Pelvic floor muscle repairing method and related equipment

Abstract

The invention relates to the technical field of pelvic floor muscle repair, in particular to a pelvic floor muscle repair method and related equipment, and the method comprises the following steps: acquiring a surface electromyogram signal collected by a surface electrode; based on bispectrum analysis, effective feature extraction is carried out on the abdominal electromyographic signals and the pelvic floor electromyographic signals, the extracted features are input into a deep learning model, and the muscle activity state is analyzed; the muscle activity state is fed back to the patient, and the patient is prompted to adjust contraction and relaxation of pelvic floor muscles through voice; through contraction and relaxation of pelvic floor muscles, the pelvic floor function condition is evaluated based on a Glazer evaluation method; and according to the pelvic floor function condition, selecting a corresponding treatment mode for pelvic floor muscle repair. The pelvic floor muscle repairing effect can be improved.

Description

Technical Field

[0001] The present application relates to the field of pelvic floor muscle repair, and in particular to a pelvic floor muscle repair method and related equipment. Background Art

[0002] Pelvic floor muscle dysfunction is a common problem in postpartum women, the elderly, and those with chronic illnesses. It often leads to complications such as urinary incontinence and pelvic organ prolapse, severely impacting quality of life. Traditional assessment methods rely on subjective symptom descriptions and manual palpation, which are limited by low accuracy and poor reproducibility.

[0003] In recent years, noninvasive detection technologies based on surface electromyography (sEMG) have become a research hotspot due to their high sensitivity and objectivity. sEMG records the bioelectrical activity of muscle contractions through skin electrodes, and its amplitude and frequency characteristics can quantitatively reflect muscle activation patterns and fatigue levels. However, the coordinated activity of the abdominal and pelvic floor muscles has complex nonlinear characteristics, making it difficult for conventional time-frequency analysis methods to effectively extract high-order statistical properties from weak signals. Furthermore, while traditional biofeedback therapy has some effectiveness, it suffers from low patient compliance and unstable efficacy, resulting in poor pelvic floor muscle repair. Summary of the Invention

[0004] Based on this, it is necessary to provide a pelvic floor muscle repair method and related equipment that can improve the pelvic floor muscle repair effect in response to the above technical problems.

[0005] In a first aspect, the present application provides a method for repairing pelvic floor muscles, the method comprising: Acquiring surface electromyographic signals collected by surface electrodes, wherein the surface electromyographic signals include abdominal electromyographic signals and pelvic floor electromyographic signals; Based on bispectral analysis, effective features are extracted from the abdominal electromyographic signals and the pelvic floor electromyographic signals, and the extracted features are input into a deep learning model to analyze muscle activity status; Feedback the muscle activity status to the patient, and prompt the patient to adjust the contraction and relaxation of the pelvic floor muscles through voice; By contracting and relaxing the pelvic floor muscles, the pelvic floor function is assessed based on the Glazer assessment method; According to the pelvic floor function status, a corresponding treatment mode is selected to repair the pelvic floor muscles. The treatment modes include electrical stimulation mode, biofeedback training mode and biofeedback triggered electrical stimulation mode.

[0006] In one embodiment, the bispectral analysis is based on extracting effective features of the abdominal electromyographic signal and the pelvic floor electromyographic signal, and the extracted features are input into a deep learning model to analyze the muscle activity state, including: Based on the bispectral analysis of the non-Gaussian AR parameter model, effective feature extraction is performed on the abdominal electromyographic signal and the pelvic floor electromyographic signal; The extracted features were subjected to dimensionality reduction through Fisher linear discriminant analysis; The reduced-dimensional features are input into the deep learning model to analyze the muscle activity status.

[0007] In one embodiment, the muscle activity state includes the abdominal muscle activity state and the pelvic floor muscle activity state; Feedback of the muscle activity status to the patient and prompting the patient to adjust the contraction and relaxation of the pelvic floor muscles through voice includes: Feedback the abdominal muscle activity status and the pelvic floor muscle activity status to the patient; judging whether the abdominal muscles are in a relaxed state according to the abdominal muscle activity state; If so, the patient is prompted by voice to adjust the contraction and relaxation of the pelvic floor muscles according to the activity status of the pelvic floor muscles.

[0008] In one embodiment, the assessing of pelvic floor function by contraction and relaxation of the pelvic floor muscles based on the Glazer assessment method includes: By contracting and relaxing the pelvic floor muscles, based on the Glazer assessment method, a pelvic floor muscle assessment index is obtained, wherein the pelvic floor muscle assessment index includes anterior / posterior resting phase, rapid contraction phase, sustained contraction phase, and endurance contraction phase; The pelvic floor function status is assessed based on the anterior / posterior resting phases, the rapid contraction phase, the sustained contraction phase, and the endurance contraction phase.

[0009] In one embodiment, the pelvic floor function status includes muscle fiber status, fast and slow muscle coordination status, and muscle activity status; The assessing of the pelvic floor function according to the anterior / posterior resting phase, the rapid contraction phase, the sustained contraction phase, and the endurance contraction phase includes: Obtain the amplitude of the rapid contraction phase, the amplitude of the endurance contraction phase, the amplitude of the front / back rest phase, and the coefficient of variation of the sustained contraction phase; The muscle fiber condition, the fast and slow muscle coordination condition, and the muscle activity condition are obtained based on the rapid contraction phase amplitude, the endurance contraction phase amplitude, the front / back rest phase amplitude, the coefficient of variation of the sustained contraction phase, and multiple preset thresholds.

[0010] In one embodiment, selecting a corresponding treatment mode for pelvic floor muscle repair according to the pelvic floor function condition includes: If the muscle fibers are weak or lack the ability to sustain contraction, select the electrical stimulation mode to stimulate the pelvic floor muscles by outputting different pulse width, pulse frequency and pulse intensity parameters; If the fast and slow muscle coordination status is dyssynergia or the muscle activity status is hyperactivity, the biofeedback training mode is selected to guide the patient to perform effective pelvic floor muscle contraction and relaxation training by collecting the patient's pelvic floor electromyography signals in real time; If the muscle fiber condition is weak and the fast and slow muscle coordination condition is synergistic disorder, the biofeedback triggered electrical stimulation mode is selected. By real-time monitoring of the patient's pelvic floor electromyographic signals, the patient is prompted to contract and relax the pelvic floor muscles and the electrical stimulation output is controlled.

[0011] In one embodiment, after selecting a corresponding treatment mode for pelvic floor muscle repair according to the pelvic floor function condition, the method further includes: Obtaining pelvic floor electromyographic signals after pelvic floor muscle repair and pelvic floor electromyographic signals before pelvic floor muscle repair; The pelvic floor muscle electromyographic signal after the pelvic floor muscle repair is compared with the pelvic floor muscle electromyographic signal before the pelvic floor muscle repair by the Glazer evaluation method to obtain the pelvic floor muscle repair effect.

[0012] In a second aspect, the present application also provides a pelvic floor muscle repair device. The device comprises: A surface electromyography signal acquisition module is used to acquire surface electromyography signals collected by surface electrodes, wherein the surface electromyography signals include abdominal electromyography signals and pelvic floor electromyography signals; A muscle activity state analysis module is used to extract effective features from the abdominal electromyographic signals and the pelvic floor electromyographic signals based on bispectral analysis, and input the extracted features into a deep learning model to analyze the muscle activity state; A muscle activity status feedback module is used to feed back the muscle activity status to the patient and prompt the patient to adjust the contraction and relaxation of the pelvic floor muscles through voice; A pelvic floor function assessment module, configured to assess the pelvic floor function based on the Glazer assessment method by contracting and relaxing the pelvic floor muscles; The treatment mode selection module is used to select a corresponding treatment mode for pelvic floor muscle repair according to the pelvic floor function status. The treatment modes include electrical stimulation mode, biofeedback training mode and biofeedback triggered electrical stimulation mode.

[0013] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are performed: Acquiring surface electromyographic signals collected by surface electrodes, wherein the surface electromyographic signals include abdominal electromyographic signals and pelvic floor electromyographic signals; Based on bispectral analysis, effective features are extracted from the abdominal electromyographic signals and the pelvic floor electromyographic signals, and the extracted features are input into a deep learning model to analyze muscle activity status; Feedback the muscle activity status to the patient, and prompt the patient to adjust the contraction and relaxation of the pelvic floor muscles through voice; By contracting and relaxing the pelvic floor muscles, the pelvic floor function is assessed based on the Glazer assessment method; According to the pelvic floor function status, a corresponding treatment mode is selected to repair the pelvic floor muscles. The treatment modes include electrical stimulation mode, biofeedback training mode and biofeedback triggered electrical stimulation mode.

[0014] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the following steps: Acquiring surface electromyographic signals collected by surface electrodes, wherein the surface electromyographic signals include abdominal electromyographic signals and pelvic floor electromyographic signals; Based on bispectral analysis, effective features are extracted from the abdominal electromyographic signals and the pelvic floor electromyographic signals, and the extracted features are input into a deep learning model to analyze muscle activity status; Feedback the muscle activity status to the patient, and prompt the patient to adjust the contraction and relaxation of the pelvic floor muscles through voice; By contracting and relaxing the pelvic floor muscles, the pelvic floor function is assessed based on the Glazer assessment method; According to the pelvic floor function status, a corresponding treatment mode is selected to repair the pelvic floor muscles. The treatment modes include electrical stimulation mode, biofeedback training mode and biofeedback triggered electrical stimulation mode.

[0015] The above-mentioned pelvic floor muscle repair method and related equipment, based on bispectral analysis, effectively extract features from abdominal electromyographic signals and pelvic floor electromyographic signals, can capture the non-Gaussian and nonlinear characteristics of sEMG, reveal the phase coupling information of muscle activity through third-order cumulative amount calculation, and significantly improve the feature characterization ability; through deep learning models, high-precision classification and identification of muscle status are achieved; through the Glazer evaluation method, the functional status of the pelvic floor is evaluated, and the functional status of the pelvic floor can be dynamically quantified in multiple dimensions, providing a basis for personalized treatment; according to the functional status of the pelvic floor, the corresponding treatment mode is selected for pelvic floor muscle repair, which can provide a precisely matched treatment plan, thereby improving the pelvic floor muscle repair effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1Schematic diagram of a process for repairing pelvic floor muscles according to an embodiment; Figure 2 This is a schematic diagram of the host; Figure 3 Schematic diagram of electrode connection line; Figure 4 Schematic diagram of the connection method for pelvic floor muscle assessment electrodes and biofeedback training mode; Figure 5 Connection method for electrical stimulation mode; Figure 6 1 is a structural block diagram of a pelvic floor muscle repair device in one embodiment.

[0017] Description of reference numerals: 101. CH2 channel electrical stimulation output indicator light; 102. CH1 channel electrical stimulation output indicator light; 103. Power on / off button; 104. CH1 interface; 105. CH2 interface; 106. Decorative light; 107. Charging interface; 108. Working status indicator light; 201. Connect to the host; 202. Connect to the reference electrode; 203. Connect to the skin electrode / vaginal electrode. DETAILED DESCRIPTION

[0018] The embodiments of the present invention provide a pelvic floor muscle repair method and related equipment.

[0019] The following describes embodiments of the present disclosure in more detail with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are for illustrative purposes only and are not intended to limit the scope of protection of the present disclosure.

[0020] In the description of the embodiments disclosed herein, the term "including" and similar terms should be understood as open inclusion, i.e., "including but not limited to." The term "based on" should be understood as "based, at least in part, on." The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment." The terms "first," "second," etc. may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0021] For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1 In one embodiment of the pelvic floor muscle repair method of the present invention, the method includes: S100, acquiring surface electromyographic signals collected by surface electrodes.

[0022] Specifically, the surface electrodes of the pelvic floor muscle therapy instrument are used to collect spontaneous surface electromyographic signals of the skin and vagina. The host structure of the pelvic floor muscle therapy instrument is as follows: Figure 2 As shown in the figure, the host has the functions of CH1 and CH2 dual-channel electromyographic signal acquisition and electrical stimulation output, as well as charging and working status indication. The USB charging port of the host only has charging function and no signal input and output function. Figure 3 As shown, one end of the electrode cable connects to the CH1 or CH2 channel port of the main unit. The other end has two short wires for connecting to skin electrodes or vaginal electrodes, and the longer wire connects to the reference electrode (skin electrode 2 or skin electrode 3). The vaginal electrode, used in conjunction with the main unit, transmits electrical signals between the main unit and the vagina. This electrode is used to collect myoelectric signals from the vagina or electrically stimulate the vaginal muscles for pelvic floor muscle function assessment, rehabilitation training, and treatment. This electrode should be placed in the vagina and connected to the CH1 port of the main unit via the electrode cable. The skin electrode, used in conjunction with the main unit, transmits electrical signals between the main unit and the skin. This electrode is applied to the skin on the lower back and is used to deliver electrical stimulation for uterine involution. Skin electrode 2, used in conjunction with the main unit, transmits electrical signals between the main unit and the skin. This electrode, applied to the skin on the abdomen, serves as a reference electrode for collecting myoelectric signals from the abdomen or as a delivery point for electrical stimulation for rectus abdominis rehabilitation. The skin electrode 3 is used in conjunction with the host to transmit signals between the host and the skin. The electrode is attached to the abdominal skin and serves as a reference electrode to collect abdominal electromyographic signals or as an electrical stimulation output for rectus abdominis rehabilitation.

[0023] After collecting abdominal electromyographic signals and pelvic floor electromyographic signals through skin electrodes or vaginal electrodes, the abdominal electromyographic signals and pelvic floor electromyographic signals are preprocessed, and the preprocessing includes denoising, amplification, filtering, and conversion.

[0024] S200, based on bispectral analysis, extracts effective features from abdominal and pelvic floor EMG signals, and inputs the extracted features into a deep learning model to analyze muscle activity status.

[0025] Specifically, the electromyographic signal is essentially a physiological signal with non-stationary and non-Gaussian characteristics. According to the characteristics of surface electromyographic signals, from the perspective of high-order statistical processing of signals, based on the "non-Gaussian AR parameter model" bispectral analysis, effective features are extracted. Among them, the bispectrum is a complex-valued spectrum with two frequencies. and . There are three methods for bispectrum estimation: direct method, indirect method and parameter model method. The direct method and the indirect method have large estimation variances and need to provide a large number of data samples. The increase in the amount of data brings a large amount of calculation and will cause non-stationarity. In order to overcome these shortcomings, the present invention uses the AR model method for bispectrum estimation. In addition to providing high-resolution bispectrum estimation and effectively extracting the phase information of the signal when the observation data is short, it can also perform bispectrum estimation when the non-Gaussian distribution of the model excitation signal is unknown. The extracted features are input into a deep learning model (such as a neural network), and the muscle activity state is analyzed by training the model, such as judging the muscle contraction state. During the model training process, it is necessary to use a labeled data set for supervised learning to ensure that the model can accurately distinguish different muscle activity patterns.

[0026] S300 provides feedback on muscle activity status to the patient and prompts the patient through voice to adjust the contraction and relaxation of the pelvic floor muscles.

[0027] Specifically, the system provides feedback to the patient regarding the analysis results, for example, by providing voice or visual prompts to inform the patient of the current muscle activity status. This feedback mechanism can improve the patient's rehabilitation outcomes and help them adjust their muscle activity. The system uses voice prompts to adjust the contraction and relaxation of the pelvic floor muscles, and collects and analyzes the patient's pelvic floor electromyographic signals in different states to assess pelvic floor function.

[0028] Before evaluating the pelvic floor function, Figure 4 Connect the electrodes correctly as shown, follow the system prompts to warm up, collect abdominal EMG signals, and collect pelvic floor EMG signals. Figure 4 As shown, connect the vaginal electrode to the two white short wire interfaces of the electrode connecting cable; insert one end of the electrode connecting cable into the CH1 channel of the host; insert the vaginal electrode into the vagina (if necessary, apply an appropriate amount of medical water-soluble lubricant on the surface of the vaginal electrode); connect the three skin electrodes 2 (or skin electrodes 3) to the three interfaces of the electrode connecting cable; insert the other end of the electrode connecting cable into the CH2 interface of the host.

[0029] S400 assesses pelvic floor function based on the Glazer assessment method by contracting and relaxing the pelvic floor muscles.

[0030] Among them, the Glazer assessment method is a surface electromyography detection method used to evaluate the function of the pelvic floor muscles. This method collects the electromyographic signals of the pelvic floor muscles in contraction and relaxation states through vaginal electrodes to evaluate muscle control, reaction speed, muscle fiber endurance and stability, etc.

[0031] Specifically, patients are instructed to contract and relax their pelvic floor muscles. Pelvic floor EMG signals are obtained in different states. Based on these signals, the Glazer assessment method is used to assess pelvic floor function. Significant differences exist in the signal characteristics of the pelvic floor muscles in contracted and relaxed states, primarily in terms of EMG amplitude, coefficient of variation, and contraction reaction time. In the contracted state, the EMG signals (such as the RMS value) are typically higher than in the relaxed state, indicating more active muscle activity. Furthermore, the coefficient of variation of the EMG signals in the contracted state is higher, indicating poor coordination during muscle contraction. In the relaxed state, the EMG signals have lower amplitudes and smaller coefficients of variation, indicating a relatively static and stable muscle state. These differences reflect the functional status and activity level of the pelvic floor muscles in different states.

[0032] S500, according to the pelvic floor function status, select the corresponding treatment mode to repair the pelvic floor muscles.

[0033] Specifically, the system selects a preset standard treatment mode for treatment and training based on the patient's pelvic floor function. In addition, patients can customize their treatment plans based on the results of pelvic floor muscle assessment. Depending on the working principle, the treatment mode can be divided into three different treatment modes: electrical stimulation, biofeedback training, and biofeedback electrical stimulation. Each standard treatment mode has fixed parameters. When entering a treatment mode, the system will provide the corresponding electrode connection method and usage instructions. The user should follow the system prompts to perform the corresponding operations.

[0034] When the electrical stimulation mode is selected, follow the Figure 5 Following the instructions, connect the vaginal electrodes to the two short white connectors on the electrode cable; plug one end of the electrode cable into CH1 on the main unit; and insert the vaginal electrodes into the vagina (apply a moderate amount of medical water-soluble lubricant to the surface of the vaginal electrodes if necessary). Enter the electrical stimulation protocol interface, set the appropriate treatment time, click "Start," and then slowly adjust the electrical stimulation intensity until it reaches the maximum tolerable intensity without causing pain. During operation, the CH1 channel's electrical stimulation output indicator will illuminate when electrical stimulation is being delivered.

[0035] When selecting the biofeedback training mode, follow Figure 4Following the instructions, connect the vaginal electrodes to the two short white connectors on the electrode cable. Plug one end of the electrode cable into the CH1 channel of the device. Insert the vaginal electrodes into the vagina (apply a generous amount of medical water-soluble lubricant to the surface of the vaginal electrodes if necessary). Connect the three skin electrodes 2 (or 3) to the three connectors on the electrode cable. Plug the other end of the electrode cable into the CH2 connector of the device. Apply the skin electrodes tightly and evenly to the abdomen. If your skin is dry, wipe it with warm water first to ensure good contact. Ensure that the three electrodes are at least 2 cm apart. Then, enter the biofeedback training program interface and set the appropriate treatment time. Before starting, you can adjust the abdominal muscle threshold according to your needs or use the system-provided settings. Click the "Start" button to begin the training phase. Pay close attention to the waveform changes on the screen and the voice prompts to ensure that you are correctly executing the pelvic floor muscle contraction and relaxation movements. During the training, it is important to relax your abdomen. If your abdominal involvement is excessive, the system will prompt you to relax your abdomen to avoid affecting the training effect.

[0036] When selecting the biofeedback training mode, follow Figure 4 Following the instructions, connect the vaginal electrodes to the two short white connectors on the electrode cable. Plug one end of the electrode cable into CH1 on the device. Insert the vaginal electrodes into the vagina (apply a generous amount of medical water-soluble lubricant to the surface of the vaginal electrodes if necessary). Connect the three skin electrodes 2 (or 3) to the three connectors on the electrode cable. Plug the other end of the electrode cable into CH2 on the device. Apply the skin electrodes tightly and evenly to the abdomen. If your skin is dry, wipe it with warm water to ensure good contact between the electrodes. Ensure that the three electrodes are at least 2 cm apart. Then, enter the biofeedback electrical stimulation plan interface and set the appropriate treatment time. Before starting, you can adjust the abdominal and pelvic floor thresholds according to your needs, or use the system-provided settings. Before starting, slowly adjust the electrical stimulation intensity until you reach the maximum you can tolerate without causing pain. Click the "Start" button to begin the training phase. Pay close attention to the waveform changes on the screen and the voice prompts to ensure that you are correctly executing the pelvic floor muscle contraction and relaxation movements. When prompted to contract and the user's pelvic floor muscle EMG value is greater than or equal to the pelvic floor muscle threshold, electrical stimulation will be output and the system will prompt the user to output electrical stimulation. During training, it is necessary to relax the abdomen. If the abdominal involvement is too high, the system will prompt the user to relax the abdomen to avoid affecting the training effect. During operation, when electrical stimulation is output, the CH1 channel electrical stimulation output indicator will light up.

[0037] In one embodiment, based on bispectral analysis, effective features are extracted from abdominal EMG signals and pelvic floor EMG signals, and the extracted features are input into a deep learning model to analyze muscle activity status, including: Based on the bispectral analysis of the non-Gaussian AR parameter model, effective features of the abdominal and pelvic floor electromyographic signals are extracted. The extracted features are reduced in dimension through Fisher linear discriminant analysis. The reduced features are input into the deep learning model to analyze the muscle activity status.

[0038] Specifically, the high-dimensional feature vector obtained by bispectral analysis is directly used as the input vector of the deep learning model, and the correct recognition rate of muscle activity status is low. In order to further improve the correct recognition rate of muscle activity status, effectively reducing the dimension becomes the key to solving the problem. In the present invention, Fisher linear discriminant analysis is adopted. This method uses the idea of dimensionality reduction to project all sample points onto a straight line, so that the ratio of the discreteness between sample classes and the discreteness within sample classes is maximized. Then, the reduced-dimensional features are input into the deep learning model to analyze the muscle activity status. Optionally, the deep learning model can adopt a convolutional neural network (CNN), which is mainly composed of a convolution layer, a pooling layer and a fully connected layer. The convolution layer uses a convolution kernel to extract local features, the pooling layer reduces the complexity of the features, and the fully connected layer is used for classification or regression tasks. CNN has the characteristics of local perception and weight sharing, which enables it to process data effectively.

[0039] In one embodiment, providing the patient with muscle activity status feedback and prompting the patient to adjust the contraction and relaxation of the pelvic floor muscles through voice prompts includes: The activity status of the abdominal muscles and pelvic floor muscles is fed back to the patient; based on the activity status of the abdominal muscles, it is determined whether the abdominal muscles are in a relaxed state; if so, the patient is prompted by voice to adjust the contraction and relaxation of the pelvic floor muscles according to the activity status of the pelvic floor muscles.

[0040] Specifically, the device will collect the patient's abdominal electromyographic signals in a naturally relaxed state. To ensure the accuracy of the collected results, the patient should keep the abdomen naturally relaxed when collecting abdominal electromyographic signals, and avoid actions such as inhaling or bulging the belly. In addition, the device will collect the patient's pelvic floor electromyographic signals. To ensure the accuracy of the collected pelvic floor electromyographic signals, the patient should focus on the contraction of the pelvic floor muscles, while relaxing the abdomen and avoiding the involvement of the abdominal muscles. Specifically, the abdominal muscle activity status is first obtained based on the abdominal electromyographic signals, and then based on the abdominal muscle activity status, it is judged whether the abdominal muscles are in a relaxed state. If the abdominal muscles are not in a relaxed state, the patient is prompted by voice to relax the abdominal muscles. If the abdominal muscles are in a relaxed state, the patient is prompted by voice to adjust the contraction and relaxation of the pelvic floor muscles based on the pelvic floor muscle activity status.

[0041] In one embodiment, the pelvic floor function is assessed based on the Glazer assessment method by contracting and relaxing the pelvic floor muscles, including: Through the contraction and relaxation of the pelvic floor muscles, based on the Glazer assessment method, pelvic floor muscle assessment indicators are obtained. The pelvic floor muscle assessment indicators include the anterior / posterior resting phase, rapid contraction phase, sustained contraction phase, and endurance contraction phase; the pelvic floor function status is assessed based on the anterior / posterior resting phase, rapid contraction phase, sustained contraction phase, and endurance contraction phase.

[0042] Specifically, pelvic floor muscle contraction and relaxation are performed, and pelvic floor electromyographic signals (EMGs) are collected and analyzed under different conditions. Pelvic floor function is assessed based on the Glazer assessment method. The Glazer assessment uses five phases (pre-rest, rapid contraction, sustained contraction, endurance contraction, and post-rest) to obtain quantitative data on pelvic floor muscle function. Specific indicators include the pre-rest phase, the rapid contraction phase, the sustained contraction phase, and the endurance contraction phase. The pre-rest phase measures resting muscle tone (mean amplitude) and coefficient of variation (coefficient of variation) (muscle stability). A high resting amplitude (>4μV) indicates muscle overactivity (such as hypertonic pain), while a high coefficient of variation reflects poor muscle control. The rapid contraction phase assesses the explosive power (maximum amplitude), contraction velocity (rise time), and relaxation ability (fall time) of type II fast-twitch fibers. A maximum amplitude <35μV indicates fast-twitch muscle weakness (such as stress urinary incontinence). The sustained contraction phase measures the strength (mean amplitude) and coordination (coefficient of variation) of type I slow-twitch fibers. An amplitude <30 μV indicates insufficient slow-twitch strength, while a high coefficient of variation indicates poor coordination. The endurance contraction phase assesses the endurance (mean amplitude) and fatigue recovery (post-resting amplitude ratio) of slow-twitch fibers. An endurance amplitude <25 μV or poor post-resting recovery indicates insufficient muscle endurance.

[0043] In one embodiment, assessing pelvic floor function based on the anterior / posterior resting phase, the rapid contraction phase, the sustained contraction phase, and the endurance contraction phase includes: The rapid contraction phase amplitude, the endurance contraction phase amplitude, the front / back resting phase amplitude, and the coefficient of variation of the sustained contraction phase are obtained; based on the rapid contraction phase amplitude, the endurance contraction phase amplitude, the front / back resting phase amplitude, the coefficient of variation of the sustained contraction phase, and multiple preset thresholds, the muscle fiber condition, the fast and slow muscle coordination condition, and the muscle activity condition are obtained.

[0044] Specifically, the rapid contraction phase amplitude, endurance contraction phase amplitude, pre- and post-resting phase amplitude, and sustained contraction phase coefficient of variation are compared with corresponding preset thresholds to determine muscle fiber status, fast and slow twitch synergy, and muscle activity. Specifically, if the rapid contraction phase amplitude is less than the preset threshold of 35 μV, it indicates weakness in type II muscle fibers; if the endurance contraction phase amplitude is less than the preset threshold of 25 μV, it indicates insufficient sustained contraction capacity in type I muscle fibers; if the sustained contraction phase coefficient of variation (>0.3) is high, it indicates fast and slow twitch synergy disorder; and if the pre-resting phase amplitude is greater than the preset threshold of 4 μV, it indicates muscle overactivity.

[0045] In one embodiment, according to the pelvic floor function status, selecting a corresponding treatment mode for pelvic floor muscle repair includes: If the muscle fiber condition is weak or the sustained contraction capacity is insufficient, the electrical stimulation mode is selected to stimulate the pelvic floor muscles by outputting different pulse width, pulse frequency and pulse intensity parameters; if the fast and slow muscle coordination condition is synergy disorder or the muscle activity condition is excessive activity, the biofeedback training mode is selected to guide the patient to perform effective pelvic floor muscle contraction and relaxation training by collecting the patient's pelvic floor electromyography signals in real time; if the muscle fiber condition is weak and the fast and slow muscle coordination condition is synergy disorder, the biofeedback triggered electrical stimulation mode is selected to prompt the patient to contract and relax the pelvic floor muscles by real-time monitoring of the patient's pelvic floor electromyography signals and control the electrical stimulation output.

[0046] Specifically, if type II muscle fibers are weak or type I muscle fibers lack the ability to sustain contractions, the electrical stimulation mode is selected. This mode stimulates the pelvic floor muscles by outputting different parameters, such as pulse width, pulse frequency, and pulse intensity, to enhance muscle strength and endurance. If the fast and slow muscle synergy is dyssynergistic or the muscle activity is hyperactive, the biofeedback training mode is selected. This interactive training method collects and visualizes the user's pelvic floor EMG signals in real time to guide the user in effective pelvic floor muscle contraction and relaxation training. Prior to biofeedback training, abdominal and pelvic floor EMG values are collected sequentially. Furthermore, games can be used to guide the user in pelvic floor muscle control training in an interactive and entertaining manner, making the training more engaging and effective. If the muscle fibers are weak and the fast and slow muscle synergy is dyssynergistic, the biofeedback-triggered electrical stimulation mode is selected. This treatment combines biofeedback and electrical stimulation. It monitors the user's pelvic floor EMG signals in real time and converts them into a visual graph, helping the patient understand muscle activity. The user can then contract and relax the pelvic floor muscles according to the waveform and verbal prompts, controlling the electrical stimulation output. During training, the system determines whether electrical stimulation is needed to assist or enhance the patient's pelvic floor muscle contractions based on preset pelvic floor muscle thresholds (rapid contraction phase amplitude threshold, endurance contraction phase amplitude threshold, sustained contraction phase coefficient of variation threshold, etc.). This method not only improves the user's awareness of pelvic floor muscle control but also promotes the recovery of muscle strength and coordination. When the user's pelvic floor EMG signal reaches or exceeds the set threshold, the system automatically delivers electrical stimulation to assist the muscles in completing the contraction. The intensity and duration of this electrical stimulation can be adjusted based on the user's actual situation to ensure comfort and effectiveness of the training.

[0047] In one embodiment, after selecting a corresponding treatment mode for pelvic floor muscle repair based on the pelvic floor function, the method further includes: Obtain the pelvic floor electromyographic signals after pelvic floor muscle repair and the pelvic floor electromyographic signals before pelvic floor muscle repair; use the Glazer evaluation method to compare the pelvic floor electromyographic signals after pelvic floor muscle repair with the pelvic floor electromyographic signals before pelvic floor muscle repair to obtain the pelvic floor muscle repair effect.

[0048] Specifically, the pelvic floor electromyographic signals after pelvic floor muscle repair and the pelvic floor electromyographic signals before pelvic floor muscle repair were obtained, and the rapid contraction phase amplitude, endurance contraction phase amplitude, anterior / posterior resting phase amplitude and sustained contraction phase coefficient of variation of the pelvic floor muscle after repair were obtained through the Glazer evaluation method. The rapid contraction phase amplitude, endurance contraction phase amplitude, anterior / posterior resting phase amplitude and sustained contraction phase coefficient of variation of the pelvic floor muscle after repair were compared with the rapid contraction phase amplitude, endurance contraction phase amplitude, anterior / posterior resting phase amplitude and sustained contraction phase coefficient of variation before pelvic floor muscle repair to obtain the pelvic floor muscle repair effect.

[0049] In one embodiment, Figure 6 As shown, a pelvic floor muscle repair device is provided, comprising: a surface electromyography signal acquisition module 10, a muscle activity state analysis module 20, a muscle activity state feedback module 30, a pelvic floor function status assessment module 40 and a treatment mode selection module 50, wherein: Surface electromyography signal acquisition module 10, used to acquire surface electromyography signals collected by surface electrodes, the surface electromyography signals including abdominal electromyography signals and pelvic floor electromyography signals; The muscle activity state analysis module 20 is used to extract effective features from the abdominal electromyographic signals and pelvic floor electromyographic signals based on bispectral analysis, and input the extracted features into the deep learning model to analyze the muscle activity state; The muscle activity status feedback module 30 is used to feed back the muscle activity status to the patient and prompt the patient to adjust the contraction and relaxation of the pelvic floor muscles through voice; A pelvic floor function assessment module 40 is used to assess the pelvic floor function based on the Glazer assessment method by contracting and relaxing the pelvic floor muscles; The treatment mode selection module 50 is used to select a corresponding treatment mode for pelvic floor muscle repair according to the pelvic floor function status. The treatment modes include electrical stimulation mode, biofeedback training mode and biofeedback triggered electrical stimulation mode.

[0050] In one embodiment, the muscle activity state analysis module 20 is also used to extract effective features of abdominal electromyographic signals and pelvic floor electromyographic signals based on bispectral analysis of a non-Gaussian AR parameter model; reduce the dimension of the extracted features through Fisher linear discriminant analysis; and input the reduced-dimensional features into a deep learning model to analyze the muscle activity state.

[0051] In one embodiment, the muscle activity status feedback module 30 is also used to feed back the abdominal muscle activity status and pelvic floor muscle activity status to the patient; based on the abdominal muscle activity status, it is determined whether the abdominal muscles are in a relaxed state; if so, based on the pelvic floor muscle activity status, the patient is prompted by voice to adjust the contraction and relaxation of the pelvic floor muscles.

[0052] In one embodiment, the pelvic floor function status assessment module 40 is also used to obtain pelvic floor muscle assessment indicators based on the Glazer assessment method through the contraction and relaxation of the pelvic floor muscles. The pelvic floor muscle assessment indicators include the front / back resting stage, the rapid contraction stage, the sustained contraction stage and the endurance contraction stage; the pelvic floor function status is assessed based on the front / back resting stage, the rapid contraction stage, the sustained contraction stage and the endurance contraction stage.

[0053] In one embodiment, the pelvic floor function status assessment module 40 is also used to obtain the rapid contraction phase amplitude, the endurance contraction phase amplitude, the anterior / posterior resting phase amplitude, and the coefficient of variation of the sustained contraction phase; based on the rapid contraction phase amplitude, the endurance contraction phase amplitude, the anterior / posterior resting phase amplitude, the coefficient of variation of the sustained contraction phase, and multiple preset thresholds, the muscle fiber status, the fast and slow muscle coordination status, and the muscle activity status are obtained.

[0054] In one embodiment, the treatment mode selection module 50 is also used to select the electrical stimulation mode if the muscle fiber condition is weakness or insufficient sustained contraction capacity, and stimulate the pelvic floor muscles by outputting different pulse widths, pulse frequencies, and pulse intensity parameters; if the fast and slow muscle coordination condition is synergism or the muscle activity condition is overactivity, then select the biofeedback training mode, and guide the patient to perform effective pelvic floor muscle contraction and relaxation training by collecting the patient's pelvic floor electromyography signals in real time; if the muscle fiber condition is weakness and the fast and slow muscle coordination condition is synergism, then select the biofeedback triggered electrical stimulation mode, and prompt the patient to contract and relax the pelvic floor muscles by real-time monitoring of the patient's pelvic floor electromyography signals, and control the electrical stimulation output.

[0055] In one embodiment, the pelvic floor muscle repair device also includes a pelvic floor muscle repair effect evaluation module, which is used to obtain the pelvic floor electromyographic signal after the pelvic floor muscle repair and the pelvic floor electromyographic signal before the pelvic floor muscle repair; through the Glazer evaluation method, the pelvic floor electromyographic signal after the pelvic floor muscle repair is compared with the pelvic floor electromyographic signal before the pelvic floor muscle repair to obtain the pelvic floor muscle repair effect.

[0056] Each module in the pelvic floor muscle repair device can be implemented in whole or in part through software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in hardware form, or stored in a computer device memory in software form, so that the processor can call and execute the corresponding operations of each module.

[0057] In one embodiment, the present application discloses a computer device including a memory and a processor. The memory is used to store a computer program that can be run on the processor. When the processor loads the computer program, it executes a pelvic floor muscle repair method of the above embodiment.

[0058] In one embodiment, the present application discloses a computer-readable storage medium, and the computer-readable storage medium stores a computer program, wherein when the computer program is loaded by a processor, a pelvic floor muscle repair method of the above embodiment is executed.

[0059] The above are all preferred embodiments of the present application, and are not intended to limit the scope of protection of the present application. Therefore, any equivalent changes made based on the structure, shape, and principle of the present application should be included in the scope of protection of the present application.

Claims

1. A pelvic floor muscle repair method, characterized in that: The method comprises: Acquiring surface electromyographic signals collected by surface electrodes, wherein the surface electromyographic signals include abdominal electromyographic signals and pelvic floor electromyographic signals; Based on bispectral analysis, effective features are extracted from the abdominal electromyographic signals and the pelvic floor electromyographic signals, and the extracted features are input into a deep learning model to analyze muscle activity status; Feedback the muscle activity status to the patient, and prompt the patient to adjust the contraction and relaxation of the pelvic floor muscles through voice; By contracting and relaxing the pelvic floor muscles, the pelvic floor function is assessed based on the Glazer assessment method; According to the pelvic floor function status, a corresponding treatment mode is selected to repair the pelvic floor muscles. The treatment modes include electrical stimulation mode, biofeedback training mode and biofeedback triggered electrical stimulation mode.

2. A pelvic floor muscle repair method according to claim 1, characterized in that: The bispectral analysis is based on extracting effective features from the abdominal electromyographic signals and the pelvic floor electromyographic signals, and the extracted features are input into the deep learning model to analyze the muscle activity state, including: Based on the bispectral analysis of the non-Gaussian AR parameter model, effective feature extraction is performed on the abdominal electromyographic signal and the pelvic floor electromyographic signal; The extracted features were subjected to dimensionality reduction through Fisher linear discriminant analysis; The reduced-dimensional features are input into the deep learning model to analyze the muscle activity status.

3. A pelvic floor muscle repair method according to claim 1, characterized in that: The muscle activity state includes the abdominal muscle activity state and the pelvic floor muscle activity state; Feedback of the muscle activity status to the patient and prompting the patient to adjust the contraction and relaxation of the pelvic floor muscles through voice includes: Feedback the abdominal muscle activity status and the pelvic floor muscle activity status to the patient; judging whether the abdominal muscles are in a relaxed state according to the abdominal muscle activity state; If so, the patient is prompted by voice to adjust the contraction and relaxation of the pelvic floor muscles according to the activity status of the pelvic floor muscles.

4. A pelvic floor muscle repair method according to claim 1, characterized in that: The contraction and relaxation of the pelvic floor muscles, based on the Glazer assessment method, assesses the pelvic floor function including: By contracting and relaxing the pelvic floor muscles, based on the Glazer assessment method, a pelvic floor muscle assessment index is obtained, wherein the pelvic floor muscle assessment index includes anterior / posterior resting phase, rapid contraction phase, sustained contraction phase, and endurance contraction phase; The pelvic floor function status is assessed based on the anterior / posterior resting phases, the rapid contraction phase, the sustained contraction phase, and the endurance contraction phase.

5. A pelvic floor muscle repair method according to claim 4, characterized in that: The pelvic floor function status includes muscle fiber status, fast and slow muscle coordination status and muscle activity status; The assessing of the pelvic floor function according to the anterior / posterior resting phase, the rapid contraction phase, the sustained contraction phase, and the endurance contraction phase includes: Obtain the amplitude of the rapid contraction phase, the amplitude of the endurance contraction phase, the amplitude of the front / back rest phase, and the coefficient of variation of the sustained contraction phase; The muscle fiber condition, the fast and slow muscle coordination condition, and the muscle activity condition are obtained based on the rapid contraction phase amplitude, the endurance contraction phase amplitude, the front / back rest phase amplitude, the coefficient of variation of the sustained contraction phase, and multiple preset thresholds.

6. A pelvic floor muscle repair method according to claim 1, characterized in that: The selection of a corresponding treatment mode for pelvic floor muscle repair according to the pelvic floor function condition includes: If the muscle fibers are weak or lack the ability to sustain contraction, select the electrical stimulation mode to stimulate the pelvic floor muscles by outputting different pulse width, pulse frequency and pulse intensity parameters; If the fast and slow muscle coordination status is dyssynergia or the muscle activity status is hyperactivity, the biofeedback training mode is selected to guide the patient to perform effective pelvic floor muscle contraction and relaxation training by collecting the patient's pelvic floor electromyography signals in real time; If the muscle fiber condition is weak and the fast and slow muscle coordination condition is synergistic disorder, the biofeedback triggered electrical stimulation mode is selected. By real-time monitoring of the patient's pelvic floor electromyographic signals, the patient is prompted to contract and relax the pelvic floor muscles and the electrical stimulation output is controlled.

7. The pelvic floor muscle repair method according to claim 1, characterized in that: After selecting a corresponding treatment mode to repair the pelvic floor muscles according to the pelvic floor function condition, the method further includes: Obtaining pelvic floor electromyographic signals after pelvic floor muscle repair and pelvic floor electromyographic signals before pelvic floor muscle repair; The pelvic floor muscle electromyographic signal after the pelvic floor muscle repair is compared with the pelvic floor muscle electromyographic signal before the pelvic floor muscle repair by the Glazer evaluation method to obtain the pelvic floor muscle repair effect.

8. A pelvic floor muscle repair device, characterized in that: The device comprises: A surface electromyography signal acquisition module is used to acquire surface electromyography signals collected by surface electrodes, wherein the surface electromyography signals include abdominal electromyography signals and pelvic floor electromyography signals; A muscle activity state analysis module is used to extract effective features from the abdominal electromyographic signals and the pelvic floor electromyographic signals based on bispectral analysis, and input the extracted features into a deep learning model to analyze the muscle activity state; A muscle activity status feedback module is used to feed back the muscle activity status to the patient and prompt the patient to adjust the contraction and relaxation of the pelvic floor muscles through voice; A pelvic floor function assessment module, configured to assess the pelvic floor function based on the Glazer assessment method by contracting and relaxing the pelvic floor muscles; The treatment mode selection module is used to select a corresponding treatment mode for pelvic floor muscle repair according to the pelvic floor function status. The treatment modes include electrical stimulation mode, biofeedback training mode and biofeedback triggered electrical stimulation mode.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images