A method and apparatus for extracting muscle tone related myoelectric signals

Abstract

The application discloses a method and equipment for extracting muscle tension related electromyographic signals, and belongs to the field of health and sanitation. The method comprises the following steps: constructing a human body muscle force confrontation mode between a detector and a reference person; collecting first confrontation mode characteristic data and a first electromyographic signal of the detector in the human body muscle force confrontation mode, and collecting second confrontation mode characteristic data and a second electromyographic signal of the reference person; constructing a muscle tension related electromyographic signal correction model based on the second confrontation mode characteristic data and the second electromyographic signal; inputting the first confrontation mode characteristic data into the muscle tension related electromyographic signal correction model, and calculating a theoretical value of the electromyographic signal of the detector; and extracting a muscle tension related electromyographic signal characteristic of the detector according to the difference between the first electromyographic signal and the theoretical value of the electromyographic signal. The application provides a numerical index for the muscle tension state.

Description

Technical Field

[0001] This invention belongs to the field of health and hygiene technology, and in particular relates to a method and device for extracting muscle tone-related electromyographic signals. Background Technology

[0002] Muscle tone assessment is a crucial aspect of neuromuscular function evaluation, traditionally relying primarily on the examiner's subjective approach. This typically includes passive range of motion testing (feeling resistance to joint movement), palpation of muscle stiffness, observation of posture, and integration with reflex examinations and standardized scales (such as the Ashworth scale or a modified Ashworth scale). The core of this type of assessment lies in ensuring the patient is fully relaxed, with the examiner applying passive movements at varying speeds, perceiving the nature and magnitude of resistance through touch, and making bilateral comparisons. Furthermore, passive movement patterns based on muscle resistance exist in practice; for example, in a contested activity similar to "arm wrestling," the examiner attempts to keep the examiner's muscles relaxed, subjectively judging the muscle tone state of the opponent's limb (e.g., the upper limb) based on their own sensations. On the other hand, existing technologies also include instrumental methods such as isokinetic muscle strength testing systems. These systems construct muscle resistance using mechanical devices, inducing uniform or uniform angular velocity movements in the limbs and analyzing mechanical parameters (such as torque) during active contraction, attempting to indirectly reflect muscle tone status. Meanwhile, some studies have explored combining electromyography (EMG) with biomechanics (such as isokinetic systems) to quantify neuromuscular efficiency and capture electrophysiological activities related to muscle tone by establishing torque-EMG amplitude relationship curves.

[0003] However, the aforementioned existing technologies have significant limitations. First, traditional methods centered on palpation and passive movement heavily rely on the examiner's experience and subjective feelings, making it difficult to quantify and standardize the results, hindering accurate objective comparisons and long-term follow-ups, and easily affected by inter-examiner differences. Second, while existing isokinetic muscle strength testing systems provide a certain degree of objective data, the antagonistic muscle strength they construct is usually constant and changes slowly, requiring the examiner to actively cooperate with the movement; when patients' muscle strength changes are uncontrollable or they cannot actively cooperate due to diseases (such as Parkinson's disease, post-stroke spasticity), the test results are prone to bias. More importantly, neither traditional subjective assessments nor isokinetic systems can effectively distinguish and eliminate the interference of muscle strength itself on the measurement signal. Although muscle strength and muscle tone are related in physiological mechanisms and functions, they are fundamentally different, and existing technologies cannot effectively separate muscle tone-related signals (especially electromyographic signals) from signals mixed with muscle strength contributions. For example, in the electromyography-mechanics combined mode, the detected electromyographic activity is still mixed with muscle force-related electrical activity components, and it is impossible to accurately and specifically extract and characterize electromyographic signal features that purely reflect muscle tone state (such as resting tension or nerve reflex activity related to passive traction resistance). Summary of the Invention

[0004] To address the aforementioned technical problems, this invention proposes a method and apparatus for extracting muscle tone-related electromyographic signals, thereby resolving the issues present in the prior art.

[0005] To achieve the above objectives, in a first aspect, the present invention provides a method for extracting muscle tone-related electromyographic signals, comprising:

[0006] Constructing a human muscle strength resistance model between the tester and the reference;

[0007] The first confrontation mode characteristic data and the first electromyographic signal of the tester under the human muscle strength confrontation mode were collected, and the second confrontation mode characteristic data and the second electromyographic signal of the reference were collected.

[0008] Based on the feature data of the second adversarial mode and the second electromyographic signal, a muscle tone-related electromyographic signal correction model is constructed.

[0009] The first adversarial mode feature data is input into the muscle tone-related electromyography signal correction model to calculate the theoretical value of the electromyography signal of the test subject.

[0010] Based on the difference between the first electromyographic signal and the theoretical value of the electromyographic signal, the muscle tone-related electromyographic signal features of the test subject are extracted.

[0011] Preferably, the process of constructing a human muscle strength resistance pattern between the tester and the reference includes:

[0012] By applying external force to the reference, the test subject is induced to produce at least one of the following human motions: uniform velocity motion, uniform angular velocity motion, zero acceleration motion, or constant acceleration motion at the test site.

[0013] The opposing parts of the human muscle strength resistance mode include the upper limbs, lower limbs, head, and trunk.

[0014] Preferably, the process of collecting the first confrontation mode feature data of the tester and the second confrontation mode feature data of the reference under the human muscle strength confrontation mode includes:

[0015] The first and second adversarial mode feature data are collected using accelerometers, gyroscopes, or levels.

[0016] The first and second adversarial mode feature data include motion speed, angular velocity, or acceleration.

[0017] Preferably, the process of acquiring the first electromyographic signal of the test subject and the second electromyographic signal of the reference subject under the human muscle strength resistance mode includes:

[0018] The first and second electromyographic signals are acquired using a surface electromyography device, a needle electromyography device, or a wearable device.

[0019] The first and second electromyographic signals include amplitude, frequency, and waveform.

[0020] Preferably, the process of constructing a muscle tone-related electromyographic signal correction model includes:

[0021] Based on the changes in electromyographic (EMG) signals of a reference under static, uniform velocity, uniform angular velocity, or constant acceleration conditions, a mapping model between adversarial mode feature data and EMG signals is established, which is the muscle tone-related EMG signal correction model.

[0022] Preferably, the formula for the muscle tone-related electromyography signal correction model is:

[0023] dY=a(Y 检测者 -Y 参照者 )+b;

[0024] dY=a(Y 检测者 / Y 参照者 )+b;

[0025] dY = aLog(Y) 检测者 / Y 参照者 )+b;

[0026] Where dY represents the muscle tone-related electromyographic signal characteristics, and a and b are correction coefficients.

[0027] Preferably, the process of extracting the muscle tone-related electromyographic signal features of the subject includes:

[0028] The difference was calculated based on the first electromyographic signal and the theoretical value of the electromyographic signal.

[0029] The muscle strength ratio is calculated by comparing the limb mass and acceleration data of the tester and the reference.

[0030] When the human body is moving with constant acceleration, the difference is compensated based on the muscle strength ratio.

[0031] In a second aspect, the present invention provides a device for extracting muscle tone-related electromyographic signals, comprising:

[0032] A memory for storing a computer program of the method described in the first aspect;

[0033] A processor for executing the computer program to extract muscle tone-related electromyographic signals;

[0034] Electromyography (EMG) signal acquisition device, used to acquire EMG signals from both the test subject and the reference subject;

[0035] A motion parameter acquisition device is used to collect motion parameters under human muscle strength resistance mode.

[0036] Thirdly, the present invention also discloses a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in the first aspect.

[0037] Fourthly, the present invention also discloses a computer program product, including a computer program that, when executed by a processor, implements the steps of the method described in the first aspect.

[0038] Compared with the prior art, the present invention has the following advantages and technical effects:

[0039] This invention provides a method for extracting muscle tone-related electromyographic (EMG) signals, comprising: first, constructing a human muscle strength resistance pattern between a tester and a reference; second, collecting first resistance pattern feature data and first EMG signals of the tester under the human muscle strength resistance pattern, and collecting second resistance pattern feature data and second EMG signals of the reference; third, constructing a muscle tone-related EMG signal correction model based on the second resistance pattern feature data and second EMG signals; further, inputting the first resistance pattern feature data into the muscle tone-related EMG signal correction model to calculate the theoretical value of the tester's EMG signal; and finally, extracting the muscle tone-related EMG signal features of the tester based on the difference between the first EMG signal and the theoretical value of the EMG signal.

[0040] This invention transforms muscle tone assessment into quantifiable numerical features (such as amplitude difference and correction coefficient) by simultaneously collecting adversarial characteristic data and electromyographic (EMG) signals from the tester and the reference, thus eliminating the subjective dependence of traditional palpation methods. It outputs standardized EMG signal feature values ​​(such as dY), providing objective numerical indicators for muscle tone status and offering users long-term, continuous, and real-time recording and analysis of muscle tone-related EMG signal features. After reducing the interference of muscle strength factors, the refined muscle tone-related EMG signal features extracted by this invention can more accurately reflect the function and state of human muscle tone.

[0041] This invention overcomes the bottleneck of existing technologies that cannot distinguish between muscle strength and muscle tone signals, and for the first time achieves the precise extraction of specific electrophysiological characteristics of muscle tone, enabling objective and accurate identification of human muscle tone-related electromyographic signals; it can be applied to scenarios such as personal rehabilitation effect assessment and remote health monitoring.

[0042] The electromyographic signal features extracted by this invention (such as increased amplitude of spastic discharge and velocity-dependent discharge in Parkinson's patients) have pathological specificity, and further provide quantifiable biomarkers for neuromuscular diseases (such as Parkinson's disease and post-stroke spasticity). Attached Figure Description

[0043] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0044] Figure 1 This is a flowchart of a method for extracting muscle tone-related electromyographic signals according to an embodiment of the present invention;

[0045] Figure 2 This is a schematic diagram of the device structure for extracting muscle tone-related electromyographic signals based on muscle force resistance, according to an embodiment of the present invention.

[0046] Figure 3 This is a flowchart of a method for extracting muscle tone-related electromyographic signals based on muscle force resistance according to an embodiment of the present invention.

[0047] Figure 4 This is an exemplary diagram of a device for collecting electromyographic signals and adversarial pattern features based on a Parkinson's disease embodiment of the present invention;

[0048] Figure 5 This invention provides an example of collecting muscle tone-related signals extracted from the limbs of normal individuals during uniform limb movement in a Parkinson's disease scenario.

[0049] Figure 6 This invention provides an example of collecting muscle tone-related signals from normal individuals during uniform angular velocity limb movement in a Parkinson's disease scenario.

[0050] Figure 7 This invention provides an embodiment of the method for collecting muscle tone-related signals extracted from the limbs of a Parkinson's disease patient during uniform limb movement in a Parkinson's disease case.

[0051] Figure 8 This invention provides an embodiment of the method for collecting muscle tone-related signals extracted from the limbs of a Parkinson's disease patient during uniform angular velocity movement in an embodiment of the Parkinson's disease.

[0052] Figure 9 This invention collects movement pattern characteristics of limb acceleration movements in a Parkinson's disease population in an embodiment of the Parkinson's disease case.

[0053] Figure 10 This invention collects movement pattern characteristics of limb acceleration movements in a Parkinson's disease population in an embodiment of the Parkinson's disease case.

[0054] Figure 11 This is another muscle tone-related electromyography model for collecting muscle tone-related signals extracted from the uniform limb movement of normal individuals in the Parkinson's disease embodiment of the present invention.

[0055] Figure 12This is another muscle tone-related electromyography model for collecting muscle tone-related signals extracted from the limbs of Parkinson's disease patients during uniform limb movement in an embodiment of the present invention. Detailed Implementation

[0056] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0057] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0058] Example 1

[0059] like Figure 1 As shown, this embodiment provides a method for extracting muscle tone-related electromyographic signals, including:

[0060] S1. Construct a human muscle strength resistance model between the tester and the reference;

[0061] Furthermore, the process of constructing a human muscle strength resistance pattern between the tester and the reference includes:

[0062] By applying external force to the reference, the test subject is induced to produce at least one of the following human motions: uniform velocity motion, uniform angular velocity motion, zero acceleration motion, or constant acceleration motion at the test site.

[0063] The opposing parts of the human muscle strength resistance mode include the upper limbs, lower limbs, head, and trunk.

[0064] Specifically, the human muscle strength resistance mode includes muscle strength resistance in the upper limbs, lower limbs, head, and trunk.

[0065] Muscle strength resistance can be achieved in two ways: without relying on external objects and with relying on external objects.

[0066] The means of muscle resistance include having both the reference and the tester in a relaxed or exerting state of muscle strength, and the reference applying external force to induce uniform velocity, uniform angular velocity, zero or constant acceleration, or similar human motion in both parties.

[0067] This embodiment uses the muscle strength countermeasures, based on the reference, to weaken the influence of the tester's muscle strength-related electrical signals and extract the tester's muscle tone-related electromyographic signal features.

[0068] S2. Collect the first confrontation mode characteristic data and the first electromyographic signal of the tester under the human muscle strength confrontation mode, and collect the second confrontation mode characteristic data and the second electromyographic signal of the reference.

[0069] Furthermore, this step specifically includes:

[0070] The first and second adversarial mode feature data are collected using accelerometers, gyroscopes, or levels.

[0071] The first and second adversarial mode feature data include motion speed, angular velocity, or acceleration.

[0072] The first and second electromyographic signals are acquired using a surface electromyography device, a needle electromyography device, or a wearable device.

[0073] The first and second electromyographic signals include amplitude, frequency, and waveform.

[0074] Specifically, muscle strength resistance mode data can be recorded by placing accelerometers, gyroscopes, levels, etc. on the moving limbs; the characteristics of muscle strength resistance mode are parameters such as speed, angular velocity, and acceleration that induce the resistance movement of both limbs; muscle strength resistance induces the test limb to generate movement-related electromyographic signals, which are influenced by two factors: muscle strength and muscle tone.

[0075] Electromyography (EMG) signal recording modes include invasive and non-invasive approaches; EMG signal detection tools include surface EMG, needle EMG, and wearable devices; EMG signal detection models include recording EMG signals of a reference or tester against a reference limb in a muscle resistance mode under static conditions, zero or constant acceleration, and different uniform or uniform angular velocities; EMG signal data includes amplitude, frequency, and waveform.

[0076] The process of obtaining the training dataset includes:

[0077] With the tester in a relaxed or exerting muscle state, a reference person induces the tester's limb to produce common or similar uniform velocity, uniform angular velocity, zero or constant acceleration, or numerically approximate motion through muscle resistance. Simultaneously, limb movement pattern characteristic data are collected, and related electromyographic signals of the moving limb are recorded to obtain a dataset. The dataset includes the velocity, uniform angular velocity, and acceleration of the moving limb, as well as the electromyographic signals including amplitude, frequency, and waveform, and the calculated changes between the tester and the reference person.

[0078] S3. Based on the feature data of the second antagonistic mode and the second electromyographic signal, construct a muscle tension-related electromyographic signal correction model;

[0079] S4. Input the first adversarial mode feature data into the muscle tone-related electromyography signal correction model to calculate the theoretical value of the electromyography signal of the tester.

[0080] Furthermore, the process of constructing a muscle tone-related electromyographic signal correction model includes:

[0081] Based on the changes in electromyographic (EMG) signals of a reference under static, uniform velocity, uniform angular velocity, or constant acceleration conditions, a mapping model between adversarial mode feature data and EMG signals is established, which is the muscle tone-related EMG signal correction model.

[0082] The formula for the muscle tone-related electromyography signal correction model is as follows:

[0083] dY=a(Y 检测者 -Y 参照者 )+b;

[0084] dY=a(Y 检测者 / Y 参照者 )+b;

[0085] dY = aLog(Y) 检测者 / Y 参照者 )+b;

[0086] Where dY represents the muscle tone-related electromyographic signal characteristics, and a and b are correction coefficients.

[0087] S5. Based on the difference between the first electromyographic signal and the theoretical value of the electromyographic signal, extract the muscle tone-related electromyographic signal features of the test subject.

[0088] Furthermore, the process of extracting the muscle tone-related electromyographic signal features of the test subject includes:

[0089] The difference was calculated based on the first electromyographic signal and the theoretical value of the electromyographic signal.

[0090] The muscle strength ratio is calculated by comparing the limb mass and acceleration data of the tester and the reference.

[0091] When the human body is moving with constant acceleration, the difference is compensated based on the muscle strength ratio.

[0092] This embodiment analyzes the relationship between limb movement pattern feature data and electromyographic signals, and further removes the influence of muscle strength-related electrical signals to extract muscle tone-related electromyographic signal features.

[0093] This embodiment uses a muscle strength resistance model and movement pattern analysis to remove the influence of muscle strength-related and movement-related electromyographic signals, and extracts the muscle tone-related electromyographic signal characteristics of the test subjects, providing a reference for the assessment and intervention of muscle tone-related factors.

[0094] Example 2

[0095] Based on the same inventive concept, this embodiment also provides a device for extracting muscle tone-related electromyographic signals, including:

[0096] A memory for storing a computer program of the method described in Embodiment 1;

[0097] A processor for executing the computer program to extract muscle tone-related electromyographic signals;

[0098] Electromyography (EMG) signal acquisition device, used to acquire EMG signals from both the test subject and the reference subject;

[0099] A motion parameter acquisition device is used to collect motion parameters under human muscle strength resistance mode.

[0100] The device for extracting muscle tone-related electromyographic signals provided in this embodiment has all the advantages of the method for extracting muscle tone-related electromyographic signals provided in Embodiment 1.

[0101] Furthermore, a computer program for the method described in Embodiment 1 is stored in a memory, such as... Figure 2 The diagram shown is one of the structural schematics of the device. This data collection device includes: an electromyography (EMG) signal collection device, an adversarial pattern feature data collection device, a memory, a processor, and optionally, a display, a power supply, and other communication interfaces (not shown).

[0102] The processor can be a central processing unit (CPU), and the memory may include random-access memory (RAM) or non-volatile memory.

[0103] Figure 2 The data collection device shown may take the form of a smart terminal, wearable device, etc.

[0104] Figure 2 The data collection device shown may not include an electromyography (EMG) signal collection device or an adversarial pattern feature (APF) data collection device; these may be replaced by a communication interface, which may include an interface for communicating with other devices. The interface may be a wired interface, a wireless interface, or a combination thereof. The communication interface is used to receive the user's APF and EMG signal writing data sent by the peripheral data collection device. The functionality of other components remains unchanged.

[0105] The following is combined with Figure 3 The specific implementation process of this application will be explained.

[0106] S301: Establish and store reference muscle strength correction related models.

[0107] For example, muscle strength correction models are established for various adversarial modes and stored in memory. The specific methods for establishing muscle strength correction models have been described above and will not be repeated here.

[0108] S302: Check if the electromyography signal correction model for the corresponding adversarial mode of the target is stored. If not, execute S303; otherwise, execute S304.

[0109] For example, the display shows the names of various electromyographic signal correction models of target adversarial modes stored in the memory, and receives selection instructions from the tester A. For example, the tester A selects the electromyographic signal correction model under uniform motion on the touch screen. Other selectable models include models with uniform angular velocity, specific acceleration, etc.

[0110] S303: Establish a corresponding electromyographic signal correction model based on the electromyographic signals and antagonistic pattern characteristics of the reference subject. The specific method for establishing the model is as described above and will not be repeated here.

[0111] In a specific adversarial mode, the electromyography signal correction model is stored in memory, and S304 continues to be executed.

[0112] S304: Electromyography (EMG) detection equipment and adversarial mode feature collection equipment collect adversarial mode feature data and corresponding EMG signal data of the test subject A.

[0113] S305: The device processor for muscle tone-related electromyography (EMG) signals corrects the EMG signals of subject A based on the muscle tone-related EMG signals and antagonistic characteristics of subject A and the corresponding EMG signal correction model.

[0114] S306: Display electromyographic signals related to the muscle tone of subject A on the monitor; display electromyographic signals of muscle tone corresponding to the subject's resistance mode on the monitor.

[0115] Figure 4 This is an example of a device for collecting electromyographic signals and antagonistic pattern features based on a Parkinson's disease embodiment.

[0116] The study collected electromyographic signal data by inducing movement in Parkinson's disease trainees through muscle resistance.

[0117] The study collected data on the characteristics of the resistance pattern by inducing movement in Parkinson's disease trainees through muscle resistance.

[0118] Taking the extraction of muscle tone-related electromyography (EMG) signals from Parkinson's disease patients as an example, the detection of muscle tone-related EMG signals in normal healthy individuals used as a control is shown in [link to relevant documentation]. Figure 5-6 ;

[0119] In one scenario, under induced uniform motion, with the normal subject's limb relaxed, and the reference subject guiding the normal subject's limb to produce uniform motion, the electromyography (EMG) of both moving limbs is recorded. During uniform motion, the muscle strength of the normal subject and the reference subject is equal. After collecting sufficient training data, a more standardized approach is to correct the normal subject's EMG curve to a straight line with an amplitude close to 0 μV. The correction formula is calculated as follows:

[0120] dY=Y 正常者 -Y 参照者 That is, parameter a is 1 and parameter b is 0.

[0121] According to the correction formula, the reference electromyography (EMG) was corrected to eliminate the influence of muscle strength-related EMG, and muscle tone-related EMG was extracted during uniform motion. (See...) Figure 5 ;

[0122] In another scenario, under induced uniform angular velocity movement, when a normal subject's limb is in a relaxed state and a reference subject guides the normal subject's limb to generate uniform angular velocity movement, the electromyography (EMG) of both moving limbs is recorded. During uniform angular velocity movement, the muscle strength of the normal subject and the reference subject is equal. After collecting sufficient training data, the correction standard is to correct the normal subject's EMG curve to a straight line with an amplitude close to 0 μV. The correction formula is calculated as follows:

[0123] dY=Y 正常者 -Y 参照者 That is, parameter a is 1 and parameter b is 0.

[0124] According to the correction formula, the reference electromyography (EMG) was corrected to eliminate the influence of muscle strength-related EMG, and muscle tone-related EMG was extracted during the uniform angular velocity motion pattern. (See...) Figure 6 ;

[0125] Taking the extraction of tone-related electromyography (EMG) signals in Parkinson's disease as an example, the detection of tone-related EMG signals in Parkinson's disease is described in [link to relevant documentation]. Figure 7-12 ;

[0126] In one scenario, under induced uniform motion, when the Parkinson's patient's limbs are relaxed and the reference and guide's limbs produce uniform motion, the electromyography (EMG) of both limbs is recorded. During uniform motion, the muscle strength of the Parkinson's patient and the reference is equal. After collecting sufficient training data, the correction standard is to correct the reference's EMG curve to a straight line with an amplitude close to 0 μV. The correction formula is calculated as follows:

[0127] dY = 0.97 × (Y 帕金森病者 -Y 参照者 )+0.21; that is, parameter a is 0.97 and parameter b is 0.21;

[0128] According to the correction formula, the electromyography (EMG) of Parkinson's patients was corrected to eliminate the influence of muscle strength-related EMG and extract the muscle tone-related EMG under the uniform motion pattern. (See...) Figure 7 ;

[0129] In another scenario, under induced uniform angular velocity movement, when the Parkinson's patient's limbs are in a relaxed state, and the reference patient guides the Parkinson's patient's limbs to generate uniform angular velocity movement, the electromyography (EMG) of both limbs is recorded. During the uniform angular velocity movement, the muscle strength of the Parkinson's patient and the reference patient is equal. After collecting sufficient training data, the correction standard is to correct the reference patient's EMG curve to a straight line with an amplitude close to 0 μV. The correction formula is calculated as follows:

[0130] dY=0.95×(Y 帕金森病者 -Y 参照者 +0.24; that is, parameter a is 0.95 and parameter b is 0.24;

[0131] According to the correction formula, the electromyography (EMG) of Parkinson's patients was corrected to eliminate the influence of muscle strength-related EMG, and muscle tone-related EMG was extracted during uniform angular velocity movement. (See...) Figure 8 ;

[0132] In another scenario, under induced acceleration, when a normal subject's limb is in a relaxed state and a reference subject guides the normal subject's limb to produce accelerated movement, the electromyography (EMG) of both subjects' moving limbs is recorded; during the accelerated movement L... 正常 At that time, the muscle strength of normal subjects and reference subjects was not equal. After collecting a sufficient training dataset, the correction standard was to correct the electromyography curve of the reference subjects to a straight line with an amplitude close to 0 μV. The correction formula was calculated as follows:

[0133] dY=0.93×(Y 正常者 -Y 参照者 )+0.38; that is, parameter a is 0.93 and parameter b is 0.38;

[0134] The recorded acceleration m is 4, see Figure 9 ,

[0135] According to the acceleration formula F 正常 =mL 正常 F can be obtained 正常 = 4m, where m is the mass;

[0136] In a relaxed limb state, a reference guides a Parkinson's patient to produce accelerated limb movements (L). 帕金森病 At the same time, the electromyography (EMG) of the moving limbs of both individuals was recorded; during the acceleration L... 帕金森病 At that time, the muscle strength of the Parkinson's patient and the control were not equal, and the recorded motor acceleration was 3.5.

[0137] According to the acceleration formula F帕金森病 =mL 帕金森病 F can be obtained 帕金森病 =3.5m

[0138] The correction formula is calculated as follows:

[0139] dY=F 帕金森病 / F 正常 [0.93×(Y 正常者 -Y 参照者 [)+0.38]; that is, parameter a is 0.93 and parameter b is 0.38; F 帕金森病 / F 正常 The value is 3.5 / 4 = 0.87;

[0140] That is, dY = 0.87[0.93 × (Y)] 正常者 -Y 参照者 )+0.38]

[0141] According to the correction formula, the electromyography (EMG) of Parkinson's patients was corrected to eliminate the influence of muscle strength-related EMG and extract the muscle tone-related EMG in the acceleration movement pattern. (See...) Figure 11 ;

[0142] Taking the extraction of tone-related electromyography (EMG) signals from Parkinson's disease as an example, other optional correction formulas include fold correction.

[0143] In other words, under induced uniform angular velocity movement, with the limbs of a Parkinson's patient in a relaxed state, and a reference subject guiding the Parkinson's patient to generate uniform angular velocity movement, the electromyography (EMG) of the moving limbs of both subjects is recorded. During the uniform angular velocity movement, the muscle strength of the Parkinson's patient and the reference subject is equal. After collecting sufficient training data, the correction standard is to correct the reference subject's EMG curve to a straight line with an amplitude close to 0 μV. The correction formula is calculated as follows:

[0144] dY = 0.98 × (Y 帕金森病 / Y 参照者 That is, parameter a is 0.98 and parameter b is 0;

[0145] According to the correction formula, the electromyography (EMG) of Parkinson's patients was corrected to reduce the influence of muscle strength-related EMG and extract the muscle tone-related EMG under the uniform angular velocity movement pattern. (See...) Figure 12 ;

[0146] The model for identifying muscle tone-related electromyography signals in Parkinson's disease is stored in memory, and S304 is executed.

[0147] S304: Electromyography signal and antagonistic pattern feature collection device collects and analyzes muscle tone-related electromyography signal and antagonistic pattern feature data of subject A;

[0148] S305: Electromyography (EMG) signal and adversarial pattern feature collection device collects and analyzes the muscle tone-related EMG signals, adversarial pattern features, and recognition model of the training set of test subject A to determine the muscle tone-related EMG signals of the test subject;

[0149] S306: Analysis of the characteristics of muscle tone-related electromyographic signals and resistance patterns in subject A.

[0150] Example 3

[0151] This embodiment also discloses a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method described in Embodiment 1.

[0152] Example 4

[0153] This embodiment also discloses a computer program product, including a computer program that, when executed by a processor, implements the steps of the method described in Embodiment 1.

[0154] The above are merely preferred embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for extracting muscle tone-related electromyographic signals, characterized in that, Includes the following steps: Constructing a human muscle strength resistance model between the tester and the reference; The process of constructing a human muscle strength resistance pattern between the tester and the reference includes: By applying external force to the reference, the test subject is induced to produce at least one of the following human motions: uniform velocity motion, uniform angular velocity motion, zero acceleration motion, or constant acceleration motion at the test site. The opposing parts of the human muscle strength resistance mode include the upper limbs, lower limbs, head, and trunk; The first confrontation mode characteristic data and the first electromyographic signal of the tester under the human muscle strength confrontation mode were collected, and the second confrontation mode characteristic data and the second electromyographic signal of the reference were collected. Based on the feature data of the second adversarial mode and the second electromyographic signal, a muscle tone-related electromyographic signal correction model is constructed. The process of constructing a muscle tone-related electromyography signal correction model includes: Based on the changes in electromyographic signals of the reference under static, uniform velocity, uniform angular velocity, or constant acceleration conditions, a mapping model between the characteristic data of the adversarial mode and the electromyographic signals is established, which is the muscle tone-related electromyographic signal correction model. The first adversarial mode feature data is input into the muscle tone-related electromyography signal correction model to calculate the theoretical value of the electromyography signal of the test subject. Based on the difference between the first electromyographic signal and the theoretical value of the electromyographic signal, the muscle tone-related electromyographic signal features of the test subject are extracted; The process of extracting muscle tone-related electromyographic signal features from the test subject includes: The difference was calculated based on the first electromyographic signal and the theoretical value of the electromyographic signal. The muscle strength ratio is calculated by comparing the limb mass and acceleration data of the tester and the reference. When the human body is moving with constant acceleration, the difference is compensated based on the muscle strength ratio.

2. The method according to claim 1, characterized in that, The process of collecting the first confrontation mode feature data of the tester and the second confrontation mode feature data of the reference under the human muscle strength confrontation mode includes: The first and second adversarial mode feature data are collected using accelerometers, gyroscopes, or levels. The first and second adversarial mode feature data include motion speed, angular velocity, or acceleration.

3. The method according to claim 1, characterized in that, The process of acquiring the first electromyographic signal of the test subject and the second electromyographic signal of the reference subject under the human muscle strength resistance mode includes: The first and second electromyographic signals are acquired using a surface electromyography device, a needle electromyography device, or a wearable device. The first and second electromyographic signals include amplitude, frequency, and waveform.

4. The method according to claim 1, characterized in that, The formula for the muscle tone-related electromyography signal correction model is as follows: dY=a(Y 检测者 -THE 参照者 )+b: dY=a(Y 检测者 / THE 参照者 )+b: dY=aLog(Y 检测者 / THE 参照者 )+b: Where dY represents the muscle tone-related electromyographic signal characteristics, and a and b are correction coefficients.

5. A device for extracting muscle tone-related electromyographic signals, characterized in that, include: A memory for storing a computer program for the method according to any one of claims 1-4; A processor for executing the computer program to extract muscle tone-related electromyographic signals; Electromyography (EMG) signal acquisition device, used to acquire EMG signals from both the test subject and the reference subject; A motion parameter acquisition device is used to collect motion parameters under human muscle strength resistance mode.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the method according to any one of claims 1-4.

7. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the steps of the method according to any one of claims 1-4.

Images