Lower limb and gravity center dynamic balance posture early warning method based on body measurement all-in-one machine

Abstract

The invention provides a lower limb and gravity center dynamic balance posture early warning method based on a body measurement all-in-one machine, and the method comprises the steps: obtaining a pelvis forward inclination angle, an arch height and a gravity center position under a standard standing posture, taking the pelvis forward inclination angle, the arch height and the gravity center position as individualized baseline data, and carrying out the statistical analysis of the baseline data, and generating a dynamic balance reference value reflecting the normal balance capability of an individual; on the basis of the dynamic balance reference value, in combination with a gravity center forward trend and a lower limb muscle compensation mode, a balance offset index in standing posture change is calculated through a time sequence analysis algorithm, and a dynamic balance risk assessment result is obtained; and if a dynamic balance risk assessment result exceeds a danger threshold, clustering analysis is carried out on associated parameters of the pelvic forward inclination angle and the arch height, grouping processing is carried out on the associated parameters, and potential reasons causing balance capability degradation are identified.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular, to a method for warning of the dynamic balance posture of the lower limbs and the center of gravity based on a body measurement integrated machine. Background Art

[0002] The research on dynamic balance ability is of crucial significance in the fields of human movement science and health management. It is directly related to the safety of individual daily activities, the optimization of sports performance, and the prevention and control of the fall risk of the elderly population. With the increase of population aging and chronic diseases, how to accurately evaluate and warn the degradation of balance ability through technical means has become an important topic for improving the quality of life and reducing the medical burden. Existing methods mostly rely on traditional physical measurement equipment or subjective observation. Although they can reflect the balance state to a certain extent, they generally have problems such as insufficient accuracy, single data, and lack of personalized analysis. These limitations make it difficult for existing solutions to capture the subtle dynamic characteristics in posture changes and also unable to provide targeted early intervention suggestions for users. In this context, the core challenges faced in the research field have gradually emerged, among which the interaction mechanism of the pelvic tilt angle, the arch height of the foot, and the static center of gravity position is particularly crucial. When analyzing these factors, current technologies often have difficulty accurately quantifying the dynamic effects of pelvic tilt changes on lumbar curvature and center of gravity forward movement, let alone revealing the complex adjustment process of the compensatory muscle activation pattern of the lower limbs. In addition, how to accurately correlate the parameters of pelvic tilt and arch height of the foot through a three-dimensional bone reconstruction algorithm and set a risk threshold based on individualized baseline data is still an unsolved technical problem. The existence of these challenges leads to the lack of real-time monitoring and scientific warning basis for the posture transition process in dynamic balance assessment, restricting the effectiveness of the technology in practical applications. Therefore, how to accurately identify the correlation parameters of the pelvic tilt angle and the arch height of the foot by using a body measurement integrated machine when the subject switches from a standard standing posture to a habitual standing posture, and analyze its effects on the center of gravity position and the lower limb muscle activation pattern through a three-dimensional bone reconstruction algorithm, and then establish a dynamic balance risk warning threshold based on individualized data has become the key problem that this research urgently needs to overcome. Summary of the Invention

[0003] The present invention provides a method for warning of the dynamic balance posture of the lower limbs and the center of gravity based on a body measurement integrated machine, mainly including:

[0004] Obtain the three-dimensional bone data of the subject in the standard standing posture and the habitual standing posture, collect the pelvic tilt angle and arch height parameters through the built-in sensors of the body measurement integrated machine, and generate an initial bone model by using a three-dimensional bone reconstruction algorithm to obtain the correlation parameters of the pelvis and the arch of the foot;

[0005] Extract the data of the change in the pelvic tilt angle from the initial bone model, combine it with the arch height parameter to determine the center of gravity position, calculate the center of gravity offset caused by the change in the pelvic position, and determine the forward trend of the center of gravity according to the change trend of the offset over time;

[0006] Predict the change in the standing posture according to the forward trend of the center of gravity. When the standing posture changes, obtain and process the electromyogram signals of the key muscle groups in the lower limbs to obtain the lower limb muscle activation pattern. Combine the relative magnitudes and coordination degrees of the activation levels of the muscles in the activation pattern to determine the lower limb muscle compensation pattern, and obtain the amplitude and distribution characteristics of the compensatory muscle activation;

[0007] Obtain the pelvic tilt angle, arch height, and center of gravity position in the standard standing posture as individualized baseline data, and generate a dynamic balance reference value reflecting the individual's normal balance ability through statistical analysis of the baseline data;

[0008] Based on the dynamic balance reference value, combine the forward trend of the center of gravity and the lower limb muscle compensation pattern, and calculate the balance offset index in the change of the standing posture through a time series analysis algorithm to obtain the dynamic balance risk assessment result;

[0009] If the dynamic balance risk assessment result exceeds the danger threshold, cluster analyze the correlation parameters of the pelvic tilt angle and the arch height, group process the correlation parameters, and identify the potential causes leading to the degradation of the balance ability;

[0010] Based on the dynamic balance risk assessment result, generate a corresponding warning signal, perform an association mapping of the warning signal with the real-time change trends of the pelvic tilt angle and the arch height, combine the potential causes of degradation, generate a balance degradation warning message, and provide real-time feedback through the physical fitness testing integrated machine system.

[0011] Furthermore, obtain the three-dimensional bone data of the subject in the standard standing posture and the habitual standing posture, collect the pelvic tilt angle and arch height parameters through the built-in sensors of the physical fitness testing integrated machine, and use a three-dimensional bone reconstruction algorithm to generate an initial bone model to obtain the correlation parameters of the pelvis and the arch, including:

[0012] Collect the three-dimensional coordinate data of the bone feature points, and perform noise reduction processing on the three-dimensional coordinate data through a median filter to obtain the filtered feature point position data; calculate the center of pressure trajectory curve and the arch area pressure value according to the filtered feature point position data and the plantar pressure distribution data collected by the pressure sensor array; collect the angular velocity data through the triaxial gyroscope sensor array arranged at the pelvis part, and obtain the pelvic spatial attitude angle according to the integral operation of the angular velocity data; establish a plantar surface equation based on the arch area pressure value, calculate the longitudinal arch height and transverse arch height parameters from the plantar surface equation, establish a geometric constraint equation set according to the pelvic spatial attitude angle and the longitudinal arch height and transverse arch height parameters, and solve the constraint equation by the least square method to obtain the position relationship parameters between the pelvis and the arch.

[0013] Further, extract the pelvic anteversion angle change data from the initial bone model, combine it with the arch height parameter to determine the center of gravity position, calculate the center of gravity offset caused by the pelvic position change, and determine the center of gravity forward movement trend according to the change trend of the offset over time, including:

[0014] Receive the pelvic angle data sent by the bone surface equation, and the pelvic angle data is processed by a Kalman filter to obtain a pelvic anteversion angle sequence; obtain the pressure distribution data collected by the plantar pressure sensor array according to the pelvic anteversion angle sequence, and calculate the center of pressure point coordinate sequence and the arch height value through the pressure distribution data; establish a human body center of mass distribution equation using the pelvic anteversion angle sequence and the arch height value, and calculate the initial coordinates of the whole body center of gravity through linear superposition; for the initial coordinates of the whole body center of gravity, use Euler angles to describe the three-dimensional attitude change of the pelvis, and calculate the center of gravity displacement vector sequence through the rigid body kinematics equation, and the center of gravity displacement vector sequence is obtained by least square fitting to obtain the horizontal plane displacement field distribution.

[0015] Further, predict the stance change according to the center of gravity forward movement trend. When the stance changes, obtain and process the electromyography signals of the key lower limb muscle groups to obtain the lower limb muscle activation pattern, and combine the relative magnitudes and coordination degrees of the activation degrees of the muscles in the activation pattern to determine the lower limb muscle compensation pattern, and obtain the amplitude and distribution characteristics of the compensatory muscle activation, including:

[0016] Construct a predicted spatial trajectory diagram based on the center-of-gravity displacement sequence data, and obtain the deviation value between the center-of-gravity projection position and the edge threshold of the support base area from the predicted spatial trajectory diagram; collect the potential signals of the muscle groups around the lower limb joints through a surface electromyography sensor, and perform filtering processing on the potential signals by using a center frequency band-pass filter to obtain a filtered electromyogram waveform diagram; perform wavelet transform on the filtered electromyogram waveform diagram to obtain a wavelet coefficient matrix, calculate the signal cross-correlation coefficient matrix between muscle groups from the wavelet coefficient matrix, and obtain the muscle group coordination eigenvector through eigenvalue decomposition of the cross-correlation coefficient matrix; calculate the Euclidean distance value according to the muscle group coordination eigenvector and the standard eigenvector constructed from the standard anthropometric data, establish a compensation metric function from the Euclidean distance value, and obtain a heat distribution diagram of muscle compensation degree.

[0017] Further, obtain the pelvic tilt angle, the arch height and the center-of-gravity position in the standard standing posture as individualized baseline data, and generate a dynamic balance reference value reflecting the individual's normal balance ability through statistical analysis of the baseline data, including:

[0018] Obtain the pelvic marker point data collected by an optical tracker, and obtain a pelvic tilt angle sequence through three-dimensional space coordinate transformation according to the marker point data; receive the pressure distribution diagram collected by a plantar pressure sensor array, and calculate the arch support index according to the pressure distribution diagram, where the arch support index is obtained by calculating the longitudinal arch height value and the transverse arch height value of the arch area pressure curve; establish a centroid position calculation equation according to the pelvic tilt angle sequence and the arch support index, and obtain the three-dimensional coordinates of the center of gravity from the calculation equation, where the calculation equation includes the segment mass ratio and the space coordinate transformation matrix; perform kernel density estimation on the three-dimensional coordinates of the center of gravity to obtain a center-of-gravity displacement envelope line, extract the displacement direction and amplitude feature sequence from the center-of-gravity displacement envelope line, and obtain a balance ability reference value by using the Bayesian estimation method according to the feature sequence.

[0019] Further, on the basis of the dynamic balance reference value, combine the center-of-gravity forward movement trend and the lower limb muscle compensation mode, and calculate the balance offset index in the standing posture change through a time series analysis algorithm to obtain a dynamic balance risk assessment result, including:

[0020] A time-frequency spectrum is obtained by performing wavelet transform on the center of gravity displacement sequence, and the dominant frequency and amplitude parameters of the center of gravity movement are extracted from the time-frequency spectrum to construct a balance state description vector; a recursive neural network is used to process the balance state description vector, and the center of gravity displacement rate curve is calculated through a sliding time window to obtain the balance offset; envelope extraction is performed on the lower limb electromyographic signal, and the compensatory activation intensity is calculated according to the envelope amplitude, and the temporal correlation between muscle groups is solved by the cross-correlation function to obtain the compensatory synergy feature; a feature fusion function is established based on the balance state description vector and the compensatory synergy feature, the fused feature is normalized, and the trend item of the risk feature sequence is extracted through time series analysis. If the trend slope exceeds the preset threshold, the balance risk level is determined.

[0021] Furthermore, if the dynamic balance risk assessment result exceeds the danger threshold, cluster analysis is performed on the parameters associated with the pelvic tilt angle and the arch height, and the associated parameters are grouped to identify potential causes of balance ability degradation, including:

[0022] A risk status curve is constructed according to the dynamic balance risk assessment data, and the piecewise linear mapping value is calculated through the risk status curve to obtain the danger threshold interval parameters; the density clustering algorithm is used to process the pelvic tilt angle data, the fluctuation amplitude and the change period are calculated from the angle data sequence, and the abnormal pelvic tilt parameters are obtained through the Euclidean distance measurement; the pressure distribution surface is calculated according to the plantar pressure sensor array data, and the arch boundary curve is extracted from the pressure distribution surface by the regional growing algorithm to obtain the abnormal arch morphology parameters; a parameter correlation matrix is ​​constructed for the abnormal pelvic tilt parameters and the abnormal arch morphology parameters, and the hierarchical clustering method is used to group and cluster the correlation matrix, and the dominant factors of balance ability degradation are obtained through intra-group variance calculation.

[0023] Furthermore, based on the dynamic balance risk assessment result, a corresponding warning signal is generated, the warning signal is correlated and mapped with the real-time change trend of the pelvic tilt angle and the arch height, and the balance degradation warning information is generated in combination with the potential cause of degradation, and real-time feedback is provided through the physical test integrated machine system, including:

[0024] A dynamic threshold range is obtained according to the distribution of historical data, and a fluctuation feature sequence is extracted from the dynamic threshold range to obtain a first warning parameter; a predictive analysis is performed on the first warning parameter through a recursive neural network, and if the rate of decrease of the predictive analysis score exceeds a preset threshold, a first warning mark of a corresponding level is generated; a support area change rate curve is obtained according to the plantar pressure sensor array, and a trend feature sequence is extracted from the change rate curve to obtain a second warning parameter; a mapping relationship table is established between the first warning mark and the second warning parameter, and the warning prompt content is matched through the mapping relationship table to generate a warning data packet to the display terminal.

[0025] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0026] The present invention discloses a method for warning of lower limb and center-of-gravity dynamic balance postures based on a physical fitness testing integrated machine. This method generates an initial bone model and extracts key parameters by collecting three-dimensional bone data, pelvic anterior tilt angle, and arch height parameters of the subject in different standing postures. Combining the change in the center-of-gravity position and the lower limb muscle activation pattern, the present invention determines the forward trend of the center of gravity and the lower limb muscle compensation pattern. By establishing individualized baseline data and dynamic balance reference values, the present invention calculates the balance deviation index to achieve dynamic balance risk assessment. When the risk exceeds the threshold, the present invention identifies the potential causes leading to the degradation of balance ability and generates corresponding warning information. This method can monitor and evaluate an individual's balance ability in real time, providing a scientific basis for preventing fall risks and improving standing posture balance. Description of the Drawings

[0027] Figure 1 It is a flowchart of a method for warning of lower limb and center-of-gravity dynamic balance postures based on a physical fitness testing integrated machine of the present invention.

[0028] Figure 2 It is a schematic diagram of a method for warning of lower limb and center-of-gravity dynamic balance postures based on a physical fitness testing integrated machine of the present invention. Detailed Embodiments

[0029] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of them. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0030] Such as Figure 1 -2, a method for warning of lower limb and center-of-gravity dynamic balance postures based on a physical fitness testing integrated machine in this embodiment may specifically include:

[0031] S101. Collect three-dimensional bone data of the subject in standard standing postures and habitual standing postures through built-in sensors, generate an initial bone model using a three-dimensional bone reconstruction algorithm, extract the correlation parameters of the pelvis and the arch, and at the same time process the collected data to ensure accuracy.

[0032] In the embodiments of the present invention, the physical fitness testing integrated machine is equipped with a variety of sensors for obtaining bone and pressure data of the subject in different standing postures. Specifically, when implemented, first collect the three-dimensional coordinate data of the bone feature points on the surface of the subject's body through an optical tracker, and introduce noise reduction means in data processing to improve the measurement accuracy.

[0033] S1011. Obtain the three-dimensional coordinate data of the skeletal feature points of the subject in the standard standing posture and the habitual standing posture through an optical tracker. Use a high-speed camera to synchronously record the movement trajectories of the marker points, and use a median filter to denoise the coordinate data to obtain smooth feature point position data. Then, register the data through a rigid body transformation method to generate an initial skeletal model.

[0034] In actual operation, reflective marker points with a diameter of 10 mm are arranged on the body surface of the subject. The marker point positions are selected as anatomical landmark points such as the anterior superior iliac spine point of the pelvis and the greater trochanter point of the femur. The optical tracker consists of 4 high-speed cameras, and the sampling frequency is set to 100 Hz. The window size of the median filter is 5 frames, which is used to eliminate the jitter caused by the deformation of the skin and soft tissues. The registration process uses a rigid body transformation method to ensure the alignment of the feature point data with the anatomical structure template, and then generates a personalized skeletal surface equation through a thin plate spline deformation function, providing a basis for subsequent analysis.

[0035] S1012. Use the pressure sensor array built into the body measurement integrated machine to collect the plantar pressure distribution data, calculate the center of pressure trajectory curve and the pressure value in the arch area, and calculate the pelvic spatial attitude angle and the anterior tilt angle value according to the angular velocity data collected by the triaxial gyroscope sensor array. Combine the arch height parameter to establish a geometric constraint equation set, and solve it by the least square method to obtain the spatial position relationship parameters between the pelvis and the arch.

[0036] Specifically, the pressure sensor array consists of 16×16 sensing units, each unit has a size of 10×10 mm, the measurement range is 0 to 600 kPa, and the sampling frequency is 50 Hz. When the displacement of the center of pressure trajectory within 20 consecutive frames is less than 5 mm, the standing posture is determined to be stable and the current pressure data is recorded. The sole is divided into three areas: the forefoot, the midfoot, and the heel. The arch boundary is identified by the difference between adjacent sensing units and the pressure value is extracted. The triaxial gyroscope sensor is arranged at the pelvic position, the measurement range is plus or minus 1000 degrees per second, the sampling frequency is 200 Hz, the Euler angle is obtained through integral operation and converted to the gravity coordinate system to calculate the pelvic anterior tilt angle. The plantar pressure data is processed by cubic spline interpolation, the node spacing is 5 mm, a plantar surface equation is generated, and the longitudinal arch height and transverse arch height parameters are extracted from it. A geometric constraint equation set is established with the pelvic center of gravity as the coordinate origin, and the solution results show that when the pelvic anterior tilt angle changes by 1 degree, the longitudinal arch height changes by 0.8 mm and the transverse arch height changes by 0.5 mm, reflecting the biomechanical relationship between the two.

[0037] In the embodiments of the present invention, the pelvic anterior tilt angle and arch height parameters obtained through the above steps can provide accurate input data for subsequent analysis of center of gravity shift and muscle activation patterns. The collaborative work of the optical tracker and the pressure sensor ensures the stability and reliability of data collection, and the application of the three-dimensional bone reconstruction algorithm enables the correlation parameters of the pelvis and the arch to be presented in a quantitative form, laying a foundation for dynamic balance assessment. It can be understood that the specific configuration of the sensor and the algorithm parameters can be adjusted by technicians according to actual needs and will not be overly limited.

[0038] In the embodiments of the present invention, a method for warning the dynamic balance posture of the lower limbs and the center of gravity based on a body measurement integrated machine is provided, and the accurate analysis and assessment of the dynamic balance ability are realized through the following specific steps.

[0039] S102. Extract the characteristics of the change in the pelvic anterior tilt angle from the initial bone model and determine the center of gravity position in combination with the arch height data. By analyzing the influence of the pelvic posture change on the center of gravity shift, calculate the time series trend of the offset amount and judge the forward direction of the center of gravity.

[0040] In the embodiments of the present invention, based on the initial bone model constructed in the early stage, the dynamic correlation between the pelvis and the arch is analyzed by using the multi-dimensional data collected by the sensor, and then the change law of the center of gravity position is deduced. When implementing, first extract the key parameters from the bone model and perform smoothing processing to ensure the accuracy and stability of the data.

[0041] S1021. Obtain the pelvic anterior tilt angle sequence from the initial bone model and perform smoothing processing through a Kalman filter. Combine the pressure distribution data collected by the plantar pressure sensor array to calculate the coordinates of the pressure center point and the arch height value. Use these parameters to establish a human body mass center distribution model, and determine the initial coordinates of the whole body center of gravity through the linear superposition method.

[0042] Specifically, the pelvic tilt angle sequence is extracted from the skeletal surface equation, which constructs a three-dimensional space model based on anatomical landmark points such as the anterior superior iliac spine point and the posterior superior iliac spine point of the pelvis. The Kalman filter is used to optimize the angle data. The measurement noise covariance is set to 0.5 degrees, and the process noise covariance is set to 0.1 degrees, reducing the fluctuation range of the filtered data from the original plus or minus 3 degrees to plus or minus 0.8 degrees, ensuring the smoothness of the angle sequence. The plantar pressure sensor array adopts a 16×16 grid layout, with each unit area of 1 square centimeter, and the measurement range covers 0 to 600 kPa. The coordinates of the pressure center point are calculated by the weighted average method, where the weights are determined by the pressure values of each sensing unit. The arch height value is obtained by cubic spline interpolation of the pressure data in the longitudinal arch area, with an interpolation node spacing of 5 mm, and the generated smooth curve can accurately reflect the morphological characteristics of the arch. The human body center of mass distribution model comprehensively considers the influence of the pelvic tilt angle on the trunk position and the role of the arch height in supporting the lower limbs. The mass distribution of each segment refers to anthropometric standards. For example, the trunk accounts for 43%, and the two lower limbs account for 32%. The coordinates of the center of mass of each segment are calculated by linear superposition, and a coordinate system is established with the pelvic center as the origin. Finally, the weighted average method is used to solve the initial coordinates of the whole body center of gravity, laying a foundation for subsequent offset analysis.

[0043] S1022. For the initial coordinates of the whole body center of gravity, use Euler angles to represent the attitude change of the pelvis in three-dimensional space. Calculate the sequence of the center of gravity displacement vectors caused by the change of the pelvis position through the rigid body kinematic equation. Perform least squares fitting on this sequence to obtain the displacement field distribution on the horizontal plane, and extract the dominant direction of the displacement field through Fourier transform to judge the trend of the center of gravity moving forward.

[0044] In this step, the Euler angles are obtained by integrating the data of the three-axis gyroscope and are used to describe the rotation angles of the pelvis in the sagittal plane, coronal plane, and transverse plane. The rigid body kinematic equation takes the pelvis as the core of the rigid body and calculates the dynamic influence of its attitude change on the center of gravity position. For example, when the pelvic tilt angle increases by 5 degrees, the center of gravity of the trunk moves forward by about 20 mm. The sequence of the center of gravity displacement vectors records the spatial movement trajectory of the center of gravity during the sampling time, and the sampling frequency is set to 100 Hz. By least squares fitting of the vector sequence, a displacement field curve on the horizontal plane is generated, and the fitting error is controlled within 2 mm to ensure accuracy. Fourier transform further analyzes the frequency characteristics of the displacement field and extracts the moving direction corresponding to the component with the largest amplitude. If the angle between this direction and the sagittal plane of the human body is less than 30 degrees, it is determined as the trend of the center of gravity moving forward. In addition, combined with the influence of the change of the arch height on the support base area, calculate the stability boundary range. For example, for every 100 square centimeter reduction in the base area, the allowable range of the center of gravity moving forward is reduced by 15 mm, thus providing a quantitative basis for the dynamic balance risk assessment.

[0045] In the embodiments of the present invention, through the smoothing process of the pelvic anterior tilt angle sequence and the accurate calculation of plantar pressure data, the characteristic of the center of gravity shift in the posture change can be effectively captured. Compared with the traditional method that only relies on a single parameter, this method significantly improves the accuracy of the center of gravity position derivation through the fusion analysis of multi-dimensional data. For example, the Kalman filter not only reduces the influence of measurement noise but also provides stable input data for subsequent trend judgment. At the same time, the application of the rigid body kinematic equation makes the calculation of the center of gravity displacement more physically based, reflecting the biomechanical relationship between pelvic posture adjustment and lower limb support.

[0046] To further enhance the comprehensiveness of the analysis, this step also considers the indirect effect of the arch height on the support stability. The support base area calculated through the human body center of mass distribution model can dynamically adjust the allowable range of the center of gravity shift. For example, when the increase in the arch height leads to a decrease in the base area, the system will update the stability boundary in real time to ensure that the judgment of the center of gravity forward movement trend is more in line with individual characteristics. It can be understood that the specific filtering parameters and fitting algorithms can be adjusted according to the actual application scenario. For example, in a high-dynamic environment, the frequency resolution of the Fourier transform can be increased to capture more subtle displacement changes.

[0047] In the embodiments of the present invention, the determination of the center of gravity forward movement trend provides a key basis for the subsequent prediction of the standing posture change. Through the analysis of the displacement field distribution, the user can intuitively understand the dynamic evolution of the body balance state. For example, when standing statically, the natural swing range of the center of gravity position is usually plus or minus 10 millimeters. When the offset exceeds this range, the system can identify it as a significant posture adjustment. This analysis method based on multi-parameter fusion not only improves the scientific nature of the evaluation but also provides reliable data support for real-time warning. The specific fitting algorithms and threshold settings can be further optimized by technicians according to the characteristics of the subject group, and no excessive limitations are imposed.

[0048] In the embodiments of the present invention, a method for warning the dynamic balance posture of the lower limbs and the center of gravity based on a body measurement integrated machine is provided, and the accurate evaluation of the lower limb muscle activation and the center of gravity change is realized through the following steps.

[0049] S103. Predict the standing posture change of the subject according to the center of gravity forward movement trend, and when the standing posture is adjusted, extract the muscle activation pattern by collecting and processing the electromyography signals of the key muscle groups of the lower limbs, and further analyze the characteristics of the activation degree and coordination relationship of each muscle, so as to determine the compensation mode of the lower limb muscles and obtain its activation amplitude and distribution characteristics.

[0050] In the embodiments of the present invention, using the previously calculated center of gravity displacement data and combining the signals collected by the surface electromyography sensors, the dynamic analysis of the standing posture change and the muscle compensation behavior is carried out. When implemented, first, the standing posture change direction is predicted through the center of gravity trajectory, and then the electromyography signals are processed at multiple levels to extract key features.

[0051] S1031. Construct a predicted spatial trajectory graph based on the center-of-gravity displacement sequence to judge the changing trend of the standing posture. Collect the potential signal of the main muscle groups in the lower limbs through surface electromyography sensors and perform preprocessing using a band-pass filter to obtain the filtered electromyogram waveform. Then, extract time-frequency features through wavelet transform to analyze the muscle activation intensity.

[0052] Specifically, the center-of-gravity displacement sequence is generated from plantar pressure data and pelvic posture parameters. The predicted spatial trajectory graph is bounded by the support base area, and the base area is approximately 300 to 400 square centimeters during normal standing. When the center-of-gravity projection point is less than 20 millimeters away from the base edge, the system determines it as a potential unstable state, and predicts the change in the standing posture through the tangent direction of the trajectory curve. For example, when the projection point moves more than 10 millimeters per second, it indicates a significant adjustment in the posture. The surface electromyography sensors are arranged for key muscle groups in the lower limbs such as the rectus femoris, biceps femoris, tibialis anterior, and gastrocnemius, and the sampling frequency is set to 1000 Hz. The original electromyography signal is often affected by 50-Hz power frequency interference and baseline drift. Therefore, a band-pass filter with a center frequency of 50 Hz and a passband range of 20 to 450 Hz is used for processing. The filtered signal retains the main frequency components of muscle contraction and effectively suppresses noise. Then, wavelet transform is applied to the filtered electromyogram waveform. The multi-Daubechies 4th-order wavelet basis function is selected for 6-layer decomposition, and wavelet coefficients are extracted to reflect the time-frequency distribution characteristics of the signal. The root mean square value is calculated from the wavelet coefficients with a window length of 200 milliseconds and an overlap rate of 50%, generating the time-domain activation curves of each muscle group. The curve amplitude is in microvolts and ranges from 50 to 200 microvolts during normal standing. This process can clearly show the dynamic intensity of muscle contraction and provide reliable data for subsequent collaborative analysis.

[0053] S1032. Extract features from the filtered electromyography signal. Obtain the collaborative eigenvectors by calculating the cross-correlation coefficient matrix between muscle groups and performing eigenvalue decomposition. Combine the standard anthropometric data to calculate the Euclidean distance value to quantify the compensation degree, and generate a heat map of muscle compensation degree to reflect the activation distribution characteristics.

[0054] In this step, the coordination between muscle groups is quantified by the cross-correlation coefficient matrix. For example, the cross-correlation coefficient between the rectus femoris and the tibialis anterior in the normal standing posture is about 0.7, indicating a strong temporal correlation between the two. The eigenvalue decomposition is performed on the coefficient matrix, and the first three eigenvectors are extracted. These vectors explain more than 80% of the covariance, forming the core features of the muscle coordination pattern. The standard anthropometric data is sourced from the normal standing EMG dataset of 100 healthy adults, including the coordination eigenvectors of each muscle group as a benchmark. The coordination eigenvectors of the current subject are compared with the standard vectors, and the difference is calculated through the Euclidean distance. The distance value is less than 0.3 in the normal standing posture, while it can reach more than 0.5 when the compensation is obvious. A compensation metric function is established based on the Euclidean distance to calculate the compensation activation intensity of each muscle group. For example, when the compensation intensity value of the tibialis anterior exceeds 2.0, it indicates that it undertakes additional balance adjustment tasks. The compensation degree heat map uses color coding, with red indicating high compensation areas and blue indicating close to the normal state, visually showing the spatial distribution characteristics. For example, when the tibialis anterior shows dark red, the adjacent gastrocnemius may only be light blue, and the compensation intensity is less than 0.5. This visualization method facilitates users to understand the regional differences in muscle compensation.

[0055] In the embodiment of the present invention, the significance of judging the standing posture change by predicting the spatial trajectory map is to provide a trigger condition for muscle activation analysis. For example, when the center of gravity projection exceeds the base edge, the system automatically starts the EMG signal acquisition to ensure that the data corresponds to the posture adjustment in real time. The combination of the band-pass filter and wavelet transform not only improves the signal quality but also extracts the deep features of muscle activation. Compared with the method that only relies on the original signal, this method can more accurately reflect the dynamic process of muscle contraction.

[0056] To further optimize the analysis effect, this step also considers the influence of the support base area on the standing posture stability. The outer contour of the sole is measured by an array of pressure sensors, and the change in the base area directly affects the threshold judgment of the center of gravity shift. For example, when the arch of the foot collapses, the base area increases and the stability is enhanced, but it may trigger compensatory adjustments of the lower limb muscle groups. Through the time-domain and frequency-domain analysis of the EMG signal, the subtle changes in these adjustments can be captured. For example, the short-term high-intensity activation of the biceps femoris at the initial stage of the posture change, and its root mean square value may increase from 100 microvolts to 300 microvolts.

[0057] In the embodiments of the present invention, the realization of the compensation mode analysis benefits from the integration of multi-level signal processing and feature extraction technologies. The calculation of the cross-correlation coefficient matrix reveals the collaborative working mechanism between muscle groups, and the generation of the heat map provides intuitive feedback for users. For example, when the tibialis anterior muscle is over-activated due to the forward shift of the center of gravity, the red area on the heat map indicating its compensation intensity prompts the user that there may be a problem of insufficient arch support. This data-based visual analysis not only improves the scientific nature of the evaluation but also provides a targeted basis for subsequent intervention measures. It can be understood that the filter parameters and feature extraction algorithms can be adjusted according to actual needs. For example, in a high-noise environment, the number of wavelet decomposition layers can be increased to improve the signal resolution.

[0058] In the embodiments of the present invention, a method for warning the dynamic balance posture of the lower limbs and the center of gravity based on a body measurement integrated machine is provided, and the individual balance ability is evaluated through the following steps.

[0059] S104. Collect the pelvic anterior tilt angle, arch height, and center of gravity position data of the subject in the standard standing posture as the individual baseline, and generate a dynamic balance reference value reflecting the normal balance ability through multi-dimensional statistical analysis.

[0060] In the embodiments of the present invention, based on the multi-source data collected by the optical tracker and the plantar pressure sensor array, combined with the space transformation and probability estimation methods, a personalized balance reference is constructed. Specifically, during implementation, key parameters are first accurately measured and processed.

[0061] S1041. Use the optical tracker to collect the pelvic marker point data and calculate the anterior tilt angle sequence. At the same time, obtain the pressure distribution map through the plantar pressure sensor array and extract the arch support index. Based on these parameters, establish a centroid position calculation model to determine the three-dimensional coordinates of the center of gravity.

[0062] Specifically, the optical tracker captures the movement trajectories of reflective markers with a diameter of 10 mm at the anterior superior iliac spine and posterior superior iliac spine of the subject through 4 high-speed cameras, and the sampling frequency is 100 Hz. The pelvic tilt angle is defined as the angle between the sagittal plane of the pelvis and the direction of gravity, and the normal value is usually in the range of 10 to 15 degrees. The system monitors the rate of angle change. When the change within 20 consecutive frames is less than 0.5 degrees per second, it is determined as a steady standing posture, and then a weighted average filter is used for smoothing processing. The filter window is set to 10 frames to ensure data stability. The plantar pressure sensor array consists of 16×16 units, each unit area is 1 square centimeter, and the measurement range is 0 to 600 kPa. The plantar contour is segmented by setting a pressure threshold of 50 kPa, and the support area is calculated to be approximately 280 square centimeters. Cubic spline interpolation is performed on the pressure data in the arch area, with a node spacing of 5 mm, to generate a smooth curve, from which the positions of the highest points of the longitudinal arch and the transverse arch are extracted, located at 40% of the foot length and the middle of the metatarsals respectively. The normal longitudinal arch height is 15 to 25 mm, and the transverse arch height is 8 to 12 mm. The arch support index is calculated by the ratio of the height value to the foot length, reflecting the foot support ability. The center of mass position calculation model is based on the segment parameter method. The human body is divided into the trunk and lower limbs, and the mass distributions are 43% and 32% respectively. The center of mass positions of each segment are adjusted by the pelvic tilt angle and the arch height. For example, an increase of 1 degree in the tilt angle causes the center of mass of the trunk to move forward by 2 mm, and a decrease of 1 mm in the arch height causes the center of gravity to drop by 0.5 mm. Finally, the three-dimensional coordinates of the center of gravity are generated through spatial coordinate transformation.

[0063] S1042. Perform kernel density estimation on the three-dimensional coordinates of the center of gravity to generate a displacement envelope and extract a displacement feature sequence. Fusion of multi-dimensional feature data is carried out through the Bayesian estimation method, and a dynamic balance reference value reflecting the individual's balance ability is calculated.

[0064] In this step, the kernel density estimation uses a Gaussian kernel function to analyze the distribution of the center of gravity position. The kernel bandwidth is set to 5 mm, and the area with a probability density greater than 0.1 is taken as the fluctuation range to form an elliptical envelope, and the major axis is usually consistent with the sagittal plane of the human body. Principal component analysis extracts the displacement features within the envelope. The first principal component reflects the displacement in the front-back direction, and the amplitude is usually less than 20 mm. The second principal component reflects the left-right direction, and the amplitude is less than 10 mm. The displacement frequency is between 0.3 and 1 Hz. These features constitute a dynamic balance state description function. Through the Bayesian estimation, the displacement amplitude and frequency parameters are fused to generate a comprehensive score, with a score range of 0 to 100. The higher the score, the stronger the balance ability. This method can effectively quantify the center of gravity stability of an individual in the standard standing posture and provide a reference for subsequent risk assessment.

[0065] In the embodiments of the present invention, through the collaborative work of an optical tracker and a pressure sensor, the accuracy of baseline data is ensured. Compared with traditional static measurements, this method more comprehensively reflects the normal performance of an individual's balance ability through dynamic feature extraction and probability analysis. For example, the introduction of the arch support index not only quantifies the support effect of the foot on the center of gravity but also reveals the potential impact of arch shape changes on overall stability.

[0066] To improve the practicality of the reference value, this step also combines steady-state data with dynamic fluctuation characteristics. The envelope generated by kernel density estimation can intuitively display the natural swing range of the center of gravity in space. For example, during normal standing, the fluctuation range in the anteroposterior direction reflects the subject's ability to adapt to gravity adjustment, while the smaller amplitude in the left-right direction indicates stronger lateral stability. Bayesian estimation improves the robustness of the scoring through probability fusion. For example, when the subject has a relatively low arch height, the system adjusts the scoring weight according to historical data, emphasizing the contribution of lower limb support to balance.

[0067] In the embodiments of the present invention, the generation process of the dynamic balance reference value fully considers the scientific nature of data collection and analysis. The high-frequency sampling of the optical tracker ensures the time resolution of the pelvic angle, while the spatial resolution of the pressure sensor provides reliable support for arch parameters. The establishment of the reference value not only provides a basis for individualized early warning but also provides a comparison standard for subsequent postural adjustment analysis. The filtering window or kernel bandwidth can be adjusted according to the characteristics of the subject group. For example, for the elderly population, the steady-state determination threshold can be appropriately relaxed to adapt to their postural fluctuation characteristics.

[0068] In the embodiments of the present invention, a method for early warning of the dynamic balance posture of the lower limbs and the center of gravity based on a body measurement integrated machine is provided, and the quantitative evaluation of dynamic balance risk is achieved through the following steps.

[0069] S105. Based on the dynamic balance reference value, combined with the trend of forward movement of the center of gravity and the lower limb muscle compensation mode, calculate the balance offset index in the change of standing posture through a time series analysis algorithm, thereby generating a dynamic balance risk assessment result.

[0070] In the embodiments of the present invention, using the previously extracted center of gravity displacement and electromyogram signal data, comprehensively evaluate the dynamic changes of the balance state through time-frequency analysis and feature fusion techniques. When implementing, first perform multi-scale decomposition on the center of gravity displacement sequence to extract key features.

[0071] S1051. Perform time-frequency analysis on the center of gravity displacement sequence through wavelet transform and construct a balance state description vector, process the displacement data using a recurrent neural network and calculate the balance offset, and at the same time perform envelope extraction on the lower limb electromyogram signals to obtain the compensation activation intensity and coordination features.

[0072] Specifically, the wavelet transform uses a multi-Bessel 4th-order wavelet basis function to decompose the center-of-gravity displacement sequence into 6 layers, and extracts the time-frequency spectrogram in the frequency band of 0.5 to 2 Hz, which corresponds to the main frequency range of human posture regulation. The time-frequency spectrogram shows that when standing normally, the dominant frequency of the center-of-gravity movement is about 0.8 Hz, and the amplitude is usually less than 10 mm. While during posture adjustment, the frequency may rise to 1.5 Hz, and the amplitude increases to more than 15 mm. The balance state description vector consists of the dominant frequency, displacement amplitude, and velocity parameter, and is processed by a recurrent neural network. This network adopts a long short-term memory structure. The input layer includes displacement coordinates and velocity components, and the hidden layer has 64 neurons, which can capture long-term dependencies in the time series. The sliding time window length is 1 second, and the overlap rate is 50%. The displacement rate curve is calculated. When the rate exceeds 20 mm per second, it indicates that the balance state has a significant deviation. The lower limb electromyography signals are collected for the rectus femoris, biceps femoris, tibialis anterior, and gastrocnemius muscles. The sampling frequency is 1000 Hz, and the waveform envelope is generated through rectification and smoothing processing. The envelope amplitude reflects the muscle contraction intensity, which is 50 to 200 μV when standing normally, and may rise to 400 μV during compensation. The cross-correlation function analyzes the temporal correlation degree between muscle groups. For example, the correlation coefficient between the rectus femoris and the tibialis anterior can rise from 0.6 to 0.9 in the compensatory state, indicating enhanced coordination. These features together constitute the compensatory coordination description, providing a basis for subsequent fusion analysis.

[0073] In the embodiment of the present invention, the balance risk is further quantified through feature fusion and time series analysis. Based on the balance state description vector and the compensatory coordination features, a weighted summation method is used to construct a feature fusion function, and the weights are determined by principal component analysis. For example, the weight of the displacement amplitude is 0.4, and the weight of the electromyography coordination is 0.3. The fused features are mapped to the range of 0 to 1 through normalization processing to generate a balance risk feature sequence. The time series analysis extracts the trend term and the periodic term. The trend term calculates the slope through polynomial fitting, and the periodic term identifies the fluctuation frequency through Fourier transform. When the trend slope exceeds 0.1 per second and lasts for more than 3 seconds, the system determines it as a high-risk state. The risk assessment result is presented in the form of a score. The score for the normal population is between 80 and 100, the mild risk is 60 to 80, the moderate risk is 40 to 60, and the severe risk is below 40.

[0074] To enhance the scientific nature of the assessment, this step also conducts multi-scale analysis on the compensatory synergy features. Wavelet decomposition divides the EMG signal into four frequency bands, corresponding to rapid reflex regulation, voluntary control regulation, posture maintenance, and slow drift respectively. In the compensatory state, the energy proportion of the rapid reflex frequency band increases from 15% in the normal state to 40%, reflecting the activity level of the muscle during emergency adjustment. The multi-scale features generate a compensatory pattern description vector through reconstruction. After being fused with the center of gravity data, it can more comprehensively depict the dynamic process of balance offset. For example, when the tibialis anterior muscle is over-activated due to the forward shift of the center of gravity, the increase in the energy of its rapid adjustment component indicates a potential risk of lower limb fatigue.

[0075] In the embodiment of the present invention, the application of the time series analysis algorithm significantly improves the real-time performance and accuracy of risk assessment. The recurrent neural network captures the minute changes in the displacement rate through a sliding window, ensuring sensitivity to the initial stage of posture adjustment. The wavelet transform provides a multi-dimensional perspective on the movement of the center of gravity. For example, through the time-frequency spectrogram, the correlation between the instant of frequency increase and the amplitude increase can be intuitively observed. This multi-level analysis method not only reveals the causal relationship between the center of gravity offset and muscle compensation, but also provides users with quantitative risk feedback. For example, when the score drops to 60, it indicates that attention should be paid to posture adjustment. The window length or network parameters can be adjusted according to the actual scenario. For example, in a dynamic environment, the window can be shortened to 0.5 seconds to improve the response speed.

[0076] In the embodiment of the present invention, a method for warning the dynamic balance posture of the lower limbs and the center of gravity based on a body measurement integrated machine is provided, and the in-depth analysis and warning of balance risks are realized through the following steps.

[0077] S106. If the result of the dynamic balance risk assessment exceeds the preset danger threshold, then by clustering and analyzing the correlation parameters of the pelvic tilt angle and the arch height, identify the potential reasons for the degradation of balance ability and generate corresponding warning signals.

[0078] In the embodiment of the present invention, when the risk assessment result shows a significant decline in balance ability, the system will automatically start the parameter analysis process and use multi-dimensional data mining technology to locate abnormal factors. When implementing, first quantify the risk state and determine the threshold range.

[0079] S1061. Construct a risk state curve based on the dynamic balance risk assessment data and calculate the dangerous threshold interval. At the same time, use the density clustering algorithm to analyze the fluctuation characteristics of the pelvic tilt angle sequence, extract abnormal parameters through Euclidean distance measurement, and generate abnormal arch shape characteristics using the plantar pressure distribution data.

[0080] Specifically, the results of the dynamic balance risk assessment are represented by scores from 0 to 100. A score higher than 80 is considered normal, 60 to 80 indicates mild risk, 40 to 60 indicates moderate risk, and below 40 indicates severe risk. The risk status curve is plotted using time series data to reflect the trend of the score over time. When the slope of the curve drops by more than 1 point per minute, it indicates an exacerbation of the decline in balance ability. Piecewise linear mapping divides the scores into risk intervals. For example, the interval below 40 points is defined as the danger threshold, triggering subsequent analysis. The anterior pelvic tilt angle is measured by an optical tracker, with marker points placed on the anterior superior iliac spine and the posterior superior iliac spine. The normal value is between 10 and 15 degrees. The density clustering algorithm takes the angle sequence as input, calculates the amplitude of fluctuations and the change period. The normal amplitude of fluctuations is less than 2 degrees, and the period is between 0.5 and 2 seconds. If the angle continuously exceeds 20 degrees or the fluctuation is greater than 5 degrees, the degree of deviation is quantified by the Euclidean distance to generate the anterior pelvic tilt abnormality parameter. The plantar pressure sensor array collects 16×16 grid data in the range of 0 to 600 kPa. The region growing algorithm is used to extract the arch boundary with a threshold of 50 kPa, extending from the forefoot to the calcaneal region. After the boundary curve is smoothed by cubic spline interpolation, the longitudinal arch and transverse arch heights are calculated. The normal longitudinal arch height is between 15 and 25 mm, and the transverse arch is between 8 and 12 mm. If the height deviates significantly, it is marked as an abnormal arch shape.

[0081] S1062. Construct a correlation matrix based on the anterior pelvic tilt abnormality parameter and the arch shape abnormality parameter, and perform a grouped analysis of the parameters through hierarchical clustering to extract the dominant abnormal factors and evaluate their contribution to the decline in balance ability.

[0082] In this step, the correlation matrix is generated by calculating the Pearson correlation coefficient between the anterior pelvic tilt abnormality and the arch shape abnormality. The coefficient ranges from -1 to 1. A positive value indicates a co-directional change, and a negative value indicates an inverse change. For example, the correlation coefficient between an increase in anterior pelvic tilt and a decrease in arch height may reach 0.8, indicating that the two jointly affect balance. Hierarchical clustering uses the complete linkage method to group the parameters with an absolute correlation coefficient greater than 0.7 into the same group, forming the combined feature of the parameters. Principal component analysis extracts the dominant component of the combined feature. The first principal component usually explains more than 65% of the variance, reflecting the core trend of the abnormal pattern. The within-group variance calculates the contribution of each parameter. For example, when the contribution of the anterior pelvic tilt abnormality reaches 60%, it indicates that it is the main influencing factor, while when the contribution of the arch abnormality is 30%, it is a secondary factor. This analysis method can clearly distinguish the effects of single abnormalities and superimposed effects.

[0083] In the embodiments of the present invention, the construction of the risk status curve provides a dynamic perspective for anomaly recognition. For example, when the score drops from 80 to 40, the sharp increase in the curve slope prompts the system to immediately analyze the potential causes. The density clustering algorithm can effectively separate normal fluctuations from abnormal fluctuations through an adaptive density threshold. For example, it can separately cluster the part exceeding 20 degrees in the angle sequence, revealing the abnormal persistence of the pelvic posture.

[0084] To further improve the analysis accuracy, this step also extracts the frequency-domain features of the arch morphology abnormality. By calculating the power spectral density of the pressure distribution surface through the fast Fourier transform, the energy of a normal arch is concentrated below 0.5 Hz, reflecting slow postural adjustments, while an increase in the high-frequency components of 2 to 5 Hz indicates unstable support. Wavelet transform further decomposes these high-frequency fluctuations and extracts the time-frequency distribution of the abnormal signals. For example, when the arch collapses, the proportion of high-frequency energy increases from 5% to 20%, providing a multi-dimensional basis for the quantification of abnormal parameters.

[0085] In the embodiments of the present invention, the combination of the parameter correlation matrix and hierarchical clustering not only reveals the biomechanical relationship between anterior pelvic tilt and arch height, but also provides intuitive causal feedback for users through contribution analysis. For example, when both are abnormal at the same time, the risk score may drop sharply below 40, indicating a significant superimposed effect. This data-based grouping analysis method can effectively locate the root cause of balance degradation and provide scientific guidance for subsequent intervention measures. The clustering threshold can be adjusted according to the characteristics of the subjects. For example, for people who are sensitive to arch abnormalities, the correlation coefficient grouping standard can be reduced to 0.6 to capture more potential associations.

[0086] In the embodiments of the present invention, a method for early warning of the dynamic balance posture of the lower limbs and the center of gravity based on a body measurement integrated machine is provided, and risk early warning and real-time feedback are realized through the following steps.

[0087] S107. Generate an early warning signal based on the dynamic balance risk assessment result, associate and map it with the real-time change trends of the anterior pelvic tilt angle and arch height, and generate detailed balance degradation early warning information in combination with the potential degradation causes, and present it to the user in real time through the system interface.

[0088] In the embodiments of the present invention, intuitive early warning feedback is generated through predictive analysis and feature mapping by using risk assessment data and multi-source sensor information. When implementing, first dynamically monitor the risk trend and extract early warning parameters.

[0089] S1071. Calculate the dynamic threshold range according to the historical data distribution and extract the fluctuation features to generate the first early warning parameter, predict the change trend of the risk score through a recurrent neural network and generate a hierarchical early warning identifier, and at the same time calculate the change rate of the support area from the data of the plantar pressure sensor array to extract the second early warning parameter.

[0090] Specifically, the dynamic balance risk assessment uses a 0 to 100 scoring system, and the historical data is calculated based on the normal distribution characteristics. The threshold range, for example, the mean pelvic tilt angle is 12 degrees, the standard deviation is 1.5 degrees, and the threshold range is set to 9 to 15 degrees. The fluctuation characteristics are extracted through a 0.2 second sliding window. When the angle change rate exceeds 5 degrees per second, it is marked as abnormal and the first warning parameter is generated. The recursive neural network takes the risk score sequence and change rate as input, the prediction window is set to 5 seconds, and the network contains 64 hidden layer neurons, which can capture short-term trends and long-term dependencies. When the predicted score decreases at a rate of more than 2 minutes per second, a yellow warning is generated, an orange warning at 5 minutes per second, and a red warning at 10 minutes per second, which are distinguished by flashing frequencies of 1, 2, and 3 times per second, respectively. The plantar pressure sensor array collects data in a 16×16 grid with a sampling frequency of 50 Hz. The support area is calculated by 50 kPa threshold segmentation, and the normal range is 250 to 300 square centimeters. The support area change rate is calculated based on 5 frames of data. If it exceeds 20 square centimeters per second, it is extracted as the second warning parameter to reflect the dynamic adjustment of the arch shape.

[0091] S1072. Construct a mapping relationship table through the first warning identifier and the second warning parameter, generate a warning data packet including abnormal parameters and prompt content, and transmit it to the display terminal through the serial communication protocol to achieve hierarchical visual feedback.

[0092] In this step, the mapping relationship table associates the warning mark with the pelvic posture and arch shape parameters. For example, the yellow warning corresponds to an angle deviation of 9 to 15 degrees or an area change rate that exceeds the standard, and the matching prompt content is such as "The pelvic tilt is slightly abnormal, please adjust your posture." The warning data packet encapsulates the timestamp, warning level, parameter value and recommended measures, and is transmitted using a serial protocol with a baud rate of 115200, 8 data bits, and 1 stop bit. After the display terminal parses the data, the interface is divided into three areas: the real-time warning status on the left is indicated by color and flashing, the pelvic angle and arch support area curve is drawn in the middle, and detailed prompts are displayed on the right. The pop-up window is displayed in the center, and the mild warning is a white translucent background, the moderate is light yellow, and the severe is light red to improve recognition.

[0093] In the embodiment of the present invention, the generation and real-time mapping of warning signals ensure that users can perceive the risk of balance degradation in a timely manner. For example, when the pelvic tilt angle exceeds 15 degrees and continues to increase, the system not only triggers an orange warning, but also displays its changing trend through a curve, intuitively indicating the source of the abnormality. The predictive ability of the recursive neural network makes the warning forward-looking, for example, a prompt is issued 5 seconds in advance before the score drops to 60, buying time for intervention.

[0094] To enhance the practicality of the feedback, this step also incorporates a heat map of plantar pressure distribution. When the support area of the arch of the foot is abnormally reduced, the heat map highlights the forefoot or heel area in red, indicating that the user may have a center of gravity shift caused by arch collapse. This visual design helps users understand the direct relationship between abnormal parameters and physical conditions. For example, an excessive area change rate may be related to fatigue from standing for a long time, and suggested measures such as "rest and elevate feet" provide specific guidance.

[0095] In the embodiments of the present invention, the real-time feedback of warning information is achieved through multi-level data processing and interface optimization. The adaptive adjustment of the dynamic threshold range ensures the individuation of parameter monitoring. For example, for users with a relatively flat arch of the foot, the system will widen the area change rate threshold to 25 square centimeters per second to reduce false alarms. The flexibility of the mapping relationship table allows the prompt content to be updated according to actual needs. For example, in a sports scenario, the suggestion of "check the insole support" is added. The partition layout and pop-up window design of the display interface not only improve the information transmission efficiency but also highlight key abnormalities through color coding and font size differences, enhancing the user experience. The prediction window or blink frequency can be adjusted according to the application scenario. For example, in a high-dynamic environment, it is shortened to 3 seconds to improve the response speed.

[0096] The above only lists some preferred embodiments of the present invention, but the present invention is not limited thereto, and many improvements and transformations can be made. As long as the improvements and transformations are made based on the basic principles of the present invention, they should be regarded as falling within the protection scope of the present invention.

Claims

1. A method for early warning of dynamic balance posture of lower limbs and center of gravity based on a physical test integrated machine, characterized in that: The method comprises: The three-dimensional bone data of the subjects in the standard standing posture and the habitual standing posture are obtained, the pelvic tilt angle and the arch height parameters are collected through the built-in sensor of the physical measurement machine, and the initial bone model is generated by the three-dimensional bone reconstruction algorithm to obtain the correlation parameters of the pelvis and the arch; Extract the pelvic tilt angle change data from the initial skeletal model, combine it with the arch height parameter to determine the center of gravity position, calculate the center of gravity offset caused by the pelvic position change, and determine the center of gravity forward shift trend based on the change trend of the offset over time; Predict the change of standing posture according to the trend of center of gravity moving forward. When the standing posture changes, obtain and process the electromyographic signals of key muscle groups of lower limbs to obtain the lower limb muscle activation pattern. Combined with the relative size and coordination of each muscle activation degree in the activation pattern, determine the lower limb muscle compensation pattern, and obtain the amplitude and distribution characteristics of compensatory muscle activation; The pelvic tilt angle, arch height and center of gravity position in the standard standing posture are obtained as individual baseline data. The baseline data are statistically analyzed to generate a dynamic balance benchmark value that reflects the individual's normal balance ability. Based on the dynamic balance benchmark value, combined with the center of gravity forward shift trend and lower limb muscle compensation pattern, the balance deviation index in the standing posture change is calculated through the time series analysis algorithm to obtain the dynamic balance risk assessment result. If the dynamic balance risk assessment result exceeds the danger threshold, cluster analysis is performed on the correlation parameters between the pelvic tilt angle and the arch height, and the correlation parameters are grouped to identify the potential causes of balance deterioration. Based on the results of dynamic balance risk assessment, corresponding warning signals are generated, and the warning signals are correlated and mapped with the real-time change trends of the pelvic tilt angle and arch height. Combined with the potential causes of degeneration, balance degradation warning information is generated and fed back in real time through the physical measurement integrated system.

2. The method according to claim 1, characterized in that The three-dimensional bone data of the subject in the standard standing posture and the habitual standing posture are obtained, the pelvic tilt angle and the arch height parameters are collected by the built-in sensor of the physical measurement integrated machine, and the initial bone model is generated by using the three-dimensional bone reconstruction algorithm to obtain the correlation parameters of the pelvis and the arch, including: Collecting three-dimensional coordinate data of bone feature points, and performing noise reduction processing on the three-dimensional coordinate data through a median filter to obtain filtered feature point position data; According to the filtered feature point position data and the plantar pressure distribution data collected by the pressure sensor array, a pressure center trajectory curve and a pressure value of the arch area are calculated; The angular velocity data is collected by a three-axis gyroscope sensor array arranged in the pelvis, and the spatial posture angle of the pelvis is obtained according to the integral operation of the angular velocity data; A plantar surface equation is established based on the pressure value of the arch area, and the longitudinal arch height and transverse arch height parameters are calculated from the plantar surface equation. A group of geometric constraint equations is established according to the pelvic spatial posture angle and the longitudinal arch height and transverse arch height parameters. The position relationship parameters between the pelvis and the arch are obtained by solving the constraint equations using the least squares method.

3. The method according to claim 1, characterized in that The method extracts the pelvic tilt angle change data from the initial skeletal model, combines the arch height parameter, determines the center of gravity position, calculates the center of gravity offset caused by the pelvic position change, and determines the center of gravity forward shift trend according to the change trend of the offset over time, including: Receiving pelvic angle data sent by the bone surface equation, wherein the pelvic angle data is processed by a Kalman filter to obtain a pelvic anteversion angle sequence; Acquire pressure distribution data collected by a plantar pressure sensor array according to the pelvic forward tilt angle sequence, and calculate a pressure center point coordinate sequence and a foot arch height value through the pressure distribution data; The pelvic tilt angle sequence and the arch height value are used to establish a human body center of mass distribution equation, and the initial coordinates of the whole body center of mass are obtained by linear superposition calculation; For the initial coordinates of the whole body center of gravity, Euler angles are used to describe the three-dimensional posture changes of the pelvis, and the center of gravity displacement vector sequence is calculated through the rigid body kinematic equation. The center of gravity displacement vector sequence is fitted by the least squares method to obtain the horizontal plane displacement field distribution.

4. The method according to claim 1, characterized in that The posture change is predicted according to the center of gravity forward shifting trend. When the posture changes, the electromyographic signals of the key muscle groups of the lower limbs are acquired and processed to obtain the lower limb muscle activation pattern. The lower limb muscle compensation pattern is determined by combining the relative size and coordination degree of each muscle activation degree in the activation pattern, and the amplitude and distribution characteristics of the compensatory muscle activation are obtained, including: Constructing a predicted space trajectory map according to the center of gravity displacement sequence data, and obtaining a deviation value between the center of gravity projection position and the support base area edge threshold from the predicted space trajectory map; The surface electromyography sensor is used to collect the potential signal of the muscle groups around the lower limb joints, and the center frequency bandpass filter is used to filter the potential signal to obtain a waveform of the filtered electromyography signal; Performing wavelet transform on the filtered electromyographic signal waveform to obtain a wavelet coefficient matrix, calculating the signal mutual correlation coefficient matrix between muscle groups from the wavelet coefficient matrix, and obtaining the muscle group coordination feature vector by eigenvalue decomposition of the mutual correlation coefficient matrix; The Euclidean distance value is calculated according to the muscle group synergy feature vector and the standard feature vector constructed by standard anthropometric data, and a compensation metric function is established from the Euclidean distance value to obtain a thermal distribution diagram of muscle compensation degree.

5. The method according to claim 1, characterized in that The pelvic tilt angle, arch height and center of gravity position in the standard standing posture are obtained as individualized baseline data, and a dynamic balance reference value reflecting the individual's normal balance ability is generated by statistically analyzing the baseline data, including: Acquire pelvic marker data collected by an optical tracker, and obtain a pelvic anterior tilt angle sequence through three-dimensional space coordinate transformation according to the marker data; Receiving a pressure distribution map collected by a plantar pressure sensor array, and calculating an arch support index according to the pressure distribution map, wherein the arch support index is obtained by calculating a longitudinal arch height value and a transverse arch height value of a pressure curve in an arch area; Establishing a center of mass position calculation equation according to the pelvic anteversion angle sequence and the arch support index, and obtaining the three-dimensional coordinates of the center of mass from the calculation equation, wherein the calculation equation includes a segment mass ratio and a spatial coordinate transformation matrix; The center of gravity displacement envelope is obtained by performing kernel density estimation on the three-dimensional coordinates of the center of gravity, a displacement direction and amplitude feature sequence is extracted from the center of gravity displacement envelope, and a balance capacity reference value is obtained according to the feature sequence using a Bayesian estimation method.

6. The method according to claim 1, characterized in that Based on the dynamic balance benchmark value, combined with the center of gravity forward shift trend and the lower limb muscle compensation pattern, the balance deviation index in the standing posture change is calculated through the time series analysis algorithm to obtain the dynamic balance risk assessment results, including: Performing wavelet transform on the center of gravity displacement sequence to obtain a time-frequency spectrum, extracting the dominant frequency and amplitude parameters of the center of gravity movement from the time-frequency spectrum, and constructing a description vector of the equilibrium state; The equilibrium state description vector is processed by using a recursive neural network, and a center of gravity displacement rate curve is calculated through a sliding time window to obtain a balance offset; The envelope of the lower limb electromyographic signal is extracted, the compensatory activation intensity is calculated according to the envelope amplitude, and the temporal correlation between muscle groups is solved by the cross-correlation function to obtain the compensatory synergy characteristics. A feature fusion function is established according to the equilibrium state description vector and the compensation synergy feature, the fusion feature is normalized, and the trend item of the risk feature sequence is extracted through time series analysis. If the trend slope exceeds a preset threshold, the balance risk level is determined.

7. The method according to claim 1, characterized in that If the dynamic balance risk assessment result exceeds the danger threshold, cluster analysis is performed on the parameters associated with the pelvic anteversion angle and the arch height, and the associated parameters are grouped to identify potential causes of balance ability degradation, including: Constructing a risk state curve according to the dynamic balance risk assessment data, calculating a piecewise linear mapping value through the risk state curve, and obtaining a danger threshold interval parameter; The density clustering algorithm was used to process the pelvic tilt angle data, the fluctuation amplitude and change period were calculated from the angle data sequence, and the abnormal pelvic tilt parameters were obtained through the Euclidean distance measurement. Calculate the pressure distribution surface according to the plantar pressure sensor array data, extract the arch boundary curve from the pressure distribution surface using a region growing algorithm, and obtain the arch morphology abnormality parameter; A parameter correlation matrix is ​​constructed for the abnormal pelvic anterior tilt parameters and the abnormal arch morphology parameters, and the correlation matrix is ​​grouped and clustered using a hierarchical clustering method. The dominant factor of balance ability degradation is obtained by calculating the intra-group variance.

8. The method according to claim 1, characterized in that Based on the dynamic balance risk assessment result, a corresponding warning signal is generated, and the warning signal is associated and mapped with the real-time change trend of the pelvic forward tilt angle and the arch height. In combination with the potential cause of degradation, balance degradation warning information is generated, and real-time feedback is provided through the physical test integrated machine system, including: Acquire a dynamic threshold range according to the historical data distribution, extract a fluctuation feature sequence from the dynamic threshold range, and obtain a first warning parameter; Performing a predictive analysis on the first warning parameter through a recursive neural network, and generating a first warning indicator of a corresponding level if a decreasing rate of the predictive analysis score exceeds a preset threshold; Acquire a support area change rate curve according to the plantar pressure sensor array, extract a trend feature sequence from the change rate curve, and obtain a second warning parameter; A mapping relationship table is established using the first warning identifier and the second warning parameter, and the warning prompt content is matched through the mapping relationship table to generate a warning data packet to the display terminal.

Images