CN117562560A

CN117562560A - Exercise effect evaluation method, device and storage medium in rehabilitation training

Info

Publication number: CN117562560A
Application number: CN202311548911.3A
Authority: CN
Inventors: 纪倩雯
Original assignee: Beijing Yishan Medical Technology Co ltd
Current assignee: Beijing Yishan Medical Technology Co ltd
Priority date: 2023-11-20
Filing date: 2023-11-20
Publication date: 2024-02-20
Anticipated expiration: 2043-11-20
Also published as: CN117562560B

Abstract

The invention discloses a motion effect evaluation method, a device and a storage medium in rehabilitation training, which are characterized in that the surface electromyographic signals of a designated part of a human body in the rehabilitation training process are collected, and the effective electromyographic signals corresponding to the rehabilitation action of the human body are extracted from the collected electromyographic signals by processing means such as denoising, effective signal extraction and the like; then, the myoelectric time domain characteristics of the appointed part of the human body are obtained by carrying out time domain analysis processing on the effective myoelectric signals; finally, the motion evaluation result of the appointed part of the human body can be obtained through myoelectric time domain characteristics; therefore, the invention does not need to manually participate in the motion evaluation process in the rehabilitation training, improves the evaluation accuracy and efficiency compared with the traditional technology, and is suitable for large-scale application and popularization in the rehabilitation medical field.

Description

Exercise effect evaluation method, device and storage medium in rehabilitation training

Technical Field

The invention belongs to the technical field of data processing, and particularly relates to a method and a device for evaluating a movement effect in rehabilitation training and a storage medium.

Background

In recent years, with the development of rehabilitation medicine, the detection of the motion state and effect of a patient in the rehabilitation training process is gaining more and more importance, which can help doctors to determine the rehabilitation state of the patient and make personalized rehabilitation schemes for the patient, thereby achieving the purpose of promoting the rehabilitation process of the patient; at present, the traditional assessment of the rehabilitation exercise training state and effect of a patient mainly depends on subjective experience of doctors, and the assessment result has strong subjectivity, and meanwhile, the assessment process is complex and tedious, and the level requirement on rehabilitation doctors is high; therefore, the traditional assessment of rehabilitation exercise training effect has the problems of inaccurate assessment and low efficiency; based on this, how to provide an evaluation method which is not affected by subjective experience and has high evaluation efficiency has become a problem to be solved.

Disclosure of Invention

The invention aims to provide a motion effect evaluation method, a motion effect evaluation device and a storage medium in rehabilitation training, which are used for solving the problems of low accuracy and low efficiency existing in the prior art that motion effects are evaluated manually.

In order to achieve the above purpose, the present invention adopts the following technical scheme:

in a first aspect, a method for evaluating exercise effects in rehabilitation training is provided, including:

Acquiring a surface electromyographic signal of a designated part of a target person in the rehabilitation training process;

denoising the surface electromyographic signals to obtain denoised electromyographic signals;

dividing the denoising electromyographic signal to obtain at least one effective signal segment in the denoising electromyographic signal, wherein each effective signal segment in the at least one effective signal segment comprises a motion unit action potential waveform;

performing time domain analysis processing on each effective signal segment in the at least one effective signal segment to obtain time domain characteristics of each effective signal segment;

and determining a motion evaluation result of the appointed position of the target person in the rehabilitation training process according to the time domain characteristics of each effective signal segment.

Based on the above disclosure, the invention acquires the surface electromyographic signals of the appointed part of the target person in the rehabilitation training process, and then obtains the motion evaluation result by analyzing the surface electromyographic signals; the method comprises the steps of firstly carrying out denoising treatment on the collected surface electromyographic signals to remove interference signals in the surface electromyographic signals, thereby reducing the influence of the interference signals on signal analysis; meanwhile, as the myoelectric signal before and after the rehabilitation action of the human body is also mixed in the complete myoelectric signal, the invention also extracts the effective signal section containing the action potential waveform of the motion unit from the collected surface myoelectric signal in order to improve the analysis accuracy; thus, the effective electromyographic signals (i.e. the electromyographic signals generated from the action start to the action end) generated when the human body makes each complete rehabilitation action are extracted from the surface electromyographic signals, and the electromyographic signals generated when other non-rehabilitation actions are removed; finally, performing time domain analysis processing on each extracted effective signal to obtain a motion evaluation result of the appointed position of the target person in the rehabilitation training process.

Through the design, the invention extracts the effective electromyographic signals corresponding to the rehabilitation actions of the human body from the collected electromyographic signals by collecting the surface electromyographic signals of the appointed part of the human body in the rehabilitation training process and through the processing means such as noise removal, effective signal extraction and the like; then, the myoelectric time domain characteristics of the appointed part of the human body are obtained by carrying out time domain analysis processing on the effective myoelectric signals; finally, the motion evaluation result of the appointed part of the human body can be obtained through myoelectric time domain characteristics; therefore, the invention does not need to manually participate in the motion evaluation process in the rehabilitation training, improves the evaluation accuracy and efficiency compared with the traditional technology, and is suitable for large-scale application and popularization in the rehabilitation medical field.

In one possible design, denoising the surface electromyographic signal to obtain a denoised electromyographic signal, including:

performing initial denoising treatment on the surface electromyographic signals to obtain initial denoising electromyographic signals;

performing empirical mode decomposition processing on the initial denoising electromyographic signals to obtain a plurality of first IMF components;

performing signal reconstruction processing by using the plurality of first IMF components to obtain a reconstructed initial denoising electromyographic signal;

And carrying out secondary denoising treatment on the reconstructed initial denoising electromyographic signal to obtain the denoising electromyographic signal after the secondary denoising treatment.

In one possible design, performing initial denoising processing on the surface electromyographic signal to obtain an initial denoised electromyographic signal, including:

performing windowed Fourier transform and Fourier transform on the surface electromyographic signals respectively to obtain a first amplitude spectrum and a second amplitude spectrum of the surface electromyographic signals;

acquiring first amplitude spectrum data at a position adjacent to a target position from the first amplitude spectrum, wherein the target position is a harmonic position in the first amplitude spectrum and a position at a preset threshold frequency;

interpolation is carried out to obtain actual amplitude spectrum data at the target position by utilizing the first amplitude spectrum data;

replacing the amplitude spectrum data at the target position in the second amplitude spectrum by using the actual amplitude spectrum data so as to obtain a processed second amplitude spectrum after replacement;

performing inverse Fourier transform processing on the processed second amplitude spectrum to obtain a preprocessed surface electromyographic signal;

sequentially performing open operation and close operation on the surface electromyographic signals after pretreatment to obtain a first electromyographic signal, and sequentially performing close operation and open operation on the surface electromyographic signals after pretreatment to obtain a second electromyographic signal;

Based on the first electromyographic signal and the second electromyographic signal, obtaining a baseline drift value of the preprocessed surface electromyographic signal;

and eliminating baseline drift in the preprocessed surface electromyographic signals by using the baseline drift value so as to obtain the initial denoising electromyographic signals.

In one possible design, the performing signal reconstruction processing using the plurality of first IMF components to obtain a reconstructed initial denoising electromyographic signal includes:

for a first IMF component in a plurality of first IMF components, performing multiple signal reconstruction on the first IMF component to obtain a plurality of reconstructed IMF components, wherein any one of the signal reconstruction of the first IMF component refers to randomly sequencing signal values at different moments in the first IMF component;

the average IMF component of the plurality of reconstructed IMF components is obtained, and the reconstructed initial denoising electromyographic signals are formed by utilizing a target IMF component and the average IMF component, wherein the target IMF component is all first IMF components except the first IMF component in the plurality of first IMF components;

correspondingly, performing secondary denoising processing on the reconstructed initial denoising electromyographic signal to obtain the denoising electromyographic signal after the secondary denoising processing, and then comprising the following steps:

Performing empirical mode decomposition on the reconstructed initial denoising electromyographic signals to obtain a plurality of second IMF components;

calculating an autocorrelation function of each second IMF component, and carrying out normalization processing on each autocorrelation function to obtain a normalized autocorrelation function of each second IMF component;

solving the variance of the normalized autocorrelation function of each second IMF component, and classifying the plurality of second IMF components based on the variance of each normalized autocorrelation function to obtain a high-noise IMF component and a low-noise IMF component;

performing wavelet denoising treatment on each high-noise IMF component to obtain each denoised IMF component;

and carrying out signal reconstruction processing by utilizing the denoised IMF component and the low-noise IMF component so as to obtain the denoised electromyographic signal after the signal reconstruction processing.

In one possible design, when any high-noise IMF component performs wavelet denoising processing, a transformation formula of a wavelet coefficient of the any high-noise IMF component is:

in the above formula (1), w _j,k Original wavelet coefficients, w ', representing a kth wavelet component of the arbitrary high noise IMF component at a jth scale' _j,k Represents any one of theTransformed wavelet coefficients, y, of the kth wavelet component of the high noise IMF component at the jth scale _j,k A denoising threshold representing a kth wavelet component of the any high noise IMF component at a jth scale,represents a denoising parameter, where j=1, 2, L, k=1, 2, D, and L represents the wavelet denoising decomposition total scale number of the any high noise IMF component, D is the total number of wavelet components of any high-noise IMF component under the j-th scale;

wherein y is _j,k The method is calculated by adopting the following formula (2);

in the above formula (2), media (|X) ⁱ _IMF I) represents the absolute median of any high-noise IMF component, N represents the length of the reconstructed initial denoising electromyographic signal, a represents an adjustable parameter, and u is a constant.

In one possible design, the denoising electromyographic signal is subjected to segmentation processing to segment at least one valid signal segment in the denoising electromyographic signal, including:

carrying out peak enhancement processing on the denoising electromyographic signals so as to obtain enhanced electromyographic signals after the peak enhancement processing;

performing smoothing filtering treatment on the enhanced electromyographic signals to obtain filtered electromyographic signals;

counting all maximum signal points in the filtered electromyographic signals, and calculating an amplitude threshold value based on all the counted maximum signal points;

For a p-th maximum signal point in the filtered electromyographic signal, acquiring a detection window of the p-th maximum signal point, wherein the center of the detection window is the p-th maximum signal point, and the width is e;

acquiring a minimum value in the detection window, and judging whether the minimum value is smaller than the amplitude threshold value;

if yes, the signal in the detection window is used as an effective signal segment;

and adding 1 to p, and re-acquiring a detection window of the p-th maximum signal point until p is equal to m, so as to obtain at least one effective signal segment in the denoising electromyographic signal, wherein the initial value of p is 1, and m is the total number of the maximum signal points in the filtering electromyographic signal.

In one possible design, the peak enhancement processing is performed on the denoising electromyographic signal to obtain an enhanced electromyographic signal after the peak enhancement processing, including:

carrying out peak enhancement processing on the denoising electromyographic signals by adopting the following formula (3) so as to obtain enhanced electromyographic signals after the peak enhancement processing;

z[q(t)]＝q(t)×q(t+g-2)-q(t+g-1)q(t-1) (3)

in the above formula (3), q (t) represents the denoising electromyographic signal, g represents an enhancement coefficient, z [ q (t) ] represents the enhancement electromyographic signal, and t represents a sampling time in the denoising electromyographic signal;

Correspondingly, the method further comprises the steps of:

if not, acquiring a minimum value in the detection window, and taking the average value between the minimum value and the p-th maximum signal point as a center point;

and updating the p-th maximum signal point to the center point, and reacquiring a detection window of the p-th maximum signal point until the minimum value in the detection window is smaller than the amplitude threshold value.

In a second aspect, there is provided a sports effect assessment device in rehabilitation training, comprising:

the acquisition unit is used for acquiring the surface electromyographic signals of the appointed part of the target person in the rehabilitation training process;

the denoising unit is used for denoising the surface electromyographic signals to obtain denoised electromyographic signals;

the signal segmentation unit is used for carrying out segmentation processing on the denoising electromyographic signals so as to segment at least one effective signal segment in the denoising electromyographic signals, wherein each effective signal segment in the at least one effective signal segment comprises a motion unit action potential waveform;

the signal analysis unit is used for carrying out time domain analysis processing on each effective signal segment in the at least one effective signal segment so as to obtain the time domain characteristics of each effective signal segment;

And the motion evaluation unit is used for determining a motion evaluation result of the appointed position of the target person in the rehabilitation training process according to the time domain characteristics of each effective signal segment.

In a third aspect, another exercise effect evaluation device in rehabilitation training is provided, taking the device as an example of an electronic device, where the device includes a memory, a processor and a transceiver that are sequentially communicatively connected, where the memory is configured to store a computer program, the transceiver is configured to send and receive a message, and the processor is configured to read the computer program, and execute the exercise effect evaluation method in rehabilitation training as in the first aspect or any one of the possible designs of the first aspect.

In a fourth aspect, there is provided a storage medium having instructions stored thereon which, when executed on a computer, perform a method of assessing athletic performance in the rehabilitation training as in the first aspect or any one of the possible designs of the first aspect.

In a fifth aspect, there is provided a computer program product comprising instructions which, when run on a computer, cause the computer to perform a method of assessing the effect of exercise in the rehabilitation training as in the first aspect or any one of the possible designs of the first aspect.

The beneficial effects are that:

(1) The invention extracts the effective electromyographic signals corresponding to the rehabilitation actions of the human body from the collected electromyographic signals by collecting the surface electromyographic signals of the appointed part of the human body in the rehabilitation training process and through the processing means such as noise removal, effective signal extraction and the like; then, the myoelectric time domain characteristics of the appointed part of the human body are obtained by carrying out time domain analysis processing on the effective myoelectric signals; finally, the motion evaluation result of the appointed part of the human body can be obtained through myoelectric time domain characteristics; therefore, the invention does not need to manually participate in the motion evaluation process in the rehabilitation training, improves the evaluation accuracy and efficiency compared with the traditional technology, and is suitable for large-scale application and popularization in the rehabilitation medical field.

Drawings

Fig. 1 is a schematic flow chart of steps of a method for evaluating exercise effects in rehabilitation training according to an embodiment of the present invention;

fig. 2 is a schematic structural diagram of a motion effect evaluation device in rehabilitation training according to an embodiment of the present invention;

fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.

Detailed Description

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the present invention will be briefly described below with reference to the accompanying drawings and the description of the embodiments or the prior art, and it is obvious that the following description of the structure of the drawings is only some embodiments of the present invention, and other drawings can be obtained according to these drawings without inventive effort to a person skilled in the art. It should be noted that the description of these examples is for aiding in understanding the present invention, but is not intended to limit the present invention.

It will be understood that, although the terms first, second, etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another element. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element, without departing from the scope of example embodiments of the present invention.

It should be understood that for the term "and/or" that may appear herein, it is merely one association relationship that describes an associated object, meaning that there may be three relationships, e.g., a and/or B, may represent: a alone, B alone, and both a and B; for the term "/and" that may appear herein, which is descriptive of another associative object relationship, it means that there may be two relationships, e.g., a/and B, it may be expressed that: a alone, a alone and B alone; in addition, for the character "/" that may appear herein, it is generally indicated that the context associated object is an "or" relationship.

Examples:

referring to fig. 1, in the exercise effect evaluation method in rehabilitation training provided by the embodiment, the effective myoelectric signals of the designated part of the human body when each complete rehabilitation training action is performed are obtained by collecting the surface myoelectric signals of the designated part of the human body in the rehabilitation training process, denoising the surface myoelectric signals and extracting the effective signals; then, the motion evaluation result of the appointed part of the human body in rehabilitation training can be obtained by carrying out time domain analysis on the effective electromyographic signals; compared with the traditional subjective evaluation of doctors, the method has higher accuracy and efficiency, and is more suitable for large-scale application and popularization in the rehabilitation medical field; the method may be executed on a rehabilitation training terminal, alternatively, the rehabilitation training terminal may be a personal computer (personal computer, PC), a tablet computer or a smart phone, and it is to be understood that the foregoing execution subject is not limited to the embodiments of the present application, and the operation steps of the method may be, but are not limited to, those shown in the following steps S1 to S5.

S1, acquiring a surface electromyographic signal of a designated part of a target person in a rehabilitation training process; in this embodiment, the specific position of the target person may be specifically set according to the rehabilitation training, for example, in the rehabilitation training of the lower limb, the surface myoelectric signals of the muscles such as gastrocnemius, gastrocnemius longus, soleus and tibialis anterior in the rehabilitation training process of the lower limb are mainly collected, and in the rehabilitation training of the upper limb, the surface myoelectric signals of the muscles such as pectoral major, supraspinatus, erector spinal and posterior deltoid of the target person may be collected; of course, the designated parts of different rehabilitation exercises are different and are not described in detail herein; meanwhile, in the rehabilitation training process, for example, a myoelectric sensor or a needle electrode can be used for collecting surface myoelectric signals of the designated part of the target person, and the collected surface myoelectric signals are sent to a rehabilitation training end.

After the surface electromyographic signals of the appointed position of the target person are obtained, signal analysis processing can be carried out so as to obtain a motion evaluation result of the appointed position of the target person in the rehabilitation training process based on the signal analysis result; since various interference signals exist in the collected surface electromyographic signals, in order to ensure the accuracy of signal analysis, a denoising step is further provided in this embodiment, as shown in step S2 below.

S2, denoising the surface electromyographic signals to obtain denoised electromyographic signals; in specific application, the embodiment is provided with a plurality of denoising processes so as to eliminate noise in the signal to the greatest extent; alternatively, the foregoing multiple denoising process may be, but is not limited to, as shown in steps S21 to S24 described below.

S21, carrying out initial denoising treatment on the surface electromyographic signals to obtain initial denoising electromyographic signals; in the specific implementation, the initial denoising mainly comprises the steps of removing power frequency interference, harmonic waves and baseline drift in signals; the following steps S21a to S21h may be used, but are not limited to, to achieve the initial denoising of the surface electromyographic signals.

S21a, performing windowed Fourier transform and Fourier transform on the surface electromyographic signals respectively to obtain a first amplitude spectrum and a second amplitude spectrum of the surface electromyographic signals; in this embodiment, the windowed fourier transform and the fourier transform are common technical means for signal processing, and the principle thereof is not described in detail.

After the first amplitude spectrum and the second amplitude spectrum of the surface electromyographic signals are obtained, the two can be used for removing power frequency interference and harmonic waves in the surface electromyographic signals; the process of removing the power frequency interference and the harmonic wave is shown in the following steps S21b to S21 e.

S21b, acquiring first amplitude spectrum data at a position adjacent to a target position from the first amplitude spectrum, wherein the target position is a harmonic position in the first amplitude spectrum and a position at a preset threshold frequency; in specific applications, the example preset threshold may be, but is not limited to, 50, that is, the target position is the position at the 50Hz frequency and the harmonic position in the first amplitude spectrum, and meanwhile, the adjacent position at any target position is the amplitude spectrum data in the window with the width as the preset value and with the any target position as the center; if it is assumed that 6 target positions (including 3 harmonic positions and 3 50Hz positions) exist, then, taking one harmonic position as a center, and acquiring first amplitude spectrum data in a window with the width as a preset value; of course, the process of acquiring the first amplitude spectrum data of the adjacent positions corresponding to the other target positions is the same as the foregoing example, and will not be described again here; the width of the window may be set specifically according to the actual use, and is not particularly limited herein.

After the first amplitude spectrum data of the target position corresponding to the adjacent position is obtained, the actual amplitude spectrum data of each target position can be obtained through an interpolation method, wherein the interpolation process is shown in the following step S21 c.

S21c, interpolating to obtain actual amplitude spectrum data at the target position by using the first amplitude spectrum data; in the present embodiment, step S21 is explained on the basis of the foregoing example, for 3 harmonic positions and 3 50Hz positions, each of which corresponds to the first amplitude spectrum data of the respective adjacent position; if the 3 harmonic positions are A, B and C in sequence, then interpolation is performed by using the first amplitude spectrum data at the adjacent position corresponding to the a to obtain the actual amplitude spectrum data at the a position, and similarly, the obtaining process of the actual amplitude spectrum data at the other harmonic positions and the 50Hz position is the same as the foregoing example, and will not be repeated herein.

After the actual amplitude spectrum data at each target position is obtained by interpolation, the amplitude spectrum data can be replaced, wherein the replacing process is as shown in the following step S21d.

S21d, replacing the amplitude spectrum data at the target position in the second amplitude spectrum by using the actual amplitude spectrum data so as to obtain a processed second amplitude spectrum after replacement; in specific application, the target position in the second amplitude spectrum is the same as the target position in the first amplitude spectrum, so that the amplitude spectrum data at the same position can be replaced; after the replacement is completed, the signal obtained by the re-acquisition can be subjected to inverse Fourier transform, so that the surface electromyographic signal after the power frequency interference and harmonic waves are removed is obtained; wherein the inverse fourier transform process of the signal is shown in step S21e below.

S21e, performing inverse Fourier transform on the processed second amplitude spectrum to obtain a preprocessed surface electromyographic signal; in this embodiment, the inverse fourier transform of the signal is a common technique for signal processing, and the principle thereof is not described in detail.

After the power frequency interference and harmonic waves in the surface electromyographic signals are removed, a subsequent baseline drift removal step can be performed; the baseline shift removal process is shown in the following steps S21f to S21 h.

S21f, sequentially performing open operation and close operation on the surface electromyographic signals after pretreatment to obtain a first electromyographic signal, and sequentially performing close operation and open operation on the surface electromyographic signals after pretreatment to obtain a second electromyographic signal; in this embodiment, the structural element is selected first, and then the structural element is used to perform morphological operation on the surface electromyographic signal; specifically, the open operation is to sequentially perform the corrosion operation and the expansion operation; the closed operation is to sequentially perform the expansion operation and the corrosion operation.

After different morphological operations are carried out on the surface electromyographic signals after pretreatment, the baseline drift value of the surface electromyographic signals can be determined according to morphological operation results; the determination of the baseline wander value is shown in step S21g below.

S21g, obtaining a baseline drift value of the preprocessed surface electromyographic signals based on the first electromyographic signals and the second electromyographic signals; in this embodiment, the first electromyographic signal and the second electromyographic signal (the signal values at each moment in the signal are added), and the average value of the summed signal is taken, so as to obtain the baseline drift value; after the baseline drift value is obtained, the baseline drift can be removed from the surface electromyographic signals after pretreatment; the baseline drift removal process is shown in step S21h below.

S21h, eliminating baseline drift in the preprocessed surface electromyographic signals by using the baseline drift value to obtain the initial denoising electromyographic signals; and in specific application, subtracting the baseline drift value from the preprocessed surface electromyographic signal to obtain the initial denoising electromyographic signal.

Thus, through the steps S21a to S21h, the initial denoising of the surface electromyographic signals at the designated part can be completed; then, in order to further improve the signal-to-noise ratio of the initial denoising electromyographic signal, the embodiment is further provided with a signal reconstruction step before performing the secondary denoising, which can weaken the noise energy, and thus is also equivalent to a denoising process, wherein the signal reconstruction process is as follows in step S22 and step S23.

S22, performing empirical mode decomposition processing on the initial denoising electromyographic signals to obtain a plurality of first IMF components; in this embodiment, the empirical mode decomposition is a common technique of signal decomposition, and the principle thereof is not described again; after the empirical mode decomposition of the initial denoising electromyographic signals is completed, a plurality of first IMF components obtained by decomposition can be utilized to reconstruct signals; the specific process of signal reconstruction is shown in the following step S23.

S23, performing signal reconstruction processing by using the plurality of first IMF components to obtain a reconstructed initial denoising electromyographic signal; in a specific application, the following steps S23a and S23b may be used, for example and without limitation, to complete the signal reconstruction of the initial denoising electromyographic signal.

S23a, carrying out signal reconstruction on a first IMF component in a plurality of first IMF components for a plurality of times to obtain a plurality of reconstructed IMF components, wherein any signal reconstruction of the first IMF component refers to random ordering of signal values at different moments in the first IMF component; in the present embodiment, it is assumed that the first IMF component is denoted as c (d), d=1, 2,3,4; and d represents the signal time; then the first IMF component contains 4 signal elements; in this way, when the signal reconstruction is performed once, c (1), c (2), c (3) and c (4) are randomly ordered, for example, the orders are c (3), c (2), c (1) and c (4), that is, when d is 1, the corresponding signal value is replaced by c (3) in the original signal, and when d is 3, the corresponding signal value is replaced by c (1) in the original signal; after random ordering, the reconstructed IMF component can be obtained; of course, the remaining signal reconstruction process of the first IMF component is the same as the above example, and will not be repeated here.

After obtaining the plurality of reconstructed IMF components, the reconstruction processing of the initial denoising electromyographic signal can be completed by using the plurality of reconstructed IMF components and the first IMF component except the first IMF component in the step S23 a; the detailed reconstruction process of the initial denoising electromyographic signal is shown in step S23b below.

S23b, obtaining an average IMF component of a plurality of reconstructed IMF components, and utilizing a target IMF component and the average IMF component to form the initial denoising electromyographic signal after reconstruction, wherein the target IMF component is all first IMF components except a first IMF component in the plurality of first IMF components; when the method is specifically applied, the average IMF component of the plurality of reconstructed IMF components is obtained, and then the average IMF component is added with each target IMF component to obtain the reconstructed initial denoising electromyographic signal.

In the present embodiment, the principle of the foregoing step S23a and step S23b is as follows: the plurality of IMF components obtained through empirical mode decomposition are arranged from high to low according to frequency, wherein the components with low orders correspond to high-frequency components of signals, and the signal-to-noise ratio is low; therefore, the first IMF component needs to be randomly ordered for a plurality of times and accumulated and averaged to improve the signal to noise ratio; and then, summing the obtained reconstructed IMF component and the residual first IMF component to obtain the reconstructed initial denoising electromyographic signal with the signal to noise ratio larger than that of the original signal.

After obtaining the reconstructed initial denoising electromyographic signal, a secondary denoising process may be performed, where the secondary denoising process may be, but is not limited to, as shown in step S24 below.

S24, performing secondary denoising treatment on the reconstructed initial denoising electromyographic signal to obtain the denoising electromyographic signal after the secondary denoising treatment, wherein in the embodiment, the secondary denoising mainly removes Gaussian white noise in the signal; optionally, an empirical mode decomposition combined wavelet denoising method can be adopted to remove Gaussian white noise in the reconstructed initial denoising electromyographic signals; further, the specific denoising process of the gaussian white noise is shown in the following steps S24a to S24 e.

S24a, performing empirical mode decomposition on the reconstructed initial denoising electromyographic signals to obtain a plurality of second IMF components.

S24b, calculating an autocorrelation function of each second IMF component, and carrying out normalization processing on each autocorrelation function to obtain a normalized autocorrelation function of each second IMF component; in this embodiment, the autocorrelation function and normalization process of each second IMF component are common techniques in empirical mode decomposition, and the principles thereof are not described in detail.

After the normalized autocorrelation function of each second IMF component is obtained, a method for obtaining each normalized autocorrelation function can be obtained, and the noise content of each second IMF component can be distinguished according to the variance; wherein the sorting process is as shown in step S24c below.

S24c, solving the variance of the normalized autocorrelation function of each second IMF component, and classifying the plurality of second IMF components based on the variance of each normalized autocorrelation function to obtain a high-noise IMF component and a low-noise IMF component; in specific application, the variance of the normalized autocorrelation function can be calculated according to the corresponding first moment and second moment, which is a common calculation technique for the variance of the autocorrelation function, and the principle is not repeated in detail; simultaneously, the variance of each normalized autocorrelation function can be used to measure the noise content of the corresponding second IMF component; specifically, the variance of each normalized autocorrelation function is compared with a preset value, if the variance of the normalized autocorrelation function of a certain second IMF component is smaller than the preset value, the second IMF component is determined to be a low-order IMF component with high noise, otherwise, the second IMF component is determined to be a high-order IMF component with low noise.

For the high-noise low-order IMF component (i.e., the high-noise IMF component), denoising is required; alternatively, the present embodiment adopts a wavelet denoising method to denoise the high-noise IMF component, and the denoising process is as shown in step S24d below.

S24d, carrying out wavelet denoising treatment on each high-noise IMF component to obtain each denoised IMF component; in this embodiment, wavelet denoising is a common denoising technique of gaussian white noise, and the principle thereof is not described again; meanwhile, in order to improve the denoising effect, this embodiment designs a new wavelet coefficient transformation formula, taking any high-noise IMF component as an example, and the transformation formula of the corresponding wavelet coefficient is shown in the following formula (1).

In the above formula (1), w _j,k Original wavelet coefficients, w ', representing a kth wavelet component of the arbitrary high noise IMF component at a jth scale' _j,k Transformed wavelet coefficients representing a kth wavelet component of said any high noise IMF component at a jth scale, y _j,k A denoising threshold representing a kth wavelet component of the any high noise IMF component at a jth scale,represents a denoising parameter, where j=1, 2, L, k=1, 2, D, and L represents the wavelet denoising decomposition total scale number of the any high noise IMF component, D is the total number of wavelet components of any high-noise IMF component under the j-th scale; in this embodiment, <' > a->The value interval of (1) is [0,1 ]]The method can be selected to be 0.5, meanwhile, a sym8 wavelet function can be selected as a wavelet basis function during wavelet denoising, and the total scale is 4; of course, the wavelet basis functions and dimensions may be specifically set according to practical use, and are not limited to the foregoing examples.

Further, e.g., y _j,k Can be calculated by, but not limited to, the following formula (2);

in the above formula (2), media (|X) ⁱ _IMF I) represents the absolute median of any high-noise IMF component, N represents the length of the reconstructed initial denoising electromyographic signal, a represents an adjustable parameter, and u is a constant; in a particular application, the adjustable parameter may be, but is not limited to, 0.3, and u may be, but is not limited to, 0.6745.

The wavelet coefficients corresponding to the high-noise IMF components can be transformed through the formula (2) and the formula (3), so that the noise signals can be removed by adjusting the wavelet coefficients; and finally, carrying out signal reconstruction on the residual wavelet components of each high-noise IMF component to obtain the denoised IMF component.

After the denoising of each high-noise IMF component is completed, the low-noise IMF component can be combined to reconstruct signals, so that surface electromyographic signals (namely denoising electromyographic signals) from which Gaussian white noise is removed are obtained through reconstruction; wherein the reconstruction process is as shown in step S24e below.

S24e, carrying out signal reconstruction processing by using the denoised IMF component and the low-noise IMF component so as to obtain the denoised electromyographic signal after the signal reconstruction processing.

Thus, the denoising processing of the collected surface electromyographic signals can be completed through the steps S21 to S24 and the sub-steps thereof, so as to obtain the electromyographic signals with low noise content; then, an effective signal is also required to be segmented from the denoising electromyographic signal so as to carry out motion estimation based on the effective signal; the signal splitting process is shown in the following step S3.

S3, carrying out segmentation processing on the denoising electromyographic signals so as to segment at least one effective signal segment in the denoising electromyographic signals, wherein each effective signal segment in the at least one effective signal segment comprises a motion unit action potential waveform.

In this embodiment, since a complete electromyographic signal includes an electromyographic signal when a human body performs a rehabilitation motion, and electromyographic signals before and after the rehabilitation motion is performed, in order to improve the accuracy of analysis, an effective signal segment including a motion unit action potential waveform needs to be extracted from the collected surface electromyographic signals; wherein, the motion unit action potential waveform refers to MUAP waveform, which represents continuous waveform formed by action potential generated by single motion unit neuron activation; therefore, extracting the effective signal segment containing the MUAP waveform is equivalent to extracting the effective electromyographic signal (i.e., the electromyographic signal generated from the motion start to the end) generated when the human body makes each complete rehabilitation motion from the surface electromyographic signal; therefore, the electromyographic signals corresponding to the rehabilitation training actions of the human body are used for carrying out the motion analysis, and the accuracy of the analysis can be further improved.

Alternatively, the following steps S31 to S37 may be used, for example and not limited thereto, to extract the effective signal segment.

S31, carrying out peak enhancement processing on the denoising electromyographic signals so as to obtain enhanced electromyographic signals after the peak enhancement processing; in the present embodiment, the enhanced electromyographic signal can be obtained by, for example, but not limited to, using the following formula (3).

z[q(t)]＝q(t)×q(t+g-2)-q(t+g-1)q(t-1) (3)

In the above formula (3), q (t) represents the denoising electromyographic signal, g represents an enhancement coefficient, z [ q (t) ] represents the enhancement electromyographic signal, and t represents a sampling time in the denoising electromyographic signal; alternatively, the example g takes a value of 2, that is, when g takes a value of 2, the foregoing formula (3) is equivalent to performing peak enhancement based on a nonlinear energy operator.

After the enhanced electromyographic signal is obtained based on the above formula (3), there may be a false peak in the signal, so in order to ensure the reliability of the peak plus the enhancement effect, in this embodiment, the enhanced electromyographic signal needs to be subjected to smoothing filtering; the smoothing filtering process is shown in the following step S32.

S32, performing smooth filtering treatment on the enhanced electromyographic signals to obtain filtered electromyographic signals; in a specific application, for example, but not limited to, a hamming window may be used to perform smoothing on the enhanced electromyographic signal, and the effective signal segment may be extracted after the smoothing.

In this embodiment, the state change speed of the motion unit is much smaller than the fluctuation speed of the electromyographic signal, so the filtered electromyographic signal can be regarded as a short-time stationary signal, based on which the determination of the effective signal segment (i.e. the active segment) can be performed by counting the maximum point in the signal and adopting the amplitude threshold, wherein the extraction process of the effective signal segment is as shown in the following steps S33 to S37.

S33, counting all maximum signal points in the filter electromyographic signals, and calculating an amplitude threshold value based on all the counted maximum signal points; in practice, the average of all maximum signal points may be used as the amplitude threshold.

After the amplitude threshold is calculated, a detection window can be selected by taking each maximum signal point as a center, and whether the signal segment in the detection window is an effective signal or not is judged according to the magnitude relation between the minimum value in the detection window and the amplitude threshold; the detection process is as follows in steps S34 to S37.

S34, for a p-th maximum signal point in the filtered electromyographic signal, acquiring a detection window of the p-th maximum signal point, wherein the center of the detection window is the p-th maximum signal point, and the width is e; in the specific implementation, assuming that the width is 4, the detection window of the p-th maximum signal point is [ pt-2, pt+2]; of course, the width of the detection window may be specifically set according to practical use, and is not specifically limited herein.

After a detection window of the p-th maximum signal point is obtained, the minimum value in the detection window can be obtained, and then the minimum value is compared with an amplitude threshold value, so that whether the signal in the detection window is a valid signal or not is judged according to a comparison result; the determination process is as follows in step S35 and step S36.

S35, obtaining the minimum value in the detection window, and judging whether the minimum value is smaller than the amplitude threshold value.

S36, if yes, taking the signal in the detection window as an effective signal segment.

The selection of the effective signal corresponding to the p-th maximum signal point can be completed through the steps S34 to S35, and then the detection of the effective signal segment corresponding to the next maximum signal point is performed according to the same principle until all the maximum signal points are polled, and all the effective signal segments can be extracted from the filtered electromyographic signals.

Wherein the looping process is shown in step S37 below.

S37, adding 1 to p, and re-acquiring a detection window of the p-th maximum signal point until p is equal to m, so as to obtain at least one effective signal segment in the denoising electromyographic signal, wherein the initial value of p is 1, and m is the total number of the maximum signal points in the filtering electromyographic signal.

In this embodiment, if in step S35, the p-th maximum signal point corresponds to the minimum value in the detection window and is greater than or equal to the amplitude threshold, then the minimum value in the detection window needs to be obtained, and the average value between the minimum value and the p-th maximum signal point is taken as the center point; and updating the p-th maximum signal point to the center point, and reacquiring a detection window of the p-th maximum signal point until the minimum value in the detection window is smaller than the amplitude threshold value.

In addition, if the signals in the detection windows corresponding to the two adjacent maximum signal points are valid signal segments and the superposition of the two signals exceeds e/2, the two valid signal segments can be combined into one valid signal segment.

Therefore, through the steps S31-S37, each effective signal segment can be extracted from the denoising electromyographic signals, so that the electromyographic signals before and after the rehabilitation action are avoided, and the interference on the motion estimation is avoided; then, motion estimation may be performed based on the valid signal segment, and the estimation process may be, but is not limited to, as shown in the following steps S4 and S5.

S4, performing time domain analysis processing on each effective signal segment in the at least one effective signal segment to obtain time domain characteristics of each effective signal segment; when the method is applied specifically, time domain analysis is carried out on each effective signal segment, and the integral myoelectricity value and the root mean square value of each effective signal segment are mainly obtained; wherein, the two time domain features can both reflect the change feature of the amplitude of the human myoelectric signal in the time dimension, and the latter depends on the muscle load and the physiological state of the muscle, so the two time domain features are commonly used for analyzing the muscle activity state in real time; based on the method, the motion evaluation result of the appointed position of the target person in the rehabilitation training process can be obtained by utilizing the integral myoelectricity value and the root mean square value of each effective signal segment; optionally, the foregoing evaluation procedure is as follows in step S5.

S5, determining a motion evaluation result of the appointed position of the target person in the rehabilitation training process according to the time domain characteristics of each effective signal segment; in this embodiment, the surface electromyographic signals (the training time and the action are the same, so as to be convenient for distinguishing from the surface electromyographic signals) in the same rehabilitation training process of the same part of a normal human body can be obtained in advance, and the surface electromyographic signals are referred to as calibrated surface electromyographic signals in the following; then, the step S2 and the step S3 are adopted to carry out the same denoising and effective signal extraction on the electromyographic signals of the calibration surface, so as to obtain each calibration effective signal segment; then, performing time domain analysis on each calibrated effective signal segment to obtain calibrated time domain characteristics (namely an integral myoelectricity value and a root mean square value when a normal person performs the same rehabilitation training); finally, comparing the integral myoelectric value of the appointed position of the target person with a calibrated integral myoelectric value, and comparing a root mean square value with a calibrated root mean square value (the average value of all integral myoelectric values of each effective signal section and the average value of the root mean square value can be calculated, and then comparing the average value with the calibrated integral myoelectric value and the calibrated root mean square value, wherein the calibrated integral myoelectric value and the calibrated root mean square value are the average value of each calibrated effective signal section, so that a motion evaluation result is obtained according to the comparison result; optionally, the smaller the difference between the integral myoelectric value and the calibrated integral myoelectric value of any designated part, the better the muscle rehabilitation effect of any designated part, the better the exercise effect of the muscle rehabilitation effect; otherwise, the larger the difference value between the integral myoelectricity value of any appointed position and the calibrated integral myoelectricity value is, the worse the muscle rehabilitation effect of any appointed position is, and the worse the movement effect of the muscle rehabilitation device in the rehabilitation training process is; furthermore, the embodiment can finally directly display the difference between each time domain feature and the calibrated time domain feature, and quantify the operation effect of the appointed position of the target person in the rehabilitation training process based on the difference.

According to the exercise effect evaluation method in the rehabilitation training described in detail in the steps S1 to S5, the effective myoelectric signals of the appointed part of the human body when all the complete rehabilitation training actions are made are obtained by collecting the surface myoelectric signals of the appointed part of the human body in the rehabilitation training process, denoising the surface myoelectric signals and extracting the effective signals; then, the motion evaluation result of the appointed part of the human body in rehabilitation training can be obtained by carrying out time domain analysis on the effective electromyographic signals; compared with the traditional subjective evaluation of doctors, the method has higher accuracy and efficiency, and is more suitable for large-scale application and popularization in the field of rehabilitation medicine.

As shown in fig. 2, a second aspect of the present embodiment provides a hardware device for implementing the exercise effect evaluation method in rehabilitation training described in the first aspect of the present embodiment, including:

the acquisition unit is used for acquiring the surface electromyographic signals of the appointed part of the target person in the rehabilitation training process.

And the denoising unit is used for denoising the surface electromyographic signals to obtain denoising electromyographic signals.

The signal segmentation unit is used for carrying out segmentation processing on the denoising electromyographic signals so as to segment at least one effective signal segment in the denoising electromyographic signals, wherein each effective signal segment in the at least one effective signal segment comprises a motion unit action potential waveform.

And the signal analysis unit is used for carrying out time domain analysis processing on each effective signal segment in the at least one effective signal segment so as to obtain the time domain characteristics of each effective signal segment.

The working process, working details and technical effects of the device provided in this embodiment may refer to the first aspect of the embodiment, and are not described herein again.

As shown in fig. 3, a third aspect of the present embodiment provides another exercise effect evaluation apparatus in rehabilitation training, taking the apparatus as an example of an electronic device, including: the device comprises a memory, a processor and a transceiver which are connected in sequence in communication, wherein the memory is used for storing a computer program, the transceiver is used for receiving and transmitting messages, and the processor is used for reading the computer program and executing the exercise effect evaluation method in rehabilitation training according to the first aspect of the embodiment.

By way of specific example, the Memory may include, but is not limited to, random access Memory (random access Memory, RAM), read Only Memory (ROM), flash Memory (Flash Memory), first-in-first-out Memory (First Input First Output, FIFO) and/or first-in-last-out Memory (First In Last Out, FILO), etc.; in particular, the processor may include one or more processing cores, such as a 4-core processor, an 8-core processor, or the like. The processor may be implemented in at least one hardware form of DSP (Digital Signal Processing ), FPGA (Field-Programmable Gate Array, field programmable gate array), PLA (Programmable Logic Array ), and may also include a main processor and a coprocessor, where the main processor is a processor for processing data in an awake state, and is also called CPU (Central Processing Unit ); a coprocessor is a low-power processor for processing data in a standby state.

In some embodiments, the processor may be integrated with a GPU (Graphics Processing Unit, image processor) for taking charge of rendering and rendering of content required to be displayed by the display screen, for example, the processor may not be limited to a microprocessor employing a model number of STM32F105 family, a reduced instruction set computer (reduced instruction set computer, RISC) microprocessor, an X86 or other architecture processor, or a processor integrating an embedded neural network processor (neural-network processing units, NPU); the transceiver may be, but is not limited to, a wireless fidelity (WIFI) wireless transceiver, a bluetooth wireless transceiver, a general packet radio service technology (General Packet Radio Service, GPRS) wireless transceiver, a ZigBee protocol (low power local area network protocol based on the ieee802.15.4 standard), a 3G transceiver, a 4G transceiver, and/or a 5G transceiver, etc. In addition, the device may include, but is not limited to, a power module, a display screen, and other necessary components.

The working process, working details and technical effects of the electronic device provided in this embodiment may refer to the first aspect of the embodiment, and are not described herein again.

A fourth aspect of the present embodiment provides a storage medium storing instructions including the exercise effect evaluation method in rehabilitation training according to the first aspect of the present embodiment, that is, the storage medium storing instructions thereon, which when executed on a computer, perform the exercise effect evaluation method in rehabilitation training according to the first aspect of the present embodiment.

The storage medium refers to a carrier for storing data, and may include, but is not limited to, a floppy disk, an optical disk, a hard disk, a flash Memory, a flash disk, and/or a Memory Stick (Memory Stick), where the computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable devices.

The working process, working details and technical effects of the storage medium provided in this embodiment may refer to the first aspect of the embodiment, and are not described herein again.

A fifth aspect of the present embodiment provides a computer program product comprising instructions which, when run on a computer, cause the computer to perform the method of assessing athletic performance in rehabilitation training according to the first aspect of the embodiment, wherein the computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable device.

Finally, it should be noted that: the foregoing description is only of the preferred embodiments of the invention and is not intended to limit the scope of the invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for evaluating a athletic performance in rehabilitation training, comprising:

2. The method of claim 1, wherein denoising the surface electromyographic signal to obtain a denoised electromyographic signal comprises:

3. The method of claim 2, wherein performing initial denoising processing on the surface electromyographic signal to obtain an initial denoised electromyographic signal comprises:

4. The method of claim 2, wherein performing a signal reconstruction process using the plurality of first IMF components to obtain a reconstructed initial de-noised electromyographic signal comprises:

5. The method according to claim 4, wherein the transform formula of the wavelet coefficients of any high-noise IMF component when performing wavelet denoising processing is:

in the above formula (1), w _j,k Representing a kth wavelet component of said any high noise IMF component at a jth scaleIs w' _j,k Transformed wavelet coefficients representing a kth wavelet component of said any high noise IMF component at a jth scale, y _j,k A denoising threshold representing a kth wavelet component of the any high noise IMF component at a jth scale,represents a denoising parameter, where j=1, 2, L, k=1, 2, D, and L represents the wavelet denoising decomposition total scale number of the any high noise IMF component, D is the total number of wavelet components of any high-noise IMF component under the j-th scale;

6. The method of claim 1, wherein the segmenting the de-noised electromyographic signal to segment at least one valid signal segment in the de-noised electromyographic signal comprises:

7. The method of claim 6, wherein peak enhancement processing is performed on the de-noised electromyographic signal to obtain an enhanced electromyographic signal after peak enhancement processing, comprising:

z[q(t)]＝q(t)×q(t+g-2)-q(t+g-1)q(t-1) (3)

correspondingly, the method further comprises the steps of:

8. A sports effect evaluation device in rehabilitation training, characterized by comprising:

9. A sports effect evaluation device in rehabilitation training, characterized by comprising: the memory, the processor and the transceiver are connected in sequence in communication, wherein the memory is used for storing a computer program, the transceiver is used for receiving and transmitting messages, and the processor is used for reading the computer program and executing the exercise effect evaluation method in rehabilitation training according to any one of claims 1 to 7.

10. A storage medium having instructions stored thereon that, when executed on a computer, perform the exercise effect assessment method in rehabilitation training of any one of claims 1 to 7.