An electrical stimulation signal generation method, device, and storage medium

Abstract

This invention provides a method, device, and storage medium for generating electrical stimulation signals, comprising: sending an instruction to a target user to perform a specified action; and acquiring an electromyographic (EMG) signal collected by an electromyography (EMG) acquisition device in the i-th acquisition cycle. i If EMG i The deviation Δd between the signal and the corresponding baseline electromyographic signal i <d1, exit the current control program; if △d i ≥d1, if EMG i and SEMG f i The deviation between them is greater than or equal to d2, using EMG i Update the current stimulus signal generation model, and obtain the electrical stimulation signal (ETS) for the i-th stimulation cycle based on the current stimulus signal query information and the current stimulus signal generation model. i ; ETS i The signal is sent to the electromyography (EMG) stimulator; the signal processing continues. This invention generates a corresponding electrical stimulation signal based on the EMG signals acquired in each acquisition cycle, thereby making the generated electrical stimulation signal more accurate.

Description

Technical Field

[0001] This invention relates to the field of signal processing, and in particular to a method, apparatus and storage medium for generating electrical stimulation signals. Background Technology

[0002] Currently, surface electromyography (SEMG) signals are widely used to regulate muscle function. In some applications, SEMG signals collected by an EEMG acquisition device placed over a user's muscle area are processed to analyze whether the muscle function is normal. In other applications, an EEMG stimulator is placed over a muscle area where the user's function is abnormal to stimulate that area and restore its normal function. However, these two applications are currently implemented separately, and the stimulation signals used to stimulate the muscle area are generally set based on the operator's experience. Furthermore, the same stimulation signal is used in each stimulation cycle, leading to long recovery periods and inaccurate stimulation signals, resulting in a poor user experience. Summary of the Invention

[0003] To address the aforementioned technical problems, the technical solution adopted by this invention is as follows:

[0004] According to a first aspect of the present invention, a method for generating an electrical stimulation signal is provided, the method comprising the following steps:

[0005] S100, send an instruction to the target user to perform a specified action; the specified muscle group of the target user is in a first state and is equipped with n electromyography (EMG) collectors and m EMG stimulators, where n≥m≥1.

[0006] S200, acquire electromyographic (EMG) signals from n EMG acquisition devices during the i-th acquisition cycle. i ={EMG i1 EMG i2 ..., EMG ir ..., EMG in};EMG ir Let r be the electromyographic signal acquired by the r-th electromyographic acquisition device in the i-th acquisition cycle, where r ranges from 1 to n and i is initially 1.

[0007] S300, if EMG ir The deviation Δd between the signal and the corresponding baseline electromyographic signal i <d1, exit the current control program; if △d i ≥d1, and if EMG ir and SEMG f irThe deviation between them is greater than or equal to d2, using EMG i Update the current stimulus signal generation model and execute S400; where d1 is the first setpoint deviation and d2 is the second setpoint deviation; SEMG f ir The electromyographic (EMG) signals are obtained based on the previous stimulation signal query information and the corresponding stimulation signal generation model. The stimulation signal query information includes at least: the target user's basic information, the EMG signals collected by the EMG stimulators on the target user's specified muscle groups in the previous acquisition cycle, and the baseline EMG signals corresponding to the EMG signals collected in the previous acquisition cycle. The target user's basic information includes: the target user's attribute parameter information, the muscle group identifier of the target user's specified muscle groups, the action identifier of the specified action performed by the target user, the number of EMG collectors on the target user's specified muscle groups, the location information of the EMG collectors on the target user's specified muscle groups, the number of EMG stimulators on the target user's specified muscle groups, and the location information of the EMG stimulators on the target user's specified muscle groups.

[0008] S400, based on the current stimulus signal query information and the current stimulus signal generation model, obtain the electrical stimulation signal (ETS) for the i-th stimulation cycle. i The current stimulus signal query information includes at least: basic information of the target user, EMG... ir and EMG ir The corresponding baseline electromyographic signal; execute S500.

[0009] S500, ETS i Send to the m electromyographic stimulators so that the m electromyographic stimulators output ETS in the i-th stimulation cycle. i ;Execute S600.

[0010] S600, set i = i + 1, execute S200.

[0011] A second aspect of the present invention provides a signal processing device, comprising: a housing and a processor and a memory disposed within the housing, the memory containing a computer program, the processor being configured to communicate with an electromyography (EMG) acquisition device and an EMG stimulator disposed on a designated muscle group of a target user, and the computer program being executed by the processor to implement the method described in the first aspect.

[0012] A third aspect of the present invention provides a non-transitory computer-readable storage medium storing a program or instructions that cause a computer to perform the steps of the method as described in the first aspect.

[0013] The present invention has at least the following beneficial effects:

[0014] The electrical stimulation signal generation method provided in this invention can generate corresponding electrical stimulation signals based on the electromyographic signals collected in each acquisition cycle during electromyographic stimulation, thereby making the generated electrical stimulation signals more accurate and providing a better user experience.

[0015] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating the electrical stimulation signal generation method provided in an embodiment of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of this invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0020] It should be noted that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts describe the steps as sequential processes, many of these steps can be performed in parallel, concurrently, or simultaneously. Furthermore, the order of the steps can be rearranged. A process can be terminated when its operation is complete, but it may also have additional steps not included in the figures. A process can correspond to a method, function, procedure, subroutine, subroutine, etc.

[0021] This invention provides a method for generating electrical stimulation signals, such as... Figure 1 As shown, the method may include the following steps:

[0022] S100, send an instruction to the target user to perform a specified action; the specified muscle group of the target user is in a first state and is equipped with n electromyography (EMG) collectors and m EMG stimulators, where n≥m≥1.

[0023] In this embodiment of the invention, the target user may be a user with muscle abnormalities, such as pain. The first state may be an abnormal state, that is, the specified muscle group may be a muscle group on the target user's body that has abnormalities. The specified action may be an action suitable for collecting electromyographic signals of the specified muscle group, such as functional movements of the fingers and wrists, which are determined based on the actual situation.

[0024] In this embodiment of the invention, the electromyography (EMG) acquisition device and the EMG stimulator can be existing EMG acquisition sensors. The sensors can use electrodes made of Ag / AgCl to reduce electrode noise interference. The EMG acquisition device and the EMG stimulator can be placed at designated locations in a specific muscle group. Electrodes can be attached using conductive gel or conductive patches to ensure good contact and reduce environmental noise interference.

[0025] In this embodiment of the invention, the number of electromyography (EMG) acquisition devices and EMG stimulators can be set according to actual needs. In one illustrative embodiment, the number of EMG acquisition devices can be 8, and the number of EMG stimulators can be 4.

[0026] S200, acquire electromyographic (EMG) signals from n EMG acquisition devices during the i-th acquisition cycle. i ={EMG i1 EMG i2 ..., EMG ir ..., EMG in};EMG ir Let r be the electromyographic signal acquired by the r-th electromyographic acquisition device in the i-th acquisition cycle, where r ranges from 1 to n and i is initially 1.

[0027] In this embodiment of the invention, the electromyographic signal captured by the electromyographic electrodes of the electromyographic acquisition device can be received by the receiver. The electromyographic signal is amplified by the amplification circuit and then converted by A / D conversion.

[0028] In this embodiment of the invention, the electromyographic signal is a voltage signal, and its characteristic parameters include at least amplitude and median frequency. The amplitude can be, for example, the root mean square of the electromyographic signal. The amplitude and median frequency of the electromyographic signal can be obtained by analyzing the acquired electromyographic signal using signal analysis software; the specific analysis method can be existing technology.

[0029] In this embodiment of the invention, the acquisition period can be determined based on actual needs. In one illustrative embodiment, the acquisition period can be 60s.

[0030] S300, if EMG ir The deviation between the signal and the corresponding reference electromyographic signal, i.e., the absolute value of the difference between the two, Δd ir <d1, exit the current control program; if △d ir ≥d1, and if EMG ir and SEMG f ir The deviation between them is greater than or equal to d2, using EMG ir Update the current stimulus signal generation model and execute S400; where d1 is the first setpoint deviation and d2 is the second setpoint deviation; SEMG f ir The electromyographic (EMG) signals are obtained based on the previous stimulation signal query information and the corresponding stimulation signal generation model. The stimulation signal query information includes at least: the target user's basic information, the EMG signals collected by the EMG stimulators on the target user's specified muscle groups in the previous acquisition cycle, and the baseline EMG signals corresponding to the EMG signals collected in the previous acquisition cycle. The target user's basic information includes: the target user's attribute parameter information, the muscle group identifier of the target user's specified muscle groups, the action identifier of the specified action performed by the target user, the number of EMG collectors on the target user's specified muscle groups, the location information of the EMG collectors on the target user's specified muscle groups, the number of EMG stimulators on the target user's specified muscle groups, and the location information of the EMG stimulators on the target user's specified muscle groups.

[0031] In this embodiment of the invention, the user's attribute parameter information may include at least age and gender.

[0032] In this embodiment of the invention, d1 and d2 can be set according to actual needs. d1 can be equal to d2 or they can be different.

[0033] In this embodiment of the invention, each action may have a designated identifier, and each muscle group may have a designated identifier. The specific identifier settings can be configured based on actual needs.

[0034] In this embodiment of the invention, if a specified muscle group belongs to the first type of muscle group, the reference electromyographic signal corresponding to the specified muscle group is determined in the following manner:

[0035] Acquire the electromyographic signals collected by the electromyography (EMG) acquisition device set on the symmetrical muscle group of the specified muscle group, and use them as candidate EMG signals. If the error between the candidate EMG signal and the EMG signal of the muscle group in the second state is less than d1, then the candidate EMG signal is used as the reference EMG signal corresponding to the specified muscle group. Otherwise, the EMG signal of the specified muscle group in the second state is used as the reference EMG signal of the muscle group in the second state.

[0036] If the specified muscle group belongs to the second type of muscle group, the reference electromyographic signal corresponding to the specified muscle group is the electromyographic signal when the specified muscle group is in the second state.

[0037] The first type of muscle group refers to the muscle groups located in the upper and lower limbs, while the second type refers to the muscle groups located in the head and trunk. The second state refers to the specified muscle groups being in a normal state, such as a healthy state.

[0038] In this embodiment of the invention, the electromyography (EMG) signal of a specified muscle group in a second state may include Q reference EMG signals, and the user attribute parameter information corresponding to any two reference EMG signals is different. The Q reference EMG signals are obtained based on different age ranges and genders. The Q reference EMG signals may include Q1 reference EMG signals corresponding to users with different age ranges but the first gender, and Q1 reference EMG signals corresponding to users with different age ranges but the second gender. The first gender may be male, and the second gender may be female. The Q reference EMG signals can be obtained by existing methods. For example, the h-th reference EMG signal in the Q1 reference EMG signals is obtained based on N users whose age range is h, whose gender is the first gender, and who are in a healthy state. Specifically, an EMG acquisition device is set on the muscle group corresponding to the specified muscle group of N users, and then the user is instructed to perform a specified action for a preset time. During this time, the EMG signals of each user are collected, and then the average value of the amplitude and the average value of the median frequency of the obtained N EMG signals are used as the reference EMG signals of the specified muscle group under the specified action.

[0039] In this embodiment of the invention, if the specified muscle group belongs to the first type of muscle group, and if the electromyographic signal of the symmetrical muscle group of the specified muscle group is normal, then the electromyographic signal of the symmetrical muscle group is used as the reference electromyographic signal. In this way, since the electromyographic signals are collected synchronously, the generated electrical stimulation signal can be more accurate.

[0040] Furthermore, in an embodiment of the present invention, if EMG ir and SEMG f ir If the deviation between the two values ​​is greater than or equal to d2, it indicates that the current stimulus signal generation model is not accurate enough, and EMG can be used to correct it. ir Update the current stimulus signal generation model, specifically by adding SEMG to the current stimulus signal generation model. f ir Updated to EMG ir This makes the generation results of the stimulus signal generation model more accurate.

[0041] S400, based on the current stimulus signal query information and the current stimulus signal generation model, obtain the electrical stimulation signal (ETS) for the i-th stimulation cycle.ir The current stimulus signal query information includes at least: basic information of the target user, EMG... ir and EMG ir The corresponding baseline electromyographic signal; execute S500.

[0042] In this embodiment of the invention, the current stimulation signal generation model is associated with an electrical stimulation signal reference list, and the electrical stimulation signal reference data for each row of the electrical stimulation signal reference list includes (U, EMG, ETS, EMG). next ), where U represents the basic information of the user corresponding to this row of data, EMG represents the electromyographic signals acquired by the electromyography (EMG) acquisition device on the specified muscle group of the user in one acquisition cycle, and ETS represents the stimulation signal applied to the electromyography stimulator on the specified muscle group of the corresponding user in one stimulation cycle based on EMG. next The electromyography (EMG) signals acquired by the EMG acquisition device on a specified muscle group for the corresponding user during the acquisition cycle after the corresponding muscle group receives ETS stimulation; wherein, ETS is determined based on EMG and the corresponding reference EMG signal.

[0043] In this embodiment of the invention, the electrical stimulation signal is a current intensity, and its characteristic parameters include at least amplitude and frequency. The amplitude corresponding to the ETS is positively correlated with the deviation between the amplitude corresponding to the EMG and the amplitude corresponding to the corresponding reference EMG signal. If the median frequency corresponding to the EMG is less than the median frequency corresponding to the corresponding reference EMG signal, the frequency corresponding to the ETS is in the mid-to-high frequency range; otherwise, the frequency corresponding to the ETS is in the low frequency range.

[0044] In this embodiment of the invention, the low, medium and high frequency bands can be determined based on the frequency of the actual acquisition equipment used, and the invention does not impose any particular limitation.

[0045] In one illustrative embodiment, the amplitude corresponding to ETS is A1 = k*(A-A2) / Amax*Imax, where A is the amplitude of the reference electromyography signal corresponding to EMG, A2 is the amplitude corresponding to EMG, Amax is the maximum amplitude of the corresponding acquisition device, and Imax is the amplitude of the maximum current intensity that the user can withstand. k is a preset coefficient, which can be an empirical value.

[0046] In one illustrative embodiment, Imax can be obtained based on the user's gender, age, identifier of a specified muscle group, functional state of a specified muscle region, and a preset current intensity amplitude lookup table. The v-th row of the preset current intensity lookup table includes at least (Sv, Agv, Mv, STv, Fu, Imaxv), where Sv is the gender corresponding to the v-th row of data, Agv is the age corresponding to the v-th row of data, Mv is the identifier of the muscle group corresponding to the v-th row of data, STv is the functional state of the muscle region corresponding to the v-th row of data, Fv is the body characteristic corresponding to the v-th row of data, and Imaxv is the amplitude of the maximum current intensity corresponding to the v-th row of data. The value of v ranges from 1 to W, and W is the number of rows in the preset current intensity amplitude lookup table. The preset current intensity amplitude lookup table can be obtained experimentally, and specific experimental methods can employ existing technologies; this invention does not impose any particular limitation.

[0047] In this embodiment of the invention, the reference data for each electrical stimulation signal can be an empirical value. For example, if an electrical stimulation signal is set based on experience during an electrical stimulation process and it is found that the electromyographic signal is improved after the electrical stimulation, then the data of this electrical stimulation process will be saved as historical experience data.

[0048] Furthermore, in this embodiment of the invention, the reference list of electrical stimulation signals can be obtained by the stimulation signal generation model based on a preset number of historical electrical stimulation signal reference data increments. In an illustrative embodiment, the stimulation signal generation model can be an incremental neural network model. Those skilled in the art will understand that any method of incrementing a preset number of historical electrical stimulation signal reference data falls within the scope of protection of this invention. By using an incremental approach, as much reference data as possible can be provided even when the sample size is insufficient.

[0049] Furthermore, in embodiments of the present invention, ETS i This can be determined based on the following steps:

[0050] S401, based on the current stimulus signal query information, obtain the target electrical stimulation signal reference data from the current electrical stimulation signal reference list using the current stimulus signal generation model; wherein, the target electrical stimulation signal reference data satisfies the following condition: the similarity between the target electrical stimulation signal reference data and the current stimulus signal query information is greater than a set similarity threshold.

[0051] In this embodiment of the invention, the similarity can be cosine similarity. The similarity threshold can be set based on actual needs.

[0052] When calculating similarity, the stimulus signal query information can be encoded according to a set format to obtain corresponding encoded data. The specific encoding method can be any existing technology. Then, the stimulus signal generation model will use its own calculation strategy to obtain the data with the highest similarity to the encoded data as the target data. It is understood by those skilled in the art that any method by which a stimulus signal generation model obtains the data with the highest similarity to the encoded data as the target data based on its own calculation strategy falls within the scope of protection of this invention.

[0053] S402, using the electrical stimulation signal in the target electrical stimulation signal reference data as the ETS i .

[0054] Those skilled in the art will know that if multiple target data exist, the electrical stimulation signal from the target data with the highest similarity can be used as the ETS. ir .

[0055] In this embodiment of the invention, the electrical stimulation signal in the target data with the highest similarity is used as the ETS. i This ensures that the stimulation time is as short as possible, thereby improving the user experience.

[0056] As is known to those skilled in the art, SEMG f ir This can be the EMG from the target electrical stimulation signal reference data obtained based on the query information of the previous stimulation signal. next .

[0057] Furthermore, in another illustrative embodiment of the present invention, if i > 1, ETS i It can also be determined based on the following steps:

[0058] If the amplitude of the currently acquired electromyographic signal is less than the amplitude of the corresponding reference electromyographic signal, the amplitude of the current electrical stimulation signal is increased by a set amplitude; if the amplitude of the currently acquired electromyographic signal is greater than the amplitude of the corresponding reference electromyographic signal, the amplitude of the current electrical stimulation signal is decreased by a set amplitude. Furthermore, if the median frequency of the currently acquired electromyographic signal is less than the median frequency of the corresponding reference electromyographic signal, the frequency of the current electrical stimulation signal is increased by a set frequency range; if the amplitude of the currently acquired electromyographic signal is greater than the amplitude of the corresponding reference electromyographic signal, the frequency of the current electrical stimulation signal is decreased by a set frequency range.

[0059] In this embodiment of the invention, the set amplitude and set frequency range can be empirical values.

[0060] S500, ETS ir Send to the m electromyographic stimulators so that the m electromyographic stimulators output ETS in the i-th stimulation cycle. ir ;Execute S600.

[0061] In this embodiment of the invention, the duration of the acquisition period is longer than the duration of the stimulation period. In one illustrative embodiment, the duration of the stimulation period may be, for example, 10 seconds.

[0062] S600, set i = i + 1, execute S200.

[0063] It should be noted that in the embodiments of the present invention, the acquisition time of the electromyography (EMG) acquisition device does not overlap with the stimulation time of the EMG stimulator. That is, the EMG acquisition device starts acquiring data only after the EMG stimulator has stimulated the data. However, the EMG acquisition device can also acquire data continuously, but the EMG signal used for analysis is the EMG signal after stimulation.

[0064] Furthermore, the method provided in this embodiment of the invention further includes the following steps:

[0065] The acquired electromyographic signals are then visualized.

[0066] Those skilled in the art will know that any method of visualizing the acquired electromyographic signals falls within the scope of protection of this invention.

[0067] Furthermore, in a preferred embodiment, the electrodes of the electromyography (EMG) collector are dry electrodes.

[0068] In this embodiment of the invention, a dry electrode refers to an electrode that directly contacts the skin using a conductive patch instead of conductive gel, while a wet electrode refers to an electrode that maintains good contact with the skin using conductive paste. Wet electrodes possess sufficient flexibility to ensure adequate contact with the skin, thereby minimizing contact impedance and improving signal quality. However, in practical gesture recognition applications, such as gesture-controlled mobile phones or machinery, the electrodes need to be durable. Electrodes with conductive paste, which are prone to failure over time, are unsuitable because users' skin conditions vary—they may be dry, oily, or sweaty. This requires the electrodes and their amplification circuitry to adapt to these complex conditions. Dry electrodes use capacitive coupling to measure signals. A thin insulating film exists between the electrode and the human body, creating a capacitance between the body and the metal electrode. The bioelectrical signal can then pass through this capacitor to the amplifier's input. This type of electrode simplifies the usage procedure, avoids the corrosion and irritation of the skin caused by conductive paste, and eliminates the risk of short circuits between electrodes. It is suitable for continuous measurement over time, but its higher electrode skin contact impedance will increase noise levels and signal instability.

[0069] Because dry electrode-acquired electromyography (EMG) signals suffer from increased noise levels and signal instability, it is necessary to correct the accuracy of these signals. Therefore, when the electrodes in the EMG acquisition device are dry electrodes, EMG... iThe corrected electromyographic signal can be achieved through a trained AI model.

[0070] Furthermore, EMG i Specifically, it is obtained through the following method: [EMG] i The corresponding raw electromyographic signal, the action identifier corresponding to the raw electromyographic signal, the muscle group identifier corresponding to the raw electromyographic signal, and the acquisition duration corresponding to the electromyographic signal are input into the trained AI model to obtain the corresponding corrected electromyographic signal.

[0071] Furthermore, the trained AI model can be obtained through the following steps:

[0072] S10, Obtain the training sample dataset; the r-th sample data in the training sample dataset includes (S1... r M r A r T r S2 r ), where r ranges from 1 to G, and G is the number of samples in the training dataset; where S1 r M represents the first electromyographic signal corresponding to the r-th sample data. r Let A be the muscle group identifier corresponding to the r-th sample data. r T is the action identifier corresponding to the r-th sample data. r S2 represents the collection time corresponding to the r-th sample data. r Let be the second electromyographic signal corresponding to the r-th sample data, wherein the first electromyographic signal is acquired through dry electrodes and the second electromyographic signal is acquired through wet electrodes.

[0073] In this embodiment of the invention, each action may have a designated identifier, and each muscle group may have a designated identifier. The specific identifier settings can be configured based on actual needs.

[0074] In this embodiment of the invention, the r-th training sample data can be obtained through the following steps:

[0075] S101, the test object U corresponding to the r-th training sample data. r and M r A preset number of electromyography (EMG) collectors are installed on the corresponding muscle groups, and the electrodes of the EMG collectors are attached to the corresponding positions of the muscle groups via conductive patches.

[0076] S102, Control U r Execute A r During the execution of the corresponding action, an electromyography (EMG) acquisition device is used to collect EMG signals for a preset duration, which are then used as the corresponding first EMG signal.

[0077] S103, the test object U corresponding to the r-th training sample data. r and M r A preset number of electromyography (EMG) collectors are installed on the corresponding muscle groups, and the electrodes of the EMG collectors are attached to the corresponding positions of the muscle groups using hydrogel.

[0078] S104, Control U r Execute A r During the execution of the corresponding action, an electromyography (EMG) acquisition device is used to collect EMG signals for a preset duration, which are then used as the corresponding second EMG signal.

[0079] S12, input the training sample data of the current batch into the current AI model for training, and obtain the corresponding prediction results; the prediction result of each training sample data is the second electromyographic signal.

[0080] S13. Based on the prediction results of the current batch and the corresponding real results, obtain the current loss function value of the current AI model, and determine whether the current loss function value meets the preset model training termination condition. If it does, execute step S15; otherwise, execute step S14.

[0081] S14, Update the parameters of the current AI model based on the current loss function value, and use the next batch of training sample data as the current batch of training sample data, and execute S12;

[0082] S15, use the current AI model as the trained AI model.

[0083] In this embodiment of the invention, the AI ​​model can be a neural network model. The loss function value can be calculated based on an existing loss function. The preset model training termination condition can be set according to actual needs. For example, the loss is less than or equal to a set loss threshold and remains unchanged within a set time period.

[0084] Since dry electrode electromyography signals can be acquired, the user experience can be further improved.

[0085] Based on the same inventive concept, this invention also provides an electromyography (EMG) signal processing device, which includes a housing and a processor and a memory disposed within the housing. The memory contains a computer program, and the processor is used to communicate with an EMG acquisition device and an EMG stimulator disposed on a designated muscle group of a target user. When the computer program is executed by the processor, it is used to implement the aforementioned method.

[0086] Furthermore, the device provided in this embodiment of the invention also includes: a signal conditioning circuit, which has functions such as signal amplification, signal filtering, and dynamic range adjustment of the signal, used to condition weak surface electromyography signals to a voltage range that matches the dynamic range of the analog-to-digital converter chip; a sampling circuit, which uses the analog-to-digital converter chip to sample and quantize analog signals into digital signals; a control unit, generally a microcontroller, FPGA, etc., used to coordinate the execution timing of various modules of the system to ensure the correct operation of the system; a communication unit, used for protocol conversion when the control unit communicates with the host computer software; and a power management unit, which consists of a series of voltage conversion chips to provide operating voltage for the components of the entire system.

[0087] This invention also provides a non-transitory computer-readable storage medium that stores a program or instructions that cause a computer to perform the steps of the aforementioned method.

[0088] It should be understood that the various forms of processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this invention can be achieved, and this is not limited herein.

[0089] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A signal processing device, characterized in that, It includes: a housing and a processor and a memory disposed within the housing. The memory contains a computer program. The processor is used to communicate with n electromyography (EMG) acquisition devices and m EMG stimulators disposed on designated muscle groups of a target user, where n ≥ m ≥ 1. When the computer program is executed by the processor, it performs the following steps: S100, sends an instruction to the target user to perform a specified action; S200, acquire electromyographic (EMG) signals from n EMG acquisition devices during the i-th acquisition cycle. i ={EMG i1 EMG i2 ..., EMG ir ..., EMG in };EMG ir Let r be the electromyographic signal acquired by the r-th electromyographic acquisition device in the i-th acquisition cycle, where r ranges from 1 to n and i is initially 1. S300, if EMG ir The deviation Δd between the signal and the corresponding baseline electromyographic signal i <d1, exit the current control program; if △d i ≥d1, and if EMG ir and SEMG f ir The deviation between them is greater than or equal to d2, using EMG i Update the current stimulus signal generation model and execute S400; where d1 is the first setpoint deviation and d2 is the second setpoint deviation; SEMG f ir The electromyographic (EMG) signals are obtained based on the previous stimulation signal query information and the corresponding stimulation signal generation model. The stimulation signal query information includes at least: the target user's basic information, the EMG signals collected by the EMG acquisition devices on the target user's specified muscle groups in the previous acquisition cycle, and the baseline EMG signals corresponding to the EMG signals collected in the previous acquisition cycle. The target user's basic information includes: the target user's attribute parameters, the muscle group identifier of the target user's specified muscle groups, the action identifier of the specified action performed by the target user, the number of EMG acquisition devices on the target user's specified muscle groups, the location information of the EMG acquisition devices on the target user's specified muscle groups, the number of EMG stimulators on the target user's specified muscle groups, and the location information of the EMG stimulators on the target user's specified muscle groups. The stimulation signal generation model is an incremental neural network model. S400, based on the current stimulus signal query information and the current stimulus signal generation model, obtain the electrical stimulation signal (ETS) for the i-th stimulation cycle. i The current stimulus signal query information includes at least: basic information of the target user, EMG... ir and EMG ir The corresponding baseline electromyographic signal; execute S500; wherein, the current stimulation signal generation model is associated with an electrical stimulation signal reference list, and the electrical stimulation signal reference data of each row of the electrical stimulation signal reference list includes (U, EMG, ETS, EMG) next ), where U represents the basic information of the user corresponding to this row of data, EMG represents the electromyographic signals acquired by the electromyography (EMG) acquisition device on the specified muscle group of the user in one acquisition cycle, and ETS represents the stimulation signal applied to the electromyography stimulator on the specified muscle group of the corresponding user in one stimulation cycle based on EMG. next The electromyography (EMG) signals acquired by the EMG acquisition device on a specified muscle group for the corresponding user during the acquisition cycle after the corresponding muscle group receives ETS stimulation; wherein, ETS is determined based on EMG and the corresponding reference EMG signal; Among them, ETS i Determined based on the following steps: S401, Based on the current stimulus signal query information, obtain the target electrical stimulation signal reference data from the current electrical stimulation signal reference list using the current stimulus signal generation model; wherein, the target electrical stimulation signal reference data satisfies the following condition: the similarity between the target electrical stimulation signal reference data and the current stimulus signal query information is greater than a set similarity threshold; S402, using the electrical stimulation signal in the target electrical stimulation signal reference data as the ETS ir ; S500, ETS i Send to the m electromyographic stimulators so that the m electromyographic stimulators output ETS in the i-th stimulation cycle. i ;Execute S600; S600, set i=i+1, execute S200.

2. The device according to claim 1, characterized in that, The characteristic parameters of the electromyography signal include at least amplitude and median frequency, and the characteristic parameters of the electrical stimulation signal include at least amplitude and frequency. The amplitude corresponding to the ETS is positively correlated with the deviation between the amplitude corresponding to the EMG and the amplitude corresponding to the corresponding reference electromyography signal. If the median frequency corresponding to the EMG is less than the median frequency corresponding to the corresponding reference electromyography signal, the frequency corresponding to the ETS is in the mid-to-high frequency range; otherwise, the frequency corresponding to the ETS is in the low frequency range.

3. The device according to claim 1, characterized in that, The reference list of electrical stimulation signals is obtained by the stimulation signal generation model based on a preset number of historical electrical stimulation signal reference data increments.

4. The device according to claim 1, characterized in that, in, The duration of the acquisition cycle is greater than the duration of the stimulation cycle.

5. The device according to claim 1, characterized in that, If the specified muscle group belongs to the first type of muscle group, the reference electromyographic signal corresponding to the specified muscle group is determined as follows: Acquire the electromyographic signals collected by the electromyographic acquisition device set on the symmetrical muscle group of the specified muscle group, and use them as candidate electromyographic signals. If the error between the candidate electromyographic signal and the electromyographic signal of the muscle group in the second state is less than d1, then the candidate electromyographic signal is used as the reference electromyographic signal corresponding to the specified muscle group. Otherwise, the electromyographic signal of the specified muscle group in the second state is used. If the specified muscle group belongs to the second type of muscle group, the reference electromyographic signal corresponding to the specified muscle group is the electromyographic signal when the specified muscle group is in the second state.

6. The device according to claim 5, characterized in that, The first type of muscle group consists of muscles located in the upper and lower limbs, while the second type consists of muscles located in the head and trunk.

7. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium stores a program or instructions that cause a computer to perform the following steps: S100, send an instruction to the target user to perform a specified action; the specified muscle group of the target user is in a first state and is equipped with n electromyography (EMG) collectors and m EMG stimulators, where n≥m≥1; S200, acquire electromyographic (EMG) signals from n EMG acquisition devices during the i-th acquisition cycle. i ={EMG i1 EMG i2 ..., EMG ir ..., EMG in };EMG ir Let r be the electromyographic signal acquired by the r-th electromyographic acquisition device in the i-th acquisition cycle, where r ranges from 1 to n and i is initially 1. S300, if EMG ir The deviation Δd between the signal and the corresponding baseline electromyographic signal i <d1, exit the current control program; if △d i ≥d1, and if EMG ir and SEMG f ir The deviation between them is greater than or equal to d2, using EMG i Update the current stimulus signal generation model and execute S400; where d1 is the first setpoint deviation and d2 is the second setpoint deviation; SEMG f ir The electromyographic (EMG) signals are obtained based on the previous stimulation signal query information and the corresponding stimulation signal generation model. The stimulation signal query information includes at least: the target user's basic information, the EMG signals collected by the EMG acquisition devices on the target user's specified muscle groups in the previous acquisition cycle, and the baseline EMG signals corresponding to the EMG signals collected in the previous acquisition cycle. The target user's basic information includes: the target user's attribute parameters, the muscle group identifier of the target user's specified muscle groups, the action identifier of the specified action performed by the target user, the number of EMG acquisition devices on the target user's specified muscle groups, the location information of the EMG acquisition devices on the target user's specified muscle groups, the number of EMG stimulators on the target user's specified muscle groups, and the location information of the EMG stimulators on the target user's specified muscle groups. The stimulation signal generation model is an incremental neural network model. S400, based on the current stimulus signal query information and the current stimulus signal generation model, obtain the electrical stimulation signal (ETS) for the i-th stimulation cycle. i The current stimulus signal query information includes at least: basic information of the target user, EMG... ir and EMG ir The corresponding baseline electromyographic signal; execute S500; wherein, the current stimulation signal generation model is associated with an electrical stimulation signal reference list, and the electrical stimulation signal reference data of each row of the electrical stimulation signal reference list includes (U, EMG, ETS, EMG) next ), where U represents the basic information of the user corresponding to this row of data, EMG represents the electromyographic signals acquired by the electromyography (EMG) acquisition device on the specified muscle group of the user in one acquisition cycle, and ETS represents the stimulation signal applied to the electromyography stimulator on the specified muscle group of the corresponding user in one stimulation cycle based on EMG. next The electromyography (EMG) signals acquired by the EMG acquisition device on a specified muscle group for the corresponding user during the acquisition cycle after the corresponding muscle group receives ETS stimulation; wherein, ETS is determined based on EMG and the corresponding reference EMG signal; Among them, ETS i Determined based on the following steps: S401, Based on the current stimulus signal query information, obtain the target electrical stimulation signal reference data from the current electrical stimulation signal reference list using the current stimulus signal generation model; wherein, the target electrical stimulation signal reference data satisfies the following condition: the similarity between the target electrical stimulation signal reference data and the current stimulus signal query information is greater than a set similarity threshold; S402, using the electrical stimulation signal in the target electrical stimulation signal reference data as the ETS ir ; S500, ETS i Send to the m electromyographic stimulators so that the m electromyographic stimulators output ETS in the i-th stimulation cycle. i ;Execute S600; S600, set i=i+1, execute S200.

Images