Magnetic control pulse type pelvic floor nerve muscle stimulator and parameter optimization system thereof

Abstract

The invention relates to the technical field of medical instruments, in particular to a magnetic control pulse type pelvic floor nerve muscle stimulator and a parameter optimization system thereof. Comprising a host, a dynamic layered pulse waveform generation technology module, a multi-target reinforcement learning parameter optimization module, a multi-modal biofeedback fusion and VR enhancement module, and a flexible wearable stimulation array and ultrasonic targeting module. The system has the advantages that real-time data and deep learning are achieved, and personalized precise treatment is achieved, in the actual use process, the dynamic layered pulse waveform generation technology module predicts the requirements of muscles in the next stage through a deep learning model on the basis of pelvic floor electromyographic signals, pressure data and infrared thermal imaging which are collected in real time, and the treatment accuracy is improved. Dynamically generating a composite pulse waveform; the multi-objective reinforcement learning parameter optimization module constructs a'state-action-reward 'reinforcement learning framework, and takes patient treatment feedback, muscle force improvement, pain score and treatment compliance as reward signals.

Description

Technical Field

[0001] This invention relates to the field of medical device technology, specifically to a magnetically controlled pulsed pelvic floor neuromuscular stimulator and its parameter optimization system. Background Technology

[0002] Pelvic floor dysfunction seriously affects patients' quality of life. Traditional treatments have certain limitations, and drug prevention is limited by the high risk of bleeding. Current mainstream mechanical prevention measures, such as graded compression stockings, intermittent pneumatic compression pumps, and plantar venous compression pumps, have problems such as large contact areas leading to patient discomfort and affecting daily activities.

[0003] Pelvic floor magnetic stimulation, as a novel treatment method, has advantages such as being non-invasive, providing deep stimulation, and covering a wide range. However, passive treatment alone has its limitations. While the combination of magnetic and electrical stimulation addresses some pain points, there is still room for improvement in areas such as active training and feedback, and personalized treatment.

[0004] Existing pelvic floor magnetic stimulation devices have three major limitations: Limited stimulation modes: Traditional devices mostly use square waves with fixed frequency and pulse width, which cannot dynamically adapt to the real-time state of the pelvic floor muscles. Parameter optimization relies on experience: doctors need to manually adjust parameters, which lacks individualized adaptation capabilities, resulting in long treatment cycles; The feedback mechanism is one-sided: relying solely on electromyographic signals or pressure feedback cannot fully reflect the functional status of the pelvic floor, such as blood flow and temperature.

[0005] Therefore, there is an urgent need for a magnetically controlled pulsed pelvic floor neuromuscular stimulator and its parameter optimization system to solve the above problems. Summary of the Invention

[0006] The purpose of this invention is to provide a magnetically controlled pulsed pelvic floor neuromuscular stimulator and its parameter optimization system, which has the advantages of real-time data and deep learning, enabling personalized and precise treatment, and solves the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a magnetically controlled pulsed pelvic floor neuromuscular stimulator and its parameter optimization system, comprising: a host, a dynamic layered pulse waveform generation technology module, a multi-target reinforcement learning parameter optimization module, a multimodal biofeedback fusion and VR enhancement module, and a flexible wearable stimulation array and ultrasound targeting module.

[0008] The main unit integrates a pulse magnetic generator, an ultrasonic generator, and a drive module.

[0009] The dynamic layered pulse waveform generation technology module includes a multi-band composite pulse module, a fatigue adaptive adjustment module, and a temperature feedback compensation module.

[0010] Multi-band composite pulse module, superimposing low-frequency and high-frequency pulses.

[0011] The fatigue-adaptive adjustment module automatically increases or decreases the intensity of stimulation.

[0012] The temperature feedback compensation module monitors the temperature of the treatment area using an infrared sensor.

[0013] The parameter optimization module for multi-objective reinforcement learning includes a multi-objective optimization algorithm module, a dynamic policy adjustment module, and a digital twin prediction module.

[0014] The multi-objective optimization algorithm module optimizes three objectives simultaneously.

[0015] The dynamic strategy adjustment module fine-tunes parameters based on patient feedback.

[0016] The digital twin prediction module uses basic patient data and treatment history data to build a digital twin model.

[0017] The multimodal biofeedback fusion and VR enhancement module includes a multi-sensor fusion feedback module, a VR gamification feedback module, and an abdominal muscle monitoring and correction module.

[0018] The VR gamified feedback module allows patients to control a virtual character by contracting their pelvic floor muscles, thus training their pelvic floor muscles.

[0019] The abdominal muscle monitoring and correction module is equipped with abdominal muscle electrodes to monitor abnormal abdominal muscle contractions in real time during treatment.

[0020] The flexible wearable stimulation array and ultrasound-targeted module includes a flexible electronic skin array module, an ultrasound-targeted stimulation module, and a dynamic body position adaptation module.

[0021] The flexible electronic skin array module includes a substrate material, electrode material, and a wearing method.

[0022] The ultrasound-targeted stimulation module uses focused ultrasound to focus energy onto the deep pelvic floor muscles.

[0023] The dynamic body position adaptation module automatically adjusts the position of the stimulation array and the ultrasound focal point according to the body position.

[0024] Furthermore, as a preferred embodiment of the present invention, in the multi-band composite pulse module, low-frequency stimulation of deep nerves and high-frequency activation of superficial muscle fibers.

[0025] Furthermore, as a preferred embodiment of the present invention, in the multi-objective optimization algorithm module, the three objectives refer to muscle strength recovery, urinary control ability, and comfort.

[0026] Furthermore, as a preferred embodiment of the present invention, the basic data in the digital twin prediction module includes age, delivery room, and pelvic floor muscle score.

[0027] Furthermore, as a preferred embodiment of the present invention, the multi-sensor fusion feedback module includes a pressure sensor and a temperature sensor.

[0028] Furthermore, as a preferred embodiment of the present invention, the substrate material is PDMS with a thickness of 0.5 mm, which can be bent up to 180°.

[0029] Furthermore, as a preferred embodiment of the present invention, the electrode material is a liquid metal, a gallium-indium alloy, which has better conductivity than traditional metal electrodes and poses no risk of allergies.

[0030] Furthermore, as a preferred embodiment of the present invention, the wearing method adopts an underwear-style design, allowing patients to receive treatment during daily activities.

[0031] Furthermore, as a preferred embodiment of the present invention, the driving module is electrically connected to the dynamic hierarchical pulse waveform generation technology module, the multi-target reinforcement learning parameter optimization module, the multimodal biofeedback fusion and VR enhancement module, the flexible wearable stimulation array, and the ultrasound targeting module.

[0032] The present invention discloses a method for using a magnetically controlled pulsed pelvic floor neuromuscular stimulator and its parameter optimization system, comprising the following steps: Step 1: Connect the main power supply, turn on the pulse magnetic generator, ultrasound generator, and drive module, connect the flexible electronic skin array module and ultrasound probe, check their electrical connections, infrared sensor temperature threshold 41℃, the ultrasound generator generates high-frequency mechanical vibration waves, and concentrates energy on the target tissue through focusing technology, producing thermal and mechanical effects, the pulse magnetic generator induces electric field and current in the pelvic floor tissue, directly stimulates the pelvic floor nerves and muscles, triggering passive muscle contraction, improving muscle strength and contraction coordination, and improving vaginal laxity and uterine prolapse. The pulse magnetic generator improves muscle contraction function through nerve modulation, and the ultrasound generator promotes tissue repair through thermal and mechanical effects. The combination of the two achieves the dual goal of "functional reconstruction + structural repair". The drive module controls the activation of the dynamic layered pulse waveform generation technology module, the multi-target reinforcement learning parameter optimization module, the multimodal biofeedback fusion and VR enhancement module, and the flexible wearable stimulation array and ultrasound targeting module, extending the ultrasound probe to the patient's perineum; Step Two: The multi-band composite pulse module superimposes low-frequency and high-frequency pulses according to the patient's condition and treatment needs. The low-frequency (1-10Hz) pulses penetrate to the deep pelvic floor muscles, activating nerve axon depolarization, while the high-frequency (50-100Hz) pulses activate superficial muscle fibers. By adjusting the pulse frequency (adjustable from 20-100Hz) and intensity (0-100mA), a composite stimulation of "deep penetration of low frequency + superficial application of high frequency" is achieved. The fatigue-adaptive adjustment module sets the initial stimulation intensity based on the patient's muscle strength grade (I-V) (e.g., an initial intensity of 30mA for a grade I muscle strength patient). A) During treatment, this module automatically increases or decreases the stimulation intensity according to the patient's muscle fatigue level to ensure the safety and effectiveness of the treatment. It monitors the muscle fatigue index (MFI) in real time through a surface electromyography sensor. When the MFI > 60%, it automatically reduces the intensity by 10%-20%. The temperature feedback compensation module monitors the temperature of the treatment area through an infrared sensor. When the temperature exceeds 41°C, it will trigger the following operations: reduce the pulse frequency to a safe range (e.g., from 80Hz to 50Hz), and connect an external intelligent liquid cooling system to control the equipment temperature below 35°C. Step 3: The multi-objective optimization algorithm module simultaneously optimizes three objectives: muscle strength recovery, urinary control, and comfort. Based on individual patient differences and treatment stages, it sets muscle strength recovery (weight 40%), urinary control (weight 35%), and comfort (weight 25%) within a parameter space (frequency 20-100Hz, pulse width 0.1-500ms, intensity 0-100mA). During treatment, the dynamic strategy adjustment module collects subjective patient scores (1-10 points) and objective data (sEMG amplitude, treatment duration) through a VR gamified feedback module. Based on these feedback parameters, the treatment strategy is fine-tuned: when the pain score is ≥7, the intensity is reduced by 15%; when treatment compliance is <80%, game reward points are increased. A digital twin model is constructed based on the patient's baseline data (age, BMI, delivery history) and historical treatment records, establishing an LSTM neural network prediction model to predict the muscle strength improvement after 8 treatments (error ±5%). Step 4: The multi-sensor fusion feedback module synchronously collects pelvic floor pressure and electromyographic signals through pressure sensors, monitors temperature distribution through temperature sensors, and combines ultrasound Doppler to detect blood flow velocity, comprehensively monitoring the patient's physiological changes during treatment and providing more accurate feedback information. The VR gamification feedback module designs games, with Mode 1 and Mode 2. Mode 1 involves contracting pelvic floor muscles to control a spaceship to avoid meteorites. Mode 2 combines Kegel training and virtual posture guidance, selecting appropriate VR games based on the patient's age and interests, and dynamically adjusting the game speed (30-120 bpm) and target threshold (pressure maintenance time 2-10 seconds) according to the patient's muscle strength level. The abdominal muscle monitoring and correction module places Ag / AgCl electrodes on both sides of the rectus abdominis muscle with a sampling rate of 1000 Hz. When the abdominal muscle sEMG amplitude exceeds 200% of the pelvic floor muscle amplitude, the module promptly prompts the patient with sound and images to correct incorrect force application, ensuring that the pelvic floor muscles receive proper training. Step 5: The main unit emits pulsed magnetic and ultrasonic signals according to the set parameters, which are applied to the pelvic floor muscles through a flexible wearable stimulation array. The flexible electronic skin array module uses PDMS (0.5mm thick), a base material that can be bent up to 180°, and liquid gallium-indium alloy, an electrode material with better conductivity than traditional metal electrodes and no risk of allergies. The underwear-like structure fits the area from the pubic symphysis to the coccyx, with an electrode contact area of ​​10cm². 2 The impedance is <500Ω to ensure good contact between the electrodes and the pelvic floor muscles; the ultrasound-targeted stimulation module adjusts the parameters of the focused ultrasound to a center frequency of 1MHz, a focal length of 5cm, and a sound intensity of 1.5W / cm². 2 This technology precisely focuses energy onto the deep pelvic floor muscles, improving the targetedness and effectiveness of treatment. During treatment, the dynamic positioning module automatically adjusts the stimulation array position and ultrasound focus point according to the patient's position changes, such as sitting upright, supine, or prone, ensuring that the treatment is always in the optimal state. Furthermore, the ultrasound probe can generate high-frequency mechanical vibration waves, which concentrate energy on the target tissue through focusing technology. This helps to stimulate the pelvic floor muscles and nerves, promoting blood circulation and tissue repair.

[0033] Beneficial effects: The technical solution of this application has the following technical effects: This invention has the advantages of real-time data and deep learning, enabling personalized and precise treatment. In actual use, the dynamic layered pulse waveform generation technology module, based on real-time acquired pelvic floor electromyographic signals, pressure data, and infrared thermal imaging, predicts the next stage of muscle demand through a deep learning model and dynamically generates composite pulse waveforms; the multi-objective reinforcement learning parameter optimization module constructs a "state-action-reward" reinforcement learning framework, using patient treatment feedback, muscle strength improvement, pain score, and treatment compliance as reward signals to optimize stimulation parameters, frequency, pulse width, and intensity in real time; the multimodal biofeedback fusion and VR enhancement module integrates pressure sensors and temperature sensors, generating more accurate treatment feedback through multimodal data, and combining VR technology to transform pelvic floor muscle contraction into actions in a virtual scene, improving patient participation; the flexible wearable stimulation array and ultrasound targeting module adopt flexible electronic skin technology to achieve non-contact deep stimulation, ensuring treatment precision. Attached Figure Description

[0034] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a system block diagram of the present invention; Figure 2 This is a block diagram of the main unit of the present invention; Figure 3 This is a block diagram of the dynamic layered pulse waveform generation technology module of the present invention; Figure 4 This is a block diagram of the parameter optimization module for multi-objective reinforcement learning in this invention; Figure 5 This is a block diagram of the multimodal biofeedback fusion and VR enhancement module of the present invention; Figure 6 This is a block diagram of the flexible wearable stimulation array and ultrasound targeting module of the present invention. Detailed Implementation

[0035] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. To better understand the technical content of the present invention, specific embodiments are provided and described in conjunction with the accompanying drawings. Various aspects of the present invention are described in this disclosure with reference to the accompanying drawings, which illustrate numerous illustrative embodiments. It should be understood that the various concepts and embodiments described above, as well as those described in more detail below, can be implemented in any of many ways. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0036] As attached Figure 1 To be continued Figure 6 As shown: This embodiment provides a magnetically controlled pulsed pelvic floor neuromuscular stimulator and its parameter optimization system, including: a host, a dynamic layered pulse waveform generation technology module, a multi-target reinforcement learning parameter optimization module, a multimodal biofeedback fusion and VR enhancement module, and a flexible wearable stimulation array and ultrasound targeting module.

[0037] The main unit integrates a pulse magnetic generator, an ultrasonic generator, and a drive module.

[0038] The dynamic layered pulse waveform generation technology module includes a multi-band composite pulse module, a fatigue adaptive adjustment module, and a temperature feedback compensation module.

[0039] Multi-band composite pulse module, superimposing low-frequency and high-frequency pulses.

[0040] The fatigue-adaptive adjustment module automatically increases or decreases the intensity of stimulation.

[0041] The temperature feedback compensation module monitors the temperature of the treatment area using an infrared sensor.

[0042] The parameter optimization module for multi-objective reinforcement learning includes a multi-objective optimization algorithm module, a dynamic policy adjustment module, and a digital twin prediction module.

[0043] The multi-objective optimization algorithm module optimizes three objectives simultaneously.

[0044] The dynamic strategy adjustment module fine-tunes parameters based on patient feedback.

[0045] The digital twin prediction module uses basic patient data and treatment history data to build a digital twin model.

[0046] The multimodal biofeedback fusion and VR enhancement module includes a multi-sensor fusion feedback module, a VR gamification feedback module, and an abdominal muscle monitoring and correction module.

[0047] The VR gamified feedback module allows patients to control a virtual character by contracting their pelvic floor muscles, thus training their pelvic floor muscles.

[0048] The abdominal muscle monitoring and correction module is equipped with abdominal muscle electrodes to monitor abnormal abdominal muscle contractions in real time during treatment.

[0049] The flexible wearable stimulation array and ultrasound-targeted module includes a flexible electronic skin array module, an ultrasound-targeted stimulation module, and a dynamic body position adaptation module.

[0050] The flexible electronic skin array module includes a substrate material, electrode material, and a wearing method.

[0051] The ultrasound-targeted stimulation module uses focused ultrasound to focus energy onto the deep pelvic floor muscles.

[0052] The dynamic body position adaptation module automatically adjusts the position of the stimulation array and the ultrasound focal point according to the body position.

[0053] Specifically, in the multi-band composite pulse module, low-frequency stimulation targets deep nerves, while high-frequency stimulation activates superficial muscle fibers.

[0054] Specifically, in the multi-objective optimization algorithm module, the three objectives are muscle strength recovery, urinary control, and comfort.

[0055] Specifically, the basic data in the digital twin prediction module includes age, delivery room, and pelvic floor muscle score.

[0056] Specifically, the multi-sensor fusion feedback module includes pressure sensors and temperature sensors.

[0057] Specifically, the base material is PDMS, 0.5mm thick, and can be bent up to 180°.

[0058] Specifically, the electrode material uses liquid metal, a gallium-indium alloy, which has better conductivity than traditional metal electrodes and poses no risk of allergies.

[0059] Specifically, the treatment is designed like underwear, allowing patients to receive treatment during their daily activities.

[0060] Specifically, the driving module is electrically connected to the dynamic hierarchical pulse waveform generation technology module, the multi-target reinforcement learning parameter optimization module, the multimodal biofeedback fusion and VR enhancement module, the flexible wearable stimulation array, and the ultrasound targeting module.

[0061] In this invention, the method of using the magnetically controlled pulsed pelvic floor neuromuscular stimulator and its parameter optimization system includes the following steps: Step 1: Connect the main power supply, turn on the pulse magnetic generator, ultrasound generator, and drive module, connect the flexible electronic skin array module and ultrasound probe, check their electrical connections, infrared sensor temperature threshold 41℃, the ultrasound generator generates high-frequency mechanical vibration waves, and concentrates energy on the target tissue through focusing technology, producing thermal and mechanical effects, the pulse magnetic generator induces electric field and current in the pelvic floor tissue, directly stimulates the pelvic floor nerves and muscles, triggering passive muscle contraction, improving muscle strength and contraction coordination, and improving vaginal laxity and uterine prolapse. The pulse magnetic generator improves muscle contraction function through nerve modulation, and the ultrasound generator promotes tissue repair through thermal and mechanical effects. The combination of the two achieves the dual goal of "functional reconstruction + structural repair". The drive module controls the activation of the dynamic layered pulse waveform generation technology module, the multi-target reinforcement learning parameter optimization module, the multimodal biofeedback fusion and VR enhancement module, and the flexible wearable stimulation array and ultrasound targeting module, extending the ultrasound probe to the patient's perineum; Step Two: The multi-band composite pulse module superimposes low-frequency and high-frequency pulses according to the patient's condition and treatment needs. The low-frequency (1-10Hz) pulses penetrate to the deep pelvic floor muscles, activating nerve axon depolarization, while the high-frequency (50-100Hz) pulses activate superficial muscle fibers. By adjusting the pulse frequency (adjustable from 20-100Hz) and intensity (0-100mA), a composite stimulation of "deep penetration of low frequency + superficial application of high frequency" is achieved. The fatigue-adaptive adjustment module sets the initial stimulation intensity based on the patient's muscle strength grade (I-V) (e.g., an initial intensity of 30mA for a grade I muscle strength patient). A) During treatment, this module automatically increases or decreases the stimulation intensity according to the patient's muscle fatigue level to ensure the safety and effectiveness of the treatment. It monitors the muscle fatigue index (MFI) in real time through a surface electromyography sensor. When the MFI > 60%, it automatically reduces the intensity by 10%-20%. The temperature feedback compensation module monitors the temperature of the treatment area through an infrared sensor. When the temperature exceeds 41°C, it will trigger the following operations: reduce the pulse frequency to a safe range (e.g., from 80Hz to 50Hz), and connect an external intelligent liquid cooling system to control the equipment temperature below 35°C. Step 3: The multi-objective optimization algorithm module simultaneously optimizes three objectives: muscle strength recovery, urinary control, and comfort. Based on individual patient differences and treatment stages, it sets muscle strength recovery (weight 40%), urinary control (weight 35%), and comfort (weight 25%) within a parameter space (frequency 20-100Hz, pulse width 0.1-500ms, intensity 0-100mA). During treatment, the dynamic strategy adjustment module collects subjective patient scores (1-10 points) and objective data (sEMG amplitude, treatment duration) through a VR gamified feedback module. Based on these feedback parameters, the treatment strategy is fine-tuned: when the pain score is ≥7, the intensity is reduced by 15%; when treatment compliance is <80%, game reward points are increased. A digital twin model is constructed based on the patient's baseline data (age, BMI, delivery history) and historical treatment records, establishing an LSTM neural network prediction model to predict the muscle strength improvement after 8 treatments (error ±5%). Step 4: The multi-sensor fusion feedback module synchronously collects pelvic floor pressure and electromyographic signals through pressure sensors, monitors temperature distribution through temperature sensors, and combines ultrasound Doppler to detect blood flow velocity, comprehensively monitoring the patient's physiological changes during treatment and providing more accurate feedback information. The VR gamification feedback module designs games, with Mode 1 and Mode 2. Mode 1 involves contracting pelvic floor muscles to control a spaceship to avoid meteorites. Mode 2 combines Kegel training and virtual posture guidance, selecting appropriate VR games based on the patient's age and interests, and dynamically adjusting the game speed (30-120 bpm) and target threshold (pressure maintenance time 2-10 seconds) according to the patient's muscle strength level. The abdominal muscle monitoring and correction module places Ag / AgCl electrodes on both sides of the rectus abdominis muscle with a sampling rate of 1000 Hz. When the abdominal muscle sEMG amplitude exceeds 200% of the pelvic floor muscle amplitude, the module promptly prompts the patient with sound and images to correct incorrect force application, ensuring that the pelvic floor muscles receive proper training. Step 5: The main unit emits pulsed magnetic and ultrasonic signals according to the set parameters, which are applied to the pelvic floor muscles through a flexible wearable stimulation array. The flexible electronic skin array module uses PDMS (0.5mm thick), a base material that can be bent up to 180°, and liquid gallium-indium alloy, an electrode material with better conductivity than traditional metal electrodes and no risk of allergies. The underwear-like structure fits the area from the pubic symphysis to the coccyx, with an electrode contact area of ​​10cm². 2 The impedance is <500Ω to ensure good contact between the electrodes and the pelvic floor muscles; the ultrasound-targeted stimulation module adjusts the parameters of the focused ultrasound to a center frequency of 1MHz, a focal length of 5cm, and a sound intensity of 1.5W / cm². 2 This technology precisely focuses energy onto the deep pelvic floor muscles, improving the targetedness and effectiveness of treatment. During treatment, the dynamic positioning module automatically adjusts the stimulation array position and ultrasound focus point according to the patient's position changes, such as sitting upright, supine, or prone, ensuring that the treatment is always in the optimal state. Furthermore, the ultrasound probe can generate high-frequency mechanical vibration waves, which concentrate energy on the target tissue through focusing technology. This helps to stimulate the pelvic floor muscles and nerves, promoting blood circulation and tissue repair.

[0062] It should be noted that in this paper, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0063] While the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the invention. Those skilled in the art can make various modifications and refinements without departing from the spirit and scope of the invention. Therefore, the scope of protection of the present invention shall be determined by the claims.

Claims

1. A magnetically controlled pulsed pelvic floor neuromuscular stimulator and its parameter optimization system, characterized in that: include: The system includes a host computer, a dynamic hierarchical pulse waveform generation technology module, a parameter optimization module for multi-target reinforcement learning, a multimodal biofeedback fusion and VR enhancement module, and a flexible wearable stimulation array and ultrasound targeting module. The main unit integrates a pulse magnetic generator, an ultrasonic generator, and a drive module; The dynamic layered pulse waveform generation technology module includes a multi-band composite pulse module, a fatigue adaptive adjustment module, and a temperature feedback compensation module. Multi-band composite pulse module, superimposing low-frequency and high-frequency pulses; The fatigue-adaptive adjustment module automatically increases or decreases the intensity of stimulation. The temperature feedback compensation module monitors the temperature of the treatment area using an infrared sensor; The parameter optimization module for multi-objective reinforcement learning includes a multi-objective optimization algorithm module, a dynamic policy adjustment module, and a digital twin prediction module; A multi-objective optimization algorithm module that optimizes three objectives simultaneously; The dynamic strategy adjustment module fine-tunes parameters based on patient feedback. The digital twin prediction module uses basic patient data and treatment history data to build a digital twin model; The multimodal biofeedback fusion and VR enhancement module includes a multi-sensor fusion feedback module, a VR gamification feedback module, and an abdominal muscle monitoring and correction module; The VR gamified feedback module allows patients to control a virtual character by contracting their pelvic floor muscles, thus training their pelvic floor muscles. The abdominal muscle monitoring and correction module is equipped with abdominal muscle electrodes to monitor abnormal abdominal muscle contractions in real time during treatment. The flexible wearable stimulation array and ultrasound-targeted module includes a flexible electronic skin array module, an ultrasound-targeted stimulation module, and a dynamic body position adaptation module. The flexible electronic skin array module includes a substrate material, electrode material, and a wearing method; The ultrasound-targeted stimulation module uses focused ultrasound to focus energy onto the deep pelvic floor muscles. The dynamic body position adaptation module automatically adjusts the position of the stimulation array and the ultrasound focal point according to the body position.

2. The magnetically controlled pulsed pelvic floor neuromuscular stimulator and its parameter optimization system according to claim 1, characterized in that: In the multi-band composite pulse module, low-frequency stimulation of deep nerves and high-frequency activation of superficial muscle fibers.

3. The magnetically controlled pulsed pelvic floor neuromuscular stimulator and its parameter optimization system according to claim 1, characterized in that: In the multi-objective optimization algorithm module, the three objectives are muscle strength recovery, urinary control ability, and comfort.

4. The magnetically controlled pulsed pelvic floor neuromuscular stimulator and its parameter optimization system according to claim 1, characterized in that: The basic data in the digital twin prediction module includes age, delivery room, and pelvic floor muscle score.

5. The magnetically controlled pulsed pelvic floor neuromuscular stimulator and its parameter optimization system according to claim 1, characterized in that: The multi-sensor fusion feedback module includes pressure sensors and temperature sensors.

6. The magnetically controlled pulsed pelvic floor neuromuscular stimulator and its parameter optimization system according to claim 1, characterized in that: The base material is PDMS, 0.5mm thick, and can be bent up to 180°.

7. The magnetically controlled pulsed pelvic floor neuromuscular stimulator and its parameter optimization system according to claim 1, characterized in that: The electrode material is made of liquid metal, gallium-indium alloy, which has better conductivity than traditional metal electrodes and poses no risk of allergies.

8. The magnetically controlled pulsed pelvic floor neuromuscular stimulator and its parameter optimization system according to claim 1, characterized in that: The device is designed like underwear, allowing patients to receive treatment during their daily activities.

9. The magnetically controlled pulsed pelvic floor neuromuscular stimulator and its parameter optimization system according to claim 1, characterized in that: The driving module is electrically connected to the dynamic hierarchical pulse waveform generation technology module, the parameter optimization module for multi-target reinforcement learning, the multimodal biofeedback fusion and VR enhancement module, the flexible wearable stimulation array, and the ultrasound targeting module.

10. The method of using a magnetically controlled pulsed pelvic floor neuromuscular stimulator and its parameter optimization system, characterized in that: The method includes the following steps: Step 1: Connect the main power supply, turn on the pulse magnetic generator, ultrasound generator, and drive module, connect the flexible electronic skin array module and ultrasound probe, check their electrical connections, infrared sensor temperature threshold 41℃, the ultrasound generator generates high-frequency mechanical vibration waves, and concentrates energy on the target tissue through focusing technology, producing thermal and mechanical effects, the pulse magnetic generator induces electric field and current in the pelvic floor tissue, directly stimulates the pelvic floor nerves and muscles, triggers passive muscle contraction, improves muscle strength and contraction coordination, and improves vaginal laxity and uterine prolapse. The pulse magnetic generator improves muscle contraction function through nerve modulation, and the ultrasound generator promotes tissue repair through thermal and mechanical effects. The combination of the two achieves the dual goal of "functional reconstruction + structural repair". The drive module controls the activation of the dynamic layered pulse waveform generation technology module, the multi-target reinforcement learning parameter optimization module, the multimodal biofeedback fusion and VR enhancement module, and the flexible wearable stimulation array and ultrasound targeting module, extending the ultrasound probe to the patient's perineum; Step Two: The multi-band composite pulse module superimposes low-frequency and high-frequency pulses according to the patient's condition and treatment needs. The low-frequency (1-10Hz) pulses penetrate to the deep pelvic floor muscles, activating nerve axon depolarization, while the high-frequency (50-100Hz) pulses activate superficial muscle fibers. By adjusting the pulse frequency (adjustable from 20-100Hz) and intensity (0-100mA), a composite stimulation of "deep penetration of low frequency + superficial application of high frequency" is achieved. The fatigue-adaptive adjustment module sets the initial stimulation intensity, setting an initial value according to the patient's muscle strength grade (I-V) (e.g., an initial intensity of 30mA for a grade I muscle strength patient). A) During treatment, this module automatically increases or decreases the stimulation intensity according to the patient's muscle fatigue level to ensure the safety and effectiveness of the treatment. It monitors the muscle fatigue index (MFI) in real time through a surface electromyography sensor. When the MFI > 60%, it automatically reduces the intensity by 10%-20%. The temperature feedback compensation module monitors the temperature of the treatment area through an infrared sensor. When the temperature exceeds 41°C, it will trigger the following operations: reduce the pulse frequency to a safe range (e.g., from 80Hz to 50Hz), and connect an external intelligent liquid cooling system to control the equipment temperature below 35°C. Step 3: The multi-objective optimization algorithm module simultaneously optimizes three objectives: muscle strength recovery, urinary control, and comfort. Based on individual patient differences and treatment stages, it sets muscle strength recovery (weight 40%), urinary control (weight 35%), and comfort (weight 25%) within a parameter space (frequency 20-100Hz, pulse width 0.1-500ms, intensity 0-100mA). During treatment, the dynamic strategy adjustment module collects subjective patient scores (1-10 points) and objective data (sEMG amplitude, treatment duration) through a VR gamified feedback module. Based on these feedback parameters, the treatment strategy is fine-tuned: when the pain score is ≥7, the intensity is reduced by 15%; when treatment compliance is <80%, game reward points are increased. A digital twin model is constructed based on the patient's baseline data (age, BMI, delivery history) and historical treatment records, establishing an LSTM neural network prediction model to predict the muscle strength improvement after 8 treatments (error ±5%). Step 4: The multi-sensor fusion feedback module synchronously collects pelvic floor pressure and electromyographic signals through pressure sensors, monitors temperature distribution through temperature sensors, and combines ultrasound Doppler to detect blood flow velocity, comprehensively monitoring the patient's physiological changes during treatment and providing more accurate feedback information. The VR gamification feedback module designs games, with Mode 1 and Mode 2. Mode 1 involves contracting pelvic floor muscles to control a spaceship to avoid meteorites. Mode 2 combines Kegel training and virtual posture guidance, selecting appropriate VR games based on the patient's age and interests, and dynamically adjusting the game speed (30-120 bpm) and target threshold (pressure maintenance time 2-10 seconds) according to the patient's muscle strength level. The abdominal muscle monitoring and correction module places Ag / AgCl electrodes on both sides of the rectus abdominis muscle with a sampling rate of 1000 Hz. When the abdominal muscle sEMG amplitude exceeds 200% of the pelvic floor muscle amplitude, the module promptly prompts the patient with sound and images to correct incorrect force application, ensuring that the pelvic floor muscles receive proper training. Step 5: The main unit emits pulsed magnetic and ultrasonic signals according to the set parameters, which are applied to the pelvic floor muscles through a flexible wearable stimulation array. The flexible electronic skin array module uses PDMS (0.5mm thick), a base material that can be bent up to 180°, and liquid gallium-indium alloy, an electrode material with better conductivity than traditional metal electrodes and no risk of allergies. The underwear-like structure fits the area from the pubic symphysis to the coccyx, with an electrode contact area of ​​10cm². 2 The impedance is <500Ω to ensure good contact between the electrodes and the pelvic floor muscles; the ultrasound-targeted stimulation module adjusts the parameters of the focused ultrasound to a center frequency of 1MHz, a focal length of 5cm, and a sound intensity of 1.5W / cm². 2 This technology precisely focuses energy onto the deep pelvic floor muscles, improving the targetedness and effectiveness of treatment. During treatment, the dynamic positioning module automatically adjusts the stimulation array position and ultrasound focus point according to the patient's position changes, such as sitting upright, supine, or prone, ensuring that the treatment is always in the optimal state. Furthermore, the ultrasound probe can generate high-frequency mechanical vibration waves, which concentrate energy on the target tissue through focusing technology. This helps to stimulate the pelvic floor muscles and nerves, promoting blood circulation and tissue repair.

Images