Pelvic floor magnetic stimulation training system and method based on brain-computer interaction

Abstract

The invention discloses a pelvic floor magnetic stimulation training system based on brain-computer interaction, which comprises an electroencephalogram acquisition module, electroencephalogram signals acquired by the electroencephalogram acquisition module are input to an electroencephalogram decoding algorithm module, and the electroencephalogram decoding algorithm module comprises an electroencephalogram evaluation unit and an electroencephalogram training unit; the control module receives a signal from the electroencephalogram decoding algorithm module or a signal from the response evaluation module, generates a control signal and transmits the control signal to the visual and auditory feedback module and the pelvic floor magnetic stimulation module; the invention further discloses a pelvic floor magnetic stimulation training method based on brain-computer interaction. The pelvic floor magnetic stimulation training method comprises an electroencephalogram training mode and an image training mode. The pelvic floor contraction intention of a training object is actively recognized through brain-computer interaction, the pelvic floor magnetic stimulation module can be triggered to work after the optimal electroencephalogram threshold value is reached, pelvic floor muscle contraction is assisted in the process that the training object actively participates in pelvic floor muscle rehabilitation training, and active rehabilitation is achieved; when good electroencephalogram signals cannot be obtained, pelvic floor muscle contraction training can be carried out in an image training mode.

Description

Technical Field

[0001] The present invention relates to the technical field of pelvic floor rehabilitation, and in particular to a pelvic floor magnetic stimulation training system and method based on brain-computer interaction. Background Art

[0002] Postpartum pelvic floor disorders are common in my country. A study of 2,386 women screened 42-60 days after delivery found that the incidence of pelvic organ prolapse was 75.2%, anterior vaginal wall bulge was 72.4%, posterior vaginal wall bulge was 39.3%, and uterine prolapse was 20.1%. Another study found that 50% of women who delivered vaginally had varying degrees of pelvic organ prolapse during routine gynecological examinations, and the incidence ranged from 33% to 79% between six weeks and one year after delivery. The incidence of postpartum urinary incontinence is 24.5% for first vaginal births and 5.2% for first cesarean sections. The incidence rate is increasing with increasing frequency of cesarean and vaginal deliveries. Urinary incontinence and pelvic organ prolapse are common pelvic floor disorders that can affect relationships, lead to urinary and reproductive tract infections, and reduce quality of life. Current treatments and exercises include electrical biofeedback, passive magnetic stimulation, Kegel exercises, and vaginal dumbbells. However, these methods are mainly passive training with external timed stimulation. At the same time, most patients find it difficult or unable to correctly control the pelvic floor muscles during training, so the rehabilitation effect is limited.

[0003] The "A pelvic floor magnetic stimulation therapeutic device" disclosed in the Chinese patent literature has a publication number of CN107569773B and a publication date of 2020-08-28. It includes a host, a treatment chair and a display. The treatment chair includes a treatment chair bracket, a backrest and a headrest on the upper part of the backrest, a seat cushion and armrests on both sides of the seat cushion. A treatment device is installed at the bottom of the seat cushion, a backrest support is provided at the back of the backrest, protective covers are provided on both sides of the backrest support, and backlight covers are provided on both sides of the armrests. The host is installed and fixed to the rear of the treatment chair. The host includes a host housing, a host bracket and a host control system. The bottom of the host is equipped with movable casters, and the display is installed on the arm connected to the side bracket of the host. The advantage of the pelvic floor magnetic stimulation therapeutic device of this technology is that the integrated structure is easier to install and maintain, the non-invasive, non-invasive and painless treatment will not cause damage or infection to the body cavity, and the ergonomic seat is more comfortable during the treatment process. However, this technology still adopts a passive training method through external timed stimulation. During the training process, it is difficult or impossible for patients to correctly control the pelvic floor muscles, resulting in limited rehabilitation effect. Summary of the Invention

[0004] The present invention aims to overcome the problem that pelvic floor magnetic stimulation in the prior art usually adopts a passive training method of external timed stimulation. During the training process, it is difficult or impossible for patients to correctly control the pelvic floor muscles, resulting in limited rehabilitation effect. A pelvic floor magnetic stimulation training system and method based on brain-computer interaction is provided to overcome the problem that

[0005] In order to achieve the above object, the present invention adopts the following technical solutions: A pelvic floor magnetic stimulation training system based on brain-computer interaction includes an EEG acquisition module. The EEG signals collected by the EEG acquisition module are input into an EEG decoding algorithm module. The EEG decoding algorithm module includes an EEG evaluation unit and an EEG training unit. After receiving the signal from the EEG decoding algorithm module or the signal from the reaction evaluation module, the control module generates a control signal and transmits it to the visual and auditory feedback module and the pelvic floor magnetic stimulation module.

[0006] The present invention combines the control mode of the brain-computer interface with pelvic floor magnetic stimulation. When the EEG information of the training subject is collected in advance, the training subject's intention to contract the pelvic floor muscles is actively identified. When the training subject wants to contract the pelvic floor muscles, the pelvic floor magnetic stimulation module is used to assist the contraction of the pelvic floor muscles, thereby changing the existing passive stimulation training mode to an active training stimulation mode, thereby achieving convenient and rapid active rehabilitation. In addition, since the EEG decoding algorithm module can perform targeted EEG signal analysis for different training subjects and generate corresponding control methods, it can be applied to different training subjects in different periods, thereby accelerating the rehabilitation process.

[0007] Preferably, in the EEG decoding algorithm module: The EEG evaluation unit calculates the optimal EEG threshold based on the EEG information of the training subject when he / she imagines the pelvic floor muscle contraction and relaxation; The EEG training unit scores the training subject's imagined state of pelvic floor muscle contraction based on the optimal EEG threshold. When the score exceeds the preset value, the control module controls the pelvic floor magnetic stimulation module to operate.

[0008] In the present invention, the EEG decoding algorithm module receives the EEG information of the training subject, decodes the training subject's movement intention of contracting and relaxing the pelvic floor muscles in real time, and then outputs the user's movement intention state in real time; the control module can control the specific work of the pelvic floor magnetic stimulation module according to the movement intention state output by the EEG decoding algorithm module.

[0009] Preferably, the reaction evaluation module calculates the average reaction time of the training subject by taking the difference between the time of prompting the training subject to contract the pelvic floor muscles and the feedback time of the training subject contracting the pelvic floor muscles, and uses it as the instruction delay time of the imagery training mode.

[0010] The reaction evaluation module in the present invention is used to collect the training subject's reaction time after receiving the pelvic floor muscle contraction prompt, such as pressing a button / mouse / joystick, and transmit the signal back to the reaction evaluation module. The evaluation system issues the pelvic floor muscle contraction instruction to the user's execution of the corresponding contraction action, which is used to set the instruction delay time in the subsequent imagery training mode.

[0011] Preferably, the EEG evaluation unit stores an EEG threshold calculation model; Based on the collected EEG signals, EEG frequency band feature data of different frequency bands are obtained, and the characteristics of the training subjects' imagination state when imagining pelvic floor muscle contraction and the rest state characteristics when relaxing are extracted; they are substituted into the EEG threshold calculation model to calculate the optimal EEG threshold for each EEG frequency band.

[0012] In the present invention, the EEG evaluation unit decodes the most suitable optimal EEG threshold by collecting the EEG information of the training subject in the imaginary pelvic floor muscle contraction state and the relaxation state to distinguish whether the training subject is in the imagination state or the relaxation state, and establishes an imagination model for use by the EEG training unit.

[0013] Preferably, the EEG training unit stores an imagination state scoring model; The EEG frequency band characteristic data of different frequency bands are obtained according to the collected EEG signals, and are substituted into the imagination state scoring model together with the optimal EEG threshold obtained in the EEG evaluation unit to calculate the imagination state score of the training object.

[0014] In the present invention, the EEG training unit decodes the EEG characteristics of the training subject during pelvic floor muscle contraction training in real time, compares it with the optimal EEG threshold obtained by the EEG evaluation unit, and transmits the comparison result to the control module, which controls the operation of the pelvic floor magnetic stimulation module based on the comparison result.

[0015] Preferably, the EEG threshold calculation model is: Where T is the optimal EEG threshold; a1 and a2 are characteristic coefficients, the sum of which is 1; F image Characteristic of the imaginary state; F rest is the rest state feature; m is the data length.

[0016] In the present invention, since the EEG frequency bands of EEG signals are of different types, the corresponding optimal EEG threshold needs to be calculated for each EEG frequency band, that is, a training object needs to calculate and analyze the optimal EEG thresholds of multiple EEG frequency bands and substitute them into the subsequent imagination state scoring model for training. When the EEG acquisition module collects EEG information, the signal collection of the imagination state or the relaxation state will continue for a certain period of time, so there will be a certain data length, so that the characteristics of different states can be obtained more completely to avoid omissions.

[0017] Preferably, the imagination state scoring model is: Where Score is the imagination state score; C is the preset auxiliary score; k0 is the score conversion coefficient; k j is the frequency band characteristic coefficient of the jth EEG frequency band, and the sum of all frequency band characteristic coefficients is 1; F j is the EEG frequency band characteristic data of the jth EEG frequency band; T j is the optimal EEG threshold of the jth EEG frequency band; q is the number of EEG frequency bands considered in the imagination state scoring model.

[0018] The preset auxiliary scores in the present invention can be set according to actual needs. In the imagination state scoring model, there will be certain corrections according to the different types of EEG frequency bands actually considered; generally, EEG signals include δ frequency bands, θ frequency bands, α frequency bands, β frequency bands and μ frequency bands, etc. When performing EEG threshold calculations and imagination state scoring, all EEG frequency bands can be taken into consideration in the calculation process, or several frequency bands that occupy a dominant part of the EEG information can be taken into consideration for calculation according to different training objects.

[0019] A pelvic floor magnetic stimulation training method based on brain-computer interaction, comprising: The EEG training mode evaluates the training subject's imagination state score when imagining the pelvic floor muscle contraction or relaxation. When the imagination state score exceeds the preset score, the control module controls the pelvic floor magnetic stimulation module to operate; In the imagery training mode, after the pelvic floor muscle contraction prompt is given, the control module waits for the instruction delay time and then controls the pelvic floor magnetic stimulation module to work.

[0020] The present invention provides two different active training methods, the main one is the EEG training mode, which actively identifies the training subject's intention to contract the pelvic floor muscles. When the training subject wants to contract the pelvic floor muscles, the pelvic floor magnetic stimulation module is used to assist the contraction of the pelvic floor muscles, thereby performing pelvic floor muscle contraction training; at the same time, if a good EEG signal cannot be obtained, the imagery training mode can be entered to perform pelvic floor muscle contraction training.

[0021] Preferably, the process of performing the EEG training mode includes: Collect EEG information of the training subjects when imagining the pelvic floor muscle contraction and relaxation, and calculate the optimal EEG threshold; collect EEG information of the training subjects in the EEG training mode, and calculate the imagination state score based on the optimal EEG threshold; When the imagination state score exceeds the preset score, the pelvic floor magnetic stimulation module works to drive the pelvic floor muscles to contract.

[0022] Preferably, the process of performing the imagery training model includes: Evaluate the average reaction time of the training subjects from receiving the pelvic floor muscle contraction prompt to contracting the pelvic floor muscles, and set the instruction delay time when completing the imagery training mode; In the imagery training mode, after a delay time in which the training subject is prompted to perform pelvic floor muscle contraction, the pelvic floor magnetic stimulation module operates to drive the pelvic floor muscles to contract.

[0023] When the EEG training mode and the imagery training mode are performed in the present invention, the training process and results can be fed back to the training subject through the audio-visual feedback module, so as to obtain the specific pelvic floor muscle contraction training situation and help the training subject to perform pelvic floor muscle contraction training better and more effectively.

[0024] The present invention has the following beneficial effects: in the EEG training mode, the pelvic floor contraction intention of the training subject is actively identified through brain-computer interaction, and the operation of the pelvic floor magnetic stimulation module can be triggered after the optimal EEG threshold is reached; when the training subject actively participates in the rehabilitation training of the pelvic floor muscles, the pelvic floor muscle contraction is assisted to achieve active rehabilitation; for users who cannot obtain good EEG signals, the reaction evaluation module can be used to obtain the instruction delay time, and imagery training can be performed to complete the pelvic floor muscle contraction training; the control method of brain-computer interaction and pelvic floor magnetic stimulation are combined to achieve convenient and rapid active rehabilitation, and it is also suitable for patients at all stages to accelerate the rehabilitation process. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 is a structural diagram of the pelvic floor magnetic stimulation training system of the present invention; Figure 2 is a flow chart of the pelvic floor magnetic stimulation training method of the present invention; In the figure: 1. EEG acquisition module; 2. EEG decoding algorithm module; 3. Control module; 4. Reaction evaluation module; 5. Visual and auditory feedback module; 6. Pelvic floor magnetic stimulation module; 21. EEG evaluation unit; 22. EEG training unit. DETAILED DESCRIPTION

[0026] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0027] like Figure 1 As shown, a pelvic floor magnetic stimulation training system based on brain-computer interaction includes an EEG acquisition module 1. The EEG signal collected by the EEG acquisition module 1 is input into an EEG decoding algorithm module 2. The EEG decoding algorithm module 2 includes an EEG evaluation unit 21 and an EEG training unit 22. After receiving the signal from the EEG decoding algorithm module 2 or the signal from the reaction evaluation module 4, the control module 3 generates a control signal and transmits it to the visual and auditory feedback module 5 and the pelvic floor magnetic stimulation module 6.

[0028] It should be noted that the present invention combines the control mode of the brain-computer interface with the pelvic floor magnetic stimulation, and actively identifies the intention of the training subject to contract the pelvic floor muscles when the training subject wants to contract the pelvic floor muscles. When the training subject wants to contract the pelvic floor muscles, the pelvic floor magnetic stimulation module is used to assist in the contraction of the pelvic floor muscles, thereby changing the existing passive stimulation training mode to an active training stimulation mode, thereby achieving convenient and rapid active rehabilitation; and because the EEG decoding algorithm module 2 can perform targeted EEG signal analysis for different training subjects and generate corresponding control methods, it can be applied to different training subjects in different periods, thereby accelerating the rehabilitation process.

[0029] As a specific embodiment, the EEG decoding algorithm module 2 includes: The EEG evaluation unit 21 calculates the optimal EEG threshold based on the EEG information of the training subject when imagining the pelvic floor muscle contraction state and relaxation state; The EEG training unit 22 scores the training subject's pelvic floor muscle contraction imaginary state based on the optimal EEG threshold. When the score exceeds a preset value, the control module controls the pelvic floor magnetic stimulation module to operate.

[0030] It should be noted that the EEG decoding algorithm module receives the EEG information of the training subject, decodes the training subject's movement intention of contracting and relaxing the pelvic floor muscles in real time, and then outputs the user's movement intention state in real time; the control module can control the specific work of the pelvic floor magnetic stimulation module according to the movement intention state output by the EEG decoding algorithm module.

[0031] It is worth noting that the EEG decoding algorithm module performs signal preprocessing through common EEG signal analysis and processing methods such as fast Fourier transform, Butterworth filter, and Chebyshev filter.

[0032] Furthermore, the reaction evaluation module 4 calculates the average reaction time of the training subject by taking the difference between the time of prompting the training subject to contract the pelvic floor muscles and the feedback time of the training subject contracting the pelvic floor muscles, and uses this as the instruction delay time of the imagery training mode.

[0033] Specifically, the system prompts the training subject to contract the pelvic floor muscles through images / voices, and the system records the time t1 when the prompt is issued; after receiving the signal, the training subject simultaneously attempts to contract the pelvic floor muscles and presses a button, and the system records the training subject's action time t2. The time difference between t1 and t2 is the training subject's reaction time for one pelvic floor contraction, Δt1 = t2 - t1. The evaluation process performs multiple reaction time tests, and the average of the multiple times is the training subject's average reaction time Δt = (Δt1 + Δt2 + ... Δt n ) / n, the system records the average reaction time of the training subjects and uses it as the instruction delay time for subsequent imagery training mode.

[0034] It should be noted that the reaction evaluation module is used to collect the training subject's reaction time after receiving the pelvic floor muscle contraction prompt, such as pressing a button / mouse / joystick, etc., and transmit the signal back to the reaction evaluation module. The evaluation system issues the pelvic floor muscle contraction instruction to the user's execution of the corresponding contraction action. The reaction time is used to set the instruction delay time in the subsequent imagery training mode.

[0035] It is worth noting that the reaction evaluation module can be a handheld button connected to the system. When the button is pressed, it will send a signal to the system. The system receives the time when the button is pressed in real time as the feedback time for the training subject to contract the pelvic floor muscles.

[0036] Furthermore, the system also includes an EEG acquisition module 1 , a control module 3 , an audiovisual feedback module 5 and a pelvic floor magnetic stimulation module 6 .

[0037] The EEG acquisition module 1 includes an EEG cap (including electrodes) and an EEG digital-to-analog conversion module. This module converts collected EEG signals into digital signals and transmits them to the system. The number of electrodes on the EEG cap can range from 2, 4, 8, or 16 leads, with 8 being the most common. The electrodes can be dry, semi-dry, or wet, with wet electrodes being the most common. The EEG digital-to-analog conversion module utilizes conventional EEG technology to convert analog signals into digital signals for transmission, while also reducing signal noise and improving the signal-to-noise ratio.

[0038] Control module 3 receives information from EEG acquisition module 1, EEG decoding algorithm module 2, and response assessment module 3. After processing, the collected information is output as a control signal to control the operation of pelvic floor magnetic stimulation module 5 and visual and auditory feedback module 5, providing feedback to the training subject. The control module can be a computer host or control motherboard, and can run operating systems such as Windows, Linux, and Android.

[0039] The audio-visual feedback module 5 includes a display, speakers (headphones), VR, etc.; the visual feedback includes images of pelvic floor contraction or relaxation, model examples, expert guidance, etc. provided by a computer, and can also be movements presented using virtual reality technology; the auditory feedback includes movement voice guidance provided by the system, start and stop prompts, success or failure feedback, etc.

[0040] The pelvic floor magnetic stimulation module 6 uses electromagnetic pulses to act on the pelvic floor neuromuscular tissue, generating nerve impulses to the peripheral central nervous system, causing the muscles it controls to contract. After treatment, it has the effects of tightening the pelvic floor muscles, relieving pelvic pain, inhibiting bladder detrusor contraction, and promoting blood circulation. An existing conventional pelvic floor magnetic stimulator can be selected to generate electromagnetic pulses to act on the pelvic floor neuromuscular tissue, generating nerve impulses to the peripheral central nervous system, causing the muscles it controls to contract for rehabilitation training.

[0041] As an optional embodiment, the EEG threshold calculation model is stored in the EEG evaluation unit 21; Obtaining EEG frequency band characteristic data of different frequency bands according to the collected EEG signals; Based on the EEG data collected over a period of time, different EEG frequency bands are distinguished and then pre-processed, including 50Hz power frequency notch filtering and 0.5-40Hz bandpass filtering, to obtain the pre-processed data. (δ / θ / α / β / μ) , calculate the EEG frequency band feature data F of each EEG frequency band (δ / θ / a / β / μ) ,The EEG frequency band feature data is the result of calculating the average value, standard deviation, power ,spectral density or asymmetry coefficient of the preprocessed EEG data.

[0042] Extract the imaginary state feature F when the training subject imagines the pelvic floor muscle contraction image and the resting state characteristics F during relaxation rest ; Substitute it into the EEG threshold calculation model to calculate the optimal EEG threshold T for each EEG frequency band (δ / θ / α / β / μ) .

[0043] It should be noted that the EEG evaluation unit decodes the most suitable optimal EEG threshold by collecting EEG information of the training subjects when they are imagining the pelvic floor muscle contraction state and relaxation state to distinguish whether the training subjects are in the imagination state or the relaxation state, and establishes an imagination model for use by the EEG training unit.

[0044] It is worth noting that the evaluation process uses the characteristic data F of each frequency band (δ / θ / α / β / μ)The system displays the training subjects and operators in real time in the form of a broken line graph. When the system prompts the training subjects to imagine the pelvic floor muscle contraction, the training subjects will imagine it. When the system prompts them to relax, the training subjects will try to remain in a resting state. After the evaluation is completed, the system will calculate the optimal EEG threshold T based on the differences in the EEG characteristics of the training subjects in the imagination state and the relaxation state. (δ / θ / α / β / μ) .

[0045] Specifically, the EEG threshold calculation model is: Where T is the optimal EEG threshold; a1 and a2 are characteristic coefficients, the sum of which is 1; F image Characteristic of the imaginary state; F rest is the rest state feature; m is the data length.

[0046] It should be noted that in the present invention, due to the different types of EEG frequency bands of EEG signals, the corresponding optimal EEG threshold needs to be calculated for each EEG frequency band, that is, a training object needs to calculate and analyze the optimal EEG thresholds of multiple EEG frequency bands and substitute them into the subsequent imagination state scoring model for training. When the EEG acquisition module collects EEG information, the signal acquisition of the imagination state or the relaxation state will continue for a certain period of time, so there will be a certain data length, so that the characteristics of different states can be obtained more completely to avoid omissions.

[0047] It is worth noting that during the EEG assessment process performed by the EEG assessment unit, the system calculates and updates the EEG feature results of the imagined state in real time, and feeds back to the training subject in the form of line graphs, energy graphs, games, etc. through the display of the audio-visual feedback module. When the system prompts the training subject to rest, the training subject remains in a resting state; when the system prompts the training subject to imagine pelvic floor muscle contraction, the training subject executes it by imagining movement or adopting different psychological strategies. The assessment ends when the time expires or the training subject believes that he has mastered the pelvic floor muscle contraction method. After the assessment is completed, the training subject's EEG characteristics and the optimal EEG threshold given by the system can be viewed. The optimal EEG threshold given by the system or the threshold can be manually adjusted based on the results. The training subject can also be recommended to undergo re-evaluation.

[0048] As an optional embodiment, an imagination state scoring model is stored in the EEG training unit 22; Obtaining EEG frequency band characteristic data of different frequency bands according to the collected EEG signals; Based on the EEG data collected during the EEG training mode, different EEG frequency bands are distinguished, and then pre-processed including 50Hz power frequency notch and 0.5-40Hz bandpass filtering to obtain the pre-processed data. (δ / θ / α / β / μ), calculate the EEG frequency band feature data F of each EEG frequency band (δ / θ / α / β / μ) ,The EEG frequency band feature data is the result of calculating the average value, standard deviation, power ,spectral density or asymmetry coefficient of the preprocessed EEG data.

[0049] The optimal EEG threshold value obtained from the EEG evaluation unit is substituted into the imagination state scoring model to calculate the imagination state score of the training subject.

[0050] It should be noted that the EEG training unit decodes the EEG characteristics of the training subject during pelvic floor muscle contraction training in real time, compares it with the optimal EEG threshold obtained by the EEG evaluation unit, and transmits the comparison result to the control module, which controls the operation of the pelvic floor magnetic stimulation module based on the comparison result.

[0051] Specifically, the imaginary state scoring model is: Where Score is the imagination state score; C is the preset auxiliary score; k0 is the score conversion coefficient; k j is the frequency band characteristic coefficient of the jth EEG frequency band, and the sum of all frequency band characteristic coefficients is 1. F j is the EEG frequency band characteristic data of the jth EEG frequency band; T j is the optimal EEG threshold of the jth EEG frequency band; q is the number of EEG frequency bands considered in the imagination state scoring model.

[0052] It should be noted that the preset auxiliary score can be set according to actual needs, and the auxiliary score can be used as a preset score to judge the imagination state score; in the imagination state scoring model, there will be certain corrections based on the different types of EEG frequency bands actually considered; generally, EEG signals include δ frequency bands, θ frequency bands, α frequency bands, β frequency bands and μ frequency bands, etc. When calculating EEG thresholds and imaginary state scores, all EEG frequency bands can be taken into consideration in the calculation process, or several frequency bands that dominate the EEG information can be taken into consideration for calculation according to the different training objects.

[0053] The delta band, with a frequency range of 0.5-4 Hz, is characterized by high-amplitude waves, typically seen in adults during deep sleep. This band typically appears in the forehead of adults and the back of the head of infants. The theta band, with a frequency range of 4-8 Hz, has been found to be more prevalent in infants and in adults and adolescents during drowsiness (or the early stages of sleep). It can also occur when the brain is idle or meditating. The alpha band, with a frequency range of 8-13 Hz, is the most prominent rhythmic brainwave, typically appearing at the back of the head and on both sides, with higher amplitude on the dominant side. Alpha band signals can be detected in EEG scans of the occipital lobe when awake individuals are relaxed or with their eyes closed. In BCI applications, a specific alpha band signal is called the mu band (8-12 Hz). This signal is present in sensorimotor areas when the subject is not moving. It decreases or disappears (controlled inhibition) when the subject is moving or imagining movement. Beta waves, with a frequency range of 13-30Hz, are low-amplitude waves that typically appear symmetrically on both sides of the brain and are most pronounced in the front of the brain, where they can be detected in the parietal and frontal lobes. When these brain waves appear, people tend to be focused, have active logical thinking, experience emotional fluctuations, and are alert or anxious.

[0054] It is worth noting that during the EEG training phase, the preset score can be a preset auxiliary score C. The trainee imagines pelvic floor muscle contraction, and the system calculates the trainee's score during the imagination process based on the imagination state score. When the calculated imagination state score exceeds the preset score, the control module issues a control command to control the pelvic floor magnetic stimulation module. The visual and auditory feedback module displays the trainee's EEG characteristics in real time, displays animations of pelvic floor muscle contraction and relaxation, and provides auditory prompts to guide the trainee in better pelvic floor training.

[0055] like Figure 2 As shown, the pelvic floor magnetic stimulation training system based on the present invention corresponds to a pelvic floor magnetic stimulation training method based on brain-computer interaction, including: The EEG training mode evaluates the training subject's imagination state score when imagining the pelvic floor muscle contraction or relaxation. When the imagination state score exceeds the preset score, the control module controls the pelvic floor magnetic stimulation module to operate; In the imagery training mode, after the pelvic floor muscle contraction prompt is given, the control module waits for the instruction delay time and then controls the pelvic floor magnetic stimulation module to work.

[0056] The process of conducting EEG training mode includes: Collect EEG information of the training subjects when imagining the pelvic floor muscle contraction and relaxation, and calculate the optimal EEG threshold; collect EEG information of the training subjects in the EEG training mode, and calculate the imagination state score based on the optimal EEG threshold; When the imagination state score exceeds the preset score, the pelvic floor magnetic stimulation module works to drive the pelvic floor muscles to contract.

[0057] The process of conducting the imagery training model includes: Evaluate the average reaction time of the training subjects from receiving the pelvic floor muscle contraction prompt to contracting the pelvic floor muscles, and set the instruction delay time when completing the imagery training mode; In the imagery training mode, after a delay time in which the training subject is prompted to perform pelvic floor muscle contraction, the pelvic floor magnetic stimulation module operates to drive the pelvic floor muscles to contract.

[0058] It should be noted that the control module is equipped with two different active training modes, the main one is the EEG training mode, which actively identifies the training subject's intention to contract the pelvic floor muscles. When the training subject wants to contract the pelvic floor muscles, the pelvic floor magnetic stimulation module is used to assist the contraction of the pelvic floor muscles, thereby performing pelvic floor muscle contraction training; at the same time, if a good EEG signal cannot be obtained, the imagery training mode can be entered to perform pelvic floor muscle contraction training. Whether the received EEG signal is good can be judged based on the integrity of the EEG signal or whether there is a signal interruption.

[0059] It is worth noting that in the EEG training mode, when the training subject imagines the pelvic floor muscles contracting, the control module controls the operation of the magnetic stimulation module and the audio-visual feedback module when the EEG imagination state score Score calculated by the EEG decoding algorithm module exceeds the preset score. In the imagery training mode, the system periodically prompts the training subject to imagine or try to contract the pelvic floor muscles. The control module starts the pelvic floor magnetic stimulation module after the instruction delay time Δt after the system prompts. After the stimulation, it rests for a few seconds. The system prompts again and stimulates after the instruction delay time Δt. The cycle continues until the training time ends. When performing the EEG training mode and the imagery training mode, the training process and results can be fed back to the training subject through the audio-visual feedback module, so as to obtain the specific pelvic floor muscle contraction training situation and help the training subject to perform pelvic floor muscle contraction training better and more effectively.

[0060] In a specific embodiment of the operation of the pelvic floor magnetic stimulation system of the present invention, an EEG training mode and an imagery training mode are included.

[0061] In EEG training mode, the therapist puts on the EEG cap for the patient, who then sits comfortably on a pelvic floor magnetic stimulation chair. The therapist activates the system's EEG assessment unit, which prompts the user to attempt to contract the pelvic floor and rest. The patient makes the corresponding attempts according to the prompts. After completion, the system displays an assessment report and recommends the optimal EEG threshold. The therapist sets the optimal EEG threshold in the EEG training unit based on the patient's assessment and activates the EEG training mode. After training begins, the patient actively imagines contracting or relaxing the pelvic floor muscles. When the imagined state score exceeds the preset value, the pelvic floor magnetic stimulation module activates, driving the pelvic floor muscles to contract, while receiving real-time feedback from the display and speakers of the audio-visual feedback module. The system can also record the patient's training records and assessment results, making it easier for doctors to better select treatment options and conduct rehabilitation for their patients.

[0062] In the imagery training mode, the patient does not need to wear an EEG cap and can sit comfortably on a pelvic floor magnetic stimulation chair; the therapist activates the reaction assessment module. During the assessment, the system prompts the user to imagine / try to contract the pelvic floor muscles. The user presses a button while imagining / trying to contract the pelvic floor muscles. After repeated tests, the system records the user's average reaction time Δt and ends the reaction assessment. The therapist then activates the imagery training mode. The system prompts the patient to imagine the contraction of the pelvic floor muscles. After an average reaction time of Δt seconds after the prompt, the system automatically triggers the pelvic floor magnetic stimulation module to contract the pelvic floor muscles, while obtaining real-time feedback from the display and speakers of the audio-visual module. The system can also record the patient's training records, evaluation results, etc., to facilitate doctors in making better plan selections and rehabilitation for patients.

[0063] The above embodiments are further elaborations and illustrations of the present invention for ease of understanding, and are not intended to limit the present invention in any way. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A pelvic floor magnetic stimulation training system based on brain-computer interaction, characterized in that: It includes an EEG acquisition module, and the EEG signals collected by the EEG acquisition module are input into the EEG decoding algorithm module, and the EEG decoding algorithm module includes an EEG evaluation unit and an EEG training unit; after the control module receives the signal from the EEG decoding algorithm module or the signal from the reaction evaluation module, it generates a control signal and transmits it to the visual and auditory feedback module and the pelvic floor magnetic stimulation module.

2. A pelvic floor magnetic stimulation training system based on brain-computer interaction according to claim 1, characterized in that: In the EEG decoding algorithm module: The EEG evaluation unit calculates the optimal EEG threshold based on the EEG information of the training subject when he / she imagines the pelvic floor muscle contraction and relaxation; The EEG training unit scores the training subject's imagined state of pelvic floor muscle contraction based on the optimal EEG threshold. When the score exceeds the preset value, the control module controls the pelvic floor magnetic stimulation module to operate.

3. A pelvic floor magnetic stimulation training system based on brain-computer interaction according to claim 1 or 2, characterized in that: The reaction evaluation module calculates the average reaction time of the training subject by taking the difference between the time of prompting the pelvic floor muscle contraction and the feedback time of the training subject contracting the pelvic floor muscle, and uses it as the instruction delay time of the imagery training mode.

4. A pelvic floor magnetic stimulation training system based on brain-computer interaction according to claim 1 or 2, characterized in that: The EEG evaluation unit stores an EEG threshold calculation model; Obtaining EEG frequency band feature data of different frequency bands based on the collected EEG signals, and extracting the characteristics of the training subjects' imaginary state when they imagine pelvic floor muscle contraction and the characteristics of their resting state when they relax; Substitute it into the EEG threshold calculation model to calculate the optimal EEG threshold for each EEG frequency band.

5. A pelvic floor magnetic stimulation training system based on brain-computer interaction according to claim 1 or 2, characterized in that: The EEG training unit stores an imagination state scoring model; The EEG frequency band characteristic data of different frequency bands are obtained according to the collected EEG signals, and are substituted into the imagination state scoring model together with the optimal EEG threshold obtained in the EEG evaluation unit to calculate the imagination state score of the training object.

6. The pelvic floor magnetic stimulation training system based on brain-computer interaction according to claim 4, characterized in that: The EEG threshold calculation model is: Where T is the optimal EEG threshold; a1 and a2 are characteristic coefficients, the sum of which is 1; F image Characteristic of the imaginary state; F rest is the rest state feature; m is the data length.

7. The pelvic floor magnetic stimulation training system based on brain-computer interaction according to claim 5, characterized in that: The imagination state scoring model is: Where Score is the imagination state score; C is the preset auxiliary score; k0 is the score conversion coefficient; k j is the frequency band characteristic coefficient of the jth EEG frequency band, and the sum of all frequency band characteristic coefficients is 1; F j is the EEG frequency band feature data of the jth EEG frequency band; T j is the optimal EEG threshold of the jth EEG frequency band; q is the number of EEG frequency bands considered in the imagination state scoring model.

8. A pelvic floor magnetic stimulation training method based on brain-computer interaction, applicable to the pelvic floor magnetic stimulation training system according to any one of claims 1 to 7, characterized in that: include: The EEG training mode evaluates the training subject's imagination state score when imagining the pelvic floor muscle contraction or relaxation. When the imagination state score exceeds the preset score, the control module controls the pelvic floor magnetic stimulation module to operate; In the imagery training mode, after the pelvic floor muscle contraction prompt is given, the control module waits for the instruction delay time and then controls the pelvic floor magnetic stimulation module to work.

9. The pelvic floor magnetic stimulation training method based on brain-computer interaction according to claim 8, characterized in that: The process of conducting EEG training mode includes: Collect EEG information of the training subjects when they imagine the contraction and relaxation of pelvic floor muscles, and calculate the optimal EEG threshold; Collect the EEG information of the training subjects during the EEG training mode, and calculate the imagination state score based on the optimal EEG threshold; When the imagination state score exceeds the preset score, the pelvic floor magnetic stimulation module works to drive the pelvic floor muscles to contract.

10. The pelvic floor magnetic stimulation training method based on brain-computer interaction according to claim 8, characterized in that: The process of conducting the imagery training model includes: Evaluate the average reaction time of the training subjects from receiving the pelvic floor muscle contraction prompt to contracting the pelvic floor muscles, and set the instruction delay time when completing the imagery training mode; In the imagery training mode, after a delay time in which the training subject is prompted to perform pelvic floor muscle contraction, the pelvic floor magnetic stimulation module operates to drive the pelvic floor muscles to contract.

Images