual during gait, avoiding the one-sidedness of a single indicator. Compared with analyzing each coordination indicator separately, the comprehensive coordination index can more accurately evaluate the individual's motor control ability.

[0018] Preferably, the energy conversion efficiency calculation in step S3 includes: The center of mass motion energy is calculated based on the body segment inertia parameters, bilateral gait cycle data and joint dynamic parameters to obtain the center of mass energy data; The potential energy-kinetic energy conversion efficiency is calculated for the center of mass energy data to obtain the energy recovery index; Perform joint power and energy transfer analysis on the joint torque data and joint angle data in the joint dynamic parameters to obtain the joint energy transfer index; Calculate the mechanical energy external power according to the center of mass energy data and the ground force data to obtain the external power data; According to the plantar dynamic parameters, the arch elastic energy storage is evaluated to obtain the arch energy data; A comprehensive energy efficiency evaluation is performed on energy recovery indices, joint energy transfer indices, external power data, and arch energy data to obtain efficiency score data.

[0019] The present invention establishes a multi-segment human body model, combines the body segment inertia parameters, joint angle data and ground force data, calculates the change curve of the position, speed, gravitational potential energy, kinetic energy and total mechanical energy of the whole body center of mass in the gait cycle over time, and obtains the center of mass energy data. These data provide a basis for in-depth analysis of energy conversion and utilization during gait. Compared with only focusing on kinematic or dynamic parameters, the center of mass energy data can more comprehensively reflect the energy state of the human body during walking, and provide an important basis for evaluating the energy efficiency of gait, identifying the weak links of energy utilization and optimizing gait. By analyzing the change trend of the center of mass gravitational potential energy and kinetic energy, the energy recovery rate (R value) between potential energy and kinetic energy is calculated to obtain the energy recovery index. The R value reflects the efficiency of the mutual conversion between potential energy and kinetic energy during gait, and is an important indicator for evaluating the energy utilization efficiency of gait. A higher R value indicates that the energy loss in the conversion between potential energy and kinetic energy is less, and the energy efficiency of gait is higher. Compared with only focusing on the change of total mechanical energy, the energy recovery index can more clearly reveal the conversion of energy between different forms, and provide a reference for optimizing gait and reducing energy consumption. For example, by improving gait posture, adjusting step frequency and step length, the energy recovery rate can be increased, thereby reducing energy consumption during walking. By calculating joint power (the product of joint torque and joint angular velocity), analyzing the power curve of each joint, identifying the positive power and negative power stages, and evaluating the energy transfer between the joints, joint energy transfer indicators are obtained. These indicators can reflect the work done by different joints during gait and the direction and magnitude of energy flow between the joints of the lower limbs. For example, in the early stance phase, the ankle joint may do negative work to absorb energy, while the hip joint may do positive work to generate energy, indicating that energy is transferred from the ground to the hip joint. Joint energy transfer analysis helps to understand the synergy of different muscle groups at different stages of the gait cycle, and provides an important basis for optimizing gait, improving movement efficiency and evaluating rehabilitation effects. By calculating the product of ground reaction force and center of mass velocity, the external power of mechanical energy is obtained, and the average external power of the entire gait cycle is calculated to obtain external power data. External power reflects the work done by the human body on the external environment and is an important indicator for evaluating gait energy consumption. Lower average external power indicates that less work is done on the outside world during walking, and the energy efficiency of gait is higher. Compared with focusing only on internal energy conversion, external power takes into account the interaction between the human body and the environment, and can more comprehensively evaluate the energy consumption of gait, providing a reference for optimizing gait and reducing energy consumption. By analyzing the plantar pressure distribution, especially the pressure-time integral of the medial and lateral arch areas, the elastic potential energy stored and released by the arch during gait is estimated, and the arch energy data is obtained. These data reflect the role of the arch in cushioning impact, storing and releasing energy, and are important indicators for evaluating gait energy efficiency. A higher arch energy recovery rate indicates that the arch can effectively use elastic deformation to store and release energy, reducing energy consumption during walking.By integrating the energy efficiency related indicators (energy recovery rate, joint energy transfer efficiency, average external power, and arch energy recovery rate) obtained from the above four aspects, a comprehensive energy efficiency scoring system is constructed to obtain efficiency scoring data. This comprehensive score can comprehensively evaluate the energy utilization efficiency of individuals during gait, avoiding the one-sidedness of a single indicator. Compared with analyzing each energy efficiency indicator separately, the comprehensive score can more accurately evaluate the energy efficiency of an individual's gait, providing a more reliable basis for identifying weak links in energy utilization, formulating personalized gait optimization plans, and evaluating rehabilitation effects. For example, the comprehensive energy efficiency score can be improved by improving gait posture, adjusting step frequency and step length, and enhancing arch function.

[0020] Preferably, the gait parameter optimization process in step S3 includes: Acquire multi-parameter gait data of subjects tested under different step frequency guidance; Energy-stability trade-off modeling was performed based on multi-parameter gait data, efficiency score data, gait stability index, and comprehensive coordination index to obtain a parameter trade-off model; Theoretical gait parameters are calculated according to the parameter trade-off model and the body segment inertia parameters to obtain a theoretical gait parameter set; Calculate the spatiotemporal parameters based on the bilateral gait cycle data and gait event markers to obtain a gait spatiotemporal parameter set; According to the gait spatiotemporal parameter set and the theoretical gait parameter set, actual parameter deviation analysis is performed to obtain a parameter deviation feature set; The individual factor data of the subjects are obtained; the individual factor correction is performed on the parameter deviation feature set according to the individual factor data to obtain the individual target parameters.

[0021] The present invention uses a metronome to guide the subject to walk at different cadences (such as 90%, 100%, 110%, and 120% of the natural cadence), and collects multimodal gait data (including plantar pressure, ground reaction force, joint angle, motion state, etc.) at each cadence, thereby obtaining a multi-parameter gait data set. These data provide a basis for analyzing the influence of cadence on gait parameters, energy consumption, stability, coordination, etc. Compared with only collecting gait data at a single speed, multi-parameter gait data can more comprehensively reflect the gait characteristics of individuals at different cadences, and provide necessary data support for establishing a personalized gait parameter optimization model. By taking gait parameters such as cadence, step length, and step width as independent variables, and taking energy consumption indicators (such as average external power) and stability indicators (such as the root mean square value of trunk acceleration) as dependent variables, a trade-off model between energy consumption and stability is established using methods such as multiple regression analysis. The model can quantitatively describe the influence of gait parameters on energy consumption and stability, and reveals the trade-off relationship between the two. Compared with adjusting gait parameters based on experience alone, the parameter trade-off model provides a more scientific and objective basis, helps to find the optimal balance between energy consumption and stability, and provides key support for the development of personalized gait optimization solutions. Based on the parameter trade-off model and body segment inertia parameters (such as leg length, body weight, etc.), the inverted pendulum model and other methods are used to calculate the theoretical optimal step frequency, step length and step width, and obtain the theoretical gait parameter set. These theoretical parameters provide a reference benchmark for gait optimization and represent the gait parameters that individuals should adopt in order to achieve optimal energy efficiency and stability under ideal conditions. Compared with relying solely on empirical values ​​or group averages, the theoretical gait parameters take into account the individual's body structure characteristics and provide a more reasonable starting point for the development of personalized gait optimization solutions. Based on the collected gait data, the actual gait spatiotemporal parameters, such as step length, step width, step frequency, support time, swing time, single / double support time, etc., are calculated to obtain the gait spatiotemporal parameter set. These parameters reflect the current gait characteristics of the individual and are the basis for evaluating the current gait status and identifying deviations from the theoretical optimal parameters. Compared with qualitative observation only, spatiotemporal parameter calculation provides an accurate quantitative description, providing data support for the objective evaluation and subsequent optimization of gait parameters. By comparing the spatiotemporal parameters of the actual gait with the theoretical optimal parameters, the deviation between the two (such as cadence deviation, stride length deviation, and stride width deviation) is calculated to obtain the parameter deviation feature set. These deviations reflect the gap between the individual's current gait and the theoretical optimal gait, indicating the direction and focus of gait optimization. Compared with subjective evaluation alone, parameter deviation analysis provides an objective quantitative evaluation, which helps to more accurately identify problems in gait and provide a basis for formulating personalized gait optimization plans. For example, if the actual cadence is significantly higher than the theoretical optimal cadence, it may be necessary to guide the subject to reduce the cadence and increase the stride length.By collecting individual factor data such as the subject's age, gender, height, weight, leg length, exercise habits, health status (such as whether there are bone and joint diseases), and adjusting the deviation between the theoretical optimal parameters and the actual parameters based on these factors, individualized target gait parameters are obtained. This individual factor correction takes into account the physiological characteristics and functional limitations of different individuals, making the target parameters more in line with the individual's actual situation and improving the feasibility and effectiveness of gait optimization. Compared with the use of a unified optimization goal, individualized target parameters can better meet the needs of different individuals, avoid discomfort or injury caused by a "one-size-fits-all" solution, and achieve more accurate and safer gait optimization. For example, for the elderly or arthritis patients, it may be necessary to appropriately reduce the target step frequency and step length to reduce joint load and maintain balance.

[0022] Preferably, the present invention further provides a health data entry system for executing the health data entry method as described above, the health data entry system comprising: The gait parameter acquisition module is used to collect the multimodal data of the subject, and perform gait event recognition to obtain gait event markers; perform bilateral gait synchronization alignment according to the gait event markers to obtain bilateral gait cycle data, wherein the bilateral gait cycle data includes plantar pressure data, ground force data, joint angle data and motion state data; The gait dynamics analysis module is used to perform plantar pressure dynamics processing based on plantar pressure data and ground force data to obtain plantar dynamics parameters; estimate the subject's body segment inertia parameters; perform joint torque energy processing based on body segment inertia parameters and joint angle data to obtain joint dynamics parameters; calculate gait acceleration based on motion state data to obtain gait stability indicators; extract phase coordination features based on joint angle data and motion state data to obtain comprehensive coordination indicators; The gait health assessment module is used to calculate the energy conversion efficiency based on the bilateral gait cycle data, plantar power parameters and joint power parameters to obtain efficiency score data; optimize the gait parameters based on the gait stability index and comprehensive coordination index of the efficiency score data to obtain individual target parameters; and perform gait health assessment based on the individual target parameters, efficiency score data and comprehensive coordination index to obtain gait health data.

[0023] Preferably, a computer-readable storage medium stores a computer program, and when the computer program is executed, the method for entering health data as described above is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A flowchart of steps of a method for entering health data; Figure 2 It is a schematic diagram of the detailed implementation steps of step S1 in the present invention.

[0025] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0026] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0027] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0028] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0029] In the embodiment of the present invention, reference Figure 1 FIG. 1 is a flow chart of the steps of the method for entering health data of the present invention. In this example, the method for entering health data includes the following steps: Step S1: collecting multimodal data of the subject, and performing gait event recognition to obtain gait event markers; performing bilateral gait synchronization alignment according to the gait event markers to obtain bilateral gait cycle data, wherein the bilateral gait cycle data includes plantar pressure data, ground force data, joint angle data and motion state data; In an embodiment of the present invention, first, a pre-calibrated pressure insole (including hundreds of capacitive pressure sensors), multiple inertial measurement units (three-axis accelerometers and three-axis gyroscopes, respectively fixed to the bilateral calves, thighs and trunks) and a two-axis electronic goniometer (fixed to the outside of the bilateral ankle joints) are used to synchronously collect multimodal data of the subject walking at a comfortable speed on a flat ground, including plantar pressure, three-dimensional ground reaction force (calculated by pressure data), lower limb joint angles and acceleration / angular velocity of various parts of the body, and the sampling frequency is uniformly set to 200Hz. Then, the collected raw data is preprocessed, including filtering (for example, using a Butterworth low-pass filter to remove high-frequency noise) and data synchronization. Next, gait event recognition is performed, mainly by analyzing the vertical ground reaction force (VGRF) and the center of plantar pressure (COP) trajectory to determine the time points of key gait events (heel landing, plantar flat touch, heel off, toe off), and form gait event markers. Finally, taking the ipsilateral heel strike as the starting point, the data between two consecutive ipsilateral heel strikes are divided into a complete gait cycle, and each cycle is time-normalized (for example, each cycle is equally divided into 100 time points). Then, the gait cycle data on the left and right sides are time-aligned to obtain bilaterally synchronized gait cycle data containing plantar pressure data, ground force data, joint angle data, and motion state data.

[0030] Step S2: performing plantar pressure dynamic processing according to the plantar pressure data and the ground force data to obtain plantar dynamic parameters; estimating the subject's body segment inertia parameters; performing joint torque energy processing according to the body segment inertia parameters and the joint angle data to obtain joint dynamic parameters; calculating gait acceleration according to the motion state data to obtain gait stability index; extracting phase coordination features according to the joint angle data and the motion state data to obtain a comprehensive coordination index; In the embodiment of the present invention, plantar pressure dynamics processing is performed based on the plantar pressure data and ground force data in the bilateral gait cycle data, including extracting the center of pressure (COP) trajectory, analyzing the COP speed, calculating the pressure distribution and pressure-time integral of each anatomical region (such as the medial heel, lateral heel, midfoot, and each metatarsal head region), extracting the peak pressure and occurrence time of each region, etc., to obtain plantar dynamics parameters. Then, according to the height, weight and other data of the subject, the human body model parameter table is consulted to estimate the mass, center of mass position and moment of inertia of each body segment (such as the foot, calf, thigh, etc.) Body segment inertia parameters such as the mass, center of mass position and moment of inertia. Then, combined with the body segment inertia parameters, joint angle data and ground force data, inverse dynamics calculation is performed, and the net torque of the ankle, knee and hip joints in the sagittal plane, frontal plane and cross-section is calculated step by step from bottom to top (from foot to hip), and joint power analysis (power = torque × angular velocity) is performed to evaluate the generation, absorption and transmission of energy between joints, calculate joint stiffness (slope of torque-angle curve), extract torque characteristics (peak torque, key moment torque, etc.), and obtain joint dynamic parameters. At the same time, statistical analysis (calculation of mean, standard deviation, root mean square value, kurtosis, skewness, etc.) and spectrum analysis (fast Fourier transform, extraction of main frequency, power spectrum density, etc.) are performed on the motion state data (acceleration data), analyze the transmission of impact in the body (calculation of the attenuation rate of the peak acceleration of different parts), calculate gait stability indicators (such as the root mean square value of trunk acceleration, gait cycle variation coefficient, left-right symmetry index, etc.), and obtain gait stability indicators. Finally, based on the joint angle data and motion state data, phase coordination features were extracted, including calculating the joint coordination angle (the angle between each point and the origin on the adjacent joint angle-angle phase diagram), analyzing the relative phase between limb segments, and evaluating the force-action coordination (calculating the mutual correlation coefficient between the vertical ground reaction force and the joint angle) and the pressure-posture coupling relationship (calculating the mutual correlation coefficient between the plantar pressure center and the trunk acceleration) to obtain a comprehensive coordination index.

[0031] Step S3: Calculate energy conversion efficiency based on bilateral gait cycle data, plantar power parameters and joint power parameters to obtain efficiency score data; optimize gait parameters based on gait stability index and comprehensive coordination index of efficiency score data to obtain individual target parameters; perform gait health assessment based on individual target parameters, efficiency score data and comprehensive coordination index to obtain gait health data; In the embodiment of the present invention, based on the bilateral gait cycle data, the body segment inertia parameters and the joint dynamic parameters, the kinetic energy of the whole body center of mass (including gravitational potential energy and kinetic energy) is calculated, the conversion efficiency between potential energy and kinetic energy is analyzed (the energy recovery rate is calculated), the transfer of joint power and energy between joints is evaluated, the external power of mechanical energy (the product of the ground reaction force and the center of mass velocity) is calculated, and the elastic energy storage of the arch of the foot is estimated (based on the plantar pressure distribution and the assumed elastic coefficient), and the energy efficiency score data is obtained by combining these indicators. Then, by guiding the subject to walk at different step frequencies (for example, 90%, 100%, 110%, 120% of the natural step frequency), multiple groups of gait data are collected, and the trade-off model between energy consumption and stability is established by combining the energy efficiency score data, the gait stability index and the comprehensive coordination index (for example, using multiple regression analysis), and the theoretical optimal gait parameters (step frequency, step length, step width) are estimated based on the inverted pendulum model and data such as height and weight, and the deviation between the actual gait parameters and the theoretical optimal parameters is calculated. Next, the theoretical optimal parameters are adjusted to obtain individualized target gait parameters by considering individual factors such as the subject's age, gender, exercise habits, and health status. Finally, gait health data is generated by integrating individual target parameters, energy efficiency scores, gait stability indicators, comprehensive coordination indicators, and evaluation results such as joint load, gait symmetry, and fall risk.

[0032] As an example of the present invention, refer to Figure 2 As shown, in this example, step S1 includes: Step S11: using a pre-calibrated pressure sensor array to perform multimodal data on the subject to obtain a multimodal signal set; In the embodiment of the present invention, a method comprising 960 (40 The insole pressure measurement system of the capacitive pressure sensor (24 matrix arrangement) is used. Each sensor has an area of ​​1 square centimeter and a measurement range of 0-100 kPa. After the subjects put on the insoles on both feet, they were connected to the data acquisition system. The model of the data acquisition system is NI-USB-6216 (National Instruments Corporation), and the sampling frequency is set to 200 Hz. Before the formal data acquisition, the pressure sensor array is pre-calibrated: standard weights (1kg, 2kg, 5kg, 10kg) are placed in different areas of the insole in turn, and the voltage value output by each pressure sensor is recorded when each weight is placed. The pressure-voltage conversion coefficient of each pressure sensor is calculated by least squares fitting, and a 960×1 calibration matrix is ​​generated. Then, the subjects walked in a straight line at a self-comfortable speed on a flat ground with a length of 10 meters. The data acquisition system synchronously collects the voltage signals output by the pressure sensor array, the signals output by six triaxial accelerometers (model Xsens MTw Awinda, measuring range ±16g) and six triaxial gyroscopes (model Xsens MTw Awinda, measuring range ±2000° / s) installed on the bilateral thighs, calves and feet, and the signals output by the biaxial electronic goniometer (model Biometrics SX230, measuring range 0-180°) installed on the outside of the bilateral ankle joints, to form a multimodal signal set.

[0033] Step S12: performing gait event recognition on the multimodal signal set to obtain gait event labels; In an embodiment of the present invention, gait event recognition is performed on a multimodal signal set. First, a second-order Butterworth low-pass filter with a cutoff frequency of 10Hz is applied to the original pressure data to filter and remove high-frequency noise. Then, the sum of the pressures of all valid pressure sensors (sensors with pressure values ​​greater than 5kPa) at each moment is calculated to obtain a vertical ground reaction force (VGRF) curve. The initial contact (IC) event is defined as the moment when the VGRF exceeds 20N. The toe off (TO) event is defined as the moment when the VGRF is less than 20N. In order to more finely identify gait events, the center of pressure (Center of Pressure, COP) trajectory is further analyzed. The calculation method of COP is: at each moment, the weighted average of the product of the pressure values ​​of all valid pressure sensors and their coordinate positions is calculated. The foot flat (Foot Flat, FF) event is defined as the moment when the COP front-to-back direction velocity crosses zero for the first time. The heel off (HO) event is defined as the moment when the COP front-to-back direction velocity crosses zero for the second time. The time point of each gait event is recorded to form a gait event marker.

[0034] Step S13: dividing the multimodal signal set into gait cycles according to the gait event markers to obtain cycle standardized data; In an embodiment of the present invention, the multimodal signal set is divided into gait cycles according to gait event markers. The signal from the left foot IC to the next left foot IC is selected as a complete gait cycle. The duration of each gait cycle is normalized, and each cycle is equally divided into 100 time points. For each time point, the pressure data, acceleration data, angular velocity data and joint angle data corresponding to the moment are calculated by linear interpolation. For example, if the duration of a gait cycle is 1.2 seconds, the time corresponding to the 25th time point is 1.2 seconds × (25 / 100) = 0.3 seconds. The measurement values ​​of all sensors at 0.3 seconds are obtained by interpolation. In this way, each gait cycle is converted into a 100×N matrix, where N is the number of all sensors. The matrices of all gait cycles are integrated to form cycle standardized data.

[0035] Step S14: performing bilateral gait synchronization alignment according to the cycle standardization data to obtain bilateral gait cycle data; In an embodiment of the present invention, bilateral gait synchronization alignment is performed according to cycle standardized data. First, the starting points of the left and right gait cycles (i.e., left foot IC and right foot IC) are found respectively. Then, the starting points of the left gait cycle and the right gait cycle are aligned. Since the gait cycle durations of the left and right feet may differ slightly, it is necessary to perform linear stretching on the time axis for the shorter gait cycle to make it consistent with the longer gait cycle length. For example, if the left foot gait cycle is 100 time points and the right foot gait cycle is 98 time points, the time interval of each data point of the right foot gait cycle is multiplied by 100 / 98 so that its total duration also becomes 100 time points. In this way, the gait cycle data on the left and right sides are fully aligned in time to form bilateral gait cycle data, including bilaterally synchronized plantar pressure data, ground force data, joint angle data, and motion state data.

[0036] Preferably, the foot pressure dynamic processing in step S2 includes: Extract the plantar pressure center trajectory according to the plantar pressure data to obtain the pressure center trajectory data; Perform pressure center velocity analysis on pressure center trajectory data to obtain pressure center dynamic parameters; The regional pressure distribution time series analysis is performed based on the ground force data to obtain the regional pressure pattern; Perform pressure integral analysis based on regional pressure patterns to obtain cumulative pressure maps; The pressure peak time series analysis is performed based on the dynamic parameters of the pressure center, the regional pressure pattern and the cumulative pressure spectrum to obtain the pressure time series characteristics.

[0037] In an embodiment of the present invention, based on the plantar pressure data in the bilateral gait cycle data, the center of pressure (COP) coordinates at each time point are calculated. The specific method is: the coordinate system of the insole pressure measurement system is defined as: the x-axis points to the inside of the foot, the y-axis points to the front of the foot, and the origin is located at the left heel corner of the insole. For each time point t, the sum of the products of the pressure values ​​of all pressure sensors P (i, t) and their x-coordinates X (i) is calculated, and the sum is divided by the sum of the pressure values ​​of all pressure sensors to obtain the x-coordinate COPx (t) of the COP; similarly, the sum of the products of the pressure values ​​of all pressure sensors P (i, t) and their y-coordinates Y (i) is calculated, and the sum is divided by the sum of the pressure values ​​of all pressure sensors to obtain the y-coordinate COPy (t) of the COP. The COP coordinates of all time points are connected to form a COP trajectory. In order to eliminate high-frequency noise, the COP trajectory is smoothed, and a moving average filter is used, and the window size is set to 5 time points. The smoothed COP trajectory data is output, that is, the pressure center trajectory data.

[0038] Based on the pressure center trajectory data, the speed of COP in the x-direction and y-direction is calculated. The speed is calculated using the central difference method: the x-direction speed of COP at time t COPvx(t)=(COPx(t+1)-COPx(t-1)) / (2×Δt), where Δt is the sampling time interval (1 / 200 seconds in this example). Similarly, the y-direction speed of COP at time t COPvy(t)=(COPy(t+1)-COPy(t-1)) / (2×Δt). The calculated COP speed curve is filtered using a second-order Butterworth low-pass filter with a cutoff frequency of 5Hz. Extract key parameters: 1. Average forward velocity of COP in the early stance phase (from heel strike to plantar flat contact); 2. Average forward velocity of COP in the middle stance phase (from plantar flat contact to heel off); 3. Average forward velocity of COP in the late stance phase (from heel off to toe off); 4. Maximum medial and lateral displacement of COP in the stance phase. These parameters are used as dynamic parameters of the center of pressure.

[0039] Based on the plantar pressure data in the bilateral gait cycle data and the gait event markers, the plantar is divided into seven anatomical regions: 1. medial heel; 2. lateral heel; 3. midfoot; 4. first metatarsal head; 5. second metatarsal head; 6. third to fifth metatarsal heads; 7. big toe. Each region contains a number of predefined pressure sensors. For each time point t, the average pressure of all pressure sensors in each region is calculated as the pressure value of the region at that moment. Then, the percentage of the pressure value of each region to the sum of the pressure values ​​of all seven regions is calculated to obtain the pressure share of the region at that moment. The curves of the pressure share of the seven regions changing with the gait cycle (0-100%) are plotted to form a regional pressure pattern.

[0040] Based on the regional pressure pattern, the pressure-time integral of each region is calculated. For each region, its pressure-ratio curve is integrated over time, and the integral interval is the entire support phase from heel strike (IC) to toe-off (TO). The trapezoidal integration method is used for numerical integration: the pressure-ratio curve is divided into multiple small trapezoids, the width of each trapezoid is the sampling time interval Δt, and the height is the average value of the pressure ratio of two adjacent time points. The sum of the areas of all small trapezoids is calculated as the pressure-time integral value of the region. For standardization, the pressure-time integral value of each region is divided by the sum of the pressure-time integral values ​​of all seven regions to obtain the relative pressure integral value of each region. These relative pressure integral values ​​are displayed in the form of a bar graph to form a cumulative pressure map.

[0041] Based on the dynamic parameters of the pressure center, the regional pressure pattern and the cumulative pressure map, the pressure peak time series characteristics are extracted. For each anatomical region, find the peak value of its pressure ratio curve, record the peak pressure ratio and its corresponding time point (percentage relative to the gait cycle). Analyze the order of occurrence of the peak pressure ratio: Under normal gait, the peak pressure ratio should appear in the medial heel, lateral heel, midfoot, first metatarsal head, second metatarsal head, third to fifth metatarsal heads, and big toe. If the order of occurrence of the peak pressure ratio is abnormal, for example, the peak pressure ratio in the midfoot area is earlier than that in the heel area, it indicates that there may be abnormal foot force line. At the same time, the product of the peak pressure ratio of each region and the average speed of the COP trajectory in the region is calculated as the pressure impact index of the region. Output the pressure peak time series characteristics, including the peak pressure ratio of each region, the peak occurrence time, the peak occurrence order and the pressure impact index.

[0042] Preferably, the joint torque energy processing in step S2 includes: Perform inverse dynamics calculation based on body segment inertia parameters, joint angle data and ground force data to obtain joint torque data; Perform joint power analysis based on joint torque data and joint angle data to obtain energy transfer data; Calculate joint stiffness based on joint torque data to obtain stiffness characteristic data; The torque feature is extracted according to the joint torque data and gait event markers to obtain the torque feature data.

[0043] In an embodiment of the present invention, inverse dynamics calculation is performed based on the joint angle data and ground force data in the bilateral gait cycle data, and the body segment inertia parameters (including the mass of each body segment, the position of the center of mass, and the moment of inertia) obtained by looking up the height and weight table. A bottom-up (foot to hip) calculation sequence is adopted. First, the net torque of the ankle joint is calculated based on the position of the center of plantar pressure (COP) and the three components of the ground reaction force (vertical force, front and rear shear force, and internal and external shear force). The ankle joint torque is equal to the cross product of the ground reaction force vector and the distance vector from the COP to the center of the ankle joint, plus the inertia force and moment of inertia generated by the foot mass. Then, the net torque of the knee joint is calculated based on the ankle joint torque, the ankle joint angle, and the angular acceleration of the foot. The knee joint torque is equal to the ankle joint torque plus the torque generated by the calf gravity, plus the calf inertia force and moment of inertia. Finally, the net torque of the hip joint is calculated based on the knee joint torque, the knee joint angle, and the calf angular acceleration. The hip joint torque is equal to the knee joint torque plus the torque generated by the thigh gravity, plus the thigh inertia force and inertia moment. All torques are decomposed into three planes: sagittal plane, frontal plane and transverse plane. The net torques of the three joints in the three planes at each moment are output to form joint torque data.

[0044] Calculate joint power based on joint torque data and joint angle data. First, differentiate the joint angle data to obtain the joint angular velocity. Calculate the angular velocity using the central difference method: the angular velocity of the joint at time t is ω(t) = (θ(t+1) - θ(t-1)) / (2 × Δt), where θ(t) is the joint angle and Δt is the sampling time interval. Then, multiply the joint torque with the corresponding joint angular velocity to obtain the instantaneous joint power. For example, the hip sagittal plane power is equal to the hip sagittal plane torque multiplied by the hip sagittal plane angular velocity. Analyze the power curve of each joint to identify the positive power phase (the joint does positive work and produces energy) and the negative power phase (the joint does negative work and absorbs energy). Calculate the total positive and negative work of each joint throughout the gait cycle. Analyze the energy transfer between the joints: For example, if the ankle joint does negative work to absorb energy and the hip joint does positive work to produce energy in the early stance phase, it indicates that energy is transferred from the ground to the hip joint. Output energy transfer data, including power curves of each joint, total positive work, total negative work, and energy transfer direction.

[0045] Calculate joint stiffness based on joint torque data. Select data from the stance phase (from heel strike to toe off) for analysis. For each joint, draw its torque-angle curve. For example, for the ankle joint, draw a curve of the ankle joint sagittal plane torque and the ankle joint dorsiflexion / plantar flexion angle. Calculate the slope of the curve in the middle of the stance phase as the dynamic stiffness of the joint. The specific method is: select the data points in the middle of the stance phase (20%-80% gait cycle), use the linear regression method to fit the torque-angle curve, and the slope of the regression line is the joint stiffness. Calculate the stiffness of the hip, knee, and ankle joints in the sagittal plane, frontal plane, and transverse plane respectively. Output stiffness characteristic data, including the stiffness values ​​of each joint in three planes.

[0046] Extract torque features based on joint torque data and gait event markers. For each plane of each joint (sagittal plane, frontal plane, transverse section), find the peak value (maximum and minimum value) of its torque curve, record the magnitude of the peak torque and the time point of occurrence (relative to the percentage of the gait cycle). For example, for the sagittal plane of the hip joint, record the peak flexion torque, the peak extension torque and the time of occurrence. In addition, calculate the torque values ​​at several key moments: 1. The torque at heel strike (IC); 2. The torque at the middle of the stance phase (50% of the gait cycle); 3. The torque at toe-off (TO). Analyze the morphological characteristics of the torque curve: for example, whether there is a double peak pattern (under normal circumstances, the ankle plantar flexion torque should have a double peak in the late stance phase), whether there is an abnormal torque mutation. Output torque feature data, including the peak torque of each joint and each plane, the time of peak occurrence, the torque value at the key moment, and the description of the morphological characteristics of the torque curve.

[0047] Preferably, the gait acceleration calculation in step S2 includes: Perform acceleration feature statistics according to the motion state data to obtain acceleration features; Spectral characteristics are analyzed based on motion state data and gait event markers to obtain acceleration spectrum characteristics; Perform shock transfer analysis based on acceleration characteristics and acceleration spectrum characteristics to obtain shock transfer characteristics; The stability index is calculated according to the impact transmission characteristics and acceleration characteristics to obtain the gait stability index.

[0048] In an embodiment of the present invention, the statistical characteristics of acceleration are calculated based on the motion state data in the bilateral gait cycle data. The motion state data includes the measurement values ​​of six triaxial accelerometers installed on the bilateral calves, bilateral thighs and trunk (waist). For each axis (x, y, z) of each accelerometer, the following statistics are calculated respectively: 1. Mean value: the arithmetic mean of the acceleration during the entire gait cycle. 2. Standard deviation: reflects the degree of fluctuation of the acceleration around the mean value. 3. Root mean square value (RMS): the square root of the mean value of the square of the acceleration, reflecting the vibration intensity. 4. Kurtosis: reflects the sharpness of the acceleration distribution, and the kurtosis of the normal distribution is 3. 5. Skewness: reflects the asymmetry of the acceleration distribution, and the skewness of the normal distribution is 0. The above statistics are calculated for the support phase and the swing phase respectively. All statistics are combined into a vector as the acceleration feature.

[0049] Based on the motion state data and gait event markers in the bilateral gait cycle data, the spectrum characteristics are analyzed. The data of the support phase are selected for analysis. For each axis of each accelerometer, a fast Fourier transform (FFT) is performed to convert the time domain signal into a frequency domain signal. The power spectral density (PSD) is calculated to reflect the distribution of signal energy at different frequencies. The peak frequency (main frequency) of the PSD curve is found, that is, the frequency component with the highest energy. Calculate the percentage of power at the main frequency to the total power. Calculate the energy proportion of the three frequency bands of 0-3Hz, 3-8Hz and above 8Hz. 0-3Hz mainly reflects the basic frequency and low-frequency movement of gait, 3-8Hz mainly reflects the rhythmicity of gait, and above 8Hz mainly reflects impact and high-frequency vibration. The main frequency, the main frequency power proportion and the energy proportion of the three frequency bands are used as acceleration spectrum characteristics.

[0050] Based on the acceleration characteristics and acceleration spectrum characteristics, the transmission of impact in the human body is analyzed. Data 0.1 seconds after heel strike (IC) is selected for analysis. For the vertical axis (z-axis) of each accelerometer, the peak acceleration within 0.1 seconds after IC is calculated. The ratio of the peak acceleration of the calf, thigh, and trunk to the peak acceleration of the foot (assuming that the position is close to the calf accelerometer) is calculated as the impact attenuation rate. For example, the torso impact attenuation rate = trunk peak acceleration / foot peak acceleration. Perform FFT on the vertical axis acceleration signal of each accelerometer to calculate the energy proportion of the frequency band above 8Hz within 0.1 seconds after IC. Compare the energy proportion of the frequency band above 8Hz of the calf, thigh, and trunk to analyze the transmission of high-frequency impact in the body. The impact attenuation rate and the change of high-frequency energy proportion are used as the impact transmission characteristics.

[0051] Based on the impact transmission characteristics and acceleration characteristics, the gait stability index is calculated. The RMS values ​​of the three axes of the trunk (waist) accelerometer are calculated as an indicator of overall stability. The larger the RMS value, the greater the trunk shaking and the worse the stability. Calculate the coefficient of variation (CV) of the gait cycle: calculate the duration of multiple consecutive gait cycles (from the left foot IC to the next left foot IC), calculate the standard deviation of these durations, and then divide them by the average to obtain the CV. The larger the CV, the worse the gait rhythm and the lower the stability. Calculate the symmetry index (SI) of the left and right calf accelerations: SI=2×|AL-AR| / (AL+AR), where AL and AR are certain characteristic values ​​of the left and right calf accelerations, respectively (such as RMS values). The closer the SI is to 0, the more symmetrical the left and right sides are, and the more stable the gait. The trunk RMS value, the gait cycle CV, and the left-right symmetry index SI are used as gait stability indicators.

[0052] Preferably, the phase coordination feature extraction in step S2 includes: calculating joint coordination angles of joint angle data according to gait event markers; Perform segmented relative phase analysis based on joint coordination angle and motion state data to obtain phase difference data; Perform force-action coordination analysis based on joint angle data and ground force data to obtain force-action coordination data; Pressure-posture coupling analysis is performed based on pressure center trajectory data, regional pressure patterns, and motion state data to obtain pressure-posture coupling features; The joint coordination angle, phase difference data, force-action coordination data and pressure-posture coupling characteristics are integrated to obtain a comprehensive coordination index.

[0053] In an embodiment of the present invention, the joint coordination angle is calculated based on the gait event markers and the joint angle data. The angles of the hip joint, knee joint and ankle joint in the sagittal plane are selected for analysis. For every two adjacent joints (hip-knee, knee-ankle), an angle-angle phase diagram is constructed. For example, for the hip-knee joint, the hip joint flexion and extension angle is used as the horizontal coordinate, and the knee joint flexion and extension angle is used as the vertical coordinate, and the points corresponding to each moment are drawn to form a phase diagram. On the phase diagram, the angle of each point and the line connecting the origin is calculated as the joint coordination angle at that moment. The specific calculation method is: coordination angle = arctan (joint B angle / joint A angle), where joint A is the proximal joint (hip joint or knee joint), and joint B is the distal joint (knee joint or ankle joint). The average coordination angle of the support phase and the swing phase is calculated respectively. The change curve of the coordination angle over time is analyzed to identify the transition point of the coordination mode: for example, the change trend of the coordination angle from the early to the middle stage of the support phase. Output the joint coordination angle data, including the average coordination angle of the support phase and the swing phase and the change curve of the coordination angle over time.

[0054] Based on the joint coordination angle and motion state data (angular velocity data), segmented relative phase analysis is performed. For each limb segment (foot, calf, thigh), its phase angle relative to the vertical direction is calculated. The phase angle is defined as: arctan (angular velocity / angular displacement). The angular displacement is obtained by processing the joint angle data. For example, the angular displacement of the calf can be obtained by subtracting the ankle angle from the knee angle. The angular velocity is obtained by derivation of the joint angle data using the central difference method. The difference in phase angles between two adjacent segments is calculated as the relative phase. For example, the calf-foot relative phase = calf phase angle - foot phase angle. Analyze the relative phase change curve with the gait cycle. Calculate the standard deviation of the relative phase to reflect the stability of coordination. Output phase difference data, including the relative phase change curve over time and the standard deviation of the relative phase.

[0055] Force-action coordination analysis is performed based on joint angle data and ground force data. The vertical ground reaction force (VGRF) and the hip, knee, and ankle joint angles in the sagittal plane are selected for analysis. The cross-correlation coefficient between VGRF and each joint angle is calculated. The cross-correlation coefficient reflects the similarity of two signals under different time delays. The specific calculation method is: time-shift the VGRF signal and the joint angle signal respectively, calculate the Pearson correlation coefficient under different time delays, and find the maximum correlation coefficient and its corresponding time delay. For example, if the maximum correlation coefficient between VGRF and the ankle joint angle occurs when VGRF leads the ankle joint angle by 100 milliseconds, it means that the change of VGRF leads the change of the ankle joint angle by 100 milliseconds. Analyze the size and time delay of the maximum correlation coefficient to evaluate the coordination degree of force and action. Output force-action coordination data, including the maximum cross-correlation coefficient between VGRF and each joint angle and its time delay.

[0056] Pressure-posture coupling analysis was performed based on the center of pressure trajectory data, regional pressure patterns, and motion state data (trunk acceleration). The relationship between the anterior-posterior position of the center of pressure (COP) and the anterior-posterior acceleration of the trunk was calculated. First, the COP position and trunk acceleration data were low-pass filtered (cutoff frequency 3Hz) to remove high-frequency noise. Then, the cross-correlation coefficient between the two was calculated to find the maximum correlation coefficient and its corresponding time delay. The time delay reflects the reaction time of the trunk posture adjustment caused by the COP change. In addition, the relationship between COP and trunk acceleration under different gait events (e.g., heel strike, flat foot strike, heel off) was analyzed. For example, when the heel strikes, the COP moves forward quickly, and the trunk may have a backward acceleration to maintain balance. The pressure-posture coupling characteristics were output, including the maximum cross-correlation coefficient between COP and trunk acceleration and its time delay, as well as the description of the coupling relationship under different gait events.

[0057] The joint coordination angle, phase difference data, force-action coordination data, and pressure-posture coupling features were integrated to construct a comprehensive coordination index. The following parameters were combined into a vector: 1. Average joint coordination angles in stance and swing phases (hip-knee, knee-ankle); 2. Relative phase standard deviation (foot-calf, calf-thigh); 3. Maximum cross-correlation coefficient and time delay between VGRF and each joint angle; 4. Maximum cross-correlation coefficient and time delay between COP and trunk acceleration. Principal component analysis (PCA) was performed on this vector to extract the first few principal components (e.g., principal components that explained more than 80% of the variance). Each principal component represents a coordination pattern, and its coefficients reflect the contribution of different parameters to the pattern. The subject's coordination parameters were projected into the principal component space to obtain their scores on each principal component. These scores can be used as a comprehensive coordination index to evaluate the overall coordination level and coordination pattern of an individual.

[0058] Preferably, the energy conversion efficiency calculation in step S3 includes: The center of mass motion energy is calculated based on the body segment inertia parameters, bilateral gait cycle data and joint dynamic parameters to obtain the center of mass energy data; The potential energy-kinetic energy conversion efficiency is calculated for the center of mass energy data to obtain the energy recovery index; Perform joint power and energy transfer analysis on the joint torque data and joint angle data in the joint dynamic parameters to obtain the joint energy transfer index; Calculate the mechanical energy external power according to the center of mass energy data and the ground force data to obtain the external power data; According to the plantar dynamic parameters, the arch elastic energy storage is evaluated to obtain the arch energy data; A comprehensive energy efficiency evaluation is performed on energy recovery indices, joint energy transfer indices, external power data, and arch energy data to obtain efficiency score data.

[0059] In an embodiment of the present invention, the kinetic energy of the center of mass of the whole body is calculated based on the body segment inertia parameters (mass of each body segment, center of mass position, moment of inertia), bilateral gait cycle data (joint angle data) and joint dynamic parameters. A 14-segment human body model (two feet, two calves, two thighs, pelvis, trunk, head, two upper arms, two forearms) is used. According to the joint angle data and the length of the body segment, the position of the center of mass of each body segment is calculated at each moment. The position of the center of mass of each body segment is multiplied by its mass, and then divided by the total mass of the whole body to obtain the position of the center of mass of the whole body. The center of mass position of the whole body is derived to obtain the center of mass velocity. The speed is calculated using the central difference method. Calculate the gravitational potential energy of the center of mass: Ep=mgh, where m is the mass of the whole body, g is the acceleration of gravity (9.81m / s²), and h is the height of the center of mass. Calculate the kinetic energy of the center of mass: Ek=1 / 2×mv², where v is the center of mass velocity (including horizontal and vertical components). Add the potential energy and kinetic energy to obtain the total mechanical energy of the center of mass. Output center of mass energy data, including the curves of center of mass position, velocity, potential energy, kinetic energy and total mechanical energy changing with time.

[0060] Based on the center of mass energy data, calculate the conversion efficiency between potential energy and kinetic energy. During the gait cycle, potential energy and kinetic energy usually show opposite trends: when potential energy decreases, kinetic energy increases (energy is converted from potential energy to kinetic energy); when potential energy increases, kinetic energy decreases (energy is converted from kinetic energy to potential energy). Calculate the change in potential energy (ΔEp) and the change in kinetic energy (ΔEk). Select a time period in which the potential energy and kinetic energy change in opposite directions (for example, from the highest point to the lowest point, or from the lowest point to the highest point). Calculate the energy recovery rate: R=(|ΔEp|+|ΔEk|-|ΔEp+ΔEk|) / (|ΔEp|+|ΔEk|). The higher the R value, the higher the conversion efficiency between potential energy and kinetic energy, and the less energy loss. Calculate the energy recovery rate of the stance phase and the swing phase respectively. Output the energy recovery index, including the R value of the stance phase and the swing phase.

[0061] Based on the joint dynamic parameters (joint torque data and joint angle data), joint power and energy transfer analysis is performed, and the calculation method is: joint power = joint torque × joint angular velocity. Analyze the power curve of each joint to identify the positive power phase (joints do positive work and generate energy) and the negative power phase (joints do negative work and absorb energy). Calculate the total positive work and total negative work of each joint during the entire gait cycle. Analyze the energy transfer between joints: If the ankle joint does negative work to absorb energy and the hip joint does positive work to generate energy in the early stance phase, it indicates that energy is transferred from the ground to the hip joint. If the hip joint does positive work and the knee joint does negative work in the swing phase, it indicates that energy is transferred from the hip joint to the knee joint. Output joint energy transfer indicators, including the power curve of each joint, the total positive work, the total negative work, and the description of the direction and magnitude of energy transfer.

[0062] Based on the center of mass energy data and ground force data, calculate the mechanical energy external power. External power refers to the work done by the human body on the external environment, or the work done by the external environment on the human body. The calculation method is: External power = ground reaction force × center of mass velocity. The ground reaction force includes three components: vertical force (Fz), front-back shear force (Fy) and internal-external shear force (Fx). The center of mass velocity also includes three components: vertical velocity (Vz), front-back velocity (Vy) and internal-external velocity (Vx). External power = Fx × Vx + Fy × Vy + Fz × Vz. Calculate the average external power of the entire gait cycle. Analyze the characteristics of the external power curve: for example, in the early stage of the stance phase, the vertical force is large. If the center of mass moves upward at this time, the external power is positive; in the late stage of the stance phase, the front-back shear force is large. If the center of mass moves forward at this time, the external power is also positive. Output external power data, including external power curve and average external power.

[0063] Based on the plantar dynamic parameters (regional pressure pattern), the elastic energy storage of the arch is estimated. The sole is divided into the medial arch region and the lateral arch region. The medial arch region mainly includes the medial heel, midfoot and first metatarsal head region. The lateral arch region mainly includes the lateral heel, midfoot and fifth metatarsal head region. The pressure-time integral of each region is calculated. In the middle of the stance phase (about 20%-50% of the gait cycle), the arch is deformed by pressure and stores elastic potential energy. Assume that the elastic deformation of the arch is proportional to the pressure-time integral. Estimate the elastic energy storage of the arch: E=k×∫Pdt, where P is pressure, t is time, and k is a proportionality coefficient that needs to be estimated based on the anatomical structure and material properties of the foot (in this example, k=0.01J / kPa·s is assumed). In the propulsion phase (about 50%-100% of the gait cycle), the arch releases elastic potential energy to assist propulsion. Calculate the energy released by the arch: Erelease = E-Eremaining, where Eremaining is the residual energy of the arch when the toe leaves the ground, which is also estimated by pressure-time integration. Output arch energy data, including arch energy stored, released energy and energy recovery rate (Erelease / E).

[0064] Integrate energy recovery index, joint energy transfer index, external power data, and arch energy data to conduct a comprehensive evaluation of energy efficiency. Construct an energy efficiency scoring system that includes the following aspects: 1. Potential energy-kinetic energy conversion efficiency (energy recovery rate); 2. Joint energy transfer efficiency (the ratio of positive and negative work in each joint, whether the direction of energy transfer is reasonable); 3. Mechanical energy external power (the size of the average external power); 4. Arch elastic energy storage utilization rate (energy recovery rate). Score each aspect, for example, energy recovery rate > 80% is rated as excellent, 60%-80% is rated as good, and < 60% is rated as poor. Calculate a comprehensive energy efficiency score based on the scores of each aspect. For example, a weighted average method can be used, and the weights of different aspects can be adjusted according to their importance. Output efficiency scoring data, including scores of each aspect and a comprehensive score.

[0065] Preferably, the gait parameter optimization process in step S3 includes: Acquire multi-parameter gait data of subjects tested under different step frequency guidance; Energy-stability trade-off modeling was performed based on multi-parameter gait data, efficiency score data, gait stability index, and comprehensive coordination index to obtain a parameter trade-off model; Theoretical gait parameters are calculated according to the parameter trade-off model and the body segment inertia parameters to obtain a theoretical gait parameter set; Calculate the spatiotemporal parameters based on the bilateral gait cycle data and gait event markers to obtain a gait spatiotemporal parameter set; According to the gait spatiotemporal parameter set and the theoretical gait parameter set, actual parameter deviation analysis is performed to obtain a parameter deviation feature set; The individual factor data of the subjects are obtained; the individual factor correction is performed on the parameter deviation feature set according to the individual factor data to obtain the individual target parameters.

[0066] In an embodiment of the present invention, a metronome is used to guide a subject to walk at different cadences. First, the subject is asked to walk at a speed that he feels comfortable with, and his natural cadence is recorded. Then, 90%, 100%, 110% and 120% of the natural cadence are used as target cadences, and a sound signal is emitted by the metronome to guide the subject to walk according to the rhythm of the metronome. At each cadence, at least 10 complete gait cycle data are collected. The collected data include multimodal data: plantar pressure data, ground force data, joint angle data and motion state data. For each cadence, the corresponding gait parameters such as step length, step width, support time, swing time, etc. are calculated. The multimodal data and gait parameters at different cadences are combined into a multi-parameter gait data set. Based on the multi-parameter gait data set, as well as efficiency score data (energy recovery rate, energy transfer efficiency, external power, arch energy recovery rate), gait stability indicators (trunk RMS, gait cycle CV, symmetry index) and comprehensive coordination indicators (joint coordination angle, phase difference, force-action coordination, pressure-posture coupling), a trade-off model between energy consumption and stability was established. The cadence, step length and step width were used as independent variables, and the energy consumption indicators (e.g., average external power) and stability indicators (e.g., trunk RMS) were used as dependent variables. The multivariate regression analysis method was used to establish the mathematical relationship between energy consumption and stability and the cadence, step length and step width, respectively. For example: energy consumption = a1×cadence + a2×step length + a3×step width + a4; stability = b1×cadence + b2×step length + b3×step width + b4, where a1-a4 and b1-b4 are regression coefficients. By adjusting the cadence, step length and step width, an optimal balance between energy consumption and stability can be found. This regression model is used as a parameter trade-off model. Based on the parameter trade-off model and the inertial parameters of the body segments (mass of each body segment, center of mass position, moment of inertia), the theoretical optimal gait parameters are calculated. The inverted pendulum model is used to estimate the theoretical optimal step frequency. The pendulum length of the inverted pendulum model is the leg length (the distance from the hip joint to the ankle joint), and the mass of the pendulum is the mass of the human body. The natural frequency of the inverted pendulum model is f=1 / (2π)×√(g / L), where g is the acceleration of gravity and L is the leg length. The theoretical optimal step frequency should be close to the natural frequency of the inverted pendulum model. The theoretical optimal step frequency is calculated based on leg length and body weight. The theoretical optimal step length is usually related to height and can be estimated by the empirical formula: theoretical optimal step length = 0.45×height. The step width is usually small, generally 10%-20% of the foot length. Based on these theoretical calculations, the theoretical gait parameter set is obtained, including the theoretical optimal step frequency, step length, and step width. Based on the bilateral gait cycle data and gait event markers, the spatiotemporal parameters of the actual gait are calculated. Stride length: Calculate the distance the heel moves forward on the ground between two consecutive heel strikes on the same side. Stride width: Calculate the lateral distance between the left and right heels. Cadence: Calculate the number of steps per minute, cadence = 60 / gait cycle duration (seconds). Stance time: The time from heel strike to toe-off on the same side.Swing time: the time from toe-off to heel-on-the-same-side landing. Single support time: the time when one lower limb supports body weight. Double support time: the time when both feet touch the ground at the same time. Combine these parameters into a gait spatiotemporal parameter set. Based on the gait spatiotemporal parameter set and the theoretical gait parameter set, calculate the deviation between the actual gait parameters and the theoretical optimal parameters. Calculate the difference between the actual step frequency and the theoretical optimal step frequency, as well as the percentage of the difference: step frequency deviation = (actual step frequency - theoretical optimal step frequency) / theoretical optimal step frequency × 100%. Similarly, calculate the step length deviation and step width deviation. Analyze the direction and size of the deviation: for example, if the actual step frequency is higher than the theoretical optimal step frequency, it indicates that the subject may tend to walk fast in small steps; if the actual step length is less than the theoretical optimal step length, it indicates that the subject may have a short stride. Combine these deviation values ​​into a parameter deviation feature set. Obtain the subject's individual factor data, including age, gender, height, weight, leg length, exercise habits, health status (for example, whether there is bone and joint disease), etc. According to these individual factors, correct the parameter deviation feature set. For example, for the elderly, due to decreased balance ability, their optimal step frequency may be slightly lower than the theoretical calculated value, while the step width may be slightly larger than the theoretical calculated value. For patients with knee arthritis, their optimal step length may need to be appropriately reduced to reduce the joint load. According to these individual factors, the theoretical optimal parameters are adjusted to obtain individualized target gait parameters. For example, the target step frequency can be set to 95% of the theoretical optimal step frequency, and the target step length can be set to 90% of the theoretical optimal step length. Output individual target parameters, including target step frequency, step length, and step width.

[0067] Preferably, the present invention further provides a health data entry system for executing the health data entry method as described above, the health data entry system comprising: The gait parameter acquisition module is used to collect the multimodal data of the subject, and perform gait event recognition to obtain gait event markers; perform bilateral gait synchronization alignment according to the gait event markers to obtain bilateral gait cycle data, wherein the bilateral gait cycle data includes plantar pressure data, ground force data, joint angle data and motion state data; The gait dynamics analysis module is used to perform plantar pressure dynamics processing based on plantar pressure data and ground force data to obtain plantar dynamics parameters; estimate the subject's body segment inertia parameters; perform joint torque energy processing based on body segment inertia parameters and joint angle data to obtain joint dynamics parameters; calculate gait acceleration based on motion state data to obtain gait stability indicators; extract phase coordination features based on joint angle data and motion state data to obtain comprehensive coordination indicators; The gait health assessment module is used to calculate the energy conversion efficiency based on the bilateral gait cycle data, plantar power parameters and joint power parameters to obtain efficiency score data; optimize the gait parameters based on the gait stability index and comprehensive coordination index of the efficiency score data to obtain individual target parameters; and perform gait health assessment based on the individual target parameters, efficiency score data and comprehensive coordination index to obtain gait health data.

[0068] Preferably, a computer-readable storage medium stores a computer program, and when the computer program is executed, the method for entering health data as described above is implemented.

[0069] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0070] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for entering health data, characterized in that: The following steps are involved: Step S1: collecting multimodal data of the subject, and performing gait event recognition to obtain gait event markers; Perform bilateral gait synchronization alignment according to gait event markers to obtain bilateral gait cycle data, wherein the bilateral gait cycle data includes plantar pressure data, ground force data, joint angle data and motion state data; Step S2: performing plantar pressure dynamic processing according to the plantar pressure data and the ground force data to obtain plantar dynamic parameters; estimating the subject's body segment inertia parameters; performing joint torque energy processing according to the body segment inertia parameters and the joint angle data to obtain joint dynamic parameters; calculating gait acceleration according to the motion state data to obtain gait stability index; extracting phase coordination features according to the joint angle data and the motion state data to obtain a comprehensive coordination index; Step S3: Calculate the energy conversion efficiency based on the bilateral gait cycle data, plantar power parameters and joint power parameters to obtain efficiency score data; optimize the gait parameters based on the gait stability index and comprehensive coordination index of the efficiency score data to obtain individual target parameters; perform gait health assessment based on the individual target parameters, efficiency score data and comprehensive coordination index to obtain gait health data.

2. The method for entering health data according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: using a pre-calibrated pressure sensor array to perform multimodal data on the subject to obtain a multimodal signal set; Step S12: performing gait event recognition on the multimodal signal set to obtain gait event labels; Step S13: dividing the multimodal signal set into gait cycles according to the gait event markers to obtain cycle standardized data; Step S14: performing bilateral gait synchronization alignment according to the cycle standardization data to obtain bilateral gait cycle data.

3. The method for entering health data according to claim 1, characterized in that: The foot pressure dynamic processing in step S2 includes: Extract the plantar pressure center trajectory according to the plantar pressure data to obtain the pressure center trajectory data; Perform pressure center velocity analysis on pressure center trajectory data to obtain pressure center dynamic parameters; The regional pressure distribution time series analysis is performed based on the ground force data to obtain the regional pressure pattern; Perform pressure integral analysis based on regional pressure patterns to obtain cumulative pressure maps; The pressure peak time series analysis is performed based on the dynamic parameters of the pressure center, the regional pressure pattern and the cumulative pressure spectrum to obtain the pressure time series characteristics.

4. The method for entering health data according to claim 1, characterized in that: The joint torque energy processing in step S2 includes: Perform inverse dynamics calculation based on body segment inertia parameters, joint angle data and ground force data to obtain joint torque data; Perform joint power analysis based on joint torque data and joint angle data to obtain energy transfer data; Calculate joint stiffness based on joint torque data to obtain stiffness characteristic data; The torque feature is extracted according to the joint torque data and gait event markers to obtain the torque feature data.

5. The method for entering health data according to claim 1, characterized in that: The gait acceleration calculation in step S2 includes: Perform acceleration feature statistics according to the motion state data to obtain acceleration features; Spectral characteristics are analyzed based on motion state data and gait event markers to obtain acceleration spectrum characteristics; Perform shock transfer analysis based on acceleration characteristics and acceleration spectrum characteristics to obtain shock transfer characteristics; The stability index is calculated according to the impact transmission characteristics and acceleration characteristics to obtain the gait stability index.

6. The method for entering health data according to claim 1, characterized in that: The phase coordination feature extraction in step S2 includes: calculating joint coordination angles of joint angle data according to gait event markers; Perform segmented relative phase analysis based on joint coordination angle and motion state data to obtain phase difference data; Perform force-action coordination analysis based on joint angle data and ground force data to obtain force-action coordination data; Pressure-posture coupling analysis is performed based on pressure center trajectory data, regional pressure patterns, and motion state data to obtain pressure-posture coupling features; The joint coordination angle, phase difference data, force-action coordination data and pressure-posture coupling characteristics are integrated to obtain a comprehensive coordination index.

7. The method for entering health data according to claim 1, characterized in that: The energy conversion efficiency calculation in step S3 includes: The center of mass motion energy is calculated based on the body segment inertia parameters, bilateral gait cycle data and joint dynamic parameters to obtain the center of mass energy data; The potential energy-kinetic energy conversion efficiency is calculated for the center of mass energy data to obtain the energy recovery index; Perform joint power and energy transfer analysis on the joint torque data and joint angle data in the joint dynamic parameters to obtain the joint energy transfer index; Calculate the mechanical energy external power according to the center of mass energy data and the ground force data to obtain the external power data; According to the plantar dynamic parameters, the arch elastic energy storage is evaluated to obtain the arch energy data; A comprehensive energy efficiency evaluation is performed on energy recovery indices, joint energy transfer indices, external power data, and arch energy data to obtain efficiency score data.

8. The method for entering health data according to claim 1, characterized in that: The gait parameter optimization process in step S3 includes: Acquire multi-parameter gait data of subjects tested under different step frequency guidance; Energy-stability trade-off modeling was performed based on multi-parameter gait data, efficiency score data, gait stability index, and comprehensive coordination index to obtain a parameter trade-off model; Theoretical gait parameters are calculated according to the parameter trade-off model and the body segment inertia parameters to obtain a theoretical gait parameter set; Calculate the spatiotemporal parameters based on the bilateral gait cycle data and gait event markers to obtain a gait spatiotemporal parameter set; According to the gait spatiotemporal parameter set and the theoretical gait parameter set, actual parameter deviation analysis is performed to obtain a parameter deviation feature set; The individual factor data of the subjects are obtained; the individual factor correction is performed on the parameter deviation feature set according to the individual factor data to obtain the individual target parameters.

9. A health data entry system, characterized in that: For executing the health data entry method according to claim 1, the health data entry system comprises: The gait parameter acquisition module is used to collect the multimodal data of the subject, and perform gait event recognition to obtain gait event markers; perform bilateral gait synchronization alignment according to the gait event markers to obtain bilateral gait cycle data, wherein the bilateral gait cycle data includes plantar pressure data, ground force data, joint angle data and motion state data; The gait dynamics analysis module is used to perform plantar pressure dynamics processing based on plantar pressure data and ground force data to obtain plantar dynamics parameters; estimate the subject's body segment inertia parameters; perform joint torque energy processing based on body segment inertia parameters and joint angle data to obtain joint dynamics parameters; calculate gait acceleration based on motion state data to obtain gait stability indicators; extract phase coordination features based on joint angle data and motion state data to obtain comprehensive coordination indicators; The gait health assessment module is used to calculate the energy conversion efficiency based on the bilateral gait cycle data, plantar power parameters and joint power parameters to obtain efficiency score data; optimize the gait parameters based on the gait stability index and comprehensive coordination index of the efficiency score data to obtain individual target parameters; and perform gait health assessment based on the individual target parameters, efficiency score data and comprehensive coordination index to obtain gait health data.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed, the method for entering health data as described in any one of claims 1 to 8 is implemented.

Images