Method for predicting same-side motion direction of different-side lower limb electromyographic signals and related device

Abstract

The invention provides a method for predicting the same-side motion direction of different-side lower limb electromyographic signals and a related device, and belongs to the technical field of human body biological signal processing and mode recognition. The method comprises the following steps: constructing a training set and a test set by utilizing corresponding combined characteristics of surface electromyogram signals of left and right lower limb muscles, left foot sole pressure and right foot three-dimensional acceleration; constructing a three-classification gait prediction model, training the three-classification gait prediction model based on the constructed training set, and testing a training result of the three-classification gait prediction model based on the constructed test set to obtain a trained three-classification gait prediction model; and inputting the corresponding combined features of the surface electromyogram signals of the left and right lower limb muscles to be predicted, the left foot sole pressure and the right foot three-dimensional acceleration into the trained three-classification gait prediction model for right foot motion direction prediction to obtain a right foot motion direction prediction result. According to the method, the problem that the prediction accuracy of the same-side motion direction of the different-side lower limb electromyographic signals is not high is solved.

Description

Technical Field

[0001] The invention belongs to the technical field of human biological signal processing and pattern recognition, and in particular relates to a method for predicting the same-side movement direction of opposite-side lower limb electromyographic signals and a related device. Background Art

[0002] According to relevant reports, as of 2015, the total number of disabled people in China exceeded 87 million, accounting for 6.4% of the total population. It is estimated that by 2050, the number of disabled people in my country will increase to 168 million, accounting for 11%. Liu Fengbo and others took adult disabled people in Henan Province as the research object and studied the rehabilitation needs and rehabilitation service development of adult disabled people. The results showed that limb disabilities accounted for 62.2% of adult disabled people, and the demand for assistive devices accounted for 49.0% of the rehabilitation needs, and the demand for functional training accounted for 20.2%. Therefore, it is urgent to design lower limb rehabilitation robots for motor dysfunction such as hemiplegia caused by stroke and spinal cord injury. Lower limb exoskeleton robots can not only help patients with motor dysfunction to walk and restore motor function, but also improve the motor ability of the elderly, and can also enhance the load strength and endurance of manual laborers and soldiers. Unlike intelligent prostheses, exoskeletons do not replace part of the human body, but need to be integrated with the human body to perceive the human body's movement intentions, so as to cooperate with the wearer to complete specific actions in a certain way. For lower limb exoskeletons, the analysis and recognition of lower limb movements is the key to sensing and predicting human movement intentions. Accurate perception of the wearer's movement intentions is the primary condition for the exoskeleton to cooperate with the wearer to complete the target movement, and is also a key technology to solve the problem of human-machine coordination.

[0003] The occurrence of lower limb movement in the human body is a complex neurophysiological process that requires the cooperation of the cerebral cortex, thalamus, basal ganglia, cerebellum, brainstem, spinal cord and muscles. The idea of ​​movement is first generated in the motor area of ​​the cerebral cortex, transmitted to the brainstem in the form of electrical signals, and the spinal cord issues movement commands, and then the conversion of electrical energy to mechanical energy is realized in the muscles, and movement is generated. At the same time, complex calculations are performed in the cerebral cortex, thalamus, basal ganglia and brainstem to regulate the generated movement, so that the behavior of the human body is more coordinated and accurate. Therefore, the analysis and identification of lower limb movement should not only consider the kinematic and dynamic parameters of the lower limbs, but also study the physiological signals of the motor cortex of the brain and the lower limb muscles. The patient controls the exoskeleton through his own intention. On the one hand, it can improve the patient's participation and enthusiasm, and on the other hand, it can also promote the plasticity of the brain. At present, the forms of movement information of lower limb exoskeletons include plantar pressure, acceleration, joint movement angle and surface electromyography (sEMG).

[0004] At present, only the Hybrid Assistive Limb (HAL) developed by the University of Tsukuba in Japan uses surface electromyographic signals of the lower limbs for lower limb movement control, and several lower limb exoskeleton products use buttons to control lower limb movement. The preset trajectory method cannot sense the wearer's movement intention, and the wearer needs to cooperate with the movement of the exoskeleton. Therefore, it is not suitable for patients with strong lower limb autonomous movement ability. The exoskeleton that detects the gait phase based on plantar pressure, joint angle and inertial measurement can detect the wearer's movement intention, but the gait it detects is the gait after the movement occurs, and it cannot predict the movement intention before the movement. On the one hand, the surface electromyographic signal occurs before the actual movement. Although it can be used to predict the movement intention, the muscle electrical signal on the moving side occurs almost at the same time as the actual movement, so the effect of advance prediction is poor. On the other hand, the gait recognition and movement intention perception technology based on sEMG is still in the experimental stage, and there are few lower limb exoskeleton products based on electromyography and other control methods on the market. Summary of the invention

[0005] The purpose of the present invention is to provide a method and related device for predicting the same-side movement direction of electromyographic signals of opposite-side lower limbs, so as to solve the problem of low accuracy in predicting the same-side movement direction of electromyographic signals of opposite-side lower limbs in the prior art.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: In a first aspect, the present invention provides a method for predicting the same-side movement direction of an opposite-side lower limb electromyographic signal, comprising the following steps: Acquire surface electromyographic signals of left and right lower limb muscles, left foot plantar pressure, and right foot three-dimensional acceleration; Resampling and preprocessing the obtained surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot, and the three-dimensional acceleration of the right foot to obtain preprocessed surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot, and the three-dimensional acceleration of the right foot; The pre-processed surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot are respectively extracted to obtain the time domain slope change characteristics and average absolute value characteristics corresponding to the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot; The obtained surface electromyographic signals of the left and right lower limb muscles, the time domain slope change characteristics and the mean absolute value characteristics corresponding to the left and right foot plantar pressure and the right foot three-dimensional acceleration are respectively combined to obtain the combined characteristics corresponding to the surface electromyographic signals of the left and right lower limb muscles, the left foot plantar pressure and the right foot three-dimensional acceleration; The training set and test set are constructed using the combined features of the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot, and the three-dimensional acceleration of the right foot; Constructing a three-classification gait prediction model, training the three-classification gait prediction model based on the constructed training set, testing the training result of the three-classification gait prediction model based on the constructed test set, and obtaining a trained three-classification gait prediction model; The combined features of the surface electromyographic signals of the left and right lower limb muscles to be predicted, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot are input into the trained three-classification gait prediction model to predict the movement direction of the right foot and obtain the prediction result of the movement direction of the right foot.

[0007] A further improvement of the present invention lies in that, in the step of resampling and preprocessing the obtained surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot to obtain the preprocessed surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot, the obtained surface electromyographic signals of the left and right lower limb muscles are also filtered and denoised preprocessed to obtain the surface electromyographic signals of the left and right lower limb muscles after filtering and denoising.

[0008] A further improvement of the present invention is that the surface electromyographic signals of the left and right lower limb muscles are filtered and pre-processed by bandpass filtering and wavelet denoising to obtain the surface electromyographic signals of the left and right lower limb muscles after filtering and denoising.

[0009] A further improvement of the present invention lies in that, in the step of respectively performing feature extraction on the preprocessed surface electromyographic signals, plantar pressure of the left foot and three-dimensional acceleration of the right foot to obtain the time domain slope change characteristics and average absolute value characteristics corresponding to the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot, a multi-scale time window method is specifically used to respectively perform feature extraction on the preprocessed surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot to obtain the time domain slope change characteristics and average absolute value characteristics corresponding to the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot.

[0010] A further improvement of the present invention lies in that, in the step of combining the obtained time domain slope change characteristics and mean absolute value characteristics corresponding to the surface electromyographic signals of the left and right lower limb muscles, the left foot plantar pressure and the right foot three-dimensional acceleration to obtain the combined characteristics corresponding to the surface electromyographic signals of the left and right lower limb muscles, the left foot plantar pressure and the right foot three-dimensional acceleration, the corresponding combined characteristics specifically obtained are characteristics in the form of vector groups.

[0011] A further improvement of the present invention is that 70% of the combined features corresponding to the surface electromyographic signals of the left and right lower limb muscles, the left foot plantar pressure and the right foot three-dimensional acceleration are used as a training set, and 30% are used as a test set.

[0012] A further improvement of the present invention is that, in the step of constructing a three-category gait prediction model, training the three-category gait prediction model based on the constructed training set to obtain the trained three-category gait prediction model, the three-category gait prediction model is specifically constructed using the LIBSVM toolbox.

[0013] In a second aspect, the present invention provides a system for predicting the same-side movement direction of the opposite-side lower limb electromyographic signal, including a data acquisition module, a data preprocessing module, a feature extraction module, a feature combination module, a data set construction module, a three-classification gait prediction model training module and a right foot movement direction prediction module; The data acquisition module is used to acquire surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot; The data preprocessing module is used to resample and preprocess the obtained surface electromyographic signals of the left and right lower limb muscles, the left foot plantar pressure and the right foot three-dimensional acceleration to obtain the preprocessed surface electromyographic signals of the left and right lower limb muscles, the left foot plantar pressure and the right foot three-dimensional acceleration; The feature extraction module is used to extract features of the preprocessed surface electromyographic signals of the left and right lower limb muscles, the left foot plantar pressure and the right foot three-dimensional acceleration, respectively, to obtain the time domain slope change features and average absolute value features corresponding to the surface electromyographic signals of the left and right lower limb muscles, the left foot plantar pressure and the right foot three-dimensional acceleration; The feature combination module is used to combine the obtained surface electromyographic signals of the left and right lower limb muscles, the time domain slope change characteristics and the mean absolute value characteristics corresponding to the left and right foot plantar pressure and the right foot three-dimensional acceleration, respectively, to obtain the combined characteristics corresponding to the surface electromyographic signals of the left and right lower limb muscles, the left foot plantar pressure and the right foot three-dimensional acceleration; The data set construction module is used to construct a training set and a test set using the combined features corresponding to the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot; The three-category gait prediction model training module is used to construct a three-category gait prediction model, train the three-category gait prediction model based on the constructed training set, and test the training result of the three-category gait prediction model based on the constructed test set to obtain a trained three-category gait prediction model; The right foot movement direction prediction module is used to input the combined features corresponding to the surface electromyographic signals of the left and right lower limb muscles to be predicted, the left foot plantar pressure and the right foot three-dimensional acceleration into the trained three-classification gait prediction model to predict the right foot movement direction and obtain the right foot movement direction prediction result.

[0014] In a third aspect, the present invention provides an electronic device including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the above-mentioned method for predicting the ipsilateral movement direction of the electromyographic signal of the contralateral lower limb are implemented.

[0015] In a fourth aspect, the present invention provides a storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the above-mentioned method for predicting the ipsilateral movement direction of the electromyographic signal of the contralateral lower limb.

[0016] Compared with the prior art, the present invention has the following beneficial effects: The present invention belongs to an improved invention. Compared with the existing method for predicting the same-side movement direction of the electromyographic signal of the contralateral lower limb, the method for predicting the same-side movement direction of the electromyographic signal of the contralateral lower limb proposed by the present invention, on the one hand, obtains the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot, which can more comprehensively understand the movement information of the human body during movement. On the other hand, the present invention uses the corresponding combined features (surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot) to predict the three movement directions (left, middle and right) of the right foot, improves the accuracy of predicting the right foot movement intention, provides a key technical foundation for the development of the right lower limb human-computer interaction system, and effectively solves the problem of low accuracy in predicting the same-side movement direction of the electromyographic signal of the contralateral lower limb in the prior art.

[0017] Furthermore, the present invention also discloses the step of combining the surface electromyographic signals of the left and right lower limb muscles, the time domain slope change characteristics and the average absolute value characteristics corresponding to the left and right lower limb pressure and the three-dimensional acceleration of the right foot, respectively, to obtain the combined characteristics corresponding to the surface electromyographic signals of the left and right lower limb muscles, the left foot pressure and the three-dimensional acceleration of the right foot, wherein the combined characteristics corresponding to the surface electromyographic signals of the left and right lower limb muscles, the left foot pressure and the three-dimensional acceleration of the right foot are characteristics in the form of vector groups. The characteristics in the form of vector groups contain rich motion feature information, such as pressure (especially the left foot pressure, which contains key information for predicting the right foot movement), acceleration and direction change rate, which are crucial for the later prediction of motion direction because they can fully and accurately describe the motion state of the human body.

[0018] Furthermore, the present invention also discloses the steps of constructing a three-classification gait prediction model, training the three-classification gait prediction model based on the constructed training set, and obtaining the trained three-classification gait prediction model, specifically using the LIBSVM toolbox to construct the three-classification gait prediction model. The LIBSVM toolbox has the ability to efficiently process high-dimensional data and has a strong generalization capability when establishing the three-classification gait prediction model, thereby improving the accuracy of the later motion direction prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a flow chart of the method for predicting the same-side movement direction of the electromyographic signal of the opposite-side lower limb of the present invention; Figure 2 It is a schematic diagram of the ipsilateral movement direction prediction system of the contralateral lower limb electromyographic signal of the present invention; Figure 3 This is a control principle diagram of the VR-based football game experimental paradigm of the present invention; Figure 4 This is a control flow chart of the VR-based football game of the present invention; Figure 5 It is a schematic diagram of the structure of the electronic device of the present invention. DETAILED DESCRIPTION

[0020] In order to further understand the content of the present invention, the present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the embodiments are only for explaining the present invention and are not intended to limit it.

[0021] The method for predicting the same-side movement direction of the electromyographic signal of the opposite lower limb proposed by the present invention inputs the combined features corresponding to the surface electromyographic signals of the left and right lower limb muscles to be predicted, the plantar pressure of the left foot, and the three-dimensional acceleration of the right foot into a trained three-classification gait prediction model to predict the movement direction of the right foot, and obtains the prediction result of the movement direction of the right foot. Compared with the prior art, the present invention effectively solves the problem of low accuracy in predicting the same-side movement direction of the electromyographic signal of the opposite lower limb.

[0022] Embodiment 1: The flowchart of the method for predicting the same-side movement direction of the opposite-side lower limb electromyographic signal of the present invention is as follows: Figure 1 As shown, the method for predicting the same-side movement direction of the electromyographic signal of the opposite-side lower limb of the present invention comprises the following steps: S1. Obtain surface electromyographic signals of the left and right lower limb muscles, plantar pressure of the left foot, and three-dimensional acceleration of the right foot.

[0023] S2. resample and preprocess the obtained surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot, and the three-dimensional acceleration of the right foot to obtain the preprocessed surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot, and the three-dimensional acceleration of the right foot.

[0024] S3. Feature extraction is performed on the preprocessed surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot, respectively, to obtain the time domain slope change characteristics and average absolute value characteristics corresponding to the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot.

[0025] S4. Combine the obtained time domain slope change characteristics and mean absolute value characteristics corresponding to the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot, respectively, to obtain the combined characteristics corresponding to the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot.

[0026] S5. The training set and test set are constructed using the combined features corresponding to the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot, and the three-dimensional acceleration of the right foot.

[0027] S6. Construct a three-category gait prediction model, train the three-category gait prediction model based on the constructed training set, test the training result of the three-category gait prediction model based on the constructed test set, and obtain a trained three-category gait prediction model.

[0028] S7. Input the combined features corresponding to the surface electromyographic signals of the left and right lower limb muscles to be predicted, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot into the trained three-category gait prediction model to predict the movement direction of the right foot, and obtain the prediction result of the movement direction of the right foot.

[0029] Embodiment 2: The schematic diagram of the ipsilateral movement direction prediction system of the contralateral lower limb electromyographic signal of the present invention is shown in FIG. Figure 2 As shown, the ipsilateral movement direction prediction system of the contralateral lower limb electromyographic signal of the present invention includes a data acquisition module, a data preprocessing module, a feature extraction module, a feature combination module, a data set construction module, a three-classification gait prediction model training module and a right foot movement direction prediction module.

[0030] The data acquisition module is used to obtain the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot.

[0031] The data preprocessing module is used to resample and preprocess the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot to obtain the preprocessed surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot.

[0032] The feature extraction module is used to extract features of the preprocessed surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot, respectively, to obtain the time domain slope change characteristics and average absolute value characteristics corresponding to the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot.

[0033] The feature combination module is used to combine the time domain slope change characteristics and mean absolute value characteristics corresponding to the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot, respectively, to obtain the combined characteristics corresponding to the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot.

[0034] The dataset construction module is used to construct training and test sets using the combined features corresponding to the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot, and the three-dimensional acceleration of the right foot.

[0035] The three-category gait prediction model training module is used to construct a three-category gait prediction model, train the three-category gait prediction model based on the constructed training set, and test the three-category gait prediction model training result based on the constructed test set to obtain a trained three-category gait prediction model.

[0036] The right foot movement direction prediction module is used to input the combined features corresponding to the surface electromyographic signals of the left and right lower limb muscles to be predicted, the left foot plantar pressure and the right foot three-dimensional acceleration into the trained three-classification gait prediction model to predict the right foot movement direction and obtain the right foot movement direction prediction result.

[0037] Embodiment 3: S1. Obtain surface electromyographic signals of the left and right lower limb muscles, plantar pressure of the left foot, and three-dimensional acceleration of the right foot.

[0038] First, the surface electromyography signals (sEMG) of the muscles of the left and right lower limbs (the muscle channels used in this embodiment include the biceps brachii, vastus medialis, gastrocnemius and tibialis anterior of the left and right legs), the plantar pressure of the left foot and the three-dimensional acceleration of the right foot. The full name of sEMG is Surface Electromyography, which means surface electromyography in Chinese.

[0039] In this step, the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot are obtained through the OPEN BCI wireless signal acquisition system, the motion signal wireless acquisition device Shadow and the Vive external head-mounted device.

[0040] OPEN BCI wireless signal acquisition system, sampling frequency is 500Hz. Each plantar pressure insole includes two sensors, one for the toe and the other for the heel. The plantar pressure insole is connected to the motion signal wireless acquisition device Shadow. The pressure signal is sent to the software system of the motion signal wireless acquisition device Shadow via the WiFi of the motion signal wireless acquisition device Shadow.

[0041] Shadow, a motion signal wireless acquisition device, is a motion capture system consisting of two parts: hardware and software. It can measure the kinematic data of seven positions: waist, thigh, calf and foot, through a three-axis accelerometer, gyroscope and magnetometer, with a sampling frequency of 100 Hz.

[0042] S2. resample and preprocess the obtained surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot, and the three-dimensional acceleration of the right foot to obtain the preprocessed surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot, and the three-dimensional acceleration of the right foot.

[0043] The obtained surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot were resampled and preprocessed to obtain the preprocessed surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot. The resampling frequency was 500 Hz.

[0044] In this step, the obtained surface electromyographic signals of the left and right lower limb muscles are also filtered and denoised to obtain the surface electromyographic signals of the left and right lower limb muscles after filtering and denoising.

[0045] In this step, bandpass filtering (specifically, a 10-350 Hz Elliptic bandpass filter is used for bandpass filtering) and wavelet denoising are used to filter and denoise the surface electromyographic signals of the left and right lower limb muscles to obtain the surface electromyographic signals of the left and right lower limb muscles after filtering and denoising.

[0046] S3. Feature extraction is performed on the preprocessed surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot, respectively, to obtain the time domain slope change characteristics and average absolute value characteristics corresponding to the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot.

[0047] In this step, the multi-scale time window method is used to extract the features of the pre-processed surface electromyographic signals of the left and right lower limb muscles, the left foot plantar pressure and the right foot three-dimensional acceleration, and the time domain slope change (SSC) features and mean absolute value (MAV) features corresponding to the surface electromyographic signals of the left and right lower limb muscles, the left foot plantar pressure and the right foot three-dimensional acceleration are obtained. The length of the time window is set to 100ms, 250ms and 500ms respectively.

[0048] S4. Combine the obtained time domain slope change characteristics and mean absolute value characteristics corresponding to the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot to obtain the combined characteristics corresponding to the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot.

[0049] The combined features corresponding to the surface electromyographic signals of the left and right lower limb muscles, the left foot plantar pressure and the right foot three-dimensional acceleration obtained in this step are features in the form of a vector group.

[0050] S5. The training set and test set are constructed using the combined features corresponding to the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot, and the three-dimensional acceleration of the right foot.

[0051] The training set and the test set are constructed using the combined features corresponding to the surface electromyography signals of the left and right lower limb muscles, the plantar pressure of the left foot, and the three-dimensional acceleration of the right foot. 70% of the combined features corresponding to the surface electromyography signals of the left and right lower limb muscles, the plantar pressure of the left foot, and the three-dimensional acceleration of the right foot are used as the training set, and 30% are used as the test set. As other implementations, 70% and 30% in this step can be adjusted according to actual needs, for example, 80% of the combined features corresponding to the surface electromyography signals of the left and right lower limb muscles, the plantar pressure of the left foot, and the three-dimensional acceleration of the right foot are used as the training set, and 20% are used as the test set.

[0052] S6. Construct a three-category gait prediction model, train the three-category gait prediction model based on the constructed training set, test the training result of the three-category gait prediction model based on the constructed test set, and obtain a trained three-category gait prediction model.

[0053] In this step, the LIBSVM toolbox is used to construct a three-category gait prediction model, the three-category gait prediction model is trained based on the constructed training set, and the training result of the three-category gait prediction model is tested based on the constructed test set to obtain a trained three-category gait prediction model.

[0054] S7. Input the combined features corresponding to the surface electromyographic signals of the left and right lower limb muscles to be predicted, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot into the trained three-category gait prediction model to predict the movement direction of the right foot, and obtain the prediction result of the movement direction of the right foot.

[0055] In order to verify the effectiveness of the method for predicting the ipsilateral movement direction of the contralateral lower limb electromyographic signal proposed in the present invention, this embodiment first designs a VR-based kicking game experimental paradigm, and then uses the method proposed in the present invention to predict the ipsilateral movement direction of the contralateral lower limb electromyographic signal. The control principle of the VR-based kicking game experimental paradigm and the control process of the VR-based kicking game are as follows: Figure 3 and Figure 4 As shown in the figure, the specific control principle and control process of the VR-based football game experimental paradigm design are described as follows: The game includes three lower limb movements: kicking to the left, kicking to the right, and kicking to the middle. The game process includes three processes: stimulus presentation, signal processing, and game control. The three processes cooperate with each other and communicate through TCP / IP at a specific time to enable game players to actively control virtual devices through their own bioelectric signals. The entire game process needs to be completed on two computers, among which system stimulus presentation, sEMG and motion data acquisition and synchronization are completed in computer 2, and signal processing and VR game control are completed in computer 1. After the game is started, the pictures and voice prompts of the action stimuli in the experimental paradigm are presented on the display of computer 2 and the VR game model of computer 1 through the human-computer interface, and the collected sEMG and motion signals are sent to computer 2 via WiFi and signal synchronization is performed. Signal processing module: Computer 1 receives synchronized data through TCP / IP protocol. When the stimulation duration exceeds 500 ms and the acceleration of the swinging foot (also called the right foot) is greater than the set threshold (set according to actual needs), it starts decoding the movement intention and converts the decoding result into a game control instruction, which is sent to Computer 2 through TCP / IP protocol and then sent to Unity by Computer 2. Game control module: When Unity detects the movement instruction, it controls the VR game according to the instruction type. When the number of stimulations reaches the set data value, the stimulation presentation module, signal processing module and game module stop running. During the game, the action within 500 ms after the instruction is issued may be a false action, so the data starts to be processed after 500 ms. Secondly, if the acceleration of the swinging foot is not detected to be greater than the set threshold within 3000 ms after the movement instruction is issued, it is considered that the subject has not made the corresponding action, and no judgment is made. The current task ends and waits for the next task to start.

[0056] The prediction method proposed in the present invention is verified based on the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot. The prediction results 100ms before the movement occur are shown in Table 1.

[0057] Table 1 Prediction results 100ms before movement

[0058] The data in Table 1 show that the prediction method proposed in the present invention can predict the movement direction of the right foot more accurately, which proves the effectiveness of the prediction method proposed in the present invention.

[0059] Embodiment 4: See also Figure 5As shown, the present invention also provides an electronic device 100 for predicting the ipsilateral movement direction of the electromyographic signal of the opposite lower limb; the electronic device 100 includes a memory 101, at least one processor 102, a computer program 103 stored in the memory 101 and executable on the at least one processor 102, and at least one communication bus 104.

[0060] The memory 101 can be used to store the computer program 103, and the processor 102 implements the steps of the method for predicting the same-side movement direction of the electromyographic signal of the opposite lower limb described in Example 1 by running or executing the computer program stored in the memory 101 and calling the data stored in the memory 101. The memory 101 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data (such as audio data) created according to the use of the electronic device 100, etc. In addition, the memory 101 can include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other non-volatile solid-state storage devices.

[0061] The at least one processor 102 may be a central processing unit (CPU), or 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 processor 102 may be a microprocessor or any conventional processor, etc. The processor 102 is the control center of the electronic device 100, and uses various interfaces and lines to connect various parts of the entire electronic device 100.

[0062] The memory 101 in the electronic device 100 stores a plurality of instructions to implement a method for predicting the same-side movement direction of the electromyographic signal of the opposite-side lower limb, and the processor 102 can execute the plurality of instructions to implement: Acquire surface electromyographic signals of left and right lower limb muscles, left foot plantar pressure, and right foot three-dimensional acceleration; Resampling and preprocessing the obtained surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot, and the three-dimensional acceleration of the right foot to obtain preprocessed surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot, and the three-dimensional acceleration of the right foot; The pre-processed surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot are respectively extracted to obtain the time domain slope change characteristics and average absolute value characteristics corresponding to the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot; The obtained surface electromyographic signals of the left and right lower limb muscles, the time domain slope change characteristics and the mean absolute value characteristics corresponding to the left and right foot plantar pressure and the right foot three-dimensional acceleration are respectively combined to obtain the combined characteristics corresponding to the surface electromyographic signals of the left and right lower limb muscles, the left foot plantar pressure and the right foot three-dimensional acceleration; The training set and test set are constructed using the combined features of the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot, and the three-dimensional acceleration of the right foot; Constructing a three-classification gait prediction model, training the three-classification gait prediction model based on the constructed training set, testing the training result of the three-classification gait prediction model based on the constructed test set, and obtaining a trained three-classification gait prediction model; The combined features of the surface electromyographic signals of the left and right lower limb muscles to be predicted, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot are input into the trained three-classification gait prediction model to predict the movement direction of the right foot and obtain the prediction result of the movement direction of the right foot.

[0063] Embodiment 5: If the module / unit integrated in the electronic device 100 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the process in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device that can carry the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory and read-only memory (ROM, Read-Only Memory).

[0064] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0065] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0066] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0067] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0068] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for predicting the same-side movement direction of the opposite-side lower limb electromyographic signal, characterized in that: The following steps are involved: Acquire surface electromyographic signals of left and right lower limb muscles, left foot plantar pressure, and right foot three-dimensional acceleration; Resampling and preprocessing the obtained surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot, and the three-dimensional acceleration of the right foot to obtain preprocessed surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot, and the three-dimensional acceleration of the right foot; The pre-processed surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot are respectively extracted to obtain the time domain slope change characteristics and average absolute value characteristics corresponding to the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot; The obtained surface electromyographic signals of the left and right lower limb muscles, the time domain slope change characteristics and the mean absolute value characteristics corresponding to the left and right foot plantar pressure and the right foot three-dimensional acceleration are respectively combined to obtain the combined characteristics corresponding to the surface electromyographic signals of the left and right lower limb muscles, the left foot plantar pressure and the right foot three-dimensional acceleration; The training set and test set are constructed using the combined features of the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot, and the three-dimensional acceleration of the right foot; Constructing a three-classification gait prediction model, training the three-classification gait prediction model based on the constructed training set, testing the training result of the three-classification gait prediction model based on the constructed test set, and obtaining a trained three-classification gait prediction model; The combined features of the surface electromyographic signals of the left and right lower limb muscles to be predicted, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot are input into the trained three-classification gait prediction model to predict the movement direction of the right foot and obtain the prediction result of the movement direction of the right foot.

2. The method for predicting the same-side movement direction of the opposite-side lower limb electromyographic signal according to claim 1, characterized in that: In the step of resampling and preprocessing the obtained surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot to obtain the preprocessed surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot, the obtained surface electromyographic signals of the left and right lower limb muscles are also filtered and denoised to obtain the surface electromyographic signals of the left and right lower limb muscles after filtering and denoising.

3. The method for predicting the same-side movement direction of the opposite-side lower limb electromyographic signal according to claim 2, characterized in that: Specifically, bandpass filtering and wavelet denoising are used to perform filtering and denoising preprocessing on the surface electromyographic signals of the left and right lower limb muscles, so as to obtain the surface electromyographic signals of the left and right lower limb muscles after filtering and denoising.

4. The method for predicting the same-side movement direction of the opposite-side lower limb electromyographic signal according to claim 1, characterized in that: In the step of respectively extracting features from the pre-processed surface electromyographic signals of the left and right lower limb muscles, the left foot plantar pressure and the right foot three-dimensional acceleration to obtain the time domain slope change features and the average absolute value features corresponding to the surface electromyographic signals of the left and right lower limb muscles, the left foot plantar pressure and the right foot three-dimensional acceleration, a multi-scale time window method is specifically used to respectively extract features from the pre-processed surface electromyographic signals of the left and right lower limb muscles, the left foot plantar pressure and the right foot three-dimensional acceleration to obtain the time domain slope change features and the average absolute value features corresponding to the surface electromyographic signals of the left and right lower limb muscles, the left foot plantar pressure and the right foot three-dimensional acceleration.

5. The method for predicting the same-side movement direction of the opposite-side lower limb electromyographic signal according to claim 1, characterized in that: In the step of combining the obtained time domain slope change characteristics and mean absolute value characteristics corresponding to the surface electromyographic signals of the left and right lower limb muscles, the left foot plantar pressure and the right foot three-dimensional acceleration, respectively, to obtain the combined characteristics corresponding to the surface electromyographic signals of the left and right lower limb muscles, the left foot plantar pressure and the right foot three-dimensional acceleration, the specifically obtained combined characteristics corresponding to the surface electromyographic signals of the left and right lower limb muscles, the left foot plantar pressure and the right foot three-dimensional acceleration are characteristics in the form of vector groups.

6. The method for predicting the same-side movement direction of the opposite-side lower limb electromyographic signal according to claim 1, characterized in that: 70% of the combined features corresponding to the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot, and the three-dimensional acceleration of the right foot are used as the training set, and 30% are used as the test set.

7. The method for predicting the same-side movement direction of the opposite-side lower limb electromyographic signal according to claim 1, characterized in that: In the step of constructing a three-category gait prediction model, training the three-category gait prediction model based on the constructed training set, and obtaining the trained three-category gait prediction model, the three-category gait prediction model is specifically constructed using the LIBSVM toolbox.

8. A system for predicting the same-side movement direction of the opposite-side lower limb electromyographic signal, characterized in that: It includes a data acquisition module, a data preprocessing module, a feature extraction module, a feature combination module, a data set construction module, a three-classification gait prediction model training module and a right foot movement direction prediction module; The data acquisition module is used to acquire surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot; The data preprocessing module is used to resample and preprocess the obtained surface electromyographic signals of the left and right lower limb muscles, the left foot plantar pressure and the right foot three-dimensional acceleration to obtain the preprocessed surface electromyographic signals of the left and right lower limb muscles, the left foot plantar pressure and the right foot three-dimensional acceleration; The feature extraction module is used to extract features of the preprocessed surface electromyographic signals of the left and right lower limb muscles, the left foot plantar pressure and the right foot three-dimensional acceleration, respectively, to obtain the time domain slope change features and average absolute value features corresponding to the surface electromyographic signals of the left and right lower limb muscles, the left foot plantar pressure and the right foot three-dimensional acceleration; The feature combination module is used to combine the obtained surface electromyographic signals of the left and right lower limb muscles, the time domain slope change characteristics and the mean absolute value characteristics corresponding to the left and right foot plantar pressure and the right foot three-dimensional acceleration, respectively, to obtain the combined characteristics corresponding to the surface electromyographic signals of the left and right lower limb muscles, the left foot plantar pressure and the right foot three-dimensional acceleration; The data set construction module is used to construct a training set and a test set using the combined features corresponding to the surface electromyographic signals of the left and right lower limb muscles, the plantar pressure of the left foot and the three-dimensional acceleration of the right foot; The three-category gait prediction model training module is used to construct a three-category gait prediction model, train the three-category gait prediction model based on the constructed training set, and test the training result of the three-category gait prediction model based on the constructed test set to obtain a trained three-category gait prediction model; The right foot movement direction prediction module is used to input the combined features corresponding to the surface electromyographic signals of the left and right lower limb muscles to be predicted, the left foot plantar pressure and the right foot three-dimensional acceleration into the trained three-classification gait prediction model to predict the right foot movement direction and obtain the right foot movement direction prediction result.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for predicting the ipsilateral movement direction of the electromyographic signal of the contralateral lower limb according to any one of claims 1 to 7 are implemented.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for predicting the ipsilateral movement direction of the electromyographic signal of the opposite lower limb according to any one of claims 1 to 7 are implemented.

Images