Brain-muscle-machine-based pelvic floor muscle noninvasive visual monitoring and rehabilitation guidance system

Abstract

The invention discloses a brain-muscle-machine-based pelvic floor muscle noninvasive visual monitoring and rehabilitation guidance system. The system comprises a myoelectricity acquisition module, a main control module, a wireless transmission module, an upper computer and an electroencephalogram acquisition module which are in communication connection, wherein the electroencephalogram acquisition module is arranged on the skin surface of the head of a patient; the myoelectricity acquisition module and the electroencephalogram acquisition module act on a human body and feed back information to the main control module, the main control module and the wireless transmission module perform information interaction, and the wireless transmission module transmits the information to the upper computer through the cloud server. The method has the advantages that errors caused by individual differences of pelvic floor parts of a patient in the pelvic floor skin surface signal acquisition process of the patient can be reduced, and the muscle electric signal characteristics are mapped into the electroencephalogram signal characteristic space; the method has the technical characteristics that the electroencephalogram signal data has application value in pelvic floor muscle tension evaluation, a myoelectricity acquisition device can be prevented from being arranged at the pelvic floor part of a patient all the time, the privacy of the patient is protected, and the patient can use the device conveniently.

Description

Technical Field

[0001] The present invention relates to a visualization monitoring and rehabilitation guidance system, and more specifically, to a non-invasive visualization monitoring and rehabilitation guidance system for pelvic floor muscles based on a brain-muscle machine, belonging to the field of medical systems. Background Art

[0002] The pelvic floor muscles are crucial to the human body. They support pelvic organs, control excretory functions, maintain vaginal tightness, and participate in the childbirth process. Disorders of the pelvic floor muscles may lead to problems such as pelvic organ prolapse, urinary incontinence, defecation difficulties, and a decline in the quality of sexual life, which will seriously affect the daily life of patients. Therefore, for patients, pelvic floor muscle rehabilitation is particularly important, which helps patients restore the function of the pelvic floor muscles, relieve related symptoms, and improve the quality of life. Pelvic floor muscle rehabilitation equipment plays a crucial role in the rehabilitation treatment process of pelvic floor muscles.

[0003] Currently, existing pelvic floor muscle monitoring and rehabilitation training equipment mainly includes invasive and non-invasive types, but there are the following problems and defects: 1) Invasive pelvic floor muscle monitoring and rehabilitation equipment needs to continuously insert a probe into the patient's vagina to collect pelvic floor muscle signals. This invasive method will cause discomfort to the patient; 2) Non-invasive pelvic floor muscle rehabilitation equipment uses electrode patches to collect electromyographic signals. Although it avoids the discomfort of the invasive method, during the non-invasive monitoring of pelvic floor muscles, the collection of electromyographic signals is limited by the number and placement position of the electrode patches. Due to individual differences in the pelvic floor area, problems such as inaccurate placement of the electrode patches or displacement of the electrode patch positions during use often occur, thereby reducing the accuracy of electromyographic signal collection and the rehabilitation training effect; 3) Existing invasive or non-invasive pelvic floor muscle rehabilitation equipment requires the probe or electrode to be continuously placed in the pelvic floor area throughout each monitoring and rehabilitation training process, and patients need to continuously expose their private parts, which not only increases the psychological burden on patients but also reduces the convenience of equipment use; 4) Existing rehabilitation equipment mainly uses the method of outputting current stimulation to treat patients' pelvic floor muscles. This passive stimulation can exercise limited muscles and cannot repair the nerve pathways that establish pelvic floor muscle movement due to nerve impulses.

[0004] The Chinese patent application publication number CN118787858A for an invention patent discloses a pelvic floor pants and a control method thereof, including: a pants body, a plurality of electrode patches, and an electrical stimulation host. The electrode patches are installed on the pants body, and the electrode patches are connected to the electrical stimulation host by wire. The electrical stimulation host includes a charging port, an electrical stimulation control module, an electrode patch connection port, and a main function module. The charging port and the electrical stimulation control module are both electrically connected to the main function module, and the electrical stimulation control module is electrically connected to the electrode patch connection port. The electrode patches include four electrode patches on the front side of the thighs and four electrode patches on the buttocks. The four electrode patches on the front side of the thighs and the four electrode patches on the buttocks form an electrical stimulation circuit. Although this technical solution adopts the method of integrating electrode patches into the pants and takes into account the privacy of patients during the treatment process, it only considers the privacy of patients during the electrical stimulation process and does not consider issues such as privacy during the state monitoring of patients' pelvic floor muscles, and there are certain limitations.

[0005] The Chinese patent application publication number CN117959630A for an invention patent discloses a vaginal treatment system and device combining ultrasonic and electrical stimulation, including: an electromyogram data acquisition module, a controller, an ultrasonic-electrical stimulation treatment head, and a treatment device body. The electromyogram data acquisition module is used to acquire the electromyogram data on the surface of the patient's pelvic floor muscles. The controller is used to analyze the electromyogram data of the patient and determine the target pelvic floor muscle symptom type and the corresponding target ultrasonic-electrical stimulation treatment mode of the patient according to the preset pelvic floor muscle symptom type determination strategy. The ultrasonic-electrical stimulation treatment head is used to apply ultrasonic and electrical stimulation treatments to the patient according to the target ultrasonic-electrical stimulation treatment mode. It can combine the thermal effect of ultrasonic and the muscle stimulation effect of electrical stimulation to synergistically reduce the patient's pain perception, promote tissue repair, improve drug permeability, and determine the patient's symptom type through the controller according to the electromyogram data on the surface of the patient's pelvic floor muscles to adopt the corresponding treatment mode. Although this technical solution can monitor the electromyogram signal and perform rehabilitation treatment on the pelvic floor part of the patient by acquiring and analyzing the electromyogram data on the surface of the patient's pelvic floor muscles, when monitoring the electromyogram signal of the patient's pelvic floor part, it uses an insertion probe type monitoring method, which is not conducive to protecting the privacy of the patient and has low convenience. When treating the patient's pelvic floor part, it uses a passive stimulation method of electrical stimulation, and the rehabilitation treatment effect is poor.

[0006] In view of this, it is particularly important and urgent to develop a pelvic floor muscle rehabilitation monitoring and guidance system that not only takes into account the privacy and convenience of patients during the treatment process, but also can reduce the monitoring data error caused by inaccurate electrode placement positions, and at the same time allows patients to perform motor imagery and helps patients actively establish the neural pathway of pelvic floor muscle movement. Summary of the Invention

[0007] To solve the above-mentioned problems in the prior art, the present invention provides a non-invasive visual monitoring and rehabilitation guidance system for pelvic floor muscles based on a brain-machine interface, which can reduce the errors caused by individual differences in the patient's pelvic floor during signal acquisition on the patient's pelvic floor skin surface, map the characteristics of the muscle electrical signals to the electroencephalogram (EEG) signal feature space, make the EEG signal data have application value in pelvic floor muscle tension assessment, and can avoid continuously placing the electromyogram (EMG) acquisition device on the patient's pelvic floor, protecting the patient's privacy and facilitating the patient's use.

[0008] To achieve the above object, the present invention is realized through the following technical solutions:

[0009] A non-invasive visual monitoring and rehabilitation guidance system for pelvic floor muscles based on a brain-machine interface of the present invention includes an EMG acquisition module, a main control module, a wireless transmission module, a host computer, and an EEG acquisition module placed on the patient's head skin surface, which are communicatively connected;

[0010] The EMG acquisition module and the EEG acquisition module act on the human body and feedback information to the main control module. Information interaction occurs between the main control module and the wireless transmission module, and the wireless transmission module transmits the information to the host computer through a cloud server (Alibaba Cloud platform).

[0011] Preferably, the EMG acquisition module includes a first electrostatic protection circuit, a first signal amplifier, a first band-pass filter, a first FIR filter, a monitoring probe, or electrodes placed on the patient's pelvic floor skin surface, which are communicatively connected;

[0012] The EEG signal acquisition module includes a second electrostatic protection circuit, a second signal amplifier, a second band-pass filter, a second FIR filter, an electrode cap, or an electrode headband.

[0013] Preferably, the electrodes placed on the patient's pelvic floor skin surface in the EMG acquisition module are a multi-channel electrode array. The first signal amplifier is an INA148 common-mode voltage differential amplifier. The low-frequency cut-off frequency of the first band-pass filter is 0.5 Hz, and the high-frequency cut-off frequency is 450 Hz. The first FIR filter is a first FIR notch filter for removing 50 Hz power frequency interference. The EMG acquisition module collects the pelvic floor muscle signals of the patient through the monitoring probe or the multi-channel electrode array. Since the pelvic floor muscle signals are weak, random, and non-stationary EMG signals, in order to improve the signal quality, the collected pelvic floor muscle signals are first shielded from the static electricity interference of the human body by the first electrostatic protection circuit, amplified to the range of 0 - 3.3 V (i.e., the range that can be collected by the main control module) by the first signal amplifier, the DC component and some high-frequency interference are removed by the first band-pass filter, and the power frequency interference is removed by the first FIR notch filter, and finally input to the main control module;

[0014] In the EEG acquisition module placed on the patient's head skin surface, the second signal amplifier is the INA148 common-mode voltage differential amplifier. The low-frequency cut-off frequency of the second band-pass filter is 0.5 Hz, and the high-frequency cut-off frequency is 450 Hz. The second FIR filter is the second FIR notch filter for removing 50 Hz power frequency interference. The EEG acquisition module placed on the patient's head skin surface collects the signals of the patient's head skin through an electrode cap or an electrode headband. Since the EEG signal is a weak, random, and non-stationary EEG signal, in order to improve the signal quality, the collected EEG signal is first shielded from the human body's static electricity interference through an electrostatic protection circuit, and then the second signal amplifier is used to amplify the weak signal to the range of 03.3V (i.e., the range interval available for the main control module to collect). The second band-pass filter is used to remove the DC component and part of the high-frequency interference, and the second FIR notch filter is used to remove the power frequency interference, and finally it is input to the main control module.

[0015] Preferably, the main control module includes a signal acquisition unit, a signal storage unit, a signal transmission unit, a voice prompt and guidance unit that are communicatively connected. The signal transmission unit is communicatively connected to other modules in the system through a wireless transmission module. The signal acquisition unit is used to obtain monitoring data.

[0016] Preferably, the main control module includes the microcontroller STM32L151. The signal acquisition unit includes a timer module, a DMA module, and an ADC module (the timer module, DMA module, and ADC module inside the microcontroller STM32L151) that are communicatively connected. The signal acquisition unit triggers ADC sampling through the timer module and transmits the collected data to the signal storage unit through the DMA module. During the acquisition process, different sampling rates of the signal are achieved by configuring the timer parameters in the timer module.

[0017] The signal storage unit includes SRAM (SRAM inside the microcontroller STM32L151). The signal storage unit stores the EMG signals of the EMG acquisition module and the EEG signals of the EEG acquisition module placed on the patient's head skin surface. The signal storage unit stores the parameters of the timer module, and then reads the sampling rate of the signal to facilitate the restoration and processing of the signal in the upper computer. The signal storage unit receives the voice rehabilitation guidance data packet transmitted from the upper computer through the wireless transmission module.

[0018] The signal transmission unit in the main control module packs the data in the signal storage unit into different data formats according to different wireless transmission modules to facilitate the signal transmission of different wireless transmission modules. At the same time, it receives the data from different wireless transmission modules and decodes different types of data.

[0019] The voice prompt and guidance unit in the main control module realizes voice reminders for patients to perform specified monitoring actions. After the voice reminder for the specified monitoring action is turned on, the timer module (timer) is triggered, and the signal acquisition unit starts working; the voice prompt and guidance unit receives the voice rehabilitation guidance data packet from the signal storage unit and conducts voice rehabilitation guidance for the patient.

[0020] Preferably, the wireless transmission module includes one or more of a WiFi module, a Bluetooth module, and an NBLoT module, and the wireless transmission module is connected to the main control module; the wireless transmission module can perform full-duplex communication and receive and send data.

[0021] Receive the signal from the main control module through the wireless transmission module, upload the signal from the main control module to the cloud server, receive the voice rehabilitation guidance data packet from the cloud server (Alibaba Cloud), and transmit the voice rehabilitation guidance data packet from the cloud server (Alibaba Cloud) to the main control module.

[0022] Preferably, the upper computer includes a cloud platform unit, a signal processing unit, a pelvic floor muscle performance evaluation and grading unit based on electroencephalogram signals, a training guidance plan formulation unit, and a data visualization unit that are communicatively connected; the cloud platform unit of the upper computer is one or more of a PC, a mobile phone, and a tablet device; the cloud platform unit in the upper computer is used to mine the data in the cloud server and feedback and upload the data to the cloud server.

[0023] Preferably, the signal processing unit in the upper computer first extracts the features of the electromyogram signal data and the electroencephalogram signal data, then performs information fusion on the electromyogram signal features and the electroencephalogram signal features extracted within the same action cycle, and maps the features of the electromyogram signal to the feature space of the electroencephalogram signal to achieve feature classification and recognition of the electroencephalogram signal; the signal processing unit in the upper computer extracts the time-domain features, frequency-domain features, and time-frequency features of the electromyogram signal data and the electroencephalogram signal data.

[0024] Preferably, the pelvic floor muscle performance evaluation and grading unit based on electroencephalogram signals in the upper computer combines the pelvic floor muscle performance evaluation and grading model based on electromyogram signals and the mapping model from electromyogram signals to electroencephalogram signals to generate a pelvic floor muscle performance evaluation and grading model based on electroencephalogram signals; the training guidance plan customization unit in the upper computer generates a personalized voice rehabilitation guidance data packet according to the pelvic floor muscle tension evaluation level of the patient. The voice rehabilitation guidance data packet includes the content of the guidance plan and the parameters of the voice prompt module in the main control module, so as to enable the patient to perform pelvic floor muscle motor imagery and help the patient actively establish the neural pathway of pelvic floor muscle movement; the data visualization unit in the upper computer displays patient information, and the patient information includes the average muscle strength at rest, the maximum muscle strength in the maximum contraction state, the pelvic floor muscle grade after AI evaluation, and rehabilitation guidance suggestions.

[0025] Preferably, the time-domain features include: mean, variance, standard deviation, maximum value, minimum value, number of zero-crossing points, difference between the maximum and minimum values; the frequency-domain features include: DC component, power spectral density; one or more of wavelet decomposition, short-time Fourier transform, and Hilbert transform are used for extracting the time-frequency features.

[0026] Advantages:

[0027] 1. Reduce the error caused by individual differences in the patient's pelvic floor during the signal acquisition on the skin surface of the patient's pelvic floor.

[0028] 2. Map the features of the electromyogram signal to the feature space of the electroencephalogram signal, so that the electroencephalogram signal data has application value in the assessment of pelvic floor muscle tension.

[0029] 3. Use the signal collected from the skin surface of the patient's brain to evaluate and treat the state of the patient's pelvic floor muscles, avoiding always placing the electromyogram acquisition device on the patient's pelvic floor, which can not only protect the patient's privacy but also facilitate the patient's use.

[0030] 4. Through motor imagery for rehabilitation guidance training, help the patient establish the neural pathway of pelvic floor muscle movement, and achieve long-term and effective treatment effects. Description of the Drawings

[0031] Figure 1 is the principle block diagram of the present invention.

[0032] Figure 2 is the signal acquisition flow chart of the present invention.

[0033] Figure 3 is the principle block diagram of the main control module of the present invention.

[0034] Figure 4 is the principle block diagram of the upper computer module of the present invention.

[0035] Figure 5 is the schematic diagram of the performance evaluation process of the present invention. Detailed Embodiments

[0036] The following further illustrates the present invention with reference to the accompanying drawings of the specification, but the present invention is not limited to the following embodiments.

[0037] Disadvantages of the prior art: 1) When monitoring the status of the pelvic floor muscles and providing rehabilitation guidance, the prior art mainly relies on the method of collecting data using an insertable probe for monitoring. In this type of method, the probe needs to be inserted into the patient's body throughout the rehabilitation training process, causing discomfort to the patient during the collection process, reducing the patient's experience during the treatment process, and reducing the rehabilitation training effect. 2) For existing non-invasive and insertable monitoring and rehabilitation devices, during the entire monitoring and treatment process, the electrodes for collecting electromyogram signals and the treatment electrodes need to be placed on the patient's pelvic floor skin surface all the time for operation. This method is not conducive to protecting the patient's privacy, and at the same time, the monitored data is easily affected by the number and position of the electrodes placed on the patient's pelvic floor, resulting in errors. 3) When monitoring the status of the pelvic floor muscles, the prior art directly relies on the data of the patient's pelvic floor area. The probe or electrode needs to be placed on the pelvic floor area all the time, with poor convenience. At the same time, the patient needs to expose their private parts all the time, increasing the patient's psychological burden. 4) When performing rehabilitation treatment on the pelvic floor muscles, the prior art adopts a passive stimulation method of outputting current. The muscles exercised and the effects are affected by the number and position of the electrodes. The muscles that can be exercised are limited, and the patient cannot perform motor imagination and actively establish the neural pathway of pelvic floor muscle movement, so a long-term and effective treatment effect cannot be achieved.

[0038] Technical principle / scheme of the present invention:: Establish a mapping model between the electromyogram signal of the pelvic floor muscles and the electroencephalogram signal; visually monitor, evaluate, grade, and provide rehabilitation guidance for the patient's pelvic floor muscle status based on the electroencephalogram signal; establish a guidance system for pelvic floor muscle rehabilitation exercises based on motor imagination.

[0039] As Figures 1-5 shown is a specific embodiment of a non-invasive visual monitoring and rehabilitation guidance system for pelvic floor muscles based on a brain-machine interface. This embodiment is a non-invasive visual monitoring and rehabilitation guidance system for pelvic floor muscles based on a brain-machine interface, which includes an electromyogram acquisition module, a main control module, a wireless transmission module, a host computer, and an electroencephalogram acquisition module placed on the patient's head skin surface that are communicatively connected; the electromyogram acquisition module and the electroencephalogram acquisition module act on the human body and feedback information to the main control module. Information is exchanged between the main control module and the wireless transmission module, and the wireless transmission module transmits to the host computer through a cloud server (Alibaba Cloud platform).

[0040] The system designed by the present invention can not only help medical staff accurately evaluate the functional status of the patient's pelvic floor muscles, but also provide accurate rehabilitation guidance for the patient through real-time monitoring and data analysis, accelerating the recovery process. Detailed explanation of the system composition of the present invention:

[0041] Electromyogram Acquisition Module: This module uses high-precision electromyogram sensors that can closely adhere to the key muscle groups of the patient's body (especially the pelvic floor muscle area) to accurately capture the weak electrical signals generated by muscle activities. These signals reflect the contraction intensity, speed, and coordination of the muscles, which are important indicators for evaluating the function of the pelvic floor muscles.

[0042] Electroencephalogram Acquisition Module: This module is placed on the skin surface of the patient's head and uses non-invasive electroencephalogram technology (such as EEG) to monitor the neural activities related to pelvic floor muscle control in the brain. By analyzing the brain waves, the ability of the brain to regulate the pelvic floor muscles can be understood, further revealing the neural mechanism of pelvic floor muscle dysfunction.

[0043] Main Control Module: As the core of the system, the main control module is responsible for receiving data from the electromyogram acquisition module and the electroencephalogram acquisition module, and performing preliminary processing and analysis. It can also identify abnormal muscle activity patterns or brain signal characteristics according to the preset algorithm, providing a basis for formulating subsequent rehabilitation plans.

[0044] Wireless Transmission Module: To ensure the real-time and security of data, the wireless transmission module uses advanced encryption technology and high-speed communication protocols to send the data processed by the main control module to the cloud server (such as Alibaba Cloud platform) via wireless methods such as WiFi or Bluetooth. This step greatly improves the flexibility of data processing and the feasibility of remote medical services.

[0045] Host Computer (Software Platform): Users access the data on the cloud server through the dedicated software on the computer or mobile device. The software interface is user-friendly and can intuitively display the visualization charts of pelvic floor muscle activities, the electroencephalogram topographic maps of brain activities, and the rehabilitation suggestions based on data analysis. Medical staff can remotely monitor the patient's status and adjust the rehabilitation plan, while patients can also self-monitor at home and participate in the rehabilitation process.

[0046] Advantages of the System of the Present Invention:

[0047] Non-invasive Monitoring: A completely non-invasive monitoring method reduces patient discomfort and is suitable for long-term follow-up monitoring. Precise Evaluation: Combining electromyogram and electroencephalogram data provides a comprehensive evaluation of pelvic floor muscle function and discovers potential problems. Personalized Rehabilitation: Based on individual data analysis, a personalized rehabilitation plan is customized to improve the treatment effect. Remote Management: With the help of the cloud service platform, remote interaction between doctors and patients is realized, facilitating follow-up consultations and adjustment of treatment plans. Visual Feedback: Intuitive data display helps patients better understand their own conditions, promotes self-management, and actively participates in rehabilitation.

[0048] In a preferred embodiment, the electromyogram acquisition module includes a first electrostatic protection circuit, a first signal amplifier, a first band-pass filter, a first FIR filter, a monitoring probe or electrodes placed on the skin surface of the patient's pelvic floor, which are communicatively connected; the electroencephalogram signal acquisition module includes a second electrostatic protection circuit, a second signal amplifier, a second band-pass filter, a second FIR filter, an electrode cap or an electrode headband. The electromyogram acquisition module and the electroencephalogram signal acquisition module accurately capture and process the electrophysiological activities of the pelvic floor muscles and the neural electrical activities related to the control of the pelvic floor muscles by the cerebral cortex respectively. The following is a further description of the internal components of these two modules:

[0049] In the electromyogram acquisition module:

[0050] The first electrostatic protection circuit: Function: As the first line of defense of the module, this circuit is designed to protect the subsequent circuits from damage caused by external electrostatic discharge (ESD). It uses a network composed of components such as capacitors, resistors, and diodes to effectively bypass or discharge the electrostatic charge to the ground, ensuring the stable transmission of signals. Design features: High reliability, fast response, and adaptability to various environmental conditions.

[0051] The first signal amplifier: Function: Since the electromyogram signal is weak (usually in the microvolt to millivolt level), the signal amplifier is responsible for amplifying it to an easily processable level while maintaining the signal fidelity. Design features: Low noise, high gain, and wide bandwidth to ensure that the signal is not distorted during amplification.

[0052] The first band-pass filter: Function: This filter is used to suppress frequency components outside the electromyogram signal, such as environmental noise and power frequency interference, and improve the signal-to-noise ratio of the signal. Design features: According to the characteristics of the electromyogram signal, a suitable passband range is set, such as 10 Hz to 500 Hz, to cover the main activity frequency bands.

[0053] The first FIR filter: Function: The FIR (finite impulse response) filter further processes the signal, reduces phase distortion through its linear phase characteristics, and removes residual high-frequency noise or artifacts at the same time. Design features: Programmable, easy to implement various filtering characteristics, such as low-pass, high-pass, band-stop, etc., to meet different signal processing requirements.

[0054] The monitoring probe or electrodes placed on the skin surface of the patient's pelvic floor: Function: Directly contact the patient's skin and capture the weak electrical signals generated during the activity of the pelvic floor muscles. Design features: High-conductivity, low-impedance materials are used to ensure the accuracy and stability of signal acquisition; the electrode shape and layout are optimized to maximize the signal capture efficiency.

[0055] In the electroencephalogram signal acquisition module:

[0056] Second Electrostatic Protection Circuit: Function: Similar to the electrostatic protection circuit in the EMG module, but the design parameters may be different to adapt to the lower level (microvolt level) and higher sensitivity of EEG signals. Design Features: Special attention is paid to the design of low-impedance paths to minimize the interference of electrostatic discharge on EEG signals.

[0057] Second Signal Amplifier: Function: In view of the weak characteristics of EEG signals, it provides a higher amplification factor while ensuring low noise and stable gain. Design Features: Adopt differential amplification technology to effectively suppress common-mode noise and improve the signal-to-noise ratio of the signal.

[0058] Second Band-Pass Filter: Function: Design a band-pass filter for the main frequency bands of EEG signals (such as alpha waves, beta waves, theta waves, etc.) to extract useful EEG information. Design Features: The passband range is adjustable to adapt to different EEG analysis requirements, such as emotion monitoring, attention analysis, etc.

[0059] Second FIR Filter: Function: Similar to the FIR filter in the EMG module, it is used to further purify the signal, reduce noise and artifacts. Design Features: May be optimized for the characteristics of EEG signals, such as more refined frequency response and phase control.

[0060] Electrode Cap or Electrode Headband: Function: Cover key areas of the head, such as the frontal lobe, occipital lobe, etc., to capture the electrical signals generated by the cerebral cortex. Design Features: Made of soft and comfortable materials to ensure comfort during long-term wearing; the electrode layout is scientifically studied to maximize the coverage and capture efficiency of brain activities.

[0061] Through the careful design and combination of the above components, the EMG acquisition module and the EEG signal acquisition module can efficiently capture and process the electrophysiological signals of the pelvic floor muscles and the cerebral cortex, providing reliable data support for subsequent monitoring, analysis, and rehabilitation guidance.

[0062] Provide some specific examples and technical parameters for the above components to better understand their roles and performances in practical applications.

[0063] Examples of Specific Components of the EMG Acquisition Module

[0064] First Electrostatic Protection Circuit: Example: Adopt a TVS (Transient Voltage Suppressor) diode array, such as the SM6T series of Semtech. These devices can provide bidirectional protection, effectively absorb the energy of electrostatic discharge, and protect the subsequent circuit from damage.

[0065] Technical Parameters: The breakdown voltage range is selected according to system requirements, such as ±15 kV (Human Body Model, HBM), and the operating voltage is compatible with the system power supply voltage.

[0066] First Signal Amplifier:

[0067] Example: Use an instrumentation amplifier, such as the AD8221 from ADI, which has low noise, high precision, and differential input characteristics, and is very suitable for amplifying EMG signals. Technical parameters: The gain is adjustable, usually between 10 and 1000 times; the input impedance is high to reduce signal loss; the common-mode rejection ratio (CMRR) is high to suppress common-mode noise.

[0068] Bandpass filter 1: Example: Adopt an active filter design, such as a second-order bandpass filter constructed using an operational amplifier (such as LM741). Technical parameters: The center frequency is set to the main frequency band of the EMG signal, such as 50 Hz to 500 Hz (preferably 450 Hz); the bandwidth is adjusted according to the signal characteristics to retain the necessary frequency components.

[0069] FIR filter 1: Example: Implement using a digital signal processing chip (DSP), such as the TMS320F28069 from Texas Instruments, which has an FIR filter library built-in and can flexibly configure filter parameters. Technical parameters: The filter order is selected according to the design requirements to balance the filtering effect and computational complexity; the sampling rate is consistent with the system sampling rate.

[0070] Monitoring probe or electrode placed on the skin surface of the patient's pelvic floor: Example: Adopt disposable Ag / AgCl gel electrodes, such as the Red Dot electrodes from 3M, which have good conductivity and skin adhesion. Technical parameters: The electrode size and shape are selected according to the monitoring requirements, such as circular, square, or strip-shaped; the conductive material is high-purity silver chloride; the electrode impedance is low to reduce signal attenuation.

[0071] Examples of specific components of the EEG signal acquisition module

[0072] ESD protection circuit 2: Example: Also use a TVS diode array, but may require a higher breakdown voltage and a lower capacitance value to adapt to the low-level characteristics of EEG signals. Technical parameters: The breakdown voltage range may be higher, such as ±30 kV (human body model, HBM), to provide a higher level of protection; the capacitance value is low to reduce interference to EEG signals.

[0073] Signal amplifier 2: Example: Adopt a low-noise, high-precision operational amplifier, such as the INA118 from TI, which is designed specifically for amplifying EEG signals. Technical parameters: The gain is adjustable, usually between 10 and 10000 times; the input noise density is low to reduce background noise; the input impedance is high to reduce signal loss.

[0074] Bandpass Filter No. 2: Example: Using digital filter design, such as a filter implemented through an FPGA (Field Programmable Gate Array), the parameters of the filter can be precisely controlled. Technical parameters: The center frequency is set to the main frequency band of EEG signals, such as 0.5 Hz to 100 Hz; the bandwidth is adjusted according to the analysis requirements.

[0075] FIR Filter No. 2: Example: Also implemented using DSP chips, such as the SHARC series from Analog Devices, which provide powerful digital signal processing capabilities. Technical parameters: The filter order and sampling rate are selected according to the system design requirements to provide sufficient frequency resolution and filtering effect.

[0076] Electrode cap or electrode headband: Example: An EEG cap using hydrogel or dry electrode technology, such as the EEG caps from EasyCap, which provide multiple electrode points covering different regions of the brain. Technical parameters: The number of electrodes is selected according to the analysis requirements, such as 16, 32, or 64 electrodes; the electrode material is a highly conductive material, such as silver / silver chloride; the electrode layout is designed according to the international 10 - 20 system or a custom layout.

[0077] Preferably, in one embodiment, the electrodes of the electromyogram acquisition module placed on the skin surface of the patient for judgment are a multi-channel electrode array. The first signal amplifier is an INA148 common-mode voltage differential amplifier. The low-frequency cut-off frequency of the first band-pass filter is 0.5 Hz, and the high-frequency cut-off frequency is 450 Hz. The first FIR filter is a first FIR notch filter for removing 50 Hz power frequency interference. The electromyogram acquisition module collects the pelvic floor muscle signals of the patient through a monitoring probe or a multi-channel electrode array. Since the pelvic floor muscle signals are weak, random, and non-stationary EMG signals, in order to improve the signal quality, the collected pelvic floor muscle signals are first shielded from the static electricity interference of the human body through a first electrostatic protection circuit, and the first signal amplifier is used to amplify the weak signals to the range of 0 to 3.3 V (i.e., the range interval available for the main control module to collect). The first band-pass filter is used to remove the DC component and part of the high-frequency interference, and the first FIR notch filter is used to remove the power frequency interference, and finally input to the main control module. In the electroencephalogram acquisition module placed on the skin surface of the patient's head, the second signal amplifier is an INA148 common-mode voltage differential amplifier. The low-frequency cut-off frequency of the second band-pass filter is 0.5 Hz, and the high-frequency cut-off frequency is 450 Hz. The second FIR filter is a second FIR notch filter for removing 50 Hz power frequency interference. The electroencephalogram acquisition module placed on the skin surface of the patient's head collects the signals of the patient's head skin through an electrode cap or an electrode headband. Since the electroencephalogram signals are weak, random, and non-stationary EEG signals, in order to improve the signal quality, the collected electroencephalogram signals are first shielded from the static electricity interference of the human body through an electrostatic protection circuit, and then the second signal amplifier is used to amplify the weak signals, amplified to the range of 0 to 3.3 V (i.e., the range interval available for the main control module to collect). The second band-pass filter is used to remove the DC component and part of the high-frequency interference, and the second FIR notch filter is used to remove the power frequency interference, and finally input to the main control module.

[0078] Electromyogram acquisition module:

[0079] 1. Multi-channel electrode array:

[0080] Function: The electrode array consists of multiple independent electrodes, and each electrode can independently collect the electrophysiological signals of the pelvic floor muscles. The multi-channel design helps to capture more comprehensive muscle activity information and improve the spatio-temporal resolution of the signals.

[0081] Technical specifications: The number, layout, and spacing of the electrodes are customized according to specific requirements to cover the key areas of the pelvic floor muscles. The electrode material is usually high-conductivity, low-impedance Ag / AgCl or gold-plated material to ensure good signal transmission and skin adhesion.

[0082] 2. INA148 common-mode voltage differential amplifier (the first signal amplifier):

[0083] Function: INA148 is a high-precision, low-noise differential amplifier, especially suitable for amplifying weak and easily interfered bioelectric signals. It can effectively suppress common-mode noise and improve the signal-to-noise ratio of the signal.

[0084] Technical specifications: The gain is adjustable, usually set between dozens and hundreds of times to adapt to pelvic floor muscle signals of different intensities. The input impedance is high to reduce signal loss; the common-mode rejection ratio (CMRR) is high to suppress common-mode noise.

[0085] 3. Band-pass filter No. 1:

[0086] Function: The band-pass filter is used to remove the DC component (DC offset) and part of the high-frequency noise in the pelvic floor muscle signal, while retaining the main frequency band of the signal (such as 0.5 Hz to 450 Hz).

[0087] Technical specifications: The low-frequency cut-off frequency and the high-frequency cut-off frequency are set to 0.5 Hz and 450 Hz respectively to cover the main frequency band of pelvic floor muscle activity. The filter type may be Butterworth, Chebyshev or elliptic filter, depending on the required frequency response characteristics.

[0088] 4. FIR notch filter No. 1:

[0089] Function: The FIR notch filter is used to remove the power frequency interference of 50 Hz (or 60 Hz, depending on the region), which is a common interference source in bioelectric signal acquisition.

[0090] Technical specifications: The notch frequency is set to 50 Hz (or 60 Hz), and the notch width is adjusted according to requirements to effectively remove the power frequency interference without affecting the main frequency band of the signal.

[0091] 5. Signal processing flow:

[0092] Electrostatic protection: Through the first electrostatic protection circuit, the electrostatic interference of the human body is effectively shielded to protect the subsequent circuits from damage.

[0093] Signal amplification: The collected weak pelvic floor muscle signal is amplified to the 0 - 3.3 V range for the main control module to collect and process.

[0094] Filtering and notching: Use the first band-pass filter and the first FIR notch filter to remove unnecessary frequency components and power frequency interference and improve the signal quality.

[0095] Data transmission: The processed signal is transmitted to the main control module through a communication interface (such as SPI, I2C or UART) for further analysis.

[0096] EEG acquisition module:

[0097] 1. Electrode cap or electrode headband:

[0098] Function: The electrode cap or electrode headband covers key areas of the patient's head and is used to collect electrophysiological signals from the cerebral cortex. They usually contain multiple electrode points, which can capture more comprehensive EEG activity information.

[0099] Technical specifications: The number, layout, and spacing of the electrodes are based on international standards (such as the 10 - 20 system) or custom designs. The electrode material is highly conductive, low - impedance Ag / AgCl or gold - plated material to ensure good signal transmission and skin adhesion.

[0100] 2. INA148 common - mode voltage differential amplifier (the second signal amplifier):

[0101] Function and technical specifications: The same as the first signal amplifier in the EMG acquisition module, used to amplify weak and easily interfered EEG signals, suppress common - mode noise, and improve the signal - to - noise ratio of the signals.

[0102] 3. The second band - pass filter and the second FIR notch filter:

[0103] Function and technical specifications: Similar to the first band - pass filter and the first FIR notch filter in the EMG acquisition module, used to remove the DC component, high - frequency noise, and power - frequency interference in the EEG signals, and improve the signal quality. The low - frequency cut - off frequency and the high - frequency cut - off frequency are set to 0.5 Hz and 450 Hz respectively to cover the main frequency bands of EEG activity. The notch frequency is set to 50 Hz (or 60 Hz) to remove power - frequency interference.

[0104] 4. Signal processing flow:

[0105] Electrostatic protection: Through the second electrostatic protection circuit, effectively shield the electrostatic interference of the human body and protect the subsequent circuits from damage.

[0106] Signal amplification: Amplify the collected weak EEG signals to the range of 0 - 3.3 V for the main control module to collect and process.

[0107] Filtering and notching: Use the second band - pass filter and the second FIR notch filter to remove unnecessary frequency components and power - frequency interference and improve the signal quality.

[0108] Data transmission: The processed signals are transmitted to the main control module through the communication interface for further analysis, such as spectral analysis, time - domain analysis, or EEG topographic mapping, etc.

[0109] Preferably, in one embodiment, the main control module includes a signal acquisition unit, a signal storage unit, a signal transmission unit, a voice prompt and guidance unit connected by communication. The signal transmission unit is communicatively connected to other modules in the system through a wireless transmission module, and the signal acquisition unit is used to obtain monitoring data.

[0110] Core functions of the signal acquisition unit: As an important part of the main control module, the signal acquisition unit is mainly responsible for receiving preprocessed and amplified monitoring data from various monitoring modules (such as electromyogram acquisition module and electroencephalogram acquisition module). These data cover key information on pelvic floor muscle activities and cerebral electrophysiological activities. Technical implementation: The signal acquisition unit usually includes an analog-to-digital converter (ADC) to convert analog signals into digital signals for subsequent processing and storage. In addition, it may also have a multiplexer (MUX) function to process data from multiple sensors simultaneously. Data format: The acquired data is stored in a specific format, including timestamp, sensor number, signal intensity, etc., for subsequent data analysis and processing.

[0111] Core functions of the signal storage unit: The signal storage unit is responsible for storing the acquired monitoring data for playback, analysis, and report generation when needed. Technical implementation: The storage unit may use internal memories (such as RAM, Flash memory) or external storage devices (such as SD cards, USB storage devices). To ensure data integrity and security, the storage unit usually has data verification and error recovery mechanisms. Storage capacity: The storage capacity depends on actual needs and may need to support the storage of data for long-term continuous monitoring.

[0112] Core functions of the signal transmission unit: The signal transmission unit communicates and connects with other modules in the system (such as electromyogram acquisition module, electroencephalogram acquisition module, and remote server or doctor workstation) through wireless transmission modules (such as WiFi, Bluetooth, Zigbee, etc.) to achieve real-time data transmission and remote monitoring. Technical implementation: The wireless transmission module uses standard communication protocols (such as TCP / IP, UDP, etc.) for data encapsulation and transmission. To ensure the reliability and security of data transmission, data encryption and verification mechanisms may be used. Communication range: The communication range depends on the characteristics of the selected wireless transmission technology and may need to support communication requirements in indoor and outdoor environments.

[0113] Core functions of the voice prompt and guidance unit: The voice prompt and guidance unit is responsible for providing real-time voice feedback and guidance to users, including the interpretation of monitoring data, prompts for rehabilitation training, and alarms for abnormal situations. Technical implementation: This unit may use text-to-speech (TTS) technology or pre-recorded voice files to achieve voice output. In addition, it can also work in coordination with other modules of the system (such as the signal analysis unit) to provide personalized voice guidance based on the analysis results. User interaction: To improve the user experience, the voice prompt and guidance unit may support multiple language selections and volume adjustment functions.

[0114] By integrating the functions of the above-mentioned various units, the main control module realizes the comprehensive monitoring and analysis of pelvic floor muscle and EEG signals. It can not only receive and store monitoring data in real time, but also communicate with a remote server or a doctor's workstation through a wireless transmission module to achieve remote monitoring and analysis of the data. At the same time, the voice prompt and guidance unit provides a convenient interaction method for users, making the monitoring process more intuitive and easy to understand. In specific applications, the main control module can be customized and optimized according to the actual needs of users. For example, in pelvic floor muscle rehabilitation training, it can provide personalized training guidance and feedback based on the collected pelvic floor muscle signals; in EEG signal monitoring, it can monitor the changes in brain activities in real time, providing strong support for neuroscience research and clinical diagnosis.

[0115] In a preferred embodiment, the main control module includes a microcontroller STM32L151. The signal acquisition unit includes a timer module, a DMA module, and an ADC module (the timer module, DMA module, and ADC module inside the microcontroller STM32L151) connected in communication. The signal acquisition unit triggers ADC sampling through the timer module and transmits the acquired data to the signal storage unit through the DMA module. During the acquisition process, different sampling rates of signals are acquired by configuring the timer parameters in the timer module.

[0116] The signal storage unit includes SRAM (the SRAM inside the microcontroller STM32L151). The signal storage unit stores the EMG signals of the EMG acquisition module and the EEG signals of the EEG acquisition module placed on the patient's head skin surface; the signal storage unit stores the parameters of the timer module, and then reads the sampling rate of the signal to facilitate the restoration and processing of the signal in the upper computer; the signal storage unit receives the voice rehabilitation guidance data packet transmitted from the upper computer through the wireless transmission module.

[0117] The signal transmission unit in the main control module packs the data in the signal storage unit into different data formats according to different wireless transmission modules to facilitate signal transmission for different wireless transmission modules. At the same time, it receives data from different wireless transmission modules and decodes different types of data.

[0118] The voice prompt and guidance unit in the main control module realizes voice reminding the patient to perform a specified monitoring action. After the voice reminder for the specified monitoring action is turned on, the timer module (timer) is triggered, and the signal acquisition unit starts to work; the voice prompt and guidance unit receives the voice rehabilitation guidance data packet from the signal storage unit and provides voice rehabilitation guidance to the patient.

[0119] In a preferred embodiment, the wireless transmission module includes one or more of a WiFi module, a Bluetooth module, and an NBLoT module. The wireless transmission module is connected to the main control module. The wireless transmission module can perform full-duplex communication and receive and send data. The wireless transmission module receives signals from the main control module, uploads the signals from the main control module to the cloud server, receives the voice rehabilitation guidance data packet from the cloud server (Alibaba Cloud), and transmits the voice rehabilitation guidance data packet from the cloud server (Alibaba Cloud) to the main control module.

[0120] In a preferred embodiment, the host computer includes a cloud platform unit, a signal processing unit, a pelvic floor muscle performance evaluation and grading unit based on electroencephalogram signals, a training guidance plan formulation unit, and a data visualization unit that are communicatively connected. The cloud platform unit of the host computer is one or more of a PC, a mobile phone, and a tablet device. The cloud platform unit in the host computer is used to mine data in the cloud server and feedback and upload the data to the cloud server.

[0121] In a preferred embodiment, the signal processing unit in the host computer first extracts the features of the electromyogram signal data and the electroencephalogram signal data, and then performs information fusion on the electromyogram signal features and the electroencephalogram signal features extracted within the same motion cycle, maps the features of the electromyogram signal to the feature space of the electroencephalogram signal, so as to realize the feature classification and recognition of the electroencephalogram signal. The signal processing unit in the host computer extracts the time-domain features, frequency-domain features, and time-frequency features of the electromyogram signal data and the electroencephalogram signal data. The pelvic floor muscle performance evaluation and grading unit based on electroencephalogram signals in the host computer generates a pelvic floor muscle performance evaluation and grading model based on electroencephalogram signals by combining the pelvic floor muscle performance evaluation and grading model based on electromyogram signals and the mapping model from electromyogram signals to electroencephalogram signals. The training guidance plan customization unit in the host computer generates a personalized voice rehabilitation guidance data packet according to the pelvic floor muscle tension evaluation level of the patient. The voice rehabilitation guidance data packet includes the content of the guidance plan and the parameters of the voice prompt module in the main control module, so as to enable the patient to perform pelvic floor muscle motor imagery and help the patient actively establish the neural pathway of pelvic floor muscle movement. The data visualization unit in the host computer displays patient information, and the patient information includes the average muscle strength at rest, the maximum muscle strength in the maximum contraction state, the pelvic floor muscle grade after AI evaluation, and the rehabilitation guidance suggestions. The time-domain features include: mean, variance, standard deviation, maximum value, minimum value, number of zero crossings, difference between the maximum and minimum values; the frequency-domain features include: DC component, power spectral density; the extraction of the time-frequency features adopts one or more of wavelet decomposition, short-time Fourier transform, and Hilbert transform.

[0122] The signal processing unit in the present invention and the upper and middle computers is the key part of the entire system, responsible for processing and analyzing the myoelectric signals and electroencephalogram signals transmitted from the lower computers (such as myoelectric acquisition modules and electroencephalogram acquisition modules). The working process of the signal processing unit mainly includes feature extraction, information fusion, feature mapping, and classification and recognition.

[0123] 1. Feature extraction:

[0124] - Time-domain features: The signal processing unit first extracts the time-domain features of the myoelectric signals and electroencephalogram signals. These features include mean value, variance, standard deviation, maximum value, minimum value, number of zero-crossing points, and difference between the maximum and minimum values, etc. These features can reflect the fluctuation situation and amplitude distribution of the signals.

[0125] - Frequency-domain features: In addition to time-domain features, the signal processing unit also extracts the frequency-domain features of the signals, such as direct current component and power spectral density. Frequency-domain features can reveal the frequency components and energy distribution of the signals, and are of great significance for understanding the generation mechanism of the signals and identifying signals in different states.

[0126] - Time-frequency features: In order to more comprehensively describe the time-frequency characteristics of the signals, the signal processing unit also uses methods such as wavelet decomposition, short-time Fourier transform (STFT), or Hilbert transform to extract the time-frequency features of the signals. These features can reflect the frequency components and energy changes of the signals in different time periods, and are very useful for the identification and analysis of complex signals.

[0127] 2. Information fusion and feature mapping:

[0128] - In the same action cycle, the signal processing unit fuses the myoelectric signal features and electroencephalogram signal features extracted. This usually involves steps such as splicing of feature vectors and application of fusion algorithms to form a joint feature vector.

[0129] - Next, the signal processing unit maps the features of the myoelectric signals into the feature space of the electroencephalogram signals. The conversion and mapping of the feature space can be achieved through machine learning methods (such as support vector machines, neural networks, etc.), so as to realize the feature classification and recognition of the electroencephalogram signals.

[0130] The pelvic floor muscle performance evaluation and grading unit based on EEG signals in the host computer is also an important part of the system. It combines the pelvic floor muscle performance evaluation and grading model based on EMG signals and the mapping model from EMG signals to EEG signals to generate a pelvic floor muscle performance evaluation and grading model based on EEG signals. This model can use the characteristics of EEG signals to evaluate the performance of the pelvic floor muscles, thereby providing patients with more accurate and personalized rehabilitation guidance. The training guidance program customization unit in the host computer generates a personalized voice rehabilitation guidance data package based on the patient's pelvic floor muscle tension assessment level. This data package includes the content of the guidance program and the parameters of the voice prompt module in the main control module. By allowing patients to imagine pelvic floor muscle movements, the training guidance program customization unit can help patients actively establish neural pathways for pelvic floor muscle movements, thereby improving the rehabilitation effect. The data visualization unit in the host computer is responsible for displaying patient information, including average muscle strength at rest, maximum muscle strength at maximum contraction, pelvic floor muscle grade assessed by AI, and rehabilitation guidance suggestions. Through intuitive data display, doctors and patients can better understand the progress and effect of rehabilitation, so as to formulate a more reasonable rehabilitation plan.

[0131] In order to realize the technical functions, the host computer in the present invention can use signal processing technology, machine learning algorithms and data visualization tools. Furthermore, in order to improve the accuracy and reliability of the system, the algorithm can be continuously optimized and adjusted. For example, feature extraction methods, fusion algorithms and classification recognition models can be used to improve the accuracy and efficiency of signal processing; deep learning and other technologies can be used to optimize the pelvic floor muscle performance evaluation grading model and mapping model; more intuitive and easy-to-use data visualization tools can also be used to improve user experience.

[0132] Finally, it should be noted that the present invention is not limited to the above embodiments, and there are many variations. All variations that can be directly derived or associated with the content disclosed by ordinary technicians in this field should be considered as the protection scope of the present invention.

Claims

1. A non-invasive visual monitoring and rehabilitation guidance system for pelvic floor muscles based on brain-muscle-machine, characterized by: It includes a communication-connected myoelectric acquisition module, a main control module, a wireless transmission module, a host computer, and an electroencephalogram acquisition module placed on the surface of the patient's head skin; The electromyography acquisition module and the electroencephalography acquisition module act on the human body and feed back information to the main control module. The main control module interacts with the wireless transmission module, and the wireless transmission module transmits information to the host computer through the cloud server.

2. The brain-muscle-machine based non-invasive visual monitoring and rehabilitation guidance system for pelvic floor muscles according to claim 1, characterized in that: The electromyography acquisition module includes a communication-connected No. 1 electrostatic protection circuit, a No. 1 signal amplifier, a No. 1 bandpass filter, a No. 1 FIR filter, a monitoring probe or an electrode placed on the surface of the patient's pelvic floor skin; the electroencephalogram signal acquisition module includes a No. 2 electrostatic protection circuit, a No. 2 signal amplifier, a No. 2 bandpass filter, a No. 2 FIR filter, an electrode cap or an electrode headband.

3. The brain-muscle-machine based non-invasive visual monitoring and rehabilitation guidance system for pelvic floor muscles according to claim 2, characterized in that: The electrodes placed on the patient's skin surface in the electromyography acquisition module are multi-channel electrode arrays, the No. 1 signal amplifier is an INA148 common-mode voltage differential amplifier, the No. 1 bandpass filter has a low-frequency cutoff frequency of 0.5 Hz, a high-frequency cutoff frequency of 450 Hz, and the No. 1 FIR filter is a No. 1 FIR notch filter for removing 50 Hz power frequency interference; the electromyography acquisition module acquires the patient's pelvic floor muscle signal through a monitoring probe or a multi-channel electrode array, the acquired pelvic floor muscle signal is first shielded from the electrostatic interference of the human body by a No. 1 electrostatic protection circuit, the No. 1 signal amplifier is used to amplify the weak signal to the range of 0-3.3 V, the No. 1 bandpass filter is used to remove the DC component and high-frequency interference, and the No. 1 FIR notch filter is used to remove the power frequency interference, and finally input it to the main control module; The No. 2 signal amplifier in the EEG acquisition module is an INA148 common-mode voltage differential amplifier, the No. 2 bandpass filter has a low-frequency cutoff frequency of 0.5 Hz and a high-frequency cutoff frequency of 450 Hz, and the No. 2 FIR filter is a No. 2 FIR notch filter for removing 50 Hz power frequency interference; the EEG acquisition module placed on the surface of the patient's head skin collects the signal of the patient's head skin through an electrode cap or an electrode headband, uses the No. 2 signal amplifier to amplify the weak signal to the range of 0-3.3 V, uses the No. 2 bandpass filter to remove DC components and high-frequency interference, and uses the No. 2 FIR notch filter to remove power frequency interference, and finally inputs it into the main control module.

4. A brain-muscle-machine based non-invasive visual monitoring and rehabilitation guidance system for pelvic floor muscles according to claim 1 or 3, characterized in that: The main control module includes a signal acquisition unit, a signal storage unit, a signal transmission unit, and a voice prompt and guidance unit that are communicatively connected. The signal transmission unit is communicatively connected with other modules in the system via a wireless transmission module. The signal acquisition unit is used to obtain monitoring data.

5. The brain-muscle-machine based non-invasive visual monitoring and rehabilitation guidance system for pelvic floor muscles according to claim 4, characterized in that: The main control module includes a microcontroller STM32L151, and the signal acquisition unit includes a timer module, a DMA module and an ADC module that are connected in communication. The signal acquisition unit triggers ADC sampling through the timer module, and transmits the collected data to the signal storage unit through the DMA module. During the acquisition process, the signal is collected at different sampling rates by configuring the timer parameters in the timer module; The signal storage unit includes an SRAM, and the signal storage unit stores the myoelectric signal of the myoelectric acquisition module and the electroencephalographic signal of the electroencephalographic acquisition module; The signal storage unit stores the parameters of the timer module and then reads the sampling rate of the signal to facilitate signal restoration and signal processing in the host computer; The signal storage unit receives the voice rehabilitation guidance data packet transmitted from the host computer through the wireless transmission module; The signal transmission unit packages the data in the signal storage unit into different data formats according to different wireless transmission modules, so as to facilitate the signal transmission of different wireless transmission modules, receive data from different wireless transmission modules at the same time, and decode different types of data; The voice prompt and guidance unit realizes voice reminders for patients to perform designated monitoring actions. After the designated monitoring action starts the voice reminder, the timer module is triggered and the signal acquisition unit starts working; The voice prompt and guidance unit receives the voice rehabilitation guidance data packet from the signal storage unit and provides voice rehabilitation guidance to the patient.

6. The brain-muscle-machine based non-invasive visual monitoring and rehabilitation guidance system for pelvic floor muscles according to claim 1 or 5, characterized in that: The wireless transmission module includes one or more of a WiFi module, a Bluetooth module, and a NB-LoT module, and the wireless transmission module is connected to the main control module; the wireless transmission module can realize full-duplex communication, and receive and send data; The signal from the main control module is received through the wireless transmission module, the signal from the main control module is uploaded to the cloud server, the voice rehabilitation guidance data packet from the cloud server is received, and the voice rehabilitation guidance data packet from the cloud server is transmitted to the main control module.

7. The brain-muscle-machine based non-invasive visual monitoring and rehabilitation guidance system for pelvic floor muscles according to claim 1 or 5, characterized in that: The host computer includes a cloud platform unit with communication connection, a signal processing unit, a pelvic floor muscle performance evaluation and grading unit based on EEG signals, a training guidance program formulation unit, and a data visualization unit; the cloud platform unit is one or more of a PC, a mobile phone, and a tablet device; the cloud platform unit is used to mine data in the cloud server and upload data feedback to the cloud server.

8. The brain-muscle-machine based non-invasive visual monitoring and rehabilitation guidance system for pelvic floor muscles according to claim 7, characterized in that: The signal processing unit in the host computer first extracts features from the electromyographic signal data and the electroencephalographic signal data, then fuses the electromyographic signal features and the electroencephalographic signal features extracted within the same action cycle, and maps the features of the electromyographic signal to the feature space of the electroencephalographic signal to achieve feature classification and recognition of the electroencephalographic signal; the signal processing unit in the host computer extracts time domain features, frequency domain features, and time-frequency features from the electromyographic signal data and the electroencephalographic signal data.

9. The brain-muscle-machine based non-invasive visual monitoring and rehabilitation guidance system for pelvic floor muscles according to claim 8, characterized in that: The pelvic floor muscle performance evaluation and grading unit based on EEG signals combines the pelvic floor muscle performance evaluation and grading model based on EMG signals and the mapping model from EMG signals to EEG signals to generate a pelvic floor muscle performance evaluation and grading model based on EEG signals; The training guidance program customization unit in the host computer generates a personalized voice rehabilitation guidance data package based on the patient's pelvic floor muscle tension assessment level. The voice rehabilitation guidance data package includes the content of the guidance plan and the parameters of the voice prompt module in the main control module; the data visualization unit in the host computer displays patient information, which includes the average muscle strength at rest, the maximum muscle strength at maximum contraction, the pelvic floor muscle level after AI assessment, and rehabilitation guidance suggestions.

10. The brain-muscle-machine based non-invasive visual monitoring and rehabilitation guidance system for pelvic floor muscles according to claim 9, characterized in that: The time domain features include: mean, variance, standard deviation, maximum value, minimum value, number of zero crossing points, and difference between maximum and minimum values; the frequency domain features include: DC component and power spectrum density; the time-frequency features are extracted by using one or more of wavelet decomposition, short-time Fourier transform, and Hilbert transform.

Images