A gait feature extraction and analysis method and device

Abstract

The application provides a gait feature extraction and analysis method and device, the method comprises the following steps: acquiring acceleration original signals collected by a three-axis accelerometer and angular velocity original signals collected by a three-axis gyroscope in the same period, the three-axis accelerometer and the three-axis gyroscope are used to be placed on a predetermined body part; cutting the acceleration original signals and the angular velocity original signals into a plurality of signal windows respectively, and taking the signal windows in the gait section as target signal windows; extracting acceleration gait features, angular velocity gait features and signal features based on the target signal windows; and generating a gait analysis feature matrix according to the acceleration gait features, the angular velocity gait features and the signal features. The application realizes efficient and low-cost gait analysis by analyzing the multi-dimensional feature data collected by low-cost collection equipment, and improves the accuracy of gait analysis.

Description

Technical Field

[0001] This invention relates to the field of biostatistics and data analysis, and more particularly to a method and apparatus for gait feature extraction and analysis. Background Technology

[0002] Gait analysis is an important tool for the early detection of neurodegenerative diseases such as Parkinson's disease and dementia. Traditional gait analysis methods mainly rely on three-dimensional motion capture systems or gait detection in a laboratory setting. While these methods provide accurate data, they have several limitations. First, the equipment is expensive, making it difficult to implement in large-scale screenings. Second, because the tests are usually conducted in a controlled environment, they cannot capture the natural gait characteristics of subjects in their daily lives. Another common detection method is through specialist examinations in hospitals, but these methods are time-consuming, labor-intensive, and difficult to track in real time the progression of the disease.

[0003] In addition, while existing gait analysis based on wearable devices has reduced costs and improved convenience, it has certain limitations in the extraction of gait features. It only uses simple acceleration data to analyze individual movement intensity, resulting in poor accuracy of gait analysis.

[0004] Therefore, existing technologies have shortcomings and need to be improved and developed. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a gait feature extraction and analysis method and apparatus to address the above-mentioned deficiencies of the prior art, which aims to solve the problem that the accuracy of gait analysis is poor because the prior art only uses simple acceleration data to analyze the intensity of individual movement.

[0006] The technical solution adopted by this invention to solve the technical problem is as follows:

[0007] A method for gait feature extraction and analysis, wherein the method includes:

[0008] The raw acceleration signal collected by a triaxial accelerometer and the raw angular velocity signal collected by a triaxial gyroscope are acquired at the same time period. The triaxial accelerometer and the triaxial gyroscope are used to be placed at a predetermined body part.

[0009] The original acceleration signal and the original angular velocity signal are respectively divided into several signal windows, and the signal window in the gait segment is used as the target signal window;

[0010] Acceleration gait features, angular velocity gait features, and signal features are extracted based on the target signal window;

[0011] A gait analysis feature matrix is ​​generated based on the acceleration gait features, angular velocity gait features, and signal features.

[0012] In one embodiment of this application, the raw acceleration signal and the raw angular velocity signal are respectively divided into several signal windows, and the signal window in the gait segment is used as the target signal window, including:

[0013] The original acceleration signal and the original angular velocity signal are respectively divided into several signal windows;

[0014] For each signal window, extract the root mean square, range, dominant frequency, spectral flatness, signal entropy, and mean crossover rate.

[0015] The root mean square, range, dominant frequency, spectral flatness, signal entropy, and mean crossover rate of each signal window are input into the trained random forest classification model to obtain the output results.

[0016] If the output result indicates that the signal window is in the gait segment, then the signal window is taken as the target signal window.

[0017] In one embodiment of this application, the extraction of acceleration gait features, angular velocity gait features, and signal features based on the target signal window includes:

[0018] Identify each gait cycle and sub-stage of each gait cycle within each of the target signal windows;

[0019] Acceleration gait features, angular velocity gait features, and signal features are extracted based on each gait cycle and sub-stage of each target signal window.

[0020] In one embodiment of this application, identifying each gait cycle and sub-stage of each of the target signal windows includes:

[0021] Based on the continuous wavelet transform algorithm, the downward peak value of smooth vertical acceleration is detected for each target signal window to determine the initial contact, and the upward peak value of integrated vertical acceleration is detected for each target signal window to determine the final contact.

[0022] Each gait cycle and sub-phase of the gait cycle are determined within the target signal window based on the initial and final contact.

[0023] The gait cycle consists of the following sub-phases: the support phase, the swing phase, the stride phase, and the half-stride phase.

[0024] In one embodiment of this application, the acceleration gait characteristics include at least one of the following: stride duration, stride length duration, swing duration, support duration, initial double support, final double support, double support, single-leg support, stride length, asymmetry of duration, asymmetry of length, walking speed, and cadence.

[0025] The angular velocity gait characteristics include at least one of the following: average absolute angular velocity, angular change, and rotation.

[0026] The signal characteristics include at least one of the following: standard deviation, coefficient of variation, sweep factor, mean absolute change, mean crossover rate, skewness, kurtosis, absolute energy, signal entropy, root mean square of the signal, and signal range.

[0027] In one embodiment of this application, the gait analysis feature matrix further includes the mean and standard deviation calculated based on each of the acceleration gait features, the angular velocity gait features, and the signal features.

[0028] In one embodiment of this application, after acquiring the raw acceleration signals collected by a triaxial accelerometer during the same time period, the method further includes:

[0029] The signal vector amplitude is calculated based on the original acceleration signal, and the signal vector amplitude is...

[0030] The signal vector amplitude is divided into several segmented windows using a sliding window method;

[0031] Convert several segmented windows into a low-dimensional space;

[0032] Obtain the trained Gaussian mixture model, and use the Gaussian mixture model to cluster several segmented windows in the low-dimensional space to obtain the clustering probability corresponding to the amplitude of the signal vector.

[0033] This application also provides a gait feature extraction and analysis device, wherein the device includes:

[0034] The acquisition module is used to acquire the raw acceleration signal collected by the triaxial accelerometer and the raw angular velocity signal collected by the triaxial gyroscope during the same time period. The triaxial accelerometer and the triaxial gyroscope are used to be placed on a predetermined body part.

[0035] The cutting module is used to cut the original acceleration signal and the original angular velocity signal into several signal windows, and to use the signal window in the gait segment as the target signal window.

[0036] The extraction module is used to extract acceleration gait features, angular velocity gait features, and signal features based on the target signal window;

[0037] The generation module is used to generate a gait analysis feature matrix based on the acceleration gait features, angular velocity gait features, and signal features.

[0038] This application also provides a terminal, comprising: a memory, a processor, and a gait feature extraction and analysis program stored in the memory and executable on the processor, wherein the gait feature extraction and analysis program, when executed by the processor, implements the steps of the gait feature extraction and analysis method as described above.

[0039] This application discloses a computer-readable storage medium storing a computer program that can be executed to implement the steps of the gait feature extraction and analysis method described above.

[0040] This invention provides a method and apparatus for gait feature extraction and analysis. The method includes: acquiring raw acceleration signals collected by a triaxial accelerometer and raw angular velocity signals collected by a triaxial gyroscope during the same time period, wherein the triaxial accelerometer and triaxial gyroscope are placed at predetermined body parts; dividing the raw acceleration and angular velocity signals into several signal windows, and using the signal windows in the gait segment as target signal windows; extracting acceleration gait features, angular velocity gait features, and signal features based on the target signal windows; and generating a gait analysis feature matrix based on the acceleration gait features, angular velocity gait features, and signal features. This invention obtains multi-dimensional feature data through data analysis using low-cost acquisition equipment, achieving efficient and low-cost gait analysis and improving the accuracy of gait analysis. Attached Figure Description

[0041] Figure 1 This is a flowchart of a preferred embodiment of the gait feature extraction and analysis method in this invention;

[0042] Figure 2 These are the original acceleration signals for different activities in this invention;

[0043] Figure 3 This is a flowchart of gait data preprocessing in this invention;

[0044] Figure 4 This is a schematic diagram of the gait period in this invention;

[0045] Figure 5 This refers to the proportion of motion intensity clusters in the training data of this invention;

[0046] Figure 6 This refers to the clustering ratio of the exercise intensity of the test subjects in this invention;

[0047] Figure 7This is a flowchart of motion intensity clustering in this invention;

[0048] Figure 8 This is a functional principle block diagram of a preferred embodiment of the gait feature extraction and analysis device of the present invention;

[0049] Figure 9 This is a functional principle block diagram of a preferred embodiment of the terminal in this invention. Detailed Implementation

[0050] To make the objectives, technical solutions, and advantages of this invention clearer and more explicit, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0051] The disadvantages of existing technologies include:

[0052] First, high cost: Traditional gait analysis systems are expensive and difficult to apply on a large scale.

[0053] Second, environmental limitations: It is impossible to obtain natural gait data in daily life, and there is a lack of analysis of gait behavior in real-world scenarios.

[0054] Third, insufficient feature extraction: Existing wearable devices mainly rely on simple acceleration data, which cannot effectively extract angular velocity features and signal features, thus limiting the accuracy of gait analysis.

[0055] Fourth, insufficient motion intensity assessment: existing systems fail to fully utilize higher-level features for accurate motion intensity analysis, resulting in poor performance in diverse motion scenarios.

[0056] This invention provides a gait analysis platform based on wearable devices, capable of collecting and analyzing gait data over extended periods in free-living environments. First, through efficient signal processing algorithms, this invention automatically extracts various gait features, particularly angular velocity and signal features, providing richer data dimensions for gait anomaly detection and early diagnosis of neurodegenerative diseases. Second, by training independent motion intensity models and utilizing features such as signal vector amplitude (SVM), this invention can accurately assess an individual's motion intensity, identify the proportion of motion in various daily activity types (such as walking, running, and standing), and performs excellently in various motion scenarios.

[0057] This invention combines low-cost, portable wearable devices with advanced signal processing and machine learning algorithms to achieve efficient and accurate gait analysis and motion intensity assessment. By collecting real gait data in natural environments, it can improve the accuracy of early disease detection and support large-scale screening and personalized health monitoring.

[0058] Please see Figure 1 , Figure 1 This is a flowchart of the gait feature extraction and analysis method in this invention. For example... Figure 1 As shown, the gait feature extraction and analysis method described in this embodiment of the invention includes:

[0059] Step S100: Obtain the raw acceleration signal collected by a triaxial accelerometer and the raw angular velocity signal collected by a triaxial gyroscope during the same time period. The triaxial accelerometer and triaxial gyroscope are used to be placed on a predetermined body part.

[0060] Specifically, the gait feature extraction and analysis method provided in this application embodiment can be implemented based on a gait analysis platform, which includes: a data acquisition module, a data preprocessing module, a gait feature extraction module, and a motion intensity clustering module. The data acquisition module is used to collect gait data using an IMU sensor; the data preprocessing module is used to perform preprocessing operations such as denoising, filtering, and signal segmentation on the collected data; the gait feature extraction module is used to extract acceleration gait features, angular velocity gait features, and signal features to generate a feature matrix; and the motion intensity clustering module is used to construct daily activity clusters based on a Gaussian mixture model.

[0061] The data acquisition module uses an Axivity 6 (AX6) sensor, an inertial measurement unit (IMU) comprising a three-axis accelerometer and a three-axis gyroscope. The three-axis accelerometer is primarily used for body motion measurement, providing the amplitude and direction of acceleration in three-dimensional space; the three-axis gyroscope measures the angular velocity of various body parts around predetermined axes. To ensure proper sensor operation, the measurement device also integrates a temperature sensor, an ambient light sensor, a real-time clock, and a lithium polymer battery. This sensor can continuously record 6-axis IMU data at a frequency of 100Hz for 7 days. The AX6 device is attached to the participant's torso via a belt. By recording the acceleration changes when the heel contacts and leaves the ground, the IMU located on the torso can capture the gait cycle and its sub-phases.

[0062] Although the embodiments of this application are mainly based on the AX6 device for data acquisition, other similar inertial measurement units (IMUs) can also be used as alternative devices, as long as the inertial measurement unit provides data from the triaxial accelerometer and triaxial gyroscope.

[0063] like Figure 1 As shown in the embodiments of the present invention, the gait feature extraction and analysis method further includes:

[0064] Step S200: Cut the original acceleration signal and the original angular velocity signal into several signal windows, and take the signal window in the gait segment as the target signal window.

[0065] In this embodiment of the application, step S200 specifically includes:

[0066] Step S210: Divide the original acceleration signal and the original angular velocity signal into several signal windows respectively;

[0067] Step S220: Extract the root mean square, range, dominant frequency, spectral flatness, signal entropy, and mean crossover rate for each signal window;

[0068] Step S230: Input the root mean square, range, dominant frequency, spectral flatness, signal entropy and mean crossover rate of each signal window into the trained random forest classification model to obtain the output results;

[0069] Step S240: If the output result is that the signal window is in the gait segment, then the signal window is taken as the target signal window.

[0070] Specifically, Figure 2 The presentation shows raw data samples of different activities recorded by the AX6. Since sensor data is frequently affected by high-frequency noise, the raw data first needs to be processed using a low-pass filter. This embodiment uses a fourth-order digital Butterworth low-pass filter with a cutoff frequency of 20Hz.

[0071] After denoising, the gait segment needs to be distinguished from other activities. First, the original signal is divided into 3-second windows, and features such as root mean square, range, dominant frequency, spectral flatness, signal entropy, and mean crossover rate are extracted within each window. These signal features are then used as input data into a trained random forest classification model to determine whether the window falls within the gait segment.

[0072] This application embodiment preprocesses the data to facilitate the extraction of gait features.

[0073] like Figure 1 As shown in the embodiments of the present invention, the gait feature extraction and analysis method further includes:

[0074] Step S300: Extract acceleration gait features, angular velocity gait features, and signal features based on the target signal window.

[0075] In one embodiment of this application, step S300 specifically includes:

[0076] Step S310: Identify each gait cycle and sub-stage of each gait cycle in each of the target signal windows;

[0077] Step S320: Extract acceleration gait features, angular velocity gait features, and signal features based on each gait cycle and sub-stage of each target signal window.

[0078] This application embodiment can accurately determine the start and end time of each step during walking by identifying each gait cycle in the target signal window, providing an accurate time reference for subsequent gait feature extraction.

[0079] In this embodiment of the application, step S310 specifically includes:

[0080] Step S311: Based on the continuous wavelet transform algorithm, detect the downward peak value of smooth vertical acceleration for each target signal window to determine the initial contact, and detect the upward peak value of integrated vertical acceleration for each target signal window to determine the final contact.

[0081] Step S312: Determine each gait cycle and sub-stage of the gait cycle in the target signal window based on the initial contact and the final contact.

[0082] The gait cycle consists of the following sub-phases: the support phase, the swing phase, the stride phase, and the half-stride phase.

[0083] Specifically, after extracting the gait segments, in order to further identify each gait cycle and its sub-phases within the gait segments and extract relevant gait features, it is necessary to detect the initial contact (IC) and final contact (FC) of each gait cycle. From Figure 2 The changes in vertical acceleration amplitude show that when a person is stationary, the accelerometer measures the inertial force of gravity; however, at the initial contact, the mass block inside AX6 experiences hypergravity due to the inertial force, leading to an increase in the vertical acceleration amplitude, reflected in the downward peak in the graph. This provides clues for detecting integrated gait (IC) and accelerometer-force (FC). Therefore, this embodiment utilizes the continuous wavelet transform (CWT) algorithm, which has good robustness in detecting IC and FC. By applying the CWT algorithm, IC can be obtained by detecting the downward peak of smoothed vertical acceleration, and FC can be obtained by detecting the upward peak of velocity (integrated vertical acceleration). Thus, the start and end of the gait cycle can be identified. The overall process of gait data preprocessing is as follows: Figure 3 As shown, the gait cycle and its sub-stages are as follows: Figure 4 As shown.

[0084] The embodiments of this application use continuous wavelet transform (CWT) to detect key moments of gait cycle. Other time-frequency analysis methods, such as short-time Fourier transform (STFT) or Hilbert transform, can also be used to improve the ability to capture gait features.

[0085] like Figure 1 As shown in the embodiments of the present invention, the gait feature extraction and analysis method further includes:

[0086] Step S400: Generate a gait analysis feature matrix based on the acceleration gait features, angular velocity gait features, and signal features.

[0087] The definitions and symbols involved in feature extraction are shown in Table 1.

[0088] Table 1

[0089] symbol describe IC Initial contact (an event in the gait cycle) FC Final contact (an event in the gait cycle) t Time index or current time step l Effective leg length h Vertical displacement of the center of gravity T Total number of samples within a given time period v t ]]> angular velocity at time t D Gait phases (stride length, stride length, swing, support) [00000D2] duration ]] Duration of the gait phase S Signal window [SA r ]] root mean square of the square root of the signal s t ]]> Signal value at time t μ S ]]> Mean of signal S <![CDATA[σ S ]]> Standard deviation of signal S

[0090] In the embodiments of this application, the acceleration gait characteristics include at least one of the following: stride duration, stride length duration, swing duration, support duration, initial double support, final double support, double support, single-leg support, stride length, asymmetry of duration, asymmetry of length, walking speed, and cadence.

[0091] Specifically, once the gait cycle and its sub-phases are identified, the duration of each gait phase can be calculated. The duration is defined as the time difference between the start and end of a phase. This calculation relies on detecting initial contact (IC) and final contact (FC) events, including:

[0092] Stride duration: The time from one IC to the next IC of the same stride: IC(t+2) - IC(t);

[0093] Step duration: The time from one IC to the next IC of the opposite foot: IC(t+1) - IC(t);

[0094] Swing duration: The time from FC to the next IC: IC(t+2) - FC(t);

[0095] Stance duration: The time from IC to FC: FC(t) - IC(t);

[0096] Initial double support: The time during which both feet are on the ground at the start of the support: FC(t-1) - IC(t);

[0097] Terminal double support: The time during which both feet are on the ground at the end of the support phase: FC(t) - IC(t+1);

[0098] Double support: The sum of the initial and final double support times;

[0099] Single limb support: The time when only one foot is on the ground: IC(t+1) - FC(t-1).

[0100] For each gait cycle and the length of its sub-phases, including:

[0101] Step length: The distance covered in a single step, calculated using the following formula: Where h is the vertical displacement of the center of gravity, and l is the effective leg length;

[0102] Stride length: The total distance covered in one gait cycle (two steps): stride length (t) + stride length (t+1).

[0103] Asymmetry measures the characteristic differences between two consecutive gait cycles. Asymmetry includes:

[0104] Asymmetry in duration: |x duration (t+1)-x duration (t)|;

[0105] Asymmetry in length: |x_length(t+1)-x_length(t)|, where x can be any of the parameters mentioned above.

[0106] For the velocity of each gait cycle and its sub-phases, including:

[0107] Walking speed: The formula is stride length divided by stride duration: stride length / stride duration;

[0108] Step frequency, steps per minute: calculated as the number of steps taken per minute: 60 / step length duration;

[0109] After extracting these spatiotemporal features, the mean and standard deviation (SD) are calculated as representative indicators to describe the overall gait characteristics and variability of the population.

[0110] The angular velocity gait characteristics include at least one of the following: mean absolute angular velocity, angular change, and rotation.

[0111] Specifically, angular velocity features capture the motion characteristics of gait, and the triaxial angular velocities can be measured using the AX6.

[0112] Mean absolute angular velocity refers to the average absolute angular velocity over a period of time. The formula for its calculation is: Among them, v t It is the angular velocity at time t within the duration D of a specific signal phase.

[0113] Angle change refers to the cumulative change in angle over time, and the calculation formula is:

[0114] Rotation refers to the total amount of rotation within a certain period, that is, the absolute value of the change in angle.

[0115] The signal characteristics include at least one of the following: standard deviation, coefficient of variation, sweep factor, mean absolute change, mean crossover rate, skewness, kurtosis, absolute energy, signal entropy, root mean square of the signal, and signal range.

[0116] Specifically, signal features describe the characteristics of signals generated during the gait cycle, achieved by extracting representative indicators from each signal phase. For a signal window S = {s1, s2, ..., sT}, where s t Here is the signal value at time t, and T is the number of target signal windows. Calculate the following characteristics:

[0117] Standard deviation is a measure of the variation or dispersion of a set of values. It quantifies the average distance between each data point and the mean of the dataset. For a signal S with T samples, the standard deviation σ is... S The calculation is as follows:

[0118]

[0119] Where, μ S It is the mean of the signal S.

[0120] Coefficient of Variation (CV): The coefficient of variation (CV) is a standardized discrete measure of a probability distribution or frequency distribution. It is defined as the standard deviation σ. S With mean μ S The ratio:

[0121]

[0122] Where, σ S It is the standard deviation, μ S It is the mean of the signal S.

[0123] Clearance Factor: The clearance factor describes the importance of impulsive or sudden interference in a signal. It is defined as the ratio of the maximum absolute signal value to the square root of the absolute signal's root mean square.

[0124]

[0125] Among them, S r The calculation formula is:

[0126] Mean of Absolute Change: The mean of absolute change measures the average magnitude of the change among consecutive samples in a signal. It is calculated as follows:

[0127]

[0128] Among them, s t It is the signal value at time t.

[0129] Mean Cross Rate: The mean cross rate is the average rate at which a signal crosses its mean. It measures the frequency of sign changes in a quantized signal. The mean cross rate is defined as:

[0130]

[0131] Where, n crossing It is the number of times the signal crosses its mean, and T is the number of target signal windows.

[0132] Skewness: Skewness is a measure of the asymmetry of the probability distribution of a real-valued random variable about its mean. For a signal S, skewness is calculated as follows:

[0133] Where μ S It is the mean of the signal S.

[0134] Kurtosis: Kurtosis is a measure of the "tailing" of the probability distribution of a real-valued random variable, describing the shape of the distribution, particularly the height and sharpness of the peaks in relation to the standard deviation. For a signal S, kurtosis is calculated using the following formula:

[0135]

[0136] Absolute Energy: Absolute energy is a measure describing the overall amplitude and vibration of a signal. It is calculated as the sum of squares of the signal values.

[0137]

[0138] Among them, s tIt is the signal value at time t, where T is the number of target signal windows.

[0139] Signal entropy: Signal entropy is a measure of the uncertainty or randomness of a signal, representing the amount of information contained in the signal. For a normalized signal S, the signal entropy H(S) is calculated as:

[0140]

[0141] Where p(s) t ) represents each value s in signal S. t The probability of occurrence, where T is the number of distinct values ​​in the signal.

[0142] Root Mean Square (RMS): The RMS value provides a measure of the signal amplitude.

[0143]

[0144] Signal Range: The signal range quantifies the span between the maximum and minimum values ​​of a signal, providing an indication of signal amplitude variations. The calculation formula is as follows:

[0145] Range(S) = max(S) - min(S).

[0146] This application's embodiments combine data from multiple sensors to achieve efficient and low-cost gait analysis, making it suitable for large-scale population screening.

[0147] In this embodiment of the application, the gait analysis feature matrix further includes the mean and standard deviation of each feature obtained based on the acceleration gait feature, the angular velocity gait feature and the signal feature.

[0148] Specifically, a total of 28 acceleration gait features can be extracted from the triaxial accelerometer. Since the gait cycle includes four phases, a total of 72 angular velocity gait features are extracted from the triaxial gyroscope. Thus, a total of 264 signal features are extracted from the triaxial accelerometer and gyroscope.

[0149] The embodiments of this application can automatically extract angular velocity and multiple signal features, providing more multi-dimensional information for detecting gait abnormalities and neurodegenerative diseases, especially in identifying motor incoordination and abnormal gait in the early stages.

[0150] In this embodiment of the application, after acquiring the raw acceleration signals collected by a triaxial accelerometer during the same time period, the method further includes:

[0151] The signal vector amplitude is calculated based on the original acceleration signal, and the signal vector amplitude is...

[0152] The signal vector amplitude is divided into several segmented windows using a sliding window method;

[0153] Convert several segmented windows into a low-dimensional space;

[0154] Obtain the trained Gaussian mixture model, and use the Gaussian mixture model to cluster several segmented windows in the low-dimensional space to obtain the clustering probability corresponding to the amplitude of the signal vector.

[0155] Specifically, this application embodiment uses a self-trained model to reflect an individual's exercise intensity. The system clusters gait data using extracted gait features, especially key indicators such as signal vector magnitude (SVM), combined with a Gaussian mixture model (GMM). The exercise intensity model can not only identify routine movements such as walking and standing, but also detect high-intensity exercise behaviors such as running and climbing stairs. This model has been optimized based on a large amount of experimental data and can accurately assess an individual's exercise intensity and daily activity level in a free living environment, thereby providing reliable data support for early disease prevention, rehabilitation assessment, and personalized health monitoring.

[0156] Data from sixty testers wearing the AX6 device for seven days was selected from the database. Data from each participant was randomly extracted and then combined.

[0157] Signal vector magnitude (SVM): The signal vector magnitude (SVM) is calculated to assess the intensity of motion. The formula for calculating SVM is:

[0158]

[0159] Where a x a y and a z These are the accelerations measured along the x-axis, y-axis, and z-axis at time t, respectively.

[0160] Segmentation of the signal vector amplitude is a key preprocessing technique used to divide accelerometer data into continuous segments (windows). A sliding window method is employed to divide the large data stream into smaller windows to more effectively capture activity patterns. For example, a window of 100 samples, corresponding to one second of accelerometer data, is used for feature extraction because it adequately captures the cycles of daily activity. To reduce information loss at window edges, a 50% overlap between adjacent windows is considered.

[0161] After segmentation, Principal Component Analysis (PCA) is performed. Specifically, PCA is used to reduce the dimensionality of the data by transforming the original data into a lower-dimensional space and minimizing information loss. For example, by randomly selecting a 15% sliding window and applying PCA to calculate the cumulative proportions of the principal components, it was found that the top 11 principal components can explain 95% of the data variation.

[0162] After dimensionality reduction, a Gaussian Mixture Model (GMM) is used to cluster different SVMs. GMM is a probabilistic model used to represent multiple Gaussian distributions, commonly used for clustering. Assuming the data comes from a linear combination of multiple Gaussian distributions, its probability density function can be expressed as:

[0163]

[0164] Where K is the number of Gaussian components, π k It is the mixing coefficient of the k-th component, satisfying It is the k-th Gaussian distribution with a mean of μ. k The covariance matrix is ​​Σ k .

[0165] Estimate the mean μ of each model from the training data. k The covariance matrix Σ k The following clustering probabilities are obtained, such as Figure 5 As shown. Category four includes movements such as jogging, squatting, and climbing stairs. Category five includes movements such as walking. Category six includes static and sitting movements. Category eight includes movements such as waist rotation.

[0166] The embodiments of this application, through self-training exercise intensity models, can accurately identify daily activities of different intensities, such as walking, running, and standing, and perform stably in free living environments.

[0167] For new test objects, follow the same data processing procedure:

[0168] First, the signal vector amplitude is segmented into a sliding window to obtain a new matrix.

[0169] Second, principal component analysis is then applied to the test data. The same principal component matrix V as the training data is used. m The dimensionality-reduced test data is obtained: X′ ts =X ts V m ;

[0170] Third, based on the Gaussian mixture model trained on the original data, calculate y for each test window. i The probability of belonging to each category k.

[0171] The formula for calculating probability is:

[0172]

[0173] Calculate the posterior probability γ for each category k. ik Then, the test sample y i Assign it to the class with the highest posterior probability, i.e., argmax. k γ ik ;

[0174] The proportion of each cluster in the test sample:

[0175] Clustering was performed on the data of a randomly selected participant, and the results are as follows: Figure 6 and Figure 7 As shown in the figure, in this test, the proportion of category four was 2.37%, category five was 17.08%, while category six accounted for a high 79.64%. These results indicate that participants mainly exhibited activities such as rest and walking during the test, especially category six, which showed a high proportion of rest. These data provide important basis for subsequent motion analysis and individual behavior pattern recognition.

[0176] In addition to gait analysis, this invention's platform can achieve more comprehensive physical activity monitoring and health assessment by introducing other sensors (such as heart rate sensors or ECG devices). Beyond early diagnosis of neurodegenerative diseases, this invention can be extended to scenarios such as rehabilitation training assessment, gait analysis during athlete training, and fall risk assessment. Specifically, it can be expanded in the following directions:

[0177] First, multidimensional data fusion: In addition to gait data, more physiological data (such as heart rate, blood oxygen saturation, etc.) can be integrated for joint analysis in the future, thereby providing patients with a more comprehensive individual health assessment and disease prevention;

[0178] Second, personalized health management: Personalized health management programs can be further developed, using long-term gait data to predict disease progression or health risks, and customized rehabilitation training plans or exercise recommendations based on individual differences.

[0179] Third, remote health monitoring: By connecting the platform with mobile devices or cloud services, remote health monitoring can be achieved, providing real-time health data feedback to help doctors or nurses intervene in a timely manner and improve patients' quality of life.

[0180] Fourth, upgrades to smart wearable devices: By combining artificial intelligence technology, the platform can be integrated into smarter wearable devices to achieve automatic data collection, analysis, and feedback, and provide real-time health advice.

[0181] This application's embodiments expand upon traditional gait analysis, which relies solely on triaxial accelerometer data, by introducing angular velocity and signal features. This feature fusion provides more comprehensive motion information, resulting in richer feature extraction and a more accurate reflection of gait dynamics. Simultaneously, Gaussian mixture models are used for cluster analysis of motion intensity, effectively identifying and classifying different motion states. Compared to traditional clustering techniques, Gaussian mixture models offer greater flexibility and accuracy.

[0182] Furthermore, by combining the function of calculating common exercise intensity indicators, this platform can more comprehensively assess users' exercise performance and health status, thereby enhancing the practicality of the analysis and enabling the platform to provide personalized exercise suggestions, thus improving users' health management experience.

[0183] In one embodiment, such as Figure 8 As shown, based on the above-mentioned gait feature extraction and analysis method, the present invention also provides a gait feature extraction and analysis device, comprising:

[0184] The acquisition module 100 is used to acquire the raw acceleration signal collected by the triaxial accelerometer and the raw angular velocity signal collected by the triaxial gyroscope during the same time period. The triaxial accelerometer and the triaxial gyroscope are used to be placed on a predetermined body part.

[0185] The cutting module 200 is used to cut the original acceleration signal and the original angular velocity signal into several signal windows respectively, and to use the signal window in the gait segment as the target signal window.

[0186] Extraction module 300 is used to extract acceleration gait features, angular velocity gait features, and signal features based on the target signal window;

[0187] The generation module 400 is used to generate a gait analysis feature matrix based on the acceleration gait features, angular velocity gait features, and signal features.

[0188] It should be noted that the foregoing explanation of the gait feature extraction and analysis method embodiment also applies to the gait feature extraction and analysis device of this embodiment, and will not be repeated here.

[0189] The gait feature extraction and analysis device proposed in this application acquires raw acceleration signals collected by a triaxial accelerometer and raw angular velocity signals collected by a triaxial gyroscope during the same time period. The triaxial accelerometer and triaxial gyroscope are placed on predetermined body parts. The raw acceleration and angular velocity signals are respectively divided into several signal windows, and the signal windows in the gait segment are used as target signal windows. Acceleration gait features, angular velocity gait features, and signal features are extracted based on the target signal windows. A gait analysis feature matrix is ​​generated based on the acceleration gait features, angular velocity gait features, and signal features. Multidimensional feature data is obtained by analyzing data acquired by low-cost acquisition equipment, achieving efficient and low-cost gait analysis and improving the accuracy of gait analysis.

[0190] Figure 9 A schematic diagram of the structure of a terminal provided in an embodiment of this application. The terminal may include:

[0191] The memory 501, the processor 502, and the computer program stored on the memory 501 and capable of running on the processor 502.

[0192] When the processor 502 executes the program, it implements the gait feature extraction and analysis method provided in the above embodiments.

[0193] Furthermore, the terminal also includes:

[0194] Communication interface 503 is used for communication between memory 501 and processor 502.

[0195] The memory 501 is used to store computer programs that can run on the processor 502.

[0196] The memory 501 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0197] If the memory 501, processor 502, and communication interface 503 are implemented independently, they can be interconnected via a bus to communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, only one line is used in the diagram, but this does not imply that there is only one bus or one type of bus.

[0198] Optionally, in a specific implementation, if the memory 501, processor 502, and communication interface 503 are integrated on a single chip, then the memory 501, processor 502, and communication interface 503 can communicate with each other through an internal interface.

[0199] Processor 502 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application.

[0200] This embodiment also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described gait feature extraction and analysis method.

[0201] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0202] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0203] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0204] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can read and execute instructions from or in conjunction with such an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by or in conjunction with an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). In addition, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically by optically scanning paper or other media, then editing, interpreting or otherwise processing them as necessary, and then storing them in computer memory.

[0205] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0206] Those skilled in the art will understand that all or part of the steps of the methods described in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it includes one or a combination of the steps of the method embodiments.

[0207] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0208] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. A method for gait feature extraction and analysis, characterized in that, The method includes: The raw acceleration signal collected by a triaxial accelerometer and the raw angular velocity signal collected by a triaxial gyroscope are acquired at the same time period. The triaxial accelerometer and the triaxial gyroscope are used to be placed at a predetermined body part. The original acceleration signal and the original angular velocity signal are respectively divided into several signal windows, and the signal window in the gait segment is used as the target signal window; Acceleration gait features, angular velocity gait features, and signal features are extracted based on the target signal window; A gait analysis feature matrix is ​​generated based on the acceleration gait features, angular velocity gait features, and signal features. The original acceleration signal and the original angular velocity signal are each divided into several signal windows, and the signal window in the gait segment is used as the target signal window, including: The original acceleration signal and the original angular velocity signal are respectively divided into several signal windows; For each signal window, extract the root mean square, range, dominant frequency, spectral flatness, signal entropy, and mean crossover rate. The root mean square, range, dominant frequency, spectral flatness, signal entropy, and mean crossover rate of each signal window are input into the trained random forest classification model to obtain the output results. If the output result indicates that the signal window is in the gait segment, then the signal window is taken as the target signal window; The acceleration gait characteristics include at least one of the following: stride duration, stride length duration, swing duration, support duration, initial double support, final double support, double support, single-leg support, stride length, asymmetry of duration, asymmetry of length, walking speed, and cadence. The angular velocity gait characteristics include at least one of the following: average absolute angular velocity, angular change, and rotation. The signal characteristics include at least one of the following: standard deviation, coefficient of variation, sweep factor, mean absolute change, mean crossover rate, skewness, kurtosis, absolute energy, signal entropy, root mean square of signal, and signal range. The gait analysis feature matrix also includes the mean and standard deviation of each feature obtained from the acceleration gait feature, the angular velocity gait feature, and the signal feature.

2. The gait feature extraction and analysis method according to claim 1, characterized in that, Based on the target signal window, acceleration gait features, angular velocity gait features, and signal features are extracted, including: Identify each gait cycle and sub-stage of each gait cycle within each of the target signal windows; Acceleration gait features, angular velocity gait features, and signal features are extracted based on each gait cycle and sub-stage of each target signal window.

3. The gait feature extraction and analysis method according to claim 2, characterized in that, Identifying each gait cycle and sub-stage of each gait cycle within each of the target signal windows, including: Based on the continuous wavelet transform algorithm, the downward peak value of smooth vertical acceleration is detected for each target signal window to determine the initial contact, and the upward peak value of integrated vertical acceleration is detected for each target signal window to determine the final contact. Each gait cycle and sub-phase of the gait cycle are determined within the target signal window based on the initial and final contact. The gait cycle consists of the following sub-phases: the support phase, the swing phase, the stride phase, and the half-stride phase.

4. The gait feature extraction and analysis method according to claim 1, characterized in that, After acquiring the raw acceleration signals collected by a triaxial accelerometer during the same time period, the process also includes: Calculate the signal vector amplitude based on the original acceleration signal; The signal vector amplitude is divided into several segmented windows using a sliding window method; Convert several segmented windows into a low-dimensional space; Obtain the trained Gaussian mixture model, and use the Gaussian mixture model to cluster several segmented windows in the low-dimensional space to obtain the clustering probability corresponding to the amplitude of the signal vector.

5. A gait feature extraction and analysis device, used to implement the gait feature extraction and analysis method as described in any one of claims 1 to 4, characterized in that, The device includes: The acquisition module is used to acquire the raw acceleration signal collected by the triaxial accelerometer and the raw angular velocity signal collected by the triaxial gyroscope during the same time period. The triaxial accelerometer and the triaxial gyroscope are used to be placed on a predetermined body part. The cutting module is used to cut the original acceleration signal and the original angular velocity signal into several signal windows, and to use the signal window in the gait segment as the target signal window. The extraction module is used to extract acceleration gait features, angular velocity gait features, and signal features based on the target signal window; The generation module is used to generate a gait analysis feature matrix based on the acceleration gait features, angular velocity gait features, and signal features.

6. A terminal, characterized in that, include: The gait feature extraction and analysis program is stored in the memory and can run on the processor, wherein when the gait feature extraction and analysis program is executed by the processor, it implements the steps of the gait feature extraction and analysis method as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that can be executed to implement the steps of the gait feature extraction and analysis method as described in any one of claims 1 to 4.

Images