Control method and device for upper and lower limb muscle training, equipment and medium

Abstract

Embodiments of the present application disclose a control method and device for upper and lower limb muscle training, equipment and medium. The present application relates to the field of artificial intelligence technology, which comprises: if an electromyographic signal sent by an electromyographic stimulator is received, the electromyographic signal is subjected to feature extraction and normalization processing to obtain an electromyographic feature; the electromyographic feature is input into a preset denoising neural network for denoising processing to obtain a denoised electromyographic feature; a correlation function in an Elman neural network receiving layer is modified, and the denoised electromyographic feature is input into the modified Elman neural network for upper and lower limb muscle fatigue detection to obtain a fatigue detection result; and the current intensity of the electromyographic stimulator and the training speed of the upper and lower limb rehabilitation training device are adjusted according to the fatigue detection result. The embodiments of the present application can improve the accuracy of upper and lower limb muscle fatigue detection, avoid excessive fatigue of the upper and lower limb muscles, and cause secondary damage.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of artificial intelligence, and in particular to a control method and device for upper and lower limb muscle training, equipment and a medium. BACKGROUND

[0002] The upper and lower limb rehabilitation training device is an auxiliary device for patients to sit on a fixed chair or an unfixed wheelchair, place both hands on the armrests, and place both feet on the foot pedals to perform limb function rehabilitation training according to the guidance. The device can use the built-in motor to pull the patient to train the limb function, including the contraction and relaxation of the upper limbs and the similar action of the lower limbs as riding a bicycle, thereby guiding the patient to gradually recover the upper and lower limb movement function. With the development of electronic signal processing, bioelectricity research and other technical disciplines, human physiological signals have gradually become an important research object of rehabilitation medicine, and the human physiological signals mainly include electroencephalogram signals, electrocardiogram signals and electromyogram signals, etc. The electromyogram signals are widely used in the detection and research of muscle fatigue state due to the advantages of convenient acquisition, good real-time performance and obvious signals, but the traditional signal analysis method such as Fourier transform analysis electromyogram signals cannot accurately and reliably detect the upper and lower limb muscle fatigue, thereby causing the muscle fatigue and causing secondary injury. SUMMARY

[0003] The embodiments of the present application provide a control method and device for upper and lower limb muscle training, equipment and a medium, aiming to solve the problem of muscle fatigue caused by the low accuracy of upper and lower limb muscle fatigue detection.

[0004] In a first aspect, the embodiments of the present application provide a control method for upper and lower limb muscle training, which comprises:

[0005] If the electromyogram signal sent by the electromyogram stimulator is received, the electromyogram signal is subjected to feature extraction and normalization processing to obtain electromyogram features;

[0006] The electromyogram features are input into a preset denoising neural network for denoising processing to obtain denoised electromyogram features;

[0007] The correlation function in the Elman neural network receiving layer is modified, and the denoised electromyogram features are input into the modified Elman neural network for upper and lower limb muscle fatigue detection to obtain a fatigue detection result;

[0008] The current intensity of the electromyogram stimulator and the training speed of the upper and lower limb rehabilitation training device are adjusted according to the fatigue detection result.

[0009] In a second aspect, the embodiments of the present application further provide a control device for upper and lower limb muscle training, which comprises:

[0010] An extraction processing unit is configured to, if receiving an electromyogram signal sent by the electromyostimulation instrument, extract and normalize the electromyogram signal to obtain electromyogram features;

[0011] A denoising fitting unit is configured to input the electromyogram features into a preset denoising neural network to perform denoising processing to obtain denoised electromyogram features;

[0012] A fatigue detection unit is configured to modify a correlation function in a receiving layer of an Elman neural network, and input the denoised electromyogram features into the modified Elman neural network to perform upper and lower limb muscle fatigue detection to obtain fatigue detection results;

[0013] A control unit is configured to adjust a current intensity of the electromyostimulation instrument and a training speed of the upper and lower limb rehabilitation training device according to the fatigue detection results.

[0014] In a third aspect, an embodiment of the present application further provides a computer device, which comprises a memory and a processor, the memory stores a computer program, and the processor implements the above method when executing the computer program.

[0015] In a fourth aspect, an embodiment of the present application further provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the above method.

[0016] The embodiment of the present application provides a control method, device, equipment and medium for upper and lower limb muscle training. The method comprises the following steps: if receiving an electromyogram signal sent by the electromyostimulation instrument, extracting and normalizing the electromyogram signal to obtain electromyogram features; inputting the electromyogram features into a preset denoising neural network to perform denoising processing to obtain denoised electromyogram features; modifying a correlation function in a receiving layer of an Elman neural network, and inputting the denoised electromyogram features into the modified Elman neural network to perform upper and lower limb muscle fatigue detection to obtain fatigue detection results; and adjusting a current intensity of the electromyostimulation instrument and a training speed of the upper and lower limb rehabilitation training device according to the fatigue detection results. The technical solution of the embodiment of the present application first inputs the electromyogram features added with Gaussian white noise into a preset denoising neural network to perform denoising processing to obtain denoised electromyogram features, which can remove noise in the electromyogram signal to a certain extent; then modifies a correlation function in a receiving layer of an Elman neural network, and inputs the denoised electromyogram features into the modified Elman neural network to perform upper and lower limb muscle fatigue detection to obtain fatigue detection results, which can improve the accuracy of upper and lower limb muscle fatigue detection; finally, the current intensity of the electromyostimulation instrument and the training speed of the upper and lower limb rehabilitation training device are adjusted according to the fatigue detection results, which can avoid excessive fatigue of the upper and lower limb muscles and cause secondary damage. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed in the embodiment description. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without any creative effort.

[0018] Figure 1 is a scene schematic diagram of a control method for upper and lower limb muscle training provided by an embodiment of the present application;

[0019] Figure 2 is a flow schematic diagram of a control method for upper and lower limb muscle training provided by an embodiment of the present application;

[0020] Figure 3 is a sub-flow schematic diagram of a control method for upper and lower limb muscle training provided by an embodiment of the present application;

[0021] Figure 4 is a sub-flow schematic diagram of a control method for upper and lower limb muscle training provided by an embodiment of the present application;

[0022] Figure 5 is a schematic block diagram of a control device for upper and lower limb muscle training provided by an embodiment of the present application; and

[0023] Figure 6 is a schematic block diagram of a computer device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0024] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without any creative effort belong to the scope of protection of the present application.

[0025] It should be understood that, when used in the specification and the appended claims, the terms “comprise” and “include” indicate the presence of described features, integers, steps, operations, elements, and / or components, but do not exclude one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0026] It should also be understood that the terms used in the present application specification are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the present application specification and the appended claims, the singular forms “a”, “an” and “the” are intended to include the plural forms, unless the context clearly indicates otherwise.

[0027] It should also be further understood that the term "and / or" as used in the specification and in the claims, if any, means any

[0028] As used in the specification and the claims, the term "if' can be interpreted as meaning "when" or "once" or "in response to a determination" or "in response to a detection" depending on the context. Similarly, the phrase "if determined" or "if detected [the described condition or event]" can be interpreted to mean "once determined" or "in response to a determination" or "once detected [the described condition or event]" or "in response to a detection [the described condition or event]" depending on the context.

[0029] Referring to Figure 1 , Figure 1 is a scene schematic diagram of a control method for upper and lower limb muscle training provided by an embodiment of the present application. The control method for upper and lower limb muscle training of the present application can be applied to a terminal, for example, the control method for upper and lower limb muscle training can be realized by a software program configured on the terminal, thereby improving the accuracy of upper and lower limb muscle fatigue detection and avoiding secondary injury caused by excessive fatigue of the upper and lower limb muscles. As Figure 1 indicated, the upper and lower limb intelligent rehabilitation system in the embodiment of the present application includes an electromyographic sensor, a nano board, an electromyographic stimulator, an upper and lower limb rehabilitation training device, and a terminal, wherein the electromyographic sensor is a myoware electromyographic sensor, the nano board is one of arduino boards, has the ability of convenience, flexibility and programmability, and the terminal is installed with an upper and lower limb intelligent fatigue detection system. It should be noted that in the embodiment of the present application, the electromyographic sensor is connected with the electromyographic stimulator through the nano board; the electromyographic stimulator is connected with the terminal through Bluetooth to send an electromyographic signal to the terminal and receive a current intensity adjustment instruction sent by the terminal; the upper and lower limb rehabilitation training device is connected with the terminal through a USB interface to receive a training speed adjustment instruction sent by the terminal. It should also be noted that in actual application, the positive electrode, the negative electrode and the reference electrode of the electromyographic sensor need to be pasted with electromyographic sensor patches, then the positive electrode and the negative electrode are attached along the direction of the muscle, and the reference electrode is attached on the center line of the positive electrode and the negative electrode, so that the best electromyographic signal acquisition effect can be achieved.

[0030] Referring to Figure 2 , Figure 2 is a flowchart of a control method for upper and lower limb muscle training provided by an embodiment of the present application. As Figure 2 shown, the method includes the following steps S100-S130.

[0031] S100, if the electromyogram signal sent by the electromyogram stimulator is received, performing feature extraction and normalization processing on the electromyogram signal to obtain electromyogram features.

[0032] In the embodiment of the present application, when the patient sits on the upper and lower limb rehabilitation training device, the electromyogram sensor is attached, and the electromyogram stimulator is held, if the terminal receives the electromyogram signal sent by the electromyogram stimulator, the electromyogram signal is subjected to feature extraction and normalization processing to obtain electromyogram features. Specifically, the electromyogram signal is first subjected to time domain and frequency domain feature extraction to obtain time domain electromyogram features and frequency domain electromyogram features, wherein the time domain electromyogram features include the root mean square amplitude, zero crossing rate, variance and slope change rate of the electromyogram signal, and the frequency domain electromyogram features include spectral entropy and spectral flux; the time domain electromyogram features and the frequency domain electromyogram features are subjected to normalization processing to obtain target time domain electromyogram features and target frequency domain electromyogram features, and the target time domain electromyogram features and the target frequency domain electromyogram features are taken as electromyogram features. It should be noted that the formulas for calculating the time domain electromyogram features and the frequency domain electromyogram features are all existing formulas, which will not be described here.

[0033] S110, inputting the electromyogram features into a preset denoising neural network for denoising processing to obtain denoised electromyogram features.

[0034] In the embodiment of the present application, the electromyogram features are input into a preset denoising neural network for denoising processing to obtain denoised electromyogram features, wherein the preset denoising neural network is obtained by training a convolutional neural network using sample electromyogram signals and sample electromyogram signals with added Gaussian white noise, and the preset denoising neural network includes a convolutional layer, a pooling layer and a fully connected layer. It should be noted that in the embodiment of the present application, the loss function when training the convolutional neural network using the sample electromyogram signals and the sample electromyogram signals with added Gaussian white noise is wherein, is a regularization coefficient, which is a constant in the range of (0, 1), E1 represents the square of the difference between the corresponding features of the sample electromyogram signal and the sample electromyogram signal with added Gaussian white noise, and E2 represents the square of the difference between the corresponding features of the output electromyogram signal of the sample electromyogram signal with added Gaussian white noise after passing through the convolutional neural network and the sample electromyogram signal. In the training process, the minimum value of the total loss value E is calculated by the Adam algorithm, and then the value of and the parameters in the convolutional neural network are determined to obtain the denoising neural network. It should be noted that in the embodiment of the present application, the denoising neural network is used to denoise the electromyogram features, which can remove the noise in the electromyogram signal to a certain extent.

[0035] Please refer to Figure 3 ​In an embodiment, for example in an embodiment of the present application, the step S110 comprises steps S111-S113.

[0036] S111, inputting the electromyographic feature into the convolution layer to obtain a first denoised electromyographic feature through convolution;

[0037] S112, inputting the first denoised electromyographic feature into the pooling layer to obtain a second denoised electromyographic feature through pooling;

[0038] S113, inputting the second denoised electromyographic feature into the fully connected layer to output a third denoised electromyographic feature, and taking the third denoised electromyographic feature as the denoised electromyographic feature.

[0039] In an embodiment of the present application, the electromyographic feature after adding Gaussian white noise is a 1x5 matrix, the 1x5 electromyographic feature is first filled into a 13x13 matrix, then the filled 13x13 matrix is input into the convolution layer to obtain a 9x9 first denoised electromyographic feature through convolution, and then the second denoised electromyographic feature is obtained through the pooling layer to obtain a 3x3 second denoised electromyographic feature, the second denoised electromyographic feature is input into the fully connected layer to output a third denoised electromyographic feature, and the third denoised electromyographic feature is taken as the denoised electromyographic feature. It should be noted that in an embodiment of the present application, the convolutional neural network comprises three fully connected layers.

[0040] S120, modifying a correlation function in an Elman neural network transfer layer, and inputting the denoised electromyographic feature into the modified Elman neural network to perform upper and lower limb muscle fatigue detection to obtain a fatigue detection result.

[0041] In an embodiment of the present application, the Elman neural network comprises an input layer, a hidden layer, a transfer layer and an output layer, for the convenience of description, it is assumed that the denoised electromyographic feature is X=x1, x2,.., xn, that is, the input feature vector in the input layer is X=x1, x2,.., xn; in the hidden layer, a preset hidden output feature formula is shown in formula (1), in formula (1), w1 is a weight matrix from the input layer to the hidden layer, X k (t) is a hidden output feature at time t; in the transfer layer, the modified correlation function is shown in formula (2), in formula (2), X e (t) is a transfer output feature at time t, sigmoid is an activation function, w2 is a weight matrix from the hidden layer to the transfer layer, X k(t-1) represents the hidden output feature at time t-1; in the output layer, the preset output formula is shown in formula (3), where w3 and w4 represent the weight matrices from the hidden layer and the connecting layer to the output layer, respectively. It should be noted that in this embodiment of the invention, during the training phase of the modified Elman neural network, the binary cross-entropy loss function is used to train the modified Elman neural network, thereby determining the weight values ​​of w1, w2, w3, and w4, as well as the network parameters. It should also be noted that the reason for modifying the correlation function in the connecting layer, i.e., taking the log before the sigmoid function, is that it can improve the detection accuracy of upper and lower limb muscle fatigue to a certain extent.

[0042] X k (t)=w1X (1)

[0043] X e (t)=log(sigmoid(w2X k (t-1))) (2)

[0044] Y=∑(w3X e (t)+w4X k (t)) (3)

[0045] S130. Adjust the current intensity of the electromyography device and the training speed of the upper and lower limb rehabilitation trainer according to the fatigue detection results.

[0046] In this embodiment of the invention, if the fatigue detection result indicates non-muscle fatigue of the upper and lower limbs, it means that the current intensity can be increased to electrically stimulate the muscles of the upper and lower limbs. In this case, the current intensity of the electromyography (EMG) stimulator is adjusted using a current intensity adjustment method. Understandably, the training speed of the upper and lower limb rehabilitation trainer can remain unchanged, i.e., the initial training speed is maintained. If the fatigue detection result indicates muscle fatigue of the upper and lower limbs, in order to prevent damage to the patient's muscles due to excessive fatigue, a deceleration command is sent to the upper and lower limb rehabilitation trainer to reduce its training speed. Understandably, the EMG stimulator will then pause the electrical stimulation operation. It should be noted that in this embodiment of the invention, the value of the fatigue detection result is in the range of 0 to 1. A fatigue detection result less than 0.5 is defined as non-muscle fatigue of the upper and lower limbs, and a fatigue detection result greater than or equal to 0.5 is defined as muscle fatigue of the upper and lower limbs.

[0047] Please see Figure 4 In one embodiment, for example, in an embodiment of the present invention, step S130 includes the following steps S131-S136.

[0048] S131, judging whether the electric stimulation times is the preset electric stimulation times, if the electric stimulation times is the preset electric stimulation times, executing step S132, otherwise executing step S133;

[0049] S132, sending current intensity instruction to the myoelectric stimulator according to the default electric stimulation parameter to adjust the current intensity of the myoelectric stimulator;

[0050] S133, when the electric stimulation times is an integer multiple of the preset stimulation times, obtaining the target current intensity in a self-increasing manner according to the initial current intensity;

[0051] S134, judging whether the target current intensity exceeds the preset current intensity threshold, if the target current intensity does not exceed the preset current intensity threshold, executing step S135, otherwise executing step S136;

[0052] S135, sending the current intensity instruction corresponding to the target current intensity to the myoelectric stimulator to adjust the current intensity of the myoelectric stimulator, and returning to execute step S133;

[0053] S136, taking the preset current intensity threshold as the target current intensity, and returning to execute step S135.

[0054] In the embodiment of the present application, the current intensity of the electromyostimulation instrument is adjusted by the current intensity adjustment method. Specifically, it is judged whether the number of electrical stimulation is the preset number of electrical stimulation, wherein the preset number of electrical stimulation is 1. If the number of electrical stimulation is 1, it indicates that the patient is performing electrical stimulation for the first time, and then the current intensity instruction is sent to the electromyostimulation instrument according to the default electrical stimulation parameters to adjust the current intensity of the electromyostimulation instrument, wherein the default electrical stimulation parameters include electrical stimulation frequency (50 Hz), electrical stimulation pulse width (100 μs) and electrical stimulation current intensity (1 mA). If the number of electrical stimulation is not 1, it indicates that the patient is not performing electrical stimulation for the first time, and then the target current intensity is obtained in the current self-increasing manner according to the initial current intensity 1 mA and the preset number of stimulation 50. Specifically, when the number of electrical stimulation is an integer multiple of 50, for example, the number of electrical stimulation is 50, the target current intensity 2 mA is obtained by adding 1 mA to the initial current intensity 1 mA. It is judged whether the target current intensity exceeds the preset current intensity threshold, wherein the preset current intensity threshold is 70 mA. If the target current intensity does not exceed the preset current intensity threshold, the current intensity instruction corresponding to the target current intensity 2 mA is sent to the electromyostimulation instrument to adjust the current intensity of the electromyostimulation instrument, and the target current intensity is continuously increased, for example, when the number of electrical stimulation is 100, the target current intensity is 3 mA, and the like. If the target current intensity exceeds the preset current intensity threshold, it indicates that the target current intensity needs to be controlled to ensure that the target current intensity is within the human acceptable 70 mA, and the preset current intensity threshold is taken as the target current intensity. It should be noted that in actual application, the current intensity of the electromyostimulation instrument and the training speed of the upper and lower limb rehabilitation training device are adjusted by sending Bluetooth instructions and USB transmission instructions.

[0055] Figure 5 is a schematic block diagram of a control device 200 for upper and lower limb muscle training provided by the embodiment of the present application. As shown in Figure 5 , corresponding to the above control method for upper and lower limb muscle training, the present application also provides a control device 200 for upper and lower limb muscle training. The control device 200 for upper and lower limb muscle training includes units for executing the above control method for upper and lower limb muscle training, and the device can be configured in a terminal. Specifically, please refer to Figure 5 , the control device 200 for upper and lower limb muscle training includes an extraction processing unit 201, a denoising fitting unit 202, a fatigue detection unit 203 and a control unit 204.

[0056] The extraction processing unit 201 is configured to perform feature extraction and normalization processing on the electromyographic signal to obtain electromyographic features if the electromyographic signal is received from the electromyographic stimulator.

[0057] In some embodiments, such as the present embodiment, the extraction processing unit 201 includes a feature extraction unit and a normalization processing unit.

[0058] The feature extraction unit is configured to perform time domain and frequency domain feature extraction on the electromyographic signal to obtain time domain electromyographic features and frequency domain electromyographic features.

[0059] In some embodiments, such as the present embodiment, the denoising fitting unit 202 includes a convolution unit, a pooling unit, and an output unit.

[0060] The convolution unit is configured to input the electromyographic features into the convolution layer to obtain first denoised electromyographic features.

[0061] In some embodiments, such as the present embodiment, the fatigue detection unit 203 includes a receiving transmission unit, a hidden output unit, an accepting output unit, and a fatigue detection subunit.

[0062] The receiving transmission unit is configured to receive the denoised electromyographic features by the input layer and transmit the denoised electromyographic features to the hidden layer; the hidden output unit is configured to obtain hidden output features by a preset hidden output feature formula according to the denoised electromyographic features by the hidden layer; the receiving output unit is configured to obtain receiving output features by the modified correlation function according to the hidden output features by the receiving layer; and the fatigue detection subunit is configured to obtain a fatigue detection result by a preset output formula according to the hidden output features and the receiving output features by the output layer.

[0063] In some embodiments, such as the present embodiment, the control unit 204 comprises a first adjusting unit and a second adjusting unit.

[0064] The first adjusting unit is configured to adjust the current intensity of the electromyographic stimulator by a current intensity adjusting method if the fatigue detection result is non-upper and lower limb muscle fatigue; and the second adjusting unit is configured to send a speed reduction instruction to the upper and lower limb rehabilitation training device to reduce the training speed of the upper and lower limb rehabilitation training device if the fatigue detection result is upper and lower limb muscle fatigue.

[0065] In some embodiments, such as the present embodiment, the first adjusting unit comprises a first judging unit, a first adjusting subunit, and a second adjusting subunit.

[0066] The first judging unit is configured to judge whether the number of electrical stimulation is a preset number of electrical stimulation; the first adjusting subunit is configured to send a current intensity instruction to the electromyographic stimulator according to a default electrical stimulation parameter to adjust the current intensity of the electromyographic stimulator if the number of electrical stimulation is the preset number of electrical stimulation; and the second adjusting subunit is configured to obtain a target current intensity in a current self-increasing manner according to an initial current intensity and a preset number of stimulations if the number of electrical stimulation is not the preset number of electrical stimulation, and send a current intensity instruction corresponding to the target current intensity to the electromyographic stimulator to adjust the current intensity of the electromyographic stimulator.

[0067] In some embodiments, such as the present embodiment, the second adjusting subunit comprises a self-increasing unit, a second judging unit, a third adjusting subunit, and a return execution unit.

[0068] The self-increasing unit is configured to obtain a target current intensity in a self-increasing manner according to an initial current intensity when the electric stimulation times are an integer multiple of the preset stimulation times; the second judging unit is configured to judge whether the target current intensity exceeds a preset current intensity threshold; the third adjusting subunit is configured to send a current intensity instruction corresponding to the target current intensity to the electromyostimulation instrument to adjust the current intensity of the electromyostimulation instrument if the target current intensity does not exceed the preset current intensity threshold, and return to execute the step of obtaining the target current intensity in a self-increasing manner according to the initial current intensity when the electric stimulation times are an integer multiple of the preset stimulation times; and the return execution unit is configured to take the preset current intensity threshold as the target current intensity if the target current intensity exceeds the preset current intensity threshold, and execute the step of sending the current intensity instruction corresponding to the target current intensity to the electromyostimulation instrument to adjust the current intensity of the electromyostimulation instrument.

[0069] The specific implementation of the control device 200 for upper and lower limb muscle training in the embodiment of the application corresponds to the control method for upper and lower limb muscle training, and will not be described here.

[0070] The control device for upper and lower limb muscle training can be implemented in the form of a computer program, which can run on a computer device as shown in the drawings. Figure 6

[0071] Please refer to Figure 6 , Figure 6 is a schematic block diagram of a computer device provided by the embodiment of the application. The computer device 300 is a terminal.

[0072] Please refer to Figure 6 , the computer device 300 includes a processor 302, a memory and a network interface 305 connected through a system bus 301, wherein the memory can include a storage medium 303 and an internal memory 304.

[0073] The storage medium 303 can store an operating system 3031 and a computer program 3032. The computer program 3032, when executed, can enable the processor 302 to execute a control method for upper and lower limb muscle training.

[0074] The processor 302 is configured to provide computing and control capabilities to support the operation of the entire computer device 300.

[0075] The internal memory 304 provides an environment for the execution of the computer program 3032 in the storage medium 303, and the computer program 3032, when executed by the processor 302, can enable the processor 302 to execute a control method for upper and lower limb muscle training.​

[0076] The network interface 305 is configured to perform network communication with other devices. Those skilled in the art can understand that the network interface 305 can be implemented by using a network card, a network adapter, or the like. Figure 6 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device 300 to which the scheme of the present application is applied. Specifically, the computer device 300 can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0077] The processor 302 is configured to run the computer program 3032 stored in the memory to implement the following steps: if an electromyographic signal sent by an electromyographic stimulator is received, performing feature extraction and normalization processing on the electromyographic signal to obtain an electromyographic feature; inputting the electromyographic feature into a preset denoising neural network to perform denoising processing to obtain a denoised electromyographic feature; modifying a correlation function in an Elman neural network receiving layer, and inputting the denoised electromyographic feature into the modified Elman neural network to perform upper and lower limb muscle fatigue detection to obtain a fatigue detection result; and adjusting a current intensity of the electromyographic stimulator and a training speed of an upper and lower limb rehabilitation trainer according to the fatigue detection result.

[0078] In some embodiments, for example in the present embodiment, when implementing the step of performing feature extraction and normalization processing on the electromyographic signal to obtain an electromyographic feature, the processor 302 specifically implements the following steps: performing time domain and frequency domain feature extraction on the electromyographic signal to obtain a time domain electromyographic feature and a frequency domain electromyographic feature; performing normalization processing on the time domain electromyographic feature and the frequency domain electromyographic feature to obtain a target time domain electromyographic feature and a target frequency domain electromyographic feature, and taking the target time domain electromyographic feature and the target frequency domain electromyographic feature as the electromyographic feature.

[0079] In some embodiments, for example in the present embodiment, when implementing the step of inputting the electromyographic feature into a preset denoising neural network to perform denoising processing to obtain a denoised electromyographic feature, the processor 302 specifically implements the following steps: inputting the electromyographic feature into the convolution layer to perform convolution to obtain a first denoised electromyographic feature; inputting the first denoised electromyographic feature into the pooling layer to perform pooling to obtain a second denoised electromyographic feature; inputting the second denoised electromyographic feature into the fully connected layer to output a third denoised electromyographic feature, and taking the third denoised electromyographic feature as the denoised electromyographic feature.

[0080] In some embodiments, such as the present embodiment, the processor 302, when implementing the step of inputting the denoised electromyography features into the modified Elman neural network to detect upper and lower limb muscle fatigue to obtain a fatigue detection result, specifically implements the following steps: the input layer receives the denoised electromyography features and transmits the denoised electromyography features to the hidden layer; the hidden layer obtains a hidden output feature according to the denoised electromyography features through a preset hidden output feature formula; the relay layer obtains a relay output feature according to the hidden output feature through the modified correlation function; and the output layer obtains a fatigue detection result according to the hidden output feature and the relay output feature through a preset output formula.

[0081] In some embodiments, such as the present embodiment, the processor 302, when implementing the step of adjusting the current intensity of the electromyography stimulator and the training speed of the upper and lower limb rehabilitation trainer according to the fatigue detection result, specifically implements the following steps: if the fatigue detection result is non-upper and lower limb muscle fatigue, determining whether the number of electrical stimulations is a preset number of electrical stimulations; if the number of electrical stimulations is the preset number of electrical stimulations, sending a current intensity instruction corresponding to a target current intensity to the electromyography stimulator to adjust the current intensity of the electromyography stimulator according to a default electrical stimulation parameter; if the number of electrical stimulations is not the preset number of electrical stimulations, obtaining a target current intensity in a self-increasing manner according to an initial current intensity when the number of electrical stimulations is an integer multiple of a preset number of stimulations; determining whether the target current intensity exceeds a preset current intensity threshold; if the target current intensity does not exceed the preset current intensity threshold, sending a current intensity instruction corresponding to the target current intensity to the electromyography stimulator to adjust the current intensity of the electromyography stimulator, and returning to execute the step of obtaining a target current intensity in a self-increasing manner according to an initial current intensity when the number of electrical stimulations is an integer multiple of a preset number of stimulations; if the target current intensity exceeds the preset current intensity threshold, taking the preset current intensity threshold as the target current intensity, and executing the step of sending a current intensity instruction corresponding to the target current intensity to the electromyography stimulator to adjust the current intensity of the electromyography stimulator; and if the fatigue detection result is upper and lower limb muscle fatigue, sending a speed reduction instruction to the upper and lower limb rehabilitation trainer to reduce the training speed of the upper and lower limb rehabilitation trainer.

[0082] It should be understood that, in the embodiments of the present application, the processor 302 can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0083] Those skilled in the art can understand that all or part of the processes in the method of the above-mentioned embodiments can be completed by instructing the relevant hardware by a computer program. The computer program can be stored in a storage medium, which is a computer-readable storage medium. The computer program is executed by at least one processor in the computer system to realize the process steps of the above-mentioned method embodiments.

[0084] Therefore, the present application also provides a storage medium. The storage medium can be a computer-readable storage medium. The storage medium stores a computer program. The computer program is executed by a processor to make the processor execute any embodiment of the above-mentioned control method for upper and lower limb muscle training.

[0085] The storage medium can be a U disk, a mobile hard disk, a read-only memory (ROM), a magnetic disk or an optical disk, and various computer-readable storage media that can store program codes.

[0086] Those skilled in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be realized in electronic hardware, computer software or a combination of both. In order to clearly illustrate the interchangeability of hardware and software, the components and steps of the examples have been described in the above description in general terms. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0087] In several embodiments provided by the present application, it should be understood that the disclosed apparatus and method can be implemented in other manners. For example, the described apparatus embodiments are merely schematic. For example, the division of the units is merely a logical function division. For example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In a possible implementation process, the steps of the described method can be performed in a different order, or can be omitted, or can be combined into another process, or can be implemented with other technical solutions.

[0088] The steps in the method embodiments of the present application can be adjusted, combined and deleted in sequence according to actual needs. The units in the apparatus embodiments of the present application can be combined, divided and deleted according to actual needs. In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit.

[0089] The integrated unit, if realized in the form of a software functional unit and sold or used as an independent product, can be stored in a storage medium. Based on such understanding, the technical solutions of the present application essentially or say the part that contributes to the prior art, or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a terminal or a network device, etc.) to execute all or part of the steps of the methods described in the embodiments of the present application.

[0090] In the above embodiments, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the relevant description of other embodiments.

[0091] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, these modifications and variations of the present application are intended to be included within the scope of the present application, and the present application is intended to include these modifications and variations.

[0092] The above description is merely specific embodiments of the present application, but the protection scope of the present application is not limited thereto. Any modifications or replacements within the technical scope disclosed by the present application can be easily thought by those skilled in the art, and these modifications or replacements should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A control device for upper and lower limb muscle training, characterized in that, The method comprises the following steps: An extraction processing unit is configured to extract and normalize the electromyography signal received from the electromyography stimulator to obtain electromyography features. A denoising fitting unit is configured to input the electromyography features into a preset denoising neural network to obtain denoised electromyography features. A fatigue detection unit is configured to modify a correlation function in an Elman neural network receiving layer and input the denoised electromyography features into the modified Elman neural network to detect upper and lower limb muscle fatigue to obtain a fatigue detection result. A control unit is configured to adjust the current intensity of the electromyography stimulator and the training speed of the upper and lower limb rehabilitation training device according to the fatigue detection result. The control unit comprises: A first judgment unit is configured to judge whether the number of electrical stimulation is a preset number of electrical stimulation if the fatigue detection result is non-upper and lower limb muscle fatigue. A first adjustment subunit is configured to send a current intensity instruction to the electromyography stimulator according to a default electrical stimulation parameter to adjust the current intensity of the electromyography stimulator if the number of electrical stimulation is the preset number of electrical stimulation. A second adjustment subunit is configured to obtain a target current intensity in a current self-increasing manner according to an initial current intensity and a preset number of stimulations if the number of electrical stimulation is not the preset number of electrical stimulation, and send a current intensity instruction corresponding to the target current intensity to the electromyography stimulator to adjust the current intensity of the electromyography stimulator. A second adjustment unit is configured to send a speed reduction instruction to the upper and lower limb rehabilitation training device to reduce the training speed of the upper and lower limb rehabilitation training device if the fatigue detection result is upper and lower limb muscle fatigue.

2. A computer device, comprising: The computer device comprises a memory and a processor, the memory stores a computer program, and the processor implements the functions of the device of claim 1 when executing the computer program.

3. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, and the processor implements the functions of the device of claim 1 when executing the computer program.

Images