Muscle state analysis method, device and equipment based on electrical impedance measurement and storage medium

Abstract

The invention relates to the technical field of signal processing, and provides a muscle state analysis method and device based on electrical impedance measurement, equipment and a storage medium. The muscle state analysis method based on electrical impedance measurement comprises the following steps: acquiring electrical impedance data and myoelectricity data of a target muscle group; extracting electrical impedance characteristics based on the electrical impedance data, and extracting myoelectricity characteristics based on the myoelectricity data; fusing the electrical impedance feature and the myoelectricity feature to obtain a fused feature vector; and inputting the fusion feature vector into a pre-constructed muscle state evaluation model for analysis to obtain a muscle state. By combining the myoelectricity data and the electrical impedance data, the muscle state is accurately evaluated from multiple dimensions, and therefore the accuracy and reliability of muscle state evaluation are improved.

Description

Technical Field

[0001] The present invention relates to the field of signal processing technology, and in particular to a muscle state analysis method, device, equipment and storage medium based on electrical impedance measurement. Background Art

[0002] Electromyographic data is widely used in muscle activity assessment and health monitoring, but traditional electromyographic data analysis methods usually use electrodes to directly obtain muscle electrical signals. Although this method has high resolution and sensitivity, its effectiveness depends on complex signal processing technology, and its reliability and accuracy are affected when facing different skin contact conditions and external noise interference. In recent years, electrical impedance measurement technology has gradually been used to monitor the characteristics of biological tissues, and the muscle state is analyzed by evaluating the impedance changes of muscles at different frequencies. However, although impedance changes can provide more bioelectric characteristic data reflecting muscle status compared to electromyographic data, the muscle status information it provides is still not comprehensive enough, and it is difficult to accurately evaluate the true state of the muscle from multiple dimensions. Summary of the invention

[0003] The embodiments of the present application provide a muscle state analysis method, apparatus, device and storage medium based on electrical impedance measurement, which aims to accurately evaluate the state of the muscle from multiple dimensions by combining electromyographic data and electrical impedance data, thereby providing accuracy and reliability of muscle state evaluation.

[0004] In a first aspect, an embodiment of the present application provides a muscle state analysis method based on electrical impedance measurement, including: acquiring electrical impedance data and electromyography data of a target muscle group; extracting electrical impedance features based on the electrical impedance data, and extracting electromyography features based on the electromyography data; fusing the electrical impedance features and electromyography features to obtain a fused feature vector; and inputting the fused feature vector into a pre-built muscle state assessment model for analysis to obtain the muscle state.

[0005] In the second aspect, an embodiment of the present application provides a muscle state analysis device based on electrical impedance measurement, including: an acquisition module, used to acquire electrical impedance data and electromyography data of a target muscle group; an extraction module, used to extract electrical impedance features based on the electrical impedance data, and to extract electromyography features based on the electromyography data; a fusion module, used to fuse the electrical impedance features and the electromyography features to obtain a fused feature vector; and an analysis module, used to input the fused feature vector into a pre-built muscle state assessment model for analysis to obtain the muscle state.

[0006] In a third aspect, an embodiment of the present application provides an electronic device, including:

[0007] Memory and processing modules;

[0008] The memory is used to store computer programs;

[0009] The processing module is used to execute the computer program and implement the steps of the muscle state analysis method based on electrical impedance measurement as described in the first aspect above when executing the computer program.

[0010] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program;

[0011] When the computer program is executed by one or more processing modules, the one or more processing modules execute the steps of the muscle state analysis method based on electrical impedance measurement in the first aspect as described above.

[0012] The embodiments of the present application provide a muscle state analysis method, apparatus, device and storage medium based on electrical impedance measurement, wherein the muscle state analysis method based on electrical impedance measurement includes: acquiring electrical impedance data and electromyography data of a target muscle group; extracting electrical impedance features based on the electrical impedance data, and extracting electromyography features based on the electromyography data; fusing the electrical impedance features and electromyography features to obtain a fused feature vector; and inputting the fused feature vector into a pre-constructed muscle state assessment model for analysis to obtain the muscle state.

[0013] By combining electromyography data and electrical impedance data, it aims to accurately evaluate the state of muscles from multiple dimensions, thereby providing accuracy and reliability in muscle state assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying any creative work.

[0015] Figure 1 A schematic diagram of a flow chart of a muscle state analysis method based on electrical impedance measurement provided in an embodiment of the present application;

[0016] Figure 2 A schematic diagram of the structure of a muscle state analysis device based on electrical impedance measurement provided in an embodiment of the present application;

[0017] Figure 3 A schematic block diagram of a muscle state analysis device based on electrical impedance measurement provided in an embodiment of the present application. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0020] It should also be understood that the terms used in this application specification are only for the purpose of describing specific embodiments and are not intended to limit the application. As used in this application specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0021] It should be further understood that the term “and / or” used in the specification and appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0022] The technical solution provided by this application will be described in detail below with reference to the accompanying drawings.

[0023] See also Figure 1 , Figure 1 The flowchart of the muscle state analysis method based on electrical impedance measurement provided in the embodiment of the present application is shown in FIG. The muscle state analysis method based on electrical impedance measurement provided in the embodiment of the present application is implemented by a muscle state monitoring device. The embodiment of the present application does not limit the muscle state detection device in any way. Specifically, Figure 1 As shown, the muscle state analysis method based on electrical impedance measurement includes steps S101 to S104. The details are as follows:

[0024] S101: Acquire electrical impedance data and electromyographic data of the target muscle group.

[0025] Obtaining electrical impedance data of the target muscle group includes: respectively obtaining electrical impedance data of the target muscle group within a preset low frequency range and a preset high frequency range.

[0026] Within the preset low-frequency range (usually tens of hertz to hundreds of hertz), the current mainly flows through the extracellular fluid in the muscle tissue, and rarely passes through the cell membrane. Therefore, the low-frequency electrical impedance data mainly reflects the conductivity of the extracellular fluid. By measuring the electrical impedance data within the preset low-frequency range, information related to the characteristics of the extracellular space, such as the external structure of the tissue and changes in the extracellular fluid, can be obtained, which helps to understand the overall health of the muscle tissue. Within the preset high-frequency range (usually hundreds of hertz to several megahertz), the current can partially penetrate the cell membrane and enter the intracellular fluid. High-frequency electrical impedance data reflects the conductivity characteristics of the fluid inside and outside the cell, as well as the resistance and capacitance characteristics of the cell membrane, which will manifest differently under muscle fatigue, injury and healthy conditions.

[0027] By measuring the electrical impedance data at low and high frequencies respectively, the difference in electrical impedance characteristics outside and inside muscle cells can be obtained, which helps to more comprehensively evaluate changes in muscle status, such as hydration levels, muscle fatigue, and changes in tissue structure. This multi-frequency electrical impedance measurement method can more accurately reflect the health status and state of muscles.

[0028] S102: extracting electrical impedance features based on the electrical impedance data, and extracting electromyographic features based on the electromyographic data.

[0029] Extracting electrical impedance features based on electrical impedance data and extracting electromyographic features based on electromyographic data include: extracting impedance values, phase angles and relative impedance change rates of electrical impedance data within a preset low-frequency range and a preset high-frequency range, respectively; extracting average potential, amplitude and spectrum density of electromyographic data.

[0030] Specifically, the impedance value represents the total impedance of the target muscle group to the current at different frequencies. The low frequency mainly reflects the extracellular characteristics, while the high frequency contains intracellular information. Extracting the impedance values ​​at these two frequencies separately helps to distinguish the internal and external structural characteristics of muscle tissue. Phase Angle represents the phase difference between current and voltage, which is related to the capacitance characteristics of the cell membrane. The phase angle at low and high frequencies can reveal the extracellular and intracellular response characteristics to the current, respectively, and the magnitude of the phase angle value is usually related to the health of the tissue. For example, a higher phase angle may indicate better integrity of the cell membrane. The relative impedance change rate is the ratio of the impedance value to the frequency. It indicates the degree of change in the impedance value at low and high frequencies, which can reflect the difference in the conductivity characteristics of muscle tissue. Muscle fatigue or damage usually leads to changes in cell structure, and the relative impedance change rate can reveal these changes. Mean Voltage is the average value of the electromyographic signal, which represents the overall electrical activity level of the muscle during the measurement period. It can be used to reflect the basic activity intensity of muscle contraction and is an indicator for evaluating the degree of muscle activation. Amplitude represents the peak intensity of the electromyographic signal, that is, the maximum electrical activity level reached when the muscle contracts. Higher amplitudes usually mean greater muscle contraction strength, which is related to muscle activation strength and nerve signal transmission. Spectral density is the frequency distribution of the electromyographic signal, reflecting the energy distribution of the signal at different frequency components. By analyzing the spectral density, we can understand the electrical activity level of the muscle at different frequencies, which can usually be used to identify fatigue, because fatigue can cause changes in the frequency components of the electromyographic signal, such as the frequency center shifting to low frequencies.

[0031] By extracting the above features, we can obtain multi-dimensional information about muscle tissue structure, electrical activity level and muscles. After fusing these features, they can be used as input feature vectors and input into the muscle state assessment model to further analyze and judge the fatigue, activity or normal state of the muscles.

[0032] S103: Fusing the electrical impedance feature and the electromyographic feature to obtain a fused feature vector.

[0033] The electrical impedance feature and the electromyographic feature are fused to obtain a fused feature vector, including: a feature splicing algorithm based on weighted fusion, which weights and fuses the impedance value, phase angle, relative impedance change rate, average potential, amplitude and spectrum density to obtain a fused feature vector.

[0034] Based on the weighted fusion feature stitching algorithm, the impedance value, phase angle, relative impedance change rate, average potential, amplitude and spectrum density are weightedly fused to obtain a fused feature vector, including: normalizing the correlation between each predetermined feature and the muscle state to form an initial weight for each feature; according to the initial weight of each feature, the impedance value, phase angle, relative impedance change rate, average potential, amplitude and spectrum density are weightedly fused to obtain a fused feature vector.

[0035] For example, assume that N = 6 features, namely impedance value, phase angle, relative impedance change rate, average potential, amplitude and spectral density. In order to ensure that the influence of each feature in the fusion feature vector corresponds to the correlation with the muscle state, the weight of each feature is first determined according to the correlation between the feature and the muscle state. i ,but

[0036] Specifically, the weight determination process for each feature requires the collection of a large amount of sample data, including electrical impedance features such as impedance value, phase angle, and relative impedance change rate, and electromyographic features such as average potential, amplitude, and spectral density, as well as corresponding muscle state labels such as fatigue, normal, and active. The collected data is standardized or normalized to ensure that the numerical ranges of different features are consistent. For each feature, its correlation with the muscle state can be calculated using algorithms such as the Pearson correlation coefficient, point biserial correlation coefficient, or chi-square test or mutual information. The above algorithm can assign a correlation score to each feature, and since the final weight α i The sum of must be equal to 1, therefore, the correlation score is normalized to obtain the initial weight of each feature. Specifically, this embodiment does not elaborate on the correlation and weight calculation process.

[0037] For each feature, assume that x i (e.g., impedance value, phase angle, etc.), according to the preset weight α i Perform weighted calculation to obtain each dimension of the final fused feature vector. The formula is:

[0038] f fusuon =[α1.x1,α2.x2,…,α N .x N ]

[0039] where x1, x2, ..., x N They represent impedance value, phase angle, relative impedance change rate, average potential, amplitude and spectrum density, α1, α2, ..., α N are the weights of these features respectively.

[0040] The fused feature vector F is a weighted combination of features and can be expressed as:

[0041]

[0042] The fused feature vector F contains the weighted fused electrical impedance features and electromyographic features, representing the multi-dimensional information of the muscle state. Through this fused feature vector, the structural characteristics, neural activation and fatigue state of the muscle can be comprehensively reflected from different dimensions.

[0043] S104: Input the fused feature vector into a pre-built muscle state assessment model for analysis to obtain the muscle state.

[0044] The fused feature vector is input into a pre-built muscle state assessment model for analysis to obtain the muscle state, including: the fused feature vector is input into the pre-built muscle state assessment model, the input layer of the muscle state assessment model receives all elements of the fused feature vector, and each element of the fused feature vector corresponds to a neuron in the input layer; in the muscle state assessment model, the data of the input layer is transmitted to the hidden layer for mapping and activation function processing to extract deeper feature information; in the output layer of the muscle state assessment model, the deeper feature information is analyzed to obtain the muscle state.

[0045] It should be noted that in the embodiment of the present application, the muscle state assessment model is a classification model given to the neural network. The input layer of the model receives all elements of the fused feature vector, and each element corresponds to the input of a mental unit. In the neural network, the data of the input layer is passed to the hidden layer and processed by a series of mapping and activation functions to extract deeper feature information. Assume that the input value z of the Kth neuron in the hidden layer is k It is expressed as:

[0046]

[0047] Among them, ωk i represents the weight between the kth hidden layer neuron and the i-th input feature, fi is the i-th element of the fused feature vector, and b k is the bias term of the kth hidden layer neuron.

[0048] After the input value passes through the activation function, the output of the hidden layer is: α k = σ (z k )

[0049] Among them, σ is the activation function, which extracts deeper feature information through nonlinear transformation.

[0050] In the output layer of the muscle state assessment model, deeper feature information is analyzed to obtain the muscle state, including: in the output layer of the muscle state assessment model, deeper feature information is linearly combined to obtain the neuron input value of the output layer; the neuron input value of the output layer is calculated using a classification function to obtain the predicted probability of the muscle state.

[0051] Specifically, in the output layer of the muscle state assessment model, the deeper feature information is linearly combined to obtain the neuron input value of the output layer:

[0052]

[0053] Among them, vjk represents the weight between the j-th output layer neuron and the k-th hidden layer neuron, ak is the output of the k-th neuron in the hidden layer, and cj is the bias term of the j-th output layer neuron.

[0054] The value of the output layer is calculated by the classification function to obtain the predicted probability of the muscle state, which is expressed as

[0055] in, is the probability of the jth category, and C is the total number of categories.

[0056] The final muscle state classification result will be the category with the highest probability value.

[0057] It should be noted that the output of the model (i.e., the probability value of each category) will be converted into muscle state data to identify the current state of the muscle. For example, if the probability value corresponding to the category "fatigue" is calculated to be the highest, then the state label representing "fatigue" will be output; if the probability value corresponding to the category "active" is calculated to be the highest, then the state label representing "active" will be output.

[0058] Through the above analysis, it can be known that the muscle state analysis method based on electrical impedance measurement provided by the embodiment of the present application includes: obtaining electrical impedance data and electromyographic data of the target muscle group; extracting electrical impedance features based on electrical impedance data, and extracting electromyographic features based on electromyographic data; fusing electrical impedance features and electromyographic features to obtain a fused feature vector; inputting the fused feature vector into a pre-built muscle state assessment model for analysis to obtain the muscle state. By combining electromyographic data and electrical impedance data, it is intended to accurately assess the state of the muscle from multiple dimensions, thereby providing the accuracy and reliability of muscle state assessment.

[0059] See also Figure 2 As shown, Figure 2 This is a schematic diagram of the structure of a muscle state analysis device based on electrical impedance measurement provided in an embodiment of the present application. Figure 2 It can be seen that the muscle state analysis device 20 based on electrical impedance measurement provided in the embodiment of the present application includes:

[0060] An acquisition module 201 is used to acquire electrical impedance data and electromyographic data of a target muscle group;

[0061] An extraction module 202, configured to extract electrical impedance features based on electrical impedance data, and to extract electromyographic features based on electromyographic data;

[0062] A fusion module 203 is used to fuse the electrical impedance feature and the electromyographic feature to obtain a fusion feature vector;

[0063] The analysis module 204 is used to input the fused feature vector into a pre-built muscle state assessment model for analysis to obtain the muscle state.

[0064] In one embodiment, the acquisition module 201 is specifically used for:

[0065] The electrical impedance data of the target muscle group within a preset low-frequency range and a preset high-frequency range are obtained respectively.

[0066] In one embodiment, the extraction module 202 includes:

[0067] A first extraction unit is used to extract the impedance value, phase angle and relative impedance change rate of the electrical impedance data within a preset low frequency range and a preset high frequency range respectively;

[0068] The second extraction unit is used to extract the average potential, amplitude and spectrum density of the electromyographic data.

[0069] In one embodiment, the fusion module 203 is specifically configured to:

[0070] Based on the weighted fusion feature stitching algorithm, the impedance value, phase angle, relative impedance change rate, average potential, amplitude and spectrum density are weightedly fused to obtain a fused feature vector.

[0071] In one embodiment, the fusion module 203 includes:

[0072] A normalization unit, used to normalize the correlation between each predetermined feature and the muscle state to form an initial weight of each feature;

[0073] The fusion unit is used to perform weighted fusion of the impedance value, phase angle, relative impedance change rate, average potential, amplitude and spectrum density according to the initial weight of each feature to obtain a fused feature vector.

[0074] In one embodiment, the analysis module 204 includes:

[0075] An input unit, used for inputting the fused feature vector into a pre-built muscle state assessment model, wherein the input layer of the muscle state assessment model receives all elements of the fused feature vector, and each element of the fused feature vector corresponds to a neuron of the input layer;

[0076] A processing unit is used to transfer the data of the input layer to the hidden layer in the muscle state assessment model for processing through mapping and activation functions to extract deeper feature information;

[0077] The analysis unit is used to analyze deeper feature information in the output layer of the muscle state evaluation model to obtain the muscle state.

[0078] In one embodiment, the analysis unit comprises:

[0079] The combination subunit is used to linearly combine the feature information of the deeper layer in the output layer of the muscle state evaluation model to obtain the neuron input value of the output layer;

[0080] The calculation subunit is used to calculate the neuron input value of the output layer using the classification function to obtain the predicted probability of the muscle state.

[0081] It should be noted that the specific implementation process of the above modules or units can refer to the specific implementation process of each step in the previous method embodiment, which will not be repeated here.

[0082] See also Figure 3 As shown, Figure 3 A schematic block diagram of a muscle state analysis device based on electrical impedance measurement provided in an embodiment of the present application.

[0083] Exemplarily, the muscle state analysis device 30 based on electrical impedance measurement includes a processing module 301 and a memory 302 .

[0084] Exemplarily, the processing module 301 and the memory 302 are connected via a bus 303 , and the bus 303 is, for example, an I2C (Inter-integrated Circuit) bus.

[0085] Specifically, the processing module 301 may be a micro-controller unit (MCU), a central processing unit (CPU) or a digital signal processing module (DSP).

[0086] Specifically, the memory 302 may be a Flash chip, a read-only memory (ROM) disk, an optical disk, a USB flash drive, or a mobile hard disk.

[0087] The processing module 301 is used to run the computer program stored in the memory 302, and implement the steps of the above-mentioned muscle state analysis method based on electrical impedance measurement when executing the computer program.

[0088] Exemplarily, the processing module 301 is used to run the computer program stored in the memory 302, and implement the following steps when executing the computer program:

[0089] Obtain electrical impedance data and electromyographic data of target muscle groups;

[0090] Extract electrical impedance features based on electrical impedance data, and extract electromyographic features based on electromyographic data;

[0091] Fusing the electrical impedance feature and the electromyographic feature to obtain a fused feature vector;

[0092] The fused feature vector is input into the pre-built muscle state assessment model for analysis to obtain the muscle state.

[0093] In one embodiment, obtaining electrical impedance data of a target muscle group includes:

[0094] The electrical impedance data of the target muscle group within a preset low-frequency range and a preset high-frequency range are obtained respectively.

[0095] In one embodiment, extracting electrical impedance features based on electrical impedance data and extracting electromyographic features based on electromyographic data include:

[0096] Extracting impedance values, phase angles, and relative impedance change rates of the electrical impedance data within a preset low-frequency range and a preset high-frequency range, respectively;

[0097] Extract the average potential, amplitude, and spectral density of the EMG data.

[0098] In one embodiment, the electrical impedance feature and the electromyographic feature are fused to obtain a fused feature vector, including:

[0099] Based on the weighted fusion feature stitching algorithm, the impedance value, phase angle, relative impedance change rate, average potential, amplitude and spectrum density are weightedly fused to obtain a fused feature vector.

[0100] In one embodiment, based on a weighted fusion feature stitching algorithm, the impedance value, phase angle, relative impedance change rate, average potential, amplitude, and spectrum density are weightedly fused to obtain a fused feature vector, including:

[0101] Normalize the correlation between each predetermined feature and the muscle state to form the initial weight of each feature;

[0102] According to the initial weight of each feature, the impedance value, phase angle, relative impedance change rate, average potential, amplitude and spectrum density are weighted fused to obtain a fused feature vector.

[0103] In one embodiment, the fused feature vector is input into a pre-built muscle state assessment model for analysis to obtain the muscle state, including:

[0104] Inputting the fused feature vector into a pre-built muscle state assessment model, the input layer of the muscle state assessment model receives all elements of the fused feature vector, and each element of the fused feature vector corresponds to a neuron of the input layer;

[0105] In the muscle status assessment model, the data of the input layer is passed to the hidden layer and processed by mapping and activation functions to extract deeper feature information;

[0106] In the output layer of the muscle state assessment model, deeper feature information is analyzed to obtain the muscle state.

[0107] In one embodiment, at the output layer of the muscle state assessment model, deeper feature information is analyzed to obtain muscle states, including:

[0108] In the output layer of the muscle state assessment model, the feature information of the deeper layer is linearly combined to obtain the neuron input value of the output layer;

[0109] The neuron input values ​​of the output layer are calculated using the classification function to obtain the predicted probability of the muscle state.

[0110] The specific principles and implementation methods of the muscle state analysis device based on electrical impedance measurement provided in the embodiment of the present application are similar to the muscle state analysis method based on electrical impedance measurement in the aforementioned embodiment, and will not be repeated here.

[0111] The embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by the processing module, the processing module implements the following steps:

[0112] Obtain electrical impedance data and electromyographic data of target muscle groups;

[0113] Extract electrical impedance features based on electrical impedance data, and extract electromyographic features based on electromyographic data;

[0114] Fusing the electrical impedance feature and the electromyographic feature to obtain a fused feature vector;

[0115] The fused feature vector is input into the pre-built muscle state assessment model for analysis to obtain the muscle state.

[0116] In one embodiment, obtaining electrical impedance data of a target muscle group includes:

[0117] The electrical impedance data of the target muscle group within a preset low-frequency range and a preset high-frequency range are obtained respectively.

[0118] In one embodiment, extracting electrical impedance features based on electrical impedance data and extracting electromyographic features based on electromyographic data include:

[0119] Extracting impedance values, phase angles, and relative impedance change rates of the electrical impedance data within a preset low-frequency range and a preset high-frequency range, respectively;

[0120] Extract the average potential, amplitude, and spectral density of the EMG data.

[0121] In one embodiment, the electrical impedance feature and the electromyographic feature are fused to obtain a fused feature vector, including:

[0122] Based on the weighted fusion feature stitching algorithm, the impedance value, phase angle, relative impedance change rate, average potential, amplitude and spectrum density are weightedly fused to obtain a fused feature vector.

[0123] In one embodiment, based on a weighted fusion feature stitching algorithm, the impedance value, phase angle, relative impedance change rate, average potential, amplitude, and spectrum density are weightedly fused to obtain a fused feature vector, including:

[0124] Normalize the correlation between each predetermined feature and the muscle state to form the initial weight of each feature;

[0125] According to the initial weight of each feature, the impedance value, phase angle, relative impedance change rate, average potential, amplitude and spectrum density are weighted fused to obtain a fused feature vector.

[0126] In one embodiment, the fused feature vector is input into a pre-built muscle state assessment model for analysis to obtain the muscle state, including:

[0127] Inputting the fused feature vector into a pre-built muscle state assessment model, the input layer of the muscle state assessment model receives all elements of the fused feature vector, and each element of the fused feature vector corresponds to a neuron of the input layer;

[0128] In the muscle status assessment model, the data of the input layer is passed to the hidden layer and processed by mapping and activation functions to extract deeper feature information;

[0129] In the output layer of the muscle state assessment model, deeper feature information is analyzed to obtain the muscle state.

[0130] In one embodiment, at the output layer of the muscle state assessment model, deeper feature information is analyzed to obtain muscle states, including:

[0131] In the output layer of the muscle state assessment model, the feature information of the deeper layer is linearly combined to obtain the neuron input value of the output layer;

[0132] The neuron input values ​​of the output layer are calculated using the classification function to obtain the predicted probability of the muscle state.

[0133] Wherein, the computer-readable storage medium can be an internal storage unit of the muscle state analysis device based on electrical impedance measurement in the aforementioned embodiment, such as a hard disk or memory of the muscle state analysis device based on electrical impedance measurement. The computer-readable storage medium can also be an external storage device of the muscle state analysis device based on electrical impedance measurement, such as a plug-in hard disk equipped on the muscle state analysis device based on electrical impedance measurement, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc.

[0134] It should be understood that the terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application.

[0135] It will also be understood that the term "and / or" used in this application and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0136] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A muscle state analysis method based on electrical impedance measurement, characterized in that: include: Obtain electrical impedance data and electromyographic data of target muscle groups; Extracting electrical impedance features based on the electrical impedance data, and extracting electromyographic features based on the electromyographic data; Fusing the electrical impedance feature and the myoelectric feature to obtain a fused feature vector; The fused feature vector is input into a pre-built muscle state assessment model for analysis to obtain the muscle state.

2. The muscle state analysis method based on electrical impedance measurement according to claim 1, characterized in that: The step of obtaining electrical impedance data of the target muscle group comprises: The electrical impedance data of the target muscle group within a preset low-frequency range and a preset high-frequency range are obtained respectively.

3. The muscle state analysis method based on electrical impedance measurement according to claim 2, characterized in that: The step of extracting electrical impedance features based on the electrical impedance data and extracting electromyographic features based on the electromyographic data comprises: Extracting the impedance value, phase angle and relative impedance change rate of the electrical impedance data within the preset low-frequency range and the preset high-frequency range respectively; The average potential, amplitude and spectral density of the electromyographic data are extracted.

4. The muscle state analysis method based on electrical impedance measurement according to claim 3, characterized in that: The step of fusing the electrical impedance feature and the electromyographic feature to obtain a fused feature vector includes: Based on the weighted fusion feature stitching algorithm, the impedance value, the phase angle, the relative impedance change rate, the average potential, the amplitude and the spectrum density are weightedly fused to obtain the fused feature vector.

5. The muscle state analysis method based on electrical impedance measurement according to claim 4, characterized in that: The feature stitching algorithm based on weighted fusion performs weighted fusion on the impedance value, the phase angle, the relative impedance change rate, the average potential, the amplitude, and the spectrum density to obtain the fused feature vector, including: Normalize the correlation between each predetermined feature and the muscle state to form the initial weight of each feature; According to the initial weight of each feature, the impedance value, the phase angle, the relative impedance change rate, the average potential, the amplitude and the spectrum density are weightedly fused to obtain the fused feature vector.

6. The muscle state analysis method based on electrical impedance measurement according to claim 1, characterized in that: The step of inputting the fused feature vector into a pre-built muscle state assessment model for analysis to obtain the muscle state includes: Inputting the fused feature vector into a pre-built muscle state assessment model, wherein an input layer of the muscle state assessment model receives all elements of the fused feature vector, and each element of the fused feature vector corresponds to a neuron of the input layer; In the muscle state assessment model, the data of the input layer is transmitted to the hidden layer and processed by mapping and activation functions to extract deeper feature information; At the output layer of the muscle state assessment model, the deeper feature information is analyzed to obtain the muscle state data.

7. The muscle state analysis method based on electrical impedance measurement according to claim 6, characterized in that: In the output layer of the muscle state assessment model, the deeper feature information is analyzed to obtain the muscle state data, including: In the output layer of the muscle state assessment model, linearly combining the feature information of the deeper layer to obtain a neuron input value of the output layer; The neuron input values ​​of the output layer are calculated using the classification function to obtain the predicted probability of the muscle state.

8. A muscle state analysis device based on electrical impedance measurement, characterized in that: include: An acquisition module, used to acquire electrical impedance data and electromyographic data of a target muscle group; An extraction module, used for extracting electrical impedance features based on the electrical impedance data, and extracting electromyographic features based on the electromyographic data; A fusion module, used for fusing the electrical impedance feature and the myoelectric feature to obtain a fusion feature vector; The analysis module is used to input the fused feature vector into a pre-built muscle state assessment model for analysis to obtain the muscle state.

9. An electronic device, characterized in that: include: Memory and processing modules; The memory is used to store computer programs; The processing module is used to execute the computer program and implement the steps of the muscle state analysis method based on electrical impedance measurement as described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer readable storage medium stores a computer program; When the computer program is executed by one or more processing modules, the one or more processing modules execute the steps of the muscle state analysis method based on electrical impedance measurement as described in any one of claims 1 to 7.

Images