Evaluation system of disuse muscle atrophy after bone and joint injury based on surface electromyography

Abstract

The present invention discloses a system for evaluating the degree of disuse muscular atrophy after bone and joint injury based on flexible electrode surface electromyography. The system comprises a data acquisition module, a data front-end module, an algorithm analysis module and a flexible electrode module. The flexible electrode module is in close contact with the position to be measured and receives electromyographic signals. The data acquisition module is responsible for the acquisition and analog-to-digital conversion of the electromyographic signals of the flexible electrode module and sends them to the data front-end module. The data front-end module receives the data and filters and sends them to the algorithm analysis module. The algorithm analysis module receives the data and performs quantitative evaluation on the degree of disuse muscular atrophy, and contains a trained classification evaluation model. The present invention effectively avoids the instability of single parameter detection, realizes early diagnosis and early warning of disuse muscular atrophy, improves the accuracy and robustness of muscular atrophy monitoring, and improves the overall portability and convenience of operation.

Description

Technical Field

[0001] The invention belongs to a medical diagnosis and evaluation system in the field of biomedical engineering, and in particular relates to a system for evaluating the degree of disuse muscle atrophy after bone and joint injury based on flexible electrode surface electromyography. Background Art

[0002] Bone and joint injuries are an ancient disease and the leading cause of disability and even death worldwide. The development of effective treatment and rehabilitation technologies for bone and joint injuries can not only significantly improve people's quality of life, but also greatly reduce the cost of medical resources, playing a huge role in improving people's health and economic benefits.

[0003] Disuse muscle atrophy refers to a series of changes in skeletal muscle function, structural morphology, biochemical indicators and molecular biology caused by decreased limb activity, limited activity, or weightlessness. The main symptoms are reduced muscle volume, reduced muscle weight, thinning or even disappearance of muscle fibers, and decreased muscle strength. For disuse muscle atrophy caused by bone and joint injuries, how to conduct accurate and scientific pathological assessment, including dynamic and accurate feedback on the efficacy during the rehabilitation process and accurate application of rehabilitation measures, has important clinical guiding significance.

[0004] At present, the clinical assessment of the physiological condition of limb muscles in disuse muscular atrophy mainly relies on external characterization methods such as subjective motor function assessment scales, circumference measurement, range of joint motion, local tension judgment of soft tissue, muscle contraction strength and endurance.

[0005] Xie Longhan et al. mentioned in "A human upper limb prosthetic device simulating muscle strength grading" (CN201822107030.9) that the difference in circumference between the affected limb and the healthy limb was used. The limb swelling grading standard divides the degree of swelling into three levels; the muscle strength grading method divides muscle strength into six levels: level 0 is no measurable muscle contraction, level 1 is slight contraction but cannot cause joint movement, level 2 is full range of joint movement under weight loss, level 3 is full range of joint movement against gravity but cannot resist resistance; level 4 is anti-gravity and can resist a certain resistance movement; level 5 is anti-gravity and can resist a certain resistance movement; level 6 is anti-gravity and can resist a certain resistance movement; level 7 is anti-gravity and can resist a certain resistance movement; level 8 is anti-gravity and can resist a certain resistance movement; level 9 is anti-gravity and can resist a certain resistance movement; level 10 is anti-gravity and can resist a certain resistance movement; level 11 is anti-gravity and can resist a certain resistance movement; level 12 is anti-gravity and can resist a certain resistance movement; level 13 is anti-gravity and can resist a certain resistance movement; level 14 is anti-gravity and can resist a certain resistance movement; level 15 is anti-gravity and can resist a certain resistance movement; level 16 is anti-gravity and can resist a certain resistance movement; level 17 is anti-gravity and can resist a certain resistance movement; level 18 is anti-gravity and can resist a certain resistance movement; level 19 is anti-gravity and can resist a certain resistance movement; level 20 is anti-gravity and can resist a certain resistance movement; level 21 is anti-gravity and can resist a certain resistance movement; level 22 is anti-gravity and can resist a certain resistance movement; level 23 is anti-gravity and can resist a certain resistance movement; level 24 is The first level is anti-gravity and sufficient resistance exercise, and the degree of muscle atrophy is assessed from three aspects: muscle contraction, joint movement, and anti-gravity resistance capacity; Gao Yunyuan mentioned in "Upper Limb Shoulder, Elbow and Wrist Joint Movement Function Evaluation Method Based on Biological Motion Information" (CN201310022821.0) that the simplified Fugl-Meyer movement function evaluation was used to quantify the upper and lower limb functions to a percentage system, and a five-level scoring standard was used to simply evaluate the effectiveness of muscle strength recovery and the quality of life score, which also evaluated the rehabilitation of disuse muscle atrophy from three aspects.

[0006] However, the existing evaluation methods have the following problems: (1) They are highly subjective and can only perform qualitative grade evaluations, rather than objective and accurate quantitative evaluations. The results largely rely on the experience and judgment of the diagnostician. (2) They are highly uncertain and are easily affected by external factors such as swelling and blood flow. They have weak anti-interference capabilities and insufficient system robustness. (3) The measurement method is relatively complex and sometimes requires the use of large instruments, making the system difficult to operate. Summary of the invention

[0007] In view of the shortcomings of the prior art, the present invention provides a system for evaluating the degree of disuse muscle atrophy after bone and joint injuries based on flexible electrode surface electromyography, which can effectively avoid the instability of single parameter detection and realize early diagnosis and early warning of disuse muscle atrophy.

[0008] The technical solution of the present invention is as follows:

[0009] It includes a data acquisition module, a data front-end module, an algorithm analysis module and at least one flexible electrode module;

[0010] The flexible electrode module is used to closely contact the position to be measured and receive the electromyographic signal;

[0011] The data acquisition module is responsible for controlling the flexible electrode module to collect and convert the electromyographic signals into analog-to-digital signals, and sending the collected and converted data to the data front-end module;

[0012] The data front-end module is used to receive and filter the data from the data acquisition module, and send the filtered data to the algorithm analysis module;

[0013] The algorithm analysis module is used to receive the filtered data from the data front-end module to quantitatively evaluate the degree of disuse muscular atrophy, and the algorithm analysis module contains a trained classification evaluation model.

[0014] The flexible electrode module is arranged on the lower surface of the flexible circuit substrate. The upper surface of the flexible circuit substrate is provided with a transparent film layer, which is in close contact with the position to be measured and senses and receives the electromyographic signal.

[0015] The position to be measured is the body surface of the human body.

[0016] The data acquisition module and the data front-end module are integrated in a multi-channel surface electromyography acquisition circuit system. The data acquisition module and the flexible electrode module, and the data acquisition module and the data front-end module are connected via wired cables.

[0017] The algorithm analysis module is a host computer with data processing and computing capabilities, and the algorithm analysis module is connected to the data front-end module via wired or wireless connections.

[0018] The evaluation system for the degree of disuse muscle atrophy after bone and joint injury is performed in the following manner:

[0019] 1) The process of evaluating muscle function in the early stage of bone and joint injury includes:

[0020] (1-1):

[0021] The flexible electrode modules were placed in close contact with the tibialis anterior and gastrocnemius muscles of the human body, and the ankle joints on both sides of the human body performed two types of movements:

[0022] While actively dorsiflexing the ankle joint, the flexible electrode module is controlled to collect sEMG signals from the tibialis anterior muscle and the gastrocnemius muscle respectively.

[0023] While actively plantar flexing the ankle joint, the flexible electrode module is controlled to collect sEMG signals from the tibialis anterior muscle and gastrocnemius muscle respectively.

[0024] Both of the above movements lasted for 60 seconds;

[0025] (1-2): The sEMG data collected above is subjected to 20-500 Hz bandpass filtering by the data front-end module and then sent to the algorithm analysis module;

[0026] (1-3): The algorithm analysis module extracts the time domain and frequency domain feature values ​​in the sEMG data, further calculates multiple parameter values ​​in the time domain and frequency domain, and evaluates the muscle function status of the human body;

[0027] 2) Quantitative assessment of the degree of postoperative muscle atrophy, including:

[0028] (2-1): Repeat the above (1-1) and (1-2), send the sEMG data to the algorithm analysis module to extract the time domain and frequency domain feature values, and further calculate multiple parameter values ​​in the time domain and frequency domain.

[0029] The trained classification evaluation model is used to quantitatively evaluate the development stage of postoperative disuse muscle atrophy.

[0030] In (1-3), in the process of evaluating muscle function in the early stage of bone and joint injury, multiple parameter values ​​in the time domain and frequency domain include a first time domain parameter and a first frequency domain parameter. The first time domain parameter is the root mean square value, the integrated electromyographic value and the variance. The first frequency domain parameter is the number of zero crossing points, the median frequency and the average power spectrum frequency. The above six parameters are used to evaluate the muscle function status in the early stage of bone and joint injury.

[0031] The specific procedures for evaluating muscle function in the early stages of bone and joint injury are as follows: taking the healthy side of the human body as the standard, if the values ​​of various parameters of the affected side of the human body drop to 80% or less of the corresponding parameter values ​​of the healthy side, it is judged that the bone and joint injury has already caused damage to muscle function.

[0032] In (2-1), in the process of quantitative evaluation of the degree of atrophy after surgery, multiple parameter values ​​in the time domain and frequency domain are calculated, including the second time domain parameter and the second frequency domain parameter. The second time domain parameter is the root mean square value and variance, and the second frequency domain parameter is the number of zero crossing points, the median frequency and the average power spectrum frequency. The above five parameters are input into the trained classification evaluation model to obtain the current stage of muscle atrophy.

[0033] In the process of evaluating muscle function in the early stage of bone and joint injury and quantitatively evaluating the degree of atrophy after surgery, the algorithm analysis module extracts the time domain and frequency domain eigenvalues ​​as follows:

[0034] For the 60-s duration, after eliminating the 5-s period of unstable force in the early and late stages of force generation, the data between the 5th and 55th seconds were extracted and obtained. Then, a window length of 0.25 seconds was set, with 1 / 4 overlap between windows. 800 samples were extracted from the data between the 5th and 55th seconds in each window period. A total of 3200 sample points were obtained for each human body based on the two muscles on the affected and healthy sides. The time domain and frequency domain eigenvalues ​​of all sample points were used to further calculate multiple parameter values ​​in the time domain and frequency domain.

[0035] The classification evaluation model is a support vector machine (SVM) based on the RBF (Radial Basis Function, RBF) kernel function. The classification evaluation model uses a cross-validation method to find the optimal c / g parameters during the training process;

[0036] Before training, we first construct labeled sample cases based on sample points according to medical standards and combined with MRI (Magnetic Resonance Imaging) of the lower leg to obtain sample case data. The label represents the current stage parameter of atrophy.

[0037] For each sample data, the maximum and minimum values ​​are normalized to the interval [0,1], and then 80% of the sample data are randomly intercepted from all the sample point data as the training set, and the support vector machine is created and trained. At the same time, cross-validation is performed on the training set to find the model parameter setting with the highest cross-validation accuracy, and the support vector machine is optimized. The remaining 20% ​​of the sample data is used as the test set.

[0038] By utilizing the present invention, the instability of single parameter detection can be effectively avoided, and early diagnosis and early warning of disuse muscular atrophy can be achieved.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] 1. The system of the present invention can be used to evaluate the early muscle function of bone and joint injuries. When traditional evaluation methods do not show obvious distinction, the comprehensive parameters have a relatively obvious attenuation, which effectively avoids the instability of single parameter detection and can achieve early diagnosis and early warning of disuse muscle atrophy.

[0041] 2. The system of the present invention can quantitatively evaluate the degree of disuse muscle atrophy after bone and joint injury surgery. Compared with traditional evaluation methods, it is more sensitive to changes and can quantitatively evaluate the development of the current atrophy degree through simple external active actions, thereby improving the accuracy of muscle atrophy monitoring;

[0042] 3. The system of the present invention is not easily affected by external factors such as swelling, blood flow, etc. during evaluation, has strong anti-interference ability, and improves the robustness of muscle atrophy monitoring;

[0043] 4. The testing actions of the system of the present invention are relatively simple during testing, and the overall operation portability and quickness are also improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 is an evaluation flow chart of the system of the present invention;

[0045] Figure 2 This is a schematic diagram of a flexible electrode module in an embodiment of the present invention;

[0046] Figure 2 In the figure, 1 is a metal wire of the flexible and extensible electrode; 2 is a metal collection electrode of the flexible and extensible electrode, which is used to contact the skin to collect surface electromyographic signals; 3 is a polyimide substrate of the flexible and extensible electrode;

[0047] Figure 3 A schematic diagram of signal processing in an embodiment of the present invention;

[0048] Figure 4 A schematic diagram of window length selection in an embodiment of the present invention;

[0049] Figure 5 This is a multi-parameter analysis diagram for early warning of muscular atrophy in an embodiment of the present invention;

[0050] Figure 6 Schematic diagram of the accuracy of muscle atrophy prediction in an embodiment of the present invention. DETAILED DESCRIPTION

[0051] The present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be pointed out that the embodiments described below are intended to facilitate the understanding of the present invention and do not have any limiting effect on the present invention.

[0052] The specific implementation system includes a data acquisition module, a data front-end module, an algorithm analysis module and at least one flexible electrode module;

[0053] A flexible electrode module, used for closely contacting the position to be measured and receiving electromyographic signals;

[0054] The data acquisition module is responsible for controlling the flexible electrode module to collect and convert the electromyographic signals into analog-to-digital signals, and sends the collected and converted data to the data front-end module;

[0055] The data front-end module is used to receive and filter the data from the data acquisition module, and send the filtered data to the algorithm analysis module;

[0056] The algorithm analysis module is used to receive the filtered data from the data front-end module to quantitatively evaluate the degree of disuse muscular atrophy. The algorithm analysis module contains a trained classification evaluation model.

[0057] The flexible electrode module is arranged on the lower surface of the flexible circuit substrate. The upper surface of the flexible circuit substrate is provided with a transparent film layer. The transparent film layer is in close contact with the position to be measured and senses and receives the electromyographic signal.

[0058] The data acquisition module and the data front-end module are integrated in a multi-channel surface electromyography acquisition circuit system with a maximum sampling rate of 16k Samples / s. The data acquisition module and the flexible electrode module, as well as the data acquisition module and the data front-end module are connected by wired cables.

[0059] The algorithm analysis module is a host computer with data processing and computing capabilities. It can also be a computer, or it can be deployed on a microcontroller to form edge computing. The algorithm analysis module and the data front-end module are connected via wired or wireless connections.

[0060] like Figure 1 As shown, the evaluation process of the disuse muscle atrophy degree evaluation system after bone and joint injury of the present invention includes sEMG acquisition, sEMG filtering preprocessing, and sEMG characteristic parameter extraction. The data after the preliminary processing is combined with multi-parameter comprehensive analysis to evaluate the early muscle function of bone and joint injury, and on the other hand, combined with the classification evaluation model to quantify the development of the late atrophy degree. Specifically including:

[0061] 1) Evaluation process of muscle function in the early stage of bone and joint injury

[0062] (1-1):

[0063] The flexible electrode modules were placed in close contact with the tibialis anterior and gastrocnemius muscles of a human body after bone and joint injury. The ankle joints on both sides of the human body were subjected to two types of movements, including the healthy side and the affected side:

[0064] While actively dorsiflexing the ankle joint, the data acquisition module controls the flexible electrode module to collect sEMG signals from the tibialis anterior muscle and the gastrocnemius muscle respectively.

[0065] While the ankle joint is actively plantar flexed, the data acquisition module controls the flexible electrode module to collect sEMG signals from the tibialis anterior and gastrocnemius channels respectively.

[0066] The above two sports actions both lasted for 60 seconds, and the sampling rate was set to 4k Samples / s;

[0067] (1-2): The sEMG data collected above is sent from the data acquisition module to the data front-end module, and then the data front-end module performs low-pass, high-pass and 20-500 Hz band-pass filtering in sequence and then sends it to the algorithm analysis module;

[0068] (1-3): The algorithm analysis module extracts the time domain and frequency domain feature values ​​from the sEMG data, further calculates multiple parameter values ​​in the time domain and frequency domain, and evaluates the muscle function status after human bone and joint injuries;

[0069] In the process of evaluating muscle function in the early stage of bone and joint injury, multiple parameter values ​​in the time domain and frequency domain include a first time domain parameter and a first frequency domain parameter. The first time domain parameter is the root mean square value, the integrated electromyographic value and the variance. The first frequency domain parameter is the number of zero crossing points, the median frequency and the average power spectrum frequency. The above six parameters are used to evaluate the muscle function status in the early stage of bone and joint injury. Specifically, taking the healthy side of the two sides of the human body as the standard, if the values ​​of various parameters of the affected side of the two sides of the human body drop to 80% or less of the corresponding parameter values ​​of the healthy side, it is judged that the bone and joint injury has damaged the muscle function.

[0070] 2) Quantitative evaluation process of the degree of postoperative muscle atrophy

[0071] (2-1): Repeat the above (1-1) and (1-2), send the sEMG data to the algorithm analysis module to extract the time domain and frequency domain feature values, further calculate multiple parameter values ​​in the time domain and frequency domain, and use the trained classification evaluation model to achieve quantitative evaluation of the development stage of postoperative disuse muscular atrophy.

[0072] In the process of quantitative evaluation of the degree of atrophy after surgery, multiple parameter values ​​in the time domain and frequency domain are calculated, including the second time domain parameter and the second frequency domain parameter. The second time domain parameter is the root mean square value and variance, and the second frequency domain parameter is the number of zero crossings, the median frequency and the average power spectrum frequency. The above five parameters are input into the trained classification evaluation model to obtain the current stage of muscle atrophy.

[0073] In this embodiment, the classification evaluation model is a SVM classifier based on the RBF kernel function.

[0074] The overall sEMG data collection process is as follows:

[0075] When the healthy and affected ankles of patients with bone and joint injuries perform active dorsiflexion ankle joint movements at the same time, sEMG signals from the two channels of the tibialis anterior and gastrocnemius are collected respectively; when active plantar flexion ankle joint movements are performed at the same time, sEMG signals from the two channels of the tibialis anterior and gastrocnemius are collected respectively; both actions last for 60 seconds, and the system sampling rate is 4kSamples / s. The system of the present invention is not easily affected by external factors such as swelling, blood flow, etc., has strong anti-interference ability, improves the robustness of atrophy monitoring, and the test action during the test is very simple, and the overall portability and quickness of the operation are also improved. Experiments have shown that the disuse muscle atrophy degree assessment system after bone and joint injury based on flexible electrode surface electromyography of the present invention can be specifically implemented on the surface of the skin and presents a good working condition.

[0076] like Figure 2 The figure shows a schematic diagram of a flexible and stretchable electrode based on a polyimide substrate used in the present invention. 1 is the metal wire composition of the flexible and stretchable electrode; 2 is the metal collection electrode of the flexible and stretchable electrode, which is used to contact the skin to collect surface electromyographic signals; 3 is the polyimide substrate of the flexible and stretchable electrode. The flexible electrode module includes a polyimide substrate 3, a metal wire 1 and a metal collection electrode 2. The metal wire 1 and the metal collection electrode 2 are arranged on the polyimide substrate 3. The metal collection electrode 2 is electrically connected to the data acquisition module via the metal wire 1.

[0077] The overall sEMG data preprocessing process is as follows:

[0078] The sEMG data collected above have a certain baseline offset, so baseline removal processing is required first; after further 20-500Hz bandpass filtering processing, the time domain and frequency domain eigenvalues ​​are extracted and calculated. Figure 3 (a) is the original collected surface electromyography signal. Figure 3 (b) is the surface electromyography signal after filtering. Figure 3 (c) is the detailed display of the electromyographic signal after filtering. Figure 3(d) is the power spectrum density and corresponding median frequency of the electromyographic signal after filtering. Figure 3 (e) is the spectrum analysis diagram of the electromyographic signal. Figure 3 (f) is the spectrum analysis diagram of the electromyographic signal after filtering.

[0079] Specifically, in order to warn of the occurrence of disuse muscle atrophy, data is extracted for the overall 60-second force time period, and the 5-second period of unstable force in the early and late stages of force is eliminated. Multiple parameter values ​​in the time domain and frequency domain are comprehensively calculated to warn of the occurrence of disuse muscle atrophy.

[0080] For the quantitative evaluation of the development stage of postoperative disuse muscle atrophy, the 5s period of unstable force in the early and late stages of force generation was eliminated, and the data between 5s and 55s were specifically extracted; Figure 4 As shown in the figure, the window length is designed to be 0.25s, the windows overlap by 1 / 4, 800 samples are extracted in each time period, and each patient has a total of 3200 sample points based on the two muscles on the affected and healthy sides. The subsequent combined machine learning algorithm realizes the quantitative evaluation of the development stage of postoperative disuse muscular atrophy.

[0081] like Figure 5 As shown in the figure, in the period when the traditional evaluation method has not shown obvious differences, both the time domain parameters and the frequency domain parameters have shown serious decline. Figure 5 From (a) to (f), we can see that the parameters RMS, iEMG, Var, ZC, MF, and MPF have all declined significantly, which can warn of the occurrence of disuse muscular atrophy. When the traditional evaluation method does not show obvious distinction, the comprehensive parameters have a more obvious attenuation, which effectively avoids the instability of single parameter detection and can achieve early warning of disuse muscular atrophy.

[0082] Machine learning is to solve real-life problems by abstracting them into models, and clearly understand the role of different parameters in the model, and then further solve the model to solve real-life problems. Specifically, the present invention uses the SVM classifier as a classification evaluation model, and uses the cross-validation method to find the optimal model parameters based on the RBF kernel function.

[0083] Before training, a sample with a label is constructed; the label represents the current stage evaluation of atrophy, and the change of atrophy over time is annotated in combination with the calf MRI. In this embodiment, the labeling category is divided into four levels according to the different review times of the patients (before surgery, after surgery, one month after surgery, and two months after surgery), namely, level one, level two, level three, and level four, and the degree of disuse muscle atrophy gradually increases.

[0084] For each sample data, the maximum and minimum values ​​are normalized to the interval [0,1], and then 80% of the sample data are randomly intercepted from all the sample point data as the training set, and the support vector machine is created and trained. At the same time, cross-validation is performed on the training set to find the model parameter setting with the highest cross-validation accuracy, and the support vector machine is optimized. The remaining 20% ​​of the sample data is used as the test set.

[0085] The specific implementation combines five cross-validations to conduct multiple repeated tests on the same batch of data, and the proposed system method can achieve a prediction accuracy of 90% or more for the degree of atrophy. Figure 6 shown.

[0086] The embodiments described above provide a detailed description of the technical solutions and beneficial effects of the present invention. It should be understood that the above are only specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, supplements and equivalent substitutions made within the scope of the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A system for evaluating the degree of disuse muscle atrophy after bone and joint injury based on surface electromyography, characterized in that: It includes a data acquisition module, a data front-end module, an algorithm analysis module and at least one flexible electrode module; The flexible electrode module is used to closely contact the position to be measured and receive the electromyographic signal; The data acquisition module is responsible for controlling the flexible electrode module to collect and convert the electromyographic signals into analog-to-digital signals, and sending the collected and converted data to the data front-end module; The data front-end module is used to receive and filter the data from the data acquisition module, and send the filtered data to the algorithm analysis module; The algorithm analysis module is used to receive the filtered data from the data front-end module to quantitatively evaluate the degree of disuse muscular atrophy, and the algorithm analysis module contains a trained classification evaluation model; The evaluation system for the degree of disuse muscle atrophy after bone and joint injury is performed in the following manner: 1) The process of evaluating muscle function in the early stage of bone and joint injury includes: （1-1）： The flexible electrode modules were placed in close contact with the tibialis anterior and gastrocnemius muscles of the human body, and the ankle joints on both sides of the human body performed two types of movements: While actively dorsiflexing the ankle joint, the flexible electrode module is controlled to collect sEMG signals from the tibialis anterior muscle and the gastrocnemius muscle respectively. While actively plantar flexing the ankle joint, the flexible electrode module is controlled to collect sEMG signals from the tibialis anterior muscle and gastrocnemius muscle respectively. Both of the above movements lasted for 60 seconds; (1-2): The sEMG data collected above is subjected to 20-500 Hz bandpass filtering by the data front-end module and then sent to the algorithm analysis module; (1-3): The algorithm analysis module extracts the time domain and frequency domain feature values ​​from the sEMG data, calculates multiple parameter values ​​in the time domain and frequency domain, and evaluates the muscle function status of the human body; 2) The quantitative assessment process of the degree of postoperative muscle atrophy includes: (2-1): Repeat the above (1-1) and (1-2), send the sEMG data to the algorithm analysis module to extract the time domain and frequency domain feature values, and calculate multiple parameter values ​​in the time domain and frequency domain. The trained classification evaluation model is used to quantitatively evaluate the development stage of postoperative disuse muscle atrophy. In (1-3), in the process of evaluating muscle function in the early stage of bone and joint injury, the multiple parameter values ​​in the time domain and the frequency domain include a first time domain parameter and a first frequency domain parameter. The first time domain parameter is the root mean square value, the integrated electromyographic value and the variance. The first frequency domain parameter is the number of zero crossing points, the median frequency and the average power spectrum frequency. The above six parameters are used to evaluate the muscle function status in the early stage of bone and joint injury. The specific procedures for evaluating muscle function in the early stages of bone and joint injury are as follows: taking the healthy side of the human body as the standard, if the values ​​of various parameters of the affected side of the human body drop to 80% or less of the corresponding parameter values ​​of the healthy side, it is judged that the bone and joint injury has already caused damage to muscle function.

2. The system for evaluating the degree of disuse muscle atrophy after bone and joint injury based on surface electromyography according to claim 1 is characterized in that: The flexible electrode module is arranged on the lower surface of the flexible circuit substrate. The upper surface of the flexible circuit substrate is provided with a transparent film layer, which is in close contact with the position to be measured and senses and receives the electromyographic signal.

3. The system for evaluating the degree of disuse muscle atrophy after bone and joint injury based on surface electromyography according to claim 1 is characterized in that: The data acquisition module and the data front-end module are integrated in a multi-channel surface electromyography acquisition circuit system. The data acquisition module and the flexible electrode module, and the data acquisition module and the data front-end module are connected via wired cables.

4. The system for evaluating the degree of disuse muscle atrophy after bone and joint injury based on surface electromyography according to claim 1 is characterized in that: The algorithm analysis module is a host computer with data processing and computing capabilities, and the algorithm analysis module is connected to the data front-end module via wired or wireless connections.

5. The system for evaluating the degree of disuse muscle atrophy after bone and joint injury based on surface electromyography according to claim 1 is characterized in that: In (2-1), in the process of quantitative evaluation of the degree of atrophy after surgery, multiple parameter values ​​in the time domain and frequency domain are calculated, including the second time domain parameter and the second frequency domain parameter. The second time domain parameter is the root mean square value and variance, and the second frequency domain parameter is the number of zero crossing points, the median frequency and the average power spectrum frequency. The above five parameters are input into the trained classification evaluation model to obtain the current stage of muscle atrophy.

6. The system for evaluating the degree of disuse muscle atrophy after bone and joint injury based on surface electromyography according to claim 1, characterized in that: In the process of evaluating muscle function in the early stage of bone and joint injury and quantitatively evaluating the degree of atrophy after surgery, the algorithm analysis module extracts the time domain and frequency domain eigenvalues ​​as follows: For the duration of 60s, after eliminating the 5s in the early and late stages, the data between the 5th and 55th seconds were extracted. Then a window length of 0.25s was set, with 1 / 4 overlap between windows. 800 samples were extracted from each window period. A total of 3200 sample points were obtained for each human body based on two muscles on the affected and healthy sides. The time domain and frequency domain eigenvalues ​​of all sample points were used to calculate multiple parameter values ​​in the time domain and frequency domain.

7. The system for evaluating the degree of disuse muscle atrophy after bone and joint injury based on surface electromyography according to claim 6 is characterized in that: The classification evaluation model is a support vector machine based on the RBF kernel function. The classification evaluation model uses a cross-validation method to find the optimal c / g parameters during the training process; Before training, we first construct labeled sample cases based on sample points combined with MRI images of the lower leg to obtain sample case data, where the labels represent the current stage parameters of atrophy. For each sample data, the maximum and minimum values ​​are normalized to the interval [0, 1], and then 80% of the sample data are randomly intercepted from all sample point data as the training set to create and train the support vector machine. At the same time, cross-validation is performed on the training set to find the model parameter setting with the highest cross-validation accuracy and optimize the support vector machine.

Images