Multi-modal data acquisition and processing system for dynamic pressure assessment

Abstract

The invention discloses a multi-modal data acquisition and processing system for dynamic pressure assessment, and the system comprises a multi-modal data acquisition module which synchronously acquires dynamic pressure signals of different modals through a plurality of high-frequency response pressure sensors; the data preprocessing module is used for carrying out real-time signal conditioning processing on the dynamic pressure signals in different modes to obtain preprocessed multi-mode dynamic pressure signals; the multi-modal data fusion module is used for carrying out space-time alignment and feature extraction on the preprocessed multi-modal dynamic pressure signals by adopting a multi-factor coupling model, and generating fusion pressure data through a fusion algorithm; and the dynamic pressure evaluation module is used for executing time-frequency domain analysis and mode recognition on the fusion pressure data through an analysis model, extracting key characteristic parameters, calculating a dynamic pressure index through a pressure evaluation algorithm, and generating an evaluation result containing the dynamic pressure index. According to the invention, high-precision and real-time dynamic pressure assessment can be realized, and the comprehensiveness and reliability of data acquisition and analysis are improved.

Description

Technical Field

[0001] This invention relates to the field of dynamic stress assessment technology, and in particular to a multimodal data acquisition and processing system for dynamic stress assessment. Background Technology

[0002] Existing technologies in the field of dynamic pressure assessment have the following shortcomings: Firstly, traditional pressure sensors have a low response speed to high-frequency dynamic pressure changes, making it difficult to capture subtle and rapidly changing pressure signals, resulting in insufficient accuracy and reliability of the acquired data. Secondly, existing systems mostly rely on single-modal data acquisition and processing, making it difficult to comprehensively reflect the pressure characteristics and multi-factor coupling effects under dynamic environments, and their data fusion and comprehensive analysis capabilities are weak. These problems limit the application effectiveness of existing technologies in high-precision, real-time dynamic pressure assessment scenarios.

[0003] Therefore, there is an urgent need to provide a technical solution to address the above problems. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a multimodal data acquisition and processing system for dynamic stress assessment.

[0005] In a first aspect, the present invention provides a multimodal data acquisition and processing system for dynamic stress assessment, the technical solution of which is as follows: A multimodal data acquisition module is used to simultaneously acquire dynamic pressure signals of different modes through multiple high-frequency response pressure sensors; wherein, the high-frequency response pressure sensors have a target sampling frequency higher than a preset threshold. The data preprocessing module is used to perform real-time signal conditioning on the dynamic pressure signals of different modes to obtain preprocessed multimodal dynamic pressure signals. The multimodal data fusion module is used to perform spatiotemporal alignment and feature extraction on the preprocessed multimodal dynamic pressure signal using a multi-factor coupling model, and to generate fused pressure data through a fusion algorithm. The dynamic stress assessment module is used to perform time-frequency domain analysis and pattern recognition on the fused stress data through an analysis model, extract key feature parameters, and calculate a dynamic stress index based on the key feature parameters using a stress assessment algorithm, thereby generating an assessment result containing the dynamic stress index.

[0006] Furthermore, the multimodal data acquisition module is specifically used for: The multiple high-frequency response pressure sensors are arranged at multiple key monitoring points of the object under test, and a uniform target sampling frequency higher than the preset threshold is configured for the high-frequency response pressure sensors arranged at each key monitoring point. A synchronous acquisition command is sent to multiple high-frequency response pressure sensors configured with the target sampling frequency, controlling each high-frequency response pressure sensor to synchronously acquire pressure distribution signals and pressure change signals according to the synchronous acquisition command; The pressure distribution signal and pressure change signal synchronously collected by each high-frequency response pressure sensor are combined into the dynamic pressure signal of the different modes.

[0007] Furthermore, the data preprocessing module is specifically used for: The pressure distribution signal and pressure change signal in the dynamic pressure signals of the different modes are filtered respectively; Noise suppression processing is performed on the filtered pressure distribution signal and pressure change signal; The noise-suppressed pressure distribution signal and pressure change signal are subjected to amplitude normalization processing; The pressure distribution signal and pressure change signal, after amplitude normalization, are combined to form the preprocessed multimodal dynamic pressure signal.

[0008] Furthermore, the multimodal data fusion module is specifically used for: The preprocessed multimodal dynamic pressure signal is time-stamped and spatial coordinate registered to obtain a spatiotemporally aligned multimodal dynamic pressure signal. Temporal and frequency domain features are extracted from the spatiotemporally aligned multimodal dynamic pressure signal to obtain a multidimensional feature set; The multidimensional feature set is input into a multi-factor coupling model, and the coupling relationship between different features is analyzed through the multi-factor coupling model to obtain the coupling feature matrix. An adaptive weighted fusion algorithm is used to perform feature-level fusion on the coupled feature matrix to generate the fused pressure data.

[0009] Furthermore, the multimodal data fusion module is specifically used for: The time-domain mean and time-domain standard deviation are calculated for the spatiotemporally aligned multimodal dynamic pressure signal; A fast Fourier transform is performed on the spatiotemporally aligned multimodal dynamic pressure signal to obtain a frequency domain representation, and the peak frequency and spectral centroid are extracted from the frequency domain representation. Based on the time-domain standard deviation, the time-domain mean, the peak frequency, and the spectral centroid, a dynamic pressure characteristic index is calculated using a feature fusion formula; wherein, the feature fusion formula is: DPI represents the dynamic pressure characteristic index. This represents the time-domain standard deviation. This represents the time-domain mean. Indicates the peak frequency, Indicates the centroid of the spectrum; The time-domain mean, the time-domain standard deviation, the peak frequency, the spectral centroid, and the dynamic pressure characteristic index are combined to form the multidimensional feature set.

[0010] Furthermore, the multimodal data fusion module is specifically used for: The multi-factor coupling model is used to normalize each feature in the multi-dimensional feature set to obtain normalized feature values. The nonlinear coupling strength between features is calculated based on the normalized feature values, and the coupling feature matrix is ​​constructed based on the nonlinear coupling strength. The nonlinear coupling strength is calculated using the coupling strength formula, which is: ; This represents the nonlinear coupling strength between the i-th feature and the j-th feature. This represents the normalized value of the i-th feature. This represents the normalized value of the j-th feature.

[0011] Furthermore, the multimodal data fusion module is specifically used for: The adaptive weighted fusion algorithm calculates the adaptive weights corresponding to each feature in the multidimensional feature set based on the coupled feature matrix; the calculated adaptive weights are then used to perform a weighted summation of the features in the multidimensional feature set to obtain the fusion pressure data. The formula for calculating the adaptive weights is as follows: ; Indicates the first element in the multidimensional feature set. Adaptive weights corresponding to each feature Represents the first in the coupling feature matrix The first feature and the second The nonlinear coupling strength between the features Indicates the total number of features.

[0012] Furthermore, the analysis model includes a time-frequency analysis unit and a pattern recognition unit; the dynamic pressure assessment module is specifically used for: The time-frequency analysis unit performs a short-time Fourier transform on the fused pressure data to obtain a time-frequency matrix; The pattern recognition unit extracts time-frequency features from the time-frequency matrix and calculates the key feature parameters based on these features using a feature parameter formula; wherein the feature parameter formula is: KFP represents key feature parameters, S(t,f) represents the complex values ​​of the time-frequency matrix at time index t and frequency index f, T represents the total number of time windows, and F represents the total number of frequency components.

[0013] Furthermore, the dynamic pressure assessment module is specifically used for: The dynamic stress index is calculated based on the stress index formula of the stress assessment algorithm and in combination with the key feature parameters; wherein the stress index formula is: DPI represents the dynamic pressure index, KFP represents the key characteristic parameter, α represents the preset pressure sensitivity coefficient, and β represents the preset pressure reference value. The dynamic pressure index is matched with a preset pressure threshold range to determine the corresponding pressure level; Generate the evaluation results that include the dynamic pressure index and the pressure level.

[0014] Furthermore, it also includes: The results display module is used to display the dynamic pressure index and the pressure level in different areas of the display interface, and to visually label the pressure level using color identifiers corresponding to the pressure level.

[0015] The technical solution of this invention solves the problems of insufficient high-frequency signal capture accuracy and weak multimodal data processing capability in the prior art by using a high-frequency response pressure sensor to synchronously acquire multimodal dynamic pressure signals and combining data preprocessing, multimodal fusion and dynamic pressure assessment modules. It achieves high-precision, real-time dynamic pressure assessment and improves the comprehensiveness and reliability of data acquisition and analysis.

[0016] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0018] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 This is a schematic diagram of an embodiment of a multimodal data acquisition and processing system for dynamic stress assessment according to the present invention. Detailed Implementation

[0019] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0020] Figure 1 A schematic diagram of an embodiment of a multimodal data acquisition and processing system for dynamic stress assessment provided by the present invention is shown. Figure 1 As shown, the system includes: The multimodal data acquisition module 110 is used to synchronously acquire dynamic pressure signals of different modes through multiple high-frequency response pressure sensors; wherein the high-frequency response pressure sensors have a target sampling frequency higher than a preset threshold.

[0021] High-frequency response pressure sensors refer to sensing devices capable of rapidly detecting pressure changes and converting pressure signals into electrical signals, with a response frequency higher than conventional pressure sensors. For example, in plantar pressure monitoring, this sensor captures transient pressure signals generated by heel impact during walking at a sampling frequency of 1000Hz. Dynamic pressure signals of different modalities refer to sets of pressure data with time-series characteristics collected from different physical dimensions. For example, in foot pressure monitoring, pressure distribution signals record pressure values ​​in various areas of the sole, while pressure change signals record pressure gradient changes at the moment the heel touches the ground. A preset threshold refers to the minimum sampling frequency limit set according to the application scenario. For example, in dynamic pressure monitoring, 500Hz is set as the minimum sampling frequency threshold to ensure effective capture of rapid pressure fluctuations during human movement. A target sampling frequency refers to the actual data acquisition frequency used, which is higher than the preset threshold. For example, in plantar pressure monitoring, a sampling frequency of 1000Hz is used to collect pressure signals, which is higher than the preset 500Hz threshold.

[0022] The data preprocessing module 120 is used to perform real-time signal conditioning processing on the dynamic pressure signals of different modes to obtain preprocessed multimodal dynamic pressure signals.

[0023] Among them, the preprocessed multimodal dynamic pressure signal refers to a standardized pressure data set that has undergone signal conditioning; for example, the original plantar pressure signal is filtered, denoised and amplitude standardized to form a standardized signal that can be used for subsequent analysis.

[0024] The multimodal data fusion module 130 is used to perform spatiotemporal alignment and feature extraction on the preprocessed multimodal dynamic pressure signal using a multi-factor coupling model, and to generate fused pressure data through a fusion algorithm.

[0025] Among them, the multi-factor coupling model refers to a mathematical model that analyzes the interrelationships between multiple features; for example, this model analyzes the coupling relationship between the time-domain standard deviation of plantar pressure and the frequency-domain peak frequency. The fusion algorithm refers to a calculation method that integrates multi-source feature data into a unified feature vector; for example, a weighted summation algorithm is used to fuse time-domain and frequency-domain features into a comprehensive pressure feature vector. Fusion pressure data refers to comprehensive pressure feature data after multi-source data fusion processing; for example, a foot pressure feature vector formed by fusing time-domain and frequency-domain features from different regions of the foot.

[0026] The dynamic pressure assessment module 140 is used to perform time-frequency domain analysis and pattern recognition on the fused pressure data through an analysis model, extract key feature parameters, and calculate a dynamic pressure index based on the key feature parameters using a pressure assessment algorithm, thereby generating an assessment result containing the dynamic pressure index.

[0027] The analysis model refers to the computational model that performs time-frequency analysis and pattern recognition on the fused data; for example, using short-time Fourier transform and pattern recognition algorithms to analyze the time-frequency characteristics of plantar pressure signals. Key feature parameters refer to the core indicators representing pressure characteristics extracted from the analysis model; for example, the spectral concentration parameter extracted from the plantar pressure time-frequency matrix. The pressure assessment algorithm refers to the calculation method that converts feature parameters into a pressure index; for example, using a logistic function to map key feature parameters to a pressure index within the range of 0-1. The dynamic pressure index refers to a numerical index that quantifies the pressure state; for example, a foot pressure index of 0.85 calculated using a pressure assessment algorithm. The assessment result refers to a comprehensive output including the pressure index and level; for example, an output report containing a pressure index of 0.85 and a high-pressure level.

[0028] The technical solution of this embodiment solves the problems of insufficient high-frequency signal capture accuracy and weak multimodal data processing capability in the prior art by using a high-frequency response pressure sensor to synchronously acquire multimodal dynamic pressure signals and combining data preprocessing, multimodal fusion and dynamic pressure assessment modules. It achieves high-precision, real-time dynamic pressure assessment and improves the comprehensiveness and reliability of data acquisition and analysis.

[0029] In one alternative embodiment, the multimodal data acquisition module 110 is specifically used for: The multiple high-frequency response pressure sensors are arranged at multiple key monitoring points of the object under test, and a uniform target sampling frequency higher than the preset threshold is configured for the high-frequency response pressure sensors arranged at each key monitoring point.

[0030] The subject of the test refers to the organism or object undergoing pressure monitoring; for example, an adult male undergoing plantar pressure monitoring. Key monitoring points refer to important measurement locations where sensors are placed; for example, in plantar pressure monitoring, the heel, arch, and metatarsal areas are selected as key monitoring points.

[0031] A synchronous acquisition command is sent to multiple high-frequency response pressure sensors configured with the target sampling frequency, controlling each high-frequency response pressure sensor to synchronously acquire pressure distribution signals and pressure change signals according to the synchronous acquisition command.

[0032] Among them, the synchronous acquisition command refers to the trigger signal that controls multiple sensors to start acquiring data simultaneously; for example, a time-synchronized start command sent to all plantar pressure sensors. The pressure distribution signal refers to pressure intensity distribution data in the spatial dimension; for example, recording the pressure value distribution in different areas of the sole at a specific moment. The pressure change signal refers to pressure fluctuation data in the time dimension; for example, recording the curve of pressure value changing over time during heel contact with the ground.

[0033] The pressure distribution signal and pressure change signal synchronously collected by each high-frequency response pressure sensor are combined into the dynamic pressure signal of the different modes.

[0034] In the above-mentioned optional methods, the accuracy and real-time performance of dynamic pressure monitoring can be further improved by simultaneously collecting pressure distribution and change signals at multiple key monitoring points and achieving data synchronization with a unified target sampling frequency.

[0035] In an alternative embodiment, the data preprocessing module 120 is specifically used for: The pressure distribution signal and pressure change signal in the dynamic pressure signals of the different modes are filtered respectively.

[0036] Among them, filtering refers to the digital signal processing process of eliminating high-frequency noise in a signal; for example, using a low-pass filter to remove noise components above 100Hz in a plantar pressure signal.

[0037] Noise suppression processing is performed on the filtered pressure distribution signal and pressure change signal.

[0038] Among them, noise suppression processing refers to the technical means of reducing random interference in the signal; for example, using a moving average algorithm to smooth random fluctuations in plantar pressure signals.

[0039] The pressure distribution signal and pressure change signal after noise suppression are subjected to amplitude normalization processing.

[0040] Amplitude normalization refers to the process of adjusting the signal amplitude to a uniform level; for example, normalizing plantar pressure signals collected by different sensors to the range of 0-1.

[0041] The pressure distribution signal and pressure change signal, after amplitude normalization, are combined to form the preprocessed multimodal dynamic pressure signal.

[0042] In the above-mentioned optional methods, the signal quality is further optimized by filtering, denoising, and amplitude standardizing the dynamic pressure signal to ensure the stability and reliability of subsequent data processing.

[0043] In one alternative embodiment, the multimodal data fusion module 130 is specifically used for: The preprocessed multimodal dynamic pressure signal is time-stamped and spatially registered to obtain a spatiotemporally aligned multimodal dynamic pressure signal.

[0044] Timestamp synchronization refers to the process of aligning the time references of multiple signals; for example, assigning a uniform timestamp to all plantar pressure sensor signals. Spatial coordinate registration refers to the process of establishing the spatial positional relationships of sensors; for example, determining the relative positional relationships of each pressure sensor based on the plantar anatomy. Spatiotemporally aligned multimodal dynamic pressure signals refer to a standardized set of signals that has undergone time and spatial alignment processing; for example, time-aligned and spatially registered plantar pressure signals from multiple regions.

[0045] Temporal and frequency domain features are extracted from the spatiotemporally aligned multimodal dynamic pressure signal to obtain a multidimensional feature set.

[0046] Among them, time-domain features refer to signal characteristic parameters extracted from the time dimension; for example, the mean and standard deviation of plantar pressure signals. Frequency-domain features refer to signal characteristic parameters extracted from the frequency dimension; for example, the peak frequency obtained after Fourier transform of the plantar pressure signal. Multidimensional feature sets refer to composite feature groups containing both time-domain and frequency-domain features; for example, a feature set containing the time-domain mean, standard deviation, peak frequency, and spectral centroid of plantar pressure.

[0047] The multidimensional feature set is input into a multi-factor coupling model, and the coupling relationship between different features is analyzed through the multi-factor coupling model to obtain the coupling feature matrix.

[0048] The coupling feature matrix refers to a mathematical matrix that represents the coupling relationship between features; for example, a 4x4 matrix that represents the nonlinear coupling strength between various plantar pressure features.

[0049] An adaptive weighted fusion algorithm is used to perform feature-level fusion on the coupled feature matrix to generate the fused pressure data.

[0050] Among them, the adaptive weighted fusion algorithm refers to a fusion method that automatically assigns weights based on the importance of features; for example, a weighted summation algorithm that automatically calculates the weights of each feature based on the coupling strength.

[0051] In the above-mentioned optional methods, the comprehensive analysis capability of multimodal data can be improved by further performing spatiotemporal alignment and feature extraction on the preprocessed signal and generating fused stress data using a multi-factor coupling model.

[0052] In one alternative embodiment, the multimodal data fusion module 130 is specifically used for: The time-domain mean and time-domain standard deviation are calculated for the spatiotemporally aligned multimodal dynamic pressure signal.

[0053] The time-domain mean refers to the average value of the signal over time; for example, the average pressure value of a plantar pressure signal within a 1-second time window. The time-domain standard deviation refers to the degree of fluctuation of the signal over time; for example, the pressure fluctuation amplitude of a plantar pressure signal within a 1-second time window.

[0054] A fast Fourier transform is performed on the spatiotemporally aligned multimodal dynamic pressure signal to obtain a frequency domain representation, and the peak frequency and spectral centroid are extracted from the frequency domain representation.

[0055] In this context, frequency domain representation refers to the way a signal is expressed in the frequency domain; for example, the spectrum obtained by performing a Fast Fourier Transform on a plantar pressure signal. Peak frequency refers to the frequency component with the highest energy in the spectrum; for example, the frequency corresponding to the highest energy peak at 10Hz in the plantar pressure spectrum. Spectral centroid refers to the center frequency of the spectral energy distribution; for example, the center point of the plantar pressure spectrum energy distribution is located at 15Hz.

[0056] Based on the time-domain standard deviation, the time-domain mean, the peak frequency, and the spectral centroid, the dynamic pressure characteristic index is calculated using a feature fusion formula.

[0057] The dynamic pressure characteristic index refers to a composite index that comprehensively characterizes the dynamic properties of pressure; for example, the plantar pressure dynamic index calculated by combining time-domain and frequency-domain features. The feature fusion formula is as follows: DPI represents the dynamic pressure characteristic index. This represents the time-domain standard deviation. This represents the time-domain mean. Indicates the peak frequency, This indicates the centroid of the spectrum.

[0058] It should be noted that the feature fusion formula creatively combines time-domain fluctuation characteristics with frequency-domain distribution characteristics. The product of the time-domain standard deviation and the peak frequency reflects the intensity of signal changes, while the product of the time-domain mean and the spectral centroid characterizes the signal stability. The ratio of these two values ​​effectively quantifies the dynamic characteristics of pressure changes. Compared with traditional single-domain feature analysis, this formula can more comprehensively capture the spatiotemporal variation patterns of dynamic pressure signals, solving the technical problem of insufficient representation capability of single features.

[0059] The time-domain mean, the time-domain standard deviation, the peak frequency, the spectral centroid, and the dynamic pressure characteristic index are combined to form the multidimensional feature set.

[0060] In the above-mentioned optional methods, the dynamic pressure characteristic index is further calculated through the feature fusion formula, and the time domain mean, standard deviation, peak frequency, etc. are combined into a multi-dimensional feature set to enhance the systematicness and comprehensiveness of feature extraction.

[0061] In one alternative embodiment, the multimodal data fusion module 130 is specifically used for: The multi-factor coupling model normalizes each feature in the multi-dimensional feature set to obtain normalized feature values. The nonlinear coupling strength between features is calculated based on the normalized feature values, and the coupling feature matrix is ​​constructed based on the nonlinear coupling strength.

[0062] Here, normalized eigenvalues ​​refer to the characteristic values ​​after standardization; for example, the values ​​after normalizing the time-domain mean to the range of 0-1. Nonlinear coupling strength refers to the measure of the strength of nonlinear interaction between features; for example, the nonlinear correlation strength between the time-domain standard deviation of plantar pressure and the peak frequency in the frequency domain.

[0063] The nonlinear coupling strength is calculated using the coupling strength formula, which is: ; This represents the nonlinear coupling strength between the i-th feature and the j-th feature. This represents the normalized value of the i-th feature. This represents the normalized value of the j-th feature.

[0064] It should be noted that the coupling strength formula adopts a nonlinear function structure with a fractional exponent. By using a denominator term of the 1.5th power, it effectively adjusts the influence of the eigenvalue magnitude on the coupling strength, overcoming the limitation of traditional linear correlation coefficients in capturing nonlinear relationships. This specific mathematical expression can more accurately characterize the complex coupling relationships between multimodal pressure features, providing a reliable basis for subsequent weighted fusion.

[0065] In the above-mentioned optional methods, the nonlinear relationship between features is further analyzed through a multi-factor coupling model, and a coupling feature matrix is ​​constructed to improve the scientificity and rationality of the collaborative analysis of different features.

[0066] In one alternative embodiment, the multimodal data fusion module 130 is specifically used for: The adaptive weighted fusion algorithm calculates the adaptive weights corresponding to each feature in the multidimensional feature set based on the coupled feature matrix; the calculated adaptive weights are then used to perform a weighted summation of the features in the multidimensional feature set to obtain the fusion pressure data.

[0067] Adaptive weights refer to weight coefficients that are automatically adjusted based on feature importance; for example, feature weights automatically calculated based on coupling strength in plantar pressure feature fusion. The formula for calculating adaptive weights is: ; Indicates the first element in the multidimensional feature set. Adaptive weights corresponding to each feature Represents the first in the coupling feature matrix The first feature and the second The nonlinear coupling strength between the features Indicates the total number of features.

[0068] It should be noted that the adaptive weight calculation formula has made a significant improvement on the traditional softmax function. It innovatively introduces the sum of the absolute values ​​of coupling strengths as input features, and amplifies the differences in weight allocation for important features through an exponential function. This design allows features with stronger coupling strengths to receive significantly higher weights, thereby highlighting the role of key features in the fusion process and improving the representational ability of fusion-pressured data.

[0069] In the above-mentioned optional methods, an adaptive weighted fusion algorithm is further used to calculate weights based on the coupling feature matrix and perform weighted summation to dynamically optimize the feature fusion ratio and improve the overall expressiveness of the fused data.

[0070] In one alternative embodiment, the analysis model includes a time-frequency analysis unit and a pattern recognition unit; the dynamic pressure assessment module 140 is specifically used for: The time-frequency analysis unit performs a short-time Fourier transform on the fused pressure data to obtain a time-frequency matrix.

[0071] The time-frequency matrix refers to a two-dimensional data matrix that contains both time and frequency information; for example, the time-frequency distribution map obtained by short-time Fourier transform of plantar pressure signal.

[0072] The pattern recognition unit extracts time-frequency features from the time-frequency matrix and calculates the key feature parameters based on the time-frequency features using the feature parameter formula.

[0073] The formula for the characteristic parameters is as follows: KFP represents key feature parameters, S(t,f) represents the complex values ​​of the time-frequency matrix at time index t and frequency index f, T represents the total number of time windows, and F represents the total number of frequency components.

[0074] It should be noted that the characteristic parameter formula creatively extracts the kurtosis characteristic of the signal as a key characteristic parameter by calculating the ratio of the fourth moment to the square of the second moment of the time-frequency matrix amplitude. Compared with traditional time-frequency features, this parameter is more sensitive to the energy concentration characteristics of the signal and can effectively distinguish the time-frequency distribution patterns under different pressure states, providing a more discriminative characteristic indicator for dynamic pressure assessment.

[0075] In the above-mentioned optional methods, the accuracy and completeness of dynamic pressure assessment can be further improved by performing short-time Fourier transform on the fused pressure data, extracting time-frequency features and calculating key feature parameters.

[0076] In an alternative embodiment, the dynamic pressure assessment module 140 is specifically used for: The dynamic pressure index is calculated based on the pressure index formula of the pressure assessment algorithm and in combination with the key feature parameters.

[0077] The pressure index formula is as follows: DPI represents the dynamic pressure index, KFP represents the key characteristic parameter, α represents the preset pressure sensitivity coefficient, and β represents the preset pressure reference value.

[0078] The parameter settings in the pressure assessment algorithm were optimized based on a large amount of experimental data. The pressure sensitivity coefficient α ranges from [0.5, 2.0], and the pressure benchmark value β ranges from [-1.0, 1.0]. In practical applications, specific values ​​are determined through cross-validation to ensure the accuracy and stability of the dynamic pressure index under different application scenarios.

[0079] It should be noted that the pressure index formula uses a combination of logistic functions and linear transformations. By coordinating the adjustment of the pressure sensitivity coefficient and the pressure benchmark value, key characteristic parameters are mapped to a standardized pressure index range. This mapping relationship maintains the relative magnitudes of the characteristic parameters while enhancing the distinguishability of different pressure ranges through nonlinear transformations, thereby improving the accuracy and interpretability of the assessment results.

[0080] The dynamic pressure index is matched with a preset pressure threshold range to determine the corresponding pressure level.

[0081] The preset pressure threshold range refers to a pre-defined range of pressure levels; for example, a pressure index of 0-0.3 is set as low pressure, 0.3-0.7 as normal, and 0.7-1.0 as high pressure. Pressure level refers to the category of pressure level based on the pressure index; for example, a pressure index of 0.85 corresponds to a high pressure level.

[0082] Generate the evaluation results that include the dynamic pressure index and the pressure level.

[0083] In the above-mentioned optional methods, a dynamic pressure index is further calculated by combining the pressure index formula with key characteristic parameters, and the pressure level is matched to generate intuitive evaluation results, thereby improving the practicality and readability of the evaluation.

[0084] In one alternative approach, it also includes: The results display module is used to display the dynamic pressure index and the pressure level in different areas of the display interface, and to visually label the pressure level using color identifiers corresponding to the pressure level.

[0085] The display interface refers to the visual interface used to present the assessment results; for example, the application interface on a tablet computer that displays the plantar pressure assessment results. Zonal display refers to displaying various types of information in different areas of the interface; for example, displaying the pressure index value on the left and the pressure level text on the right. Color identifiers refer to visual markers that use different colors to represent different states; for example, using red to indicate a high-pressure level and green to indicate a normal level. Visual annotation refers to techniques that use graphical elements to enhance the expressiveness of information; for example, adding a color block of the corresponding color to the background of the pressure level text for highlighting.

[0086] In the above-mentioned optional methods, the dynamic pressure index and pressure level are further displayed in separate sections through the results display module, and the pressure level is indicated by color to enhance the visualization effect of the assessment results and the user experience.

[0087] To better illustrate the technical solution of this embodiment, the following example is used for complete explanation: S10: Multiple high-frequency response pressure sensors are placed at key monitoring points of the heel, arch, and metatarsal bones of the subject under test. A unified target sampling frequency higher than a preset threshold is configured. A synchronous acquisition command is sent to control the sensors to synchronously acquire pressure distribution signals and pressure change signals, and these signals are combined into dynamic pressure signals of different modes. S20: Filter, suppress noise, and normalize the dynamic pressure signals of different modes to obtain the preprocessed multimodal dynamic pressure signals; S30: Perform timestamp synchronization and spatial coordinate registration on the preprocessed multimodal dynamic pressure signal to obtain a spatiotemporally aligned multimodal dynamic pressure signal, and extract time-domain features and frequency-domain features from it, and combine them into a multidimensional feature set; S40: Normalize the multidimensional feature set through a multi-factor coupling model, calculate the nonlinear coupling strength between features, and construct the coupling feature matrix; S50: An adaptive weighted fusion algorithm is used to calculate adaptive weights based on the coupled feature matrix, and the features in the multi-dimensional feature set are weighted and summed to generate fusion pressure data; S60: The time-frequency matrix is ​​obtained by performing a short-time Fourier transform on the fused pressure data through the analysis model, and the time-frequency features are extracted from it to calculate key feature parameters; S70: Calculate the dynamic pressure index based on key characteristic parameters using a pressure assessment algorithm, and match the dynamic pressure index with a preset pressure threshold range to determine the pressure level; S80: Generates assessment results including dynamic pressure index and pressure level, and displays them in different areas of the display interface through the results display module, using color identifiers to visually annotate the pressure level.

[0088] Furthermore, the system provided in the above embodiments is only illustrated by the division of the above functional modules when implementing its functions. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the system can be divided into different functional modules according to the actual situation to complete all or part of the functions described above.

[0089] The above description is merely a preferred embodiment of the present invention and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this invention is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-disclosed concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this invention.

[0090] It should be noted that the terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and represent a limitation on a specific order or sequence. Where appropriate, the order of use for similar objects can be interchanged so that the embodiments of this application described herein can be implemented in an order other than that shown or described.

[0091] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A multimodal data acquisition and processing system for dynamic stress assessment, characterized in that, The system includes: A multimodal data acquisition module is used to simultaneously acquire dynamic pressure signals of different modes through multiple high-frequency response pressure sensors; wherein, the high-frequency response pressure sensors have a target sampling frequency higher than a preset threshold. The data preprocessing module is used to perform real-time signal conditioning on the dynamic pressure signals of different modes to obtain preprocessed multimodal dynamic pressure signals. The multimodal data fusion module is used to perform spatiotemporal alignment and feature extraction on the preprocessed multimodal dynamic pressure signal using a multi-factor coupling model, and to generate fused pressure data through a fusion algorithm. The dynamic stress assessment module is used to perform time-frequency domain analysis and pattern recognition on the fused stress data through an analysis model, extract key feature parameters, and calculate a dynamic stress index based on the key feature parameters using a stress assessment algorithm, thereby generating an assessment result containing the dynamic stress index.

2. The multimodal data acquisition and processing system for dynamic pressure assessment according to claim 1, characterized in that, The multimodal data acquisition module is specifically used for: The multiple high-frequency response pressure sensors are arranged at multiple key monitoring points of the object under test, and a uniform target sampling frequency higher than the preset threshold is configured for the high-frequency response pressure sensors arranged at each key monitoring point. A synchronous acquisition command is sent to multiple high-frequency response pressure sensors configured with the target sampling frequency, controlling each high-frequency response pressure sensor to synchronously acquire pressure distribution signals and pressure change signals according to the synchronous acquisition command; The pressure distribution signal and pressure change signal synchronously collected by each high-frequency response pressure sensor are combined into the dynamic pressure signal of the different modes.

3. The multimodal data acquisition and processing system for dynamic pressure assessment according to claim 2, characterized in that, The data preprocessing module is specifically used for: The pressure distribution signal and pressure change signal in the dynamic pressure signals of the different modes are filtered respectively; Noise suppression processing is performed on the filtered pressure distribution signal and pressure change signal; The noise-suppressed pressure distribution signal and pressure change signal are subjected to amplitude normalization processing; The pressure distribution signal and pressure change signal, after amplitude normalization, are combined to form the preprocessed multimodal dynamic pressure signal.

4. The multimodal data acquisition and processing system for dynamic pressure assessment according to claim 1, characterized in that, The multimodal data fusion module is specifically used for: The preprocessed multimodal dynamic pressure signal is time-stamped and spatial coordinate registered to obtain a spatiotemporally aligned multimodal dynamic pressure signal. Temporal and frequency domain features are extracted from the spatiotemporally aligned multimodal dynamic pressure signal to obtain a multidimensional feature set; The multidimensional feature set is input into a multi-factor coupling model, and the coupling relationship between different features is analyzed through the multi-factor coupling model to obtain the coupling feature matrix. An adaptive weighted fusion algorithm is used to perform feature-level fusion on the coupled feature matrix to generate the fused pressure data.

5. The multimodal data acquisition and processing system for dynamic pressure assessment according to claim 4, characterized in that, The multimodal data fusion module is specifically used for: The time-domain mean and time-domain standard deviation are calculated for the spatiotemporally aligned multimodal dynamic pressure signal; A fast Fourier transform is performed on the spatiotemporally aligned multimodal dynamic pressure signal to obtain a frequency domain representation, and the peak frequency and spectral centroid are extracted from the frequency domain representation. Based on the time-domain standard deviation, the time-domain mean, the peak frequency, and the spectral centroid, a dynamic pressure characteristic index is calculated using a feature fusion formula; wherein, the feature fusion formula is: DPI represents the dynamic pressure characteristic index. This represents the time-domain standard deviation. This represents the time-domain mean. Indicates the peak frequency, Indicates the centroid of the spectrum; The time-domain mean, the time-domain standard deviation, the peak frequency, the spectral centroid, and the dynamic pressure characteristic index are combined to form the multidimensional feature set.

6. The multimodal data acquisition and processing system for dynamic pressure assessment according to claim 5, characterized in that, The multimodal data fusion module is specifically used for: The multi-factor coupling model is used to normalize each feature in the multi-dimensional feature set to obtain normalized feature values. The nonlinear coupling strength between features is calculated based on the normalized feature values, and the coupling feature matrix is ​​constructed based on the nonlinear coupling strength. The nonlinear coupling strength is calculated using the coupling strength formula, which is: ; This represents the nonlinear coupling strength between the i-th feature and the j-th feature. This represents the normalized value of the i-th feature. This represents the normalized value of the j-th feature.

7. The multimodal data acquisition and processing system for dynamic pressure assessment according to claim 6, characterized in that, The multimodal data fusion module is specifically used for: The adaptive weighted fusion algorithm calculates the adaptive weights corresponding to each feature in the multidimensional feature set based on the coupled feature matrix; the calculated adaptive weights are then used to perform a weighted summation of the features in the multidimensional feature set to obtain the fusion pressure data. The formula for calculating the adaptive weights is as follows: ; Indicates the first element in the multidimensional feature set. Adaptive weights corresponding to each feature Represents the first in the coupling feature matrix The first feature and the second The nonlinear coupling strength between the features This represents the total number of features.

8. The multimodal data acquisition and processing system for dynamic pressure assessment according to claim 7, characterized in that, The analysis model includes a time-frequency analysis unit and a pattern recognition unit; the dynamic pressure assessment module is specifically used for: The time-frequency analysis unit performs a short-time Fourier transform on the fused pressure data to obtain a time-frequency matrix; The pattern recognition unit extracts time-frequency features from the time-frequency matrix and calculates the key feature parameters based on these features using a feature parameter formula; wherein the feature parameter formula is: KFP represents key feature parameters, S(t,f) represents the complex values ​​of the time-frequency matrix at time index t and frequency index f, T represents the total number of time windows, and F represents the total number of frequency components.

9. The multimodal data acquisition and processing system for dynamic pressure assessment according to claim 8, characterized in that, The dynamic stress assessment module is specifically used for: The dynamic stress index is calculated based on the stress index formula of the stress assessment algorithm and in combination with the key feature parameters; wherein the stress index formula is: DPI represents the dynamic pressure index, KFP represents the key characteristic parameter, α represents the preset pressure sensitivity coefficient, and β represents the preset pressure reference value. The dynamic pressure index is matched with a preset pressure threshold range to determine the corresponding pressure level; Generate the evaluation results that include the dynamic pressure index and the pressure level.

10. The multimodal data acquisition and processing system for dynamic pressure assessment according to claim 9, characterized in that, Also includes: The results display module is used to display the dynamic pressure index and the pressure level in different areas of the display interface, and to visually label the pressure level using color identifiers corresponding to the pressure level.

Images