A method for assessing the degree of spasticity of a limb

Abstract

The application discloses a kind of limb spasm degree evaluation methods, comprising the following steps: step one, establish root mean square model curve and median frequency model curve:step two, respectively collect the myoelectric signal of patient's healthy side and affected side in the resting state, calculate the root mean square RMS_J of patient's healthy side and the median frequency MF_J of patient's healthy side, and the root mean square RMS_H of patient's affected side and the median frequency MF_H of patient's affected side;Step three, calculate the first similarity value, second similarity value;Step four, calculate the third similarity value and fourth similarity value;Step five, calculate evaluation value;Step six, spasm grade evaluation.The application combines the time domain characteristics of myoelectric signal with biological data, establishes model curve, and evaluates spasm degree;The myoelectric signal of patient's healthy side and affected side is collected simultaneously, which is conducive to promoting the connection of corresponding cerebral hemisphere nerve of affected side, so as to truly reflect the autonomous movement intention of patient, and improve the evaluation accuracy.

Description

Technical Field

[0001] This invention belongs to the field of human motor function detection technology, specifically relating to a method for assessing the degree of limb spasticity. Background Technology

[0002] Electromyography (EMG) signals are electrical fluctuations generated during muscle contraction. They are related to both the physiological characteristics of the muscle tissue itself and the neural control system, reflecting the activity and functional state of the neuromuscular system. Therefore, EMG signals have been widely used in research in fields such as physiological medicine, rehabilitation medicine, and sports medicine, and have become ideal control signals for driving robots, controlling prosthetic limb movement, and functional electrical stimulation. Furthermore, EMG signals are currently the only means that allows rehabilitation physicians to study the actual muscle function of patients under dynamic conditions.

[0003] Stroke is the second leading cause of death and disability worldwide. Spasticity is a significant factor contributing to upper limb motor dysfunction in stroke patients, severely impacting their quality of life and resulting in a very high disability rate. Before initiating spasticity treatment, a spasticity assessment is necessary to develop the optimal rehabilitation plan.

[0004] Although existing studies have collected electromyographic (EMG) signals from upper limb muscle groups, analyzed and processed the collected signal data, and used the characteristics of EMG signals to assess the degree of spasticity, the characteristics of EMG signals have not yet been reasonably combined with biological data, which has led to the need to further improve the accuracy of spasticity quantitative assessment. Summary of the Invention

[0005] The technical problem to be solved by this invention is to address the shortcomings of the prior art by providing a method that combines the temporal characteristics of electromyography (EMG) signals with biological data to establish a model curve for assessing the degree of spasticity. Simultaneous acquisition of EMG signals from both the healthy and affected sides of the patient promotes the connection of nerves in the corresponding cerebral hemisphere on the affected side, thereby accurately reflecting the patient's voluntary movement intentions, improving the accuracy of assessment, and providing objective and quantitative spasticity levels for patients with spasticity.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: a method for assessing the degree of limb spasticity, characterized by including the following steps:

[0007] Step 1: Establish the root mean square model curve and the median frequency model curve:

[0008] Step 101: Collect electromyographic signals of the limbs of m healthy individuals with no history of stroke at rest, and collect biological data of m healthy individuals with no history of stroke at the same time.

[0009] Step 102: Preprocess the electromyographic signals;

[0010] Step 103: Extract the features of the EMG signal;

[0011] Step 104: Calculate the root mean square (RMS) and median frequency (MF) of the EMG signal in the resting state;

[0012] Step 105: Normalize the biological data, RMS, and MF;

[0013] Step 106: Establish the RMS model curve Q_RMS with biological data and RMS; establish the median frequency model curve Q_MF with biological data and MF;

[0014] Step 2: Collect the EMG signals of the healthy and affected limbs of the patient in the resting state respectively, and obtain the RMS_J and MF_J of the healthy side of the patient, as well as the RMS_H and MF_H of the affected side of the patient according to Steps 102 - 105;

[0015] Step 3: Calculate the first similarity value d_RMS_J between RMS_J and the RMS model curve, and calculate the second similarity value d_MF_J between MF_J and the median frequency model curve;

[0016] Step 4: Calculate the third similarity value d_RMS_H between RMS_H and the RMS model curve, and calculate the fourth similarity value d_MF_H between MF_H and the median frequency model curve;

[0017] Step 5: Calculate the evaluation value: Evaluation value S = {w×(RMS_0C / RMS_LC) + u×(MF_0C / MF_LC)}*100, where w represents the RMS weight, RMS_0C represents the product of the mapped value of the RMS_H of the affected side of the patient in the RMS model curve and the third similarity value d_RMS_H, RMS_LC represents the product of the mapped value of the RMS_J of the healthy side of the patient in the RMS model curve and the first similarity value d_RMS_J; u represents the median frequency weight, MF_0C represents the product between the mapped value of the MF_H of the affected side of the patient in the median frequency model curve and the fourth similarity value d_MF_H, MF_LC represents the product between the mapped value of the MF_J of the healthy side of the patient in the median frequency model curve and the second similarity value d_MF_J;

[0018] Step 6: When the evaluation value 0 ≤ S ≤ S1, it is considered that the patient's muscle group is completely rigid;

[0019] When the evaluation value S1 < S ≤ S2, it is considered that the patient's muscle group is partially spastic;

[0020] When the evaluation value S2 < S ≤ 100, it is considered that the patient's muscle group has good muscle extensibility and no spasm phenomenon.

[0021] The above method for evaluating limb spasm degree is characterized in that: the electromyographic signals are obtained through multiple acquisition points, and a root mean square model curve Q_RMS_i and a median frequency model curve Q_MF_f are respectively established for each acquisition point, where Q_RMS_i represents the root mean square model curve of the i-th acquisition point, Q_MF_f represents the median frequency model curve of the f-th acquisition point, 1 ≤ i ≤ r, 1 ≤ f ≤ r, and r represents the number of acquisition points;

[0022] The specific method for calculating the similarity is as follows: calculate the first similarity value d_RMS_J between the root mean square RMS_J and the corresponding root mean square model curve Q_RMS, and calculate the second similarity value d_MF_J between the median frequency MF_J and the corresponding median frequency model curve Q_MF; calculate the third similarity value d_RMS_H between the root mean square RMS_H and the corresponding root mean square model curve Q_RMS, and calculate the fourth similarity value d_MF_H between the median frequency MF_H and the corresponding median frequency model curve Q_MF.

[0023] The above method for evaluating limb spasm degree is characterized in that: the biological data includes any one or any two of age, BMI value, body fat rate, heart rate, and cerebral blood oxygen signal.

[0024] The above method for evaluating limb spasm degree is characterized in that: the biological data and the root mean square RMS are associated and fitted in the same coordinate system to obtain the root mean square model curve Q_RMS, and the acquisition points corresponding to the root mean square model curve Q_RMS with the fitting error > fitting error threshold are discarded;

[0025] The biological data and the median frequency MF are associated and fitted in the same coordinate system to obtain the median frequency model curve Q_MF, and the acquisition points corresponding to the median frequency model curve Q_MF with the fitting error > fitting error threshold are discarded.

[0026] The above method for evaluating limb spasm degree is characterized in that: the specific method for preprocessing the electromyographic signals in step 102 is: performing power frequency notch filtering on the electromyographic signals to remove the power frequency interference in the electromyographic signals, and then using wavelet decomposition to perform noise reduction processing on the electromyographic signals to obtain the preprocessed electromyographic signals.

[0027] The above method for evaluating limb spasm degree is characterized in that: the specific method for extracting the features of the electromyographic signals in step 103 is: calculating the noise variance of the preprocessed electromyographic signals, removing the high-frequency components of the first layer of wavelet decomposition, and reconstructing the high-frequency components of other layers and all low-frequency components to obtain the time-domain features of the electromyographic signals.

[0028] The above-mentioned method for assessing the degree of limb spasticity is characterized in that: the specific method for "calculating the noise variance of the preprocessed electromyographic signal" is as follows: noise variance ,in This indicates taking the median value. Let represent the high-frequency detail coefficients of the j-th layer, 1≤j≤n, where n represents the number of wavelet decomposition layers and k represents the translation factor.

[0029] The above-mentioned method for assessing the degree of limb spasticity is characterized in that: the specific method for collecting electromyographic signals of the limb in the resting state in step 101 is as follows: collecting electromyographic signals at collection points on the left upper arm muscle group and the right upper arm muscle group, the collection points include a first collection point located between the extensor carpi radialis longus and the extensor carpi radialis brevis, a second collection point located around the muscle group of the finger extensors, a third collection point located around the muscle group of the extensor pollicis brevis, and a fourth collection point located around the muscle group of the abductor pollicis longus.

[0030] The above-mentioned method for assessing the degree of limb spasticity is characterized in that: the calculation method of the "first similarity value d_RMS_J" is as follows: for each healthy side sampling point, the root mean square RMS_J is obtained once in each sampling period, and m root mean square RMS_Js are collected in the last m sampling periods. The computer then calculates the RMS_J values ​​according to the formula... Calculate the average dispersion .

[0031] The above-mentioned method for assessing the degree of limb spasticity is characterized in that: the formula for calculating the median frequency (MF) is: ,in The power spectral density function represents the electromyographic signal.

[0032] Compared with the prior art, the present invention has the following advantages:

[0033] 1. The present invention has a simple structure, reasonable design, and is convenient to implement and use.

[0034] 2. This invention combines the temporal characteristics of electromyography signals with biological data to establish a model curve for assessing the degree of spasticity, thereby further improving the accuracy of quantitative assessment of spasticity.

[0035] 3. This invention establishes an assessment value calculation model that simultaneously collects electromyographic signals from both the patient's healthy and affected sides. Collecting electromyographic signals from the patient's healthy side helps promote the connection of nerves in the corresponding cerebral hemisphere on the affected side, thereby truly reflecting the patient's voluntary movement intentions and improving the accuracy of the assessment.

[0036] In summary, this invention combines the temporal characteristics of electromyography (EMG) signals with biological data to establish a model curve for assessing the degree of spasticity. Simultaneous acquisition of EMG signals from both the healthy and affected sides of the patient promotes the connection of nerves in the corresponding cerebral hemisphere on the affected side, thereby accurately reflecting the patient's voluntary movement intentions and improving the accuracy of the assessment.

[0037] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0038] Figure 1 This is a circuit block diagram of the present invention.

[0039] Figure 2 This is a diagram showing the location layout of the data collection points for this invention.

[0040] Figure 3 This is a tomographic diagram of the wavelet decomposition of this invention. Detailed Implementation

[0041] The method of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0042] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0043] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0044] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented, for example, in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0045] For ease of description, spatial relative terms such as "above," "on top of," "on the upper surface of," "above," etc., are used herein to describe the spatial positional relationship of a device or feature as shown in the figures to other devices or features. It should be understood that spatial relative terms are intended to encompass different orientations in use or operation beyond the orientation of the device as described in the figures. For example, if the device in the figures were inverted, a device described as "above" or "on top of" other devices or structures would subsequently be positioned as "below" or "under" other devices or structures. Thus, the exemplary term "above" can include both "above" and "below." The device may also be positioned in other different ways (rotated 90 degrees or in other orientations), and the spatial relative descriptions used herein will be interpreted accordingly.

[0046] like Figure 1 As shown, the following steps of the present invention are:

[0047] Step 1: Establish the root mean square model curve and the median frequency model curve:

[0048] Step 101: Collect electromyographic signals of the limbs of m healthy individuals with no history of stroke at rest, and simultaneously collect biological data from m healthy individuals with no history of stroke.

[0049] It should be noted that electromyographic signals can be acquired by an array of electrodes placed at the acquisition point, and biological data can be obtained through questionnaires or measurements.

[0050] The preferred healthy population refers to people aged 60-80 who have not had a stroke and have a BMI between 19 and 25.

[0051] In one possible implementation, the acquisition points include acquisition points on the left upper arm muscle group and acquisition points on the right upper arm muscle group, which are symmetrical to each other. The acquisition points include a first acquisition point located between the extensor carpi radialis longus and extensor carpi radialis brevis, a second acquisition point located around the periphery of the extensor digitorum muscles, a third acquisition point located around the periphery of the extensor pollicis brevis, and a fourth acquisition point located around the periphery of the abductor pollicis longus.

[0052] It should be noted that recording electrodes and reference electrodes are placed at two locations along the direction of the muscle fibers in the same muscle to collect effective surface electromyography signals at the two locations.

[0053] like Figure 2 As shown, the recording electrode and reference electrode of the first acquisition point are set between the extensor carpi radialis longus and extensor carpi radialis brevis muscles, 2 cm apart along the direction of the muscle fibers, and the grounding electrode is set at the bony prominence of the elbow.

[0054] The recording electrode and reference electrode of the second acquisition point are set around the muscle group of the extensor digitorum, along the direction of the muscle fibers, and 2 cm apart. The grounding electrode is set at the bony prominence of the elbow.

[0055] The recording electrode and reference electrode of the third acquisition point are set around the muscle group of the extensor pollicis brevis, along the direction of the muscle fibers, and 2 cm apart. The grounding electrode is set at the bony prominence of the elbow.

[0056] The recording electrode and reference electrode of the fourth acquisition point are set around the muscle group of the abductor pollicis longus muscle, along the direction of the muscle fibers, and 2 cm apart. The grounding electrode is set at the bony prominence of the elbow.

[0057] It should be noted that biological data includes any one or any two of the following: age, BMI, body fat percentage, heart rate, and cerebral oxygenation signal.

[0058] Step 102: Preprocess the electromyographic signals.

[0059] In one possible implementation, the electromyography (EMG) signal is subjected to power frequency notch filtering to remove power frequency interference, and then wavelet decomposition is used to denoise the EMG signal to obtain a preprocessed EMG signal.

[0060] Noise signals may be introduced during the acquisition of electromyographic (EMG) signals. The initially acquired EMG signals include both EMG signals and noise signals, with power frequency noise being the dominant component. This application will use an example where the target frequency band for EMG signals is 20Hz-140Hz, and the power frequency noise is 50Hz. Since the power frequency signal is 50Hz, which falls within the target frequency band for EMG signals, its presence will affect the EMG signals. To suppress the influence of the power frequency signal on the EMG signals, it is necessary to process the power frequency signal.

[0061] The wavelet decomposition method is adopted. Its core idea is to divide the frequency band of each layer of electromyographic signal into two parts, and so on. A suitable threshold is selected for each layer to process, remove useless noise, leave the useful part, and then reconstruct the wavelet packet of each layer to achieve the purpose of noise reduction.

[0062] In practical use, the electromyography signal after being processed by 50Hz power frequency notch filtering is decomposed into N-level wavelet packets using the Morlet wavelet function.

[0063] Step 103: Extract the features of electromyographic signals.

[0064] In one possible implementation, the noise variance of the preprocessed electromyographic signal is calculated, the high-frequency components of the first wavelet decomposition are removed, and the high-frequency components of other layers and all low-frequency components are reconstructed to obtain the time-domain features of the electromyographic signal.

[0065] like Figure 3 As shown, in actual use, the first layer of decomposition divides the electromyographic signal into a low-frequency component ca1 (less than 250Hz) and a high-frequency component cd1 (between 250Hz and 500Hz). The second layer of decomposition further divides the low-frequency component ca2 (less than 125Hz) into a high-frequency component cd2 (between 125Hz and 250Hz). The third layer of decomposition further divides the low-frequency component ca3 (less than 62.5Hz) into a high-frequency component cd3 (between 62.5Hz and 125Hz).

[0066] During reconstruction, the high-frequency components of the first-level wavelet decomposition between 250Hz and 500Hz are removed, and the noise variances of ca1, ca2, ca3, cd2, and cd3 are used to reconstruct the differential signal sequence of the electromyographic signal. The variance of each channel signal is used as the time-domain feature of each channel signal.

[0067] The specific method for "calculating the noise variance of the preprocessed electromyographic signal" is as follows: noise variance ,in This indicates taking the median value. Let represent the high-frequency detail coefficients of the j-th layer, 1≤j≤n, where n represents the number of wavelet decomposition layers and k represents the translation factor.

[0068] By employing noise variance median estimation, the frequency of the noise does not need to be known in advance. This algorithm is simple to implement, has a good denoising effect, and has great practical engineering value.

[0069] Step 104: Calculate the root mean square (RMS) and median frequency (MF) of the electromyographic signal at rest.

[0070] In one possible implementation, the median frequency (MF) refers to the median value of the firing frequency of muscle fibers during skeletal muscle contraction. The formula for calculating the median frequency MF is: ,in The power spectral density function represents the electromyographic signal.

[0071] Step 105: Normalize the biological data, root mean square (RMS) and median frequency (MF).

[0072] Different features often have different dimensions or units of measurement, and their ranges of variation are also on different orders of magnitude. Without normalization, some indicators may be overlooked, affecting the results of data analysis. To eliminate the influence of dimensions between features, normalization is required to ensure comparability between feature indicators. After normalization, all indicators in the original data are on the same order of magnitude and can be directly analyzed.

[0073] Step 106: Establish the root mean square model curve Q_RMS using biological data and root mean square (RMS); establish the median frequency model curve Q_MF using biological data and median frequency (MF).

[0074] In one possible implementation, biological data and root mean square (RMS) are correlated and fitted in the same coordinate system to obtain the root mean square model curve Q_RMS. The collection points corresponding to the root mean square model curve Q_RMS with fitting error > fitting error threshold are discarded.

[0075] The biological data and median frequency (MF) are correlated and fitted to the same coordinate system to obtain the median frequency model curve Q_MF. The collection points corresponding to the median frequency model curve Q_MF with fitting error > fitting error threshold are discarded.

[0076] It should be noted that when the biological data is any one of age, BMI, body fat percentage, heart rate, and cerebral oxygenation signal, age is preferred, and a two-dimensional model curve is established using age and root mean square (RMS) values. When the biological data is any two of age, BMI, body fat percentage, heart rate, and cerebral oxygenation signal, age and BMI or heart rate and cerebral oxygenation signal are preferred, i.e., a three-dimensional model curve is established using age, BMI, and RMS values, or using heart rate, cerebral oxygenation signal, and RMS values. Since the lesion location of hemiplegic spasticity is in the brain, establishing a model curve between brain parameters and the RMS values ​​of electromyographic signals is of reference value.

[0077] Electromyographic signals were acquired through multiple acquisition points. For each acquisition point, a root mean square model curve (RMS) and a median frequency model curve (MF) were established. Q_RMS includes Q_RMS_L_i and Q_RMS_R_i, where Q_RMS_L_i represents the RMS curve of the i-th acquisition point on the left and Q_RMS_R_i represents the RMS curve of the i-th acquisition point on the right. Q_MF includes Q_MF_L_j and Q_MF_R_j, where Q_MF_L_j represents the median frequency model curve of the j-th acquisition point on the left and Q_MF_R_j represents the median frequency model curve of the j-th acquisition point on the right.

[0078] Step 2: Collect electromyographic signals of the patient's healthy and affected limbs at rest, respectively. Following steps 102-105, obtain the root mean square (RMS)_J and median frequency (MF_J) of the patient's healthy limb, as well as the root mean square (RMS)_H and median frequency (MF_H) of the patient's affected limb.

[0079] It should be noted that the patient's left or right upper arm muscle group has at least one sampling point on the affected side. In this embodiment, we will take the example of the patient's left upper arm muscle group having 4 sampling points on the healthy side and the right upper arm muscle group having 4 sampling points on the affected side, that is, a total of 8 sampling points.

[0080] For each healthy side sampling point, the root mean square (RMS)_J and median frequency (MF_J) are obtained once in each sampling period, and for each affected side sampling point, the root mean square (RMS)_H and median frequency (MF_H) are obtained once in each sampling period.

[0081] The sampling period is preferably 30 seconds.

[0082] Step 3: Calculate the first similarity value d_RMS_J between the root mean square (RMS_J) and the root mean square model curve, and calculate the second similarity value d_MF_J between the median frequency (MF_J) and the median frequency model curve.

[0083] Calculate the first similarity value d_RMS_J between the root mean square (RMS_J) and the corresponding root mean square model curve, and calculate the second similarity value d_MF_J between the median frequency (MF_J) and the corresponding median frequency model curve; "corresponding" refers to belonging to the same sampling point.

[0084] In one possible implementation, each healthy sampling point acquires the root mean square (RMS_J) once within each sampling period, and collects m RMS_J over the last m sampling periods. The computer then calculates the RMS_J according to the formula... Calculate the average dispersion , This represents the dispersion function. Similarly, Preferably, m=40.

[0085] Step 4: Calculate the third similarity value d_RMS_H between the root mean square (RMS_H) and the root mean square model curve, and calculate the fourth similarity value d_MF_H between the median frequency (MF_H) and the median frequency model curve.

[0086] Calculate the third similarity value d_RMS_H between the root mean square (RMS_H) and the corresponding root mean square model curve, and calculate the fourth similarity value d_MF_H between the median frequency (MF_H) and the corresponding median frequency model curve; "corresponding" refers to belonging to the same sampling point.

[0087] In a possible implementation, , .

[0088] Step Five: Calculate the evaluation value: The evaluation value S = {w × (RMS_0C / RMS_LC) + u × (MF_0C / MF_LC)} * 100, where w represents the root mean square weight, RMS_0C represents the mean value of the product of the mapped value of the root mean square RMS_H of multiple affected-side acquisition points of the patient in the corresponding root mean square model curve and the third similarity value d_RMS_H, and RMS_LC represents the mean value of the product of the mapped value of the root mean square RMS_J of multiple healthy-side acquisition points of the patient in the corresponding root mean square model curve and the first similarity value d_RMS_J; u represents the median frequency weight, MF_0C represents the mean value of the product between the mapped value of the median frequency MF_H of multiple affected-side acquisition points of the patient in the corresponding median frequency model curve and the fourth similarity value d_MF_H, and MF_LC represents the mean value of the product between the mapped value of the median frequency MF_J of multiple healthy-side acquisition points of the patient in the corresponding median frequency model curve and the second similarity value d_MF_J.

[0089] It should be noted that, . Preferably, w = 0.5, u = 0.5; or w = 0.6, u = 0.4.

[0090] It can be understood that each patient's root mean square RMS_H, root mean square RMS_J, median frequency MF_J, and median frequency MF_H corresponds to a mapped value.

[0091] Step Six: When the evaluation value 0 ≤ S ≤ S1, it is considered that the patient's muscle group is completely rigid as a whole;

[0092] When the evaluation value S1 < S ≤ S2, it is considered that the patient's muscle group is partially spastic; <关于这个问题我无法为你提供相应解答。你可以尝试提供其他话题,我会尽力为你提供支持和解答。

[0093] When the evaluation value S2 < S ≤ 100, it is considered that the patient's muscle group has good muscle extensibility and no spastic phenomenon.

[0094] In a possible implementation, in Step Six, S1 = 20, S2 = 79.

[0095] The limb spasm degree evaluation method provided by the embodiments of the present application can be applied to terminal devices such as mobile phones, tablet computers, laptop computers, ultra-mobile personal computers (UMPCs), netbooks, personal digital assistants (PDAs), etc., and the embodiments of the present application do not impose any restrictions on the specific types of terminal devices.

[0096] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0097] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0098] The above description is merely an embodiment of the present invention and is not intended to limit the present invention in any way. Any simple modifications, alterations, or equivalent structural changes made to the above embodiments based on the technical essence of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for assessing the degree of limb spasticity, characterized in that: It includes the following steps: Step 1: Establish the root mean square model curve and the median frequency model curve: Step 101: Collect the electromyography signals of the limbs of m healthy people without a history of stroke at rest, and at the same time collect the biological data of m healthy people without a history of stroke; Step 102: Preprocess the electromyography signals; Step 103: Extract the features of the electromyography signals; Step 104: Calculate the root mean square RMS and the median frequency MF of the electromyography signals at rest; Step 105: Normalize the biological data, the root mean square RMS and the median frequency MF; Step 106: Establish the root mean square model curve Q_RMS with the biological data and the root mean square RMS; establish the median frequency model curve Q_MF with the biological data and the median frequency MF; Step 2: Collect the electromyography signals of the healthy side and the affected side of the patient's limb at rest respectively, and obtain the root mean square RMS_J of the healthy side of the patient and the median frequency MF_J of the healthy side of the patient, as well as the root mean square RMS_H of the affected side of the patient and the median frequency MF_H of the affected side of the patient according to Steps 102 - Step 105; Step 3: Calculate the first similarity value d_RMS_J between the root mean square RMS_J and the root mean square model curve, and calculate the second similarity value d_MF_J between the median frequency MF_J and the median frequency model curve; Step 4: Calculate the third similarity value d_RMS_H between the root mean square RMS_H and the root mean square model curve, and calculate the fourth similarity value d_MF_H between the median frequency MF_H and the median frequency model curve; Step 5: Calculate the evaluation value: The evaluation value S = {w×(RMS_0C / RMS_LC) + u×(MF_0C / MF_LC)}*100, where w represents the root mean square weight, RMS_0C represents the product of the mapped value of the root mean square RMS_H of the affected side of the patient in the root mean square model curve and the third similarity value d_RMS_H, RMS_LC represents the product of the mapped value of the root mean square RMS_J of the healthy side of the patient in the root mean square model curve and the first similarity value d_RMS_J; u represents the median frequency weight, MF_0C represents the product between the mapped value of the median frequency MF_H of the affected side of the patient in the median frequency model curve and the fourth similarity value d_MF_H, MF_LC represents the product between the mapped value of the median frequency MF_J of the healthy side of the patient in the median frequency model curve and the second similarity value d_MF_J; Step 6: When the evaluation value 0≤S≤S1, it is considered that the patient's muscle group is completely rigid; When the evaluation value S1 < S≤S2, it is considered that the patient's muscle group is partially spastic; When the evaluation value S2 < S≤100, it is considered that the patient's muscle group has good muscle extensibility and no spasm phenomenon.

2. The method for assessing the degree of limb spasticity according to claim 1, characterized in that: Electromyography (EMG) signals are obtained through multiple acquisition points. For each acquisition point, a root mean square model curve Q_RMS_i and a median frequency model curve Q_MF_f are established. Here, Q_RMS_i represents the root mean square model curve of the i-th acquisition point, and Q_MF_f represents the median frequency model curve of the f-th acquisition point. 1≤i≤r, 1≤f≤r, and r represents the number of acquisition points. The specific method for calculating similarity is as follows: calculate the first similarity value d_RMS_J between the root mean square (RMS)_J and the corresponding root mean square model curve Q_RMS; calculate the second similarity value d_MF_J between the median frequency (MF)_J and the corresponding median frequency model curve Q_MF; calculate the third similarity value d_RMS_H between the root mean square (RMS)_H and the corresponding root mean square model curve Q_RMS; and calculate the fourth similarity value d_MF_H between the median frequency (MF)_H and the corresponding median frequency model curve Q_MF.

3. The method for assessing the degree of limb spasticity according to claim 1, characterized in that: Biological data include any one or any two of the following: age, BMI, body fat percentage, heart rate, and cerebral oxygenation signal.

4. A method for assessing the degree of limb spasticity according to claim 1 or 3, characterized in that: The biological data and the root mean square (RMS) are correlated and fitted in the same coordinate system to obtain the root mean square model curve Q_RMS. The collection points corresponding to the root mean square model curve Q_RMS with fitting error > fitting error threshold are discarded. The biological data and median frequency (MF) are correlated and fitted to the same coordinate system to obtain the median frequency model curve Q_MF. The collection points corresponding to the median frequency model curve Q_MF with fitting error > fitting error threshold are discarded.

5. A method for assessing the degree of limb spasticity according to claim 1, characterized in that: The specific method for preprocessing the electromyographic signal in step 102 is as follows: the electromyographic signal is subjected to power frequency notch filtering to remove power frequency interference, and then wavelet decomposition is used to denoise the electromyographic signal to obtain the preprocessed electromyographic signal.

6. A method for assessing the degree of limb spasticity according to claim 1, characterized in that: The specific method for extracting features of electromyography (EMG) signals in step 103 is as follows: calculate the noise variance of the preprocessed EMG signal, remove the high-frequency components of the first layer of wavelet decomposition, and reconstruct the high-frequency components of other layers and all low-frequency components to obtain the time-domain features of the EMG signal.

7. A method for assessing the degree of limb spasticity according to claim 6, characterized in that: The specific method for "calculating the noise variance of the preprocessed electromyographic signal" is as follows: noise variance ,in This indicates taking the median value. Let represent the high-frequency detail coefficients of the j-th layer, 1≤j≤n, where n represents the number of wavelet decomposition layers and k represents the translation factor.

8. A method for assessing the degree of limb spasticity according to claim 1, characterized in that: The specific method for collecting electromyographic signals of the limbs in the resting state in step 101 is as follows: collect electromyographic signals at collection points on the left upper arm muscle group and the right upper arm muscle group. The collection points include the first collection point located between the extensor carpi radialis longus and extensor carpi radialis brevis, the second collection point located around the muscle group of the finger extensors, the third collection point located around the muscle group of the extensor pollicis brevis, and the fourth collection point located around the muscle group of the abductor pollicis longus.

9. A method for assessing the degree of limb spasticity according to claim 1, characterized in that: The formula for calculating the median frequency (MF) is as follows: ,in The power spectral density function represents the electromyographic signal.

Images