A breath-synchronized dual-mode non-invasive intelligent electrical stimulation therapy system

Abstract

The present application relates to the field of medical equipment, in particular to a breath-synchronous dual-mode noninvasive intelligent electric stimulation treatment system, comprising a breath sensing subsystem, an electric stimulation subsystem, a synchronous control subsystem, an AI intelligent adjustment subsystem, a multi-modal comfort subsystem, a wireless power supply subsystem and a data management subsystem; the present application constructs seven modules of the breath sensing subsystem, the electric stimulation subsystem, the synchronous control subsystem, the AI intelligent adjustment subsystem, the multi-modal comfort subsystem, the wireless power supply subsystem and the data management subsystem, forms a complete treatment closed loop through clear electrical connection and communication link, realizes full-process automatic operation from physiological signal collection, stimulation timing control, parameter dynamic adjustment to energy supply and data management, and solves the technical problems of scattered functions and poor module collaboration of existing equipment.

Description

Technical Field

[0001] This invention belongs to the field of medical device technology, and in particular relates to a respiratory-synchronous dual-mode non-invasive intelligent electrical stimulation therapy system. Background Technology

[0002] Obstructive sleep apnea (OSA) is a common sleep-disordered breathing disorder, primarily characterized by repeated upper airway collapse during sleep, leading to intermittent hypoxia. Long-term OSA can induce hypertension, arrhythmias, and cognitive decline. Furthermore, swallowing dysfunction is common in stroke patients, increasing the risk of aspiration pneumonia. Currently, the treatment for OSA is continuous positive airway pressure (CPAP), but its adherence is poor; for swallowing dysfunction, traditional surface electrical stimulation is often used, lacking precise synchronization with physiological movements.

[0003] The existing technology has the following main technical problems, which limit the therapeutic effect: Synchronization lag and false triggering: Traditional respiratory sensing relies on a single sensor, and the signal delay is usually more than 200ms. It cannot accurately intervene in the early stage of the inspiratory phase, resulting in a misalignment between stimulation and respiratory rhythm, and even aggravating airway obstruction.

[0004] Rigid stimulation patterns: Existing devices mostly use open-loop stimulation with fixed frequency and intensity, without introducing an electromyographic (EMG) feedback mechanism. When the stimulation intensity approaches the muscle tetanic contraction threshold, it can easily trigger patient awakening or jaw pain, leading to treatment interruption.

[0005] Low wearing comfort: Traditional electrodes are mostly made of hard conductive gel, which can easily cause skin allergies if applied for a long time; and most devices use wired power supply, and the tangled wires seriously affect sleep quality.

[0006] Limited functionality: The existing system is designed for a single disease and cannot switch between "sleep breathing support" and "wakeful swallowing rehabilitation" modes on the same hardware platform. Summary of the Invention

[0007] This invention aims to at least solve one of the technical problems existing in the prior art. To this end, this invention proposes a respiratory-synchronized dual-mode non-invasive intelligent electrical stimulation therapy system.

[0008] To achieve the above objectives, the present invention provides the following technical solution: a respiratory synchronous dual-mode non-invasive intelligent electrical stimulation therapy system, comprising a respiratory sensing subsystem, an electrical stimulation subsystem, a synchronous control subsystem, an AI intelligent adjustment subsystem, a multimodal comfort subsystem, a wireless power supply system, and a data management subsystem; The respiratory sensing subsystem is used to collect the patient's respiratory signals; The electrical stimulation subsystem is used to output biphasic square wave electrical stimulation pulses; The synchronization control subsystem is electrically connected to the respiratory sensing subsystem and the electrical stimulation subsystem, respectively, and is used to receive respiratory signals and synchronously control the pulse output and stop of the electrical stimulation subsystem. The AI ​​intelligent adjustment subsystem is electrically connected to the synchronization control subsystem and the electrical stimulation subsystem, respectively, and is used to collect electromyographic signals and adjust the output parameters of the electrical stimulation subsystem. The multimodal comfort subsystem is integrated into the electrical stimulation subsystem to improve wearing comfort. The wireless power supply system is connected to the power supply of each subsystem to provide working power to each module of the system. The data management subsystem is communicatively connected to the respiratory sensing subsystem and the AI ​​intelligent adjustment subsystem, respectively, and is used for the transmission, storage and management of treatment data.

[0009] In one embodiment, the respiratory sensing subsystem includes a chest / abdomen fitted respiratory sensing band, on which an accelerometer, an impedance sensor, and an airflow sensor are integrated. The accelerometer, impedance sensor, and airflow sensor are all electrically connected to the signal processing unit of the respiratory sensing subsystem. The signal processing unit identifies the inspiratory phase, expiratory phase, and abnormal respiratory events through a multi-sensor fusion algorithm, and outputs the identified respiratory phase signals to the synchronization control subsystem.

[0010] In one embodiment, the electrical stimulation subsystem includes a flexible gel patch and a dual-electrode array; the dual-electrode array is disposed on the surface of the flexible gel patch, the dual-electrode array is electrically connected to a pulse generation circuit, and the control terminal of the pulse generation circuit is electrically connected to a synchronization control subsystem for outputting biphasic square wave pulses to the dual-electrode array according to control commands.

[0011] In one embodiment, the parameters of the biphasic square wave pulse output by the pulse generating circuit are: frequency 20-50Hz, pulse width 200-300μs, and current intensity 10-30mA.

[0012] In one embodiment, the synchronization control subsystem has a built-in millisecond-level signal processing program; after receiving the inspiratory phase start signal from the respiratory sensing subsystem, the synchronization control subsystem generates a trigger command within a 0-50ms time window and sends it to the electrical stimulation subsystem, controlling the duration of a single pulse output to be 1-2s; after receiving the expiratory phase start signal from the respiratory sensing subsystem, the synchronization control subsystem generates a stop command and sends it to the electrical stimulation subsystem; the synchronization control subsystem also receives electromyographic signals fed back from the AI ​​intelligent adjustment subsystem, which are used to limit the output current intensity of the electrical stimulation to a range below the critical current value that triggers muscle tetanic contraction.

[0013] In one embodiment, the AI ​​intelligent adjustment subsystem includes an EMG electromyography sensor and a deep learning model processing unit; the EMG electromyography sensor is disposed on a flexible gel patch of the electrical stimulation subsystem for real-time acquisition of electromyographic signals of the genioglossus muscle and transmission to the deep learning model processing unit; the deep learning model processing unit is electrically connected to the synchronization control subsystem and the electrical stimulation subsystem respectively, for fusing respiratory signals, electromyographic signals and sleep stage data, and dynamically adjusting the frequency, pulse width and intensity of electrical stimulation.

[0014] In one embodiment, the multimodal comfort subsystem is a microcapsule fragrance module; the microcapsule fragrance module is embedded inside the flexible gel patch, and the microcapsule fragrance module uses temperature-sensitive natural essential oil microcapsules to release soothing ingredients in response to human body surface temperature.

[0015] In one embodiment, the system further includes a mode switching unit, the output of which is electrically connected to the control terminal of the electrical stimulation subsystem, for switching between an OSA treatment mode and a swallowing function improvement mode; in the swallowing function improvement mode, the electrical stimulation subsystem outputs a biphasic square wave pulse with a frequency of 30-60Hz, a pulse width of 300-500μs, and a current intensity of 15-40mA, and the stimulation timing is synchronized with the swallowing action.

[0016] In one embodiment, the wireless power supply system includes a portable power supply host and a wireless receiving coil; the portable power supply host has a built-in lithium battery and a wireless transmitting circuit, and the wireless receiving coil is integrated on a flexible gel patch of the electrical stimulation subsystem; the portable power supply host transmits electrical energy to the wireless receiving coil via magnetic resonance coupling to power various functional modules on the patch side.

[0017] In one embodiment, the data management subsystem includes a mobile APP and a cloud server; the mobile APP communicates with the respiratory sensing subsystem and the AI ​​intelligent adjustment subsystem via Bluetooth to display real-time treatment data and generate efficacy reports; the cloud server communicates with the mobile APP via a wireless network to store treatment data and issue remote parameter adjustment commands.

[0018] Compared with the prior art, the beneficial effects of the present invention are: (1) The present invention constructs seven major modules: respiratory sensing subsystem, electrical stimulation subsystem, synchronous control subsystem, AI intelligent adjustment subsystem, multimodal comfort subsystem, wireless power supply system and data management subsystem. Through clear electrical connections and communication links, a complete treatment closed loop is formed, realizing the full-process automated operation from physiological signal acquisition, stimulation timing control, parameter dynamic adjustment to energy supply and data management, solving the technical problems of fragmented functions and poor module coordination of existing equipment.

[0019] (2) This invention integrates a respiratory sensing subsystem comprising an accelerometer, an impedance sensor, and an airflow sensor, and combines this with a multi-sensor fusion algorithm to accurately identify the inspiratory and expiratory phases. Combined with the millisecond-level signal processing program of the synchronization control subsystem, electrical stimulation is triggered within a 0-50ms window after the start of the inspiratory phase, ensuring that electrical stimulation is applied only to the genioglossus muscle during the inspiratory phase. This fundamentally solves the problem of misalignment between stimulation and respiration caused by signal delay in traditional devices, significantly improving the physiological effectiveness of the treatment.

[0020] (3) This invention uses an AI intelligent adjustment subsystem to collect electromyographic signals of the genioglossus muscle in real time, and uses a deep learning model processing unit to fuse respiratory signals, electromyographic signals and sleep stage data. In particular, the synchronous control subsystem dynamically limits the output current intensity to a range below the critical current value that causes muscle tetany based on the amplitude of the electromyographic signal. This avoids pain and awakening caused by excessive current, and also prevents ineffective treatment caused by insufficient current, thus achieving true adaptive closed-loop control.

[0021] (4) By setting up a mode switching unit, this invention creatively integrates OSA treatment mode and swallowing function improvement mode on a single hardware platform. The two modes use different electrical stimulation parameters and timing logic, which enables the device to serve both sleep apnea departments and neurorehabilitation departments in the treatment of swallowing disorders, greatly expanding the product's applicability and market value.

[0022] (5) This invention uses a flexible gel patch to support a dual-electrode array, which improves skin adhesion; at the same time, it integrates a microcapsule fragrance module, which uses temperature-sensitive natural essential oil microcapsules to release soothing ingredients in response to human body temperature. This multimodal comfort design effectively alleviates patient anxiety, solves the problems of easy allergies and strong foreign body sensation caused by traditional rigid electrode patches, and significantly improves long-term wear compliance.

[0023] (6) The present invention employs a wireless power supply system using magnetic resonance coupling, which transmits electrical energy to the wireless receiving coil integrated within the patch via a portable power supply host. This technology achieves complete cable-free treatment of the treatment site, avoids wire tangling, and ensures energy transmission efficiency and treatment duration, meeting the needs of overnight treatment.

[0024] (7) The "mobile APP + cloud server" data management subsystem constructed in this invention not only realizes the visualization and local storage of treatment data, but more importantly, it supports the issuance of remote parameter adjustment commands by the cloud server. This provides a convenient follow-up method for elderly patients with limited mobility, reduces medical costs, and improves the efficiency of chronic disease management. Attached Figure Description

[0025] Figure 1 This is a schematic diagram of the respiratory synchronization dual-mode non-invasive intelligent electrical stimulation therapy system of the present invention; Figure 2 This is a flowchart of the respiratory sensing subsystem of the present invention; Figure 3 This is a flowchart of the electrical stimulation subsystem of the present invention; Figure 4 This is a flowchart of the synchronization control subsystem of the present invention; Figure 5 This is a flowchart of the AI ​​intelligent adjustment subsystem of the present invention; Figure 6 This is a flowchart of the multimodal comfort subsystem of the present invention; Figure 7 This is a flowchart of the mode switching unit of the present invention; Figure 8 This is a flowchart of the wireless power supply system of the present invention; Figure 9 This is a flowchart of the data management subsystem of the present invention; Detailed Implementation To make the above-mentioned objectives, features, and advantages of this application more apparent and understandable, the specific embodiments of this application are described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of this application. However, this application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of this application. Therefore, this application is not limited to the specific embodiments disclosed below.

[0026] Please see Figure 1-9 This embodiment provides a respiratory-synchronized dual-mode non-invasive intelligent electrical stimulation therapy system, including a respiratory sensing subsystem, an electrical stimulation subsystem, a synchronization control subsystem, an AI intelligent adjustment subsystem, a multimodal comfort subsystem, a wireless power supply system, and a data management subsystem. The respiratory sensing subsystem is used to collect the patient's respiratory signals; the electrical stimulation subsystem is used to output biphasic square wave pulses; the synchronization control subsystem is electrically connected to both the respiratory sensing subsystem and the electrical stimulation subsystem; the AI ​​intelligent adjustment subsystem is electrically connected to both the synchronization control subsystem and the electrical stimulation subsystem; the multimodal comfort subsystem is integrated into the electrical stimulation subsystem; the wireless power supply system is used to power each subsystem; and the data management subsystem is communicatively connected to the respiratory sensing subsystem and the AI ​​intelligent adjustment subsystem.

[0027] The respiratory sensing subsystem includes a chest / abdomen fitted respiratory sensing band, which integrates an accelerometer, an impedance sensor, and an airflow sensor. The respiratory sensing subsystem identifies the inspiratory phase, expiratory phase, and abnormal respiratory events through a multi-sensor fusion algorithm, with an identification error ≤50ms and a sleep apnea detection sensitivity ≥95%.

[0028] In practice, the breathing sensing band is made of medical-grade flexible elastic fabric, and its tightness can be adjusted via Velcro, allowing it to fit patients with different chest and abdominal circumferences. A triaxial MEMS accelerometer is used to detect changes in displacement and acceleration of chest and abdominal respiratory movements, capturing the mechanical characteristics of these movements. A bipolar impedance sensor collects periodic impedance fluctuations in chest and abdominal tissues, serving as the core basis for respiratory phase identification. A miniature thermal airflow sensor can be attached under the patient's nose to collect airflow signals from the mouth and nose, used for cross-validation of apnea and hypoventilation events.

[0029] The system employs a weighted fusion signal processing strategy, using impedance signals as the primary signal and accelerometer signals as a secondary signal to filter sleep-related body movement interference. It combines airflow signals to confirm abnormal breathing events and ultimately outputs the inspiratory phase start point, expiratory phase start point, and trigger signals for abnormal breathing events, providing a precise timing reference for subsequent synchronous control.

[0030] The electrical stimulation subsystem includes a flexible gel patch and a dual-electrode array disposed on the patch; the flexible gel patch is 0.3-0.5 mm thick and is applied to the patient's submandibular midline area; the dual-electrode array includes two main electrodes with an electrode diameter of 10-15 mm and an electrode spacing of 20-30 mm.

[0031] The biphasic square wave pulse output by the electrical stimulation subsystem has the following parameters: frequency 20-50Hz, pulse width 200-300μs, and current intensity 10-30mA.

[0032] In practice, the flexible gel patch base is made of medical-grade polyurethane flexible gel material, which has excellent skin adhesion and biocompatibility, conforming to the deformation of the mandibular skin and is not easily dislodged when the patient turns over or changes position at night. The patch is applied to the surface projection area of ​​the genioglossus muscle, ensuring that the electrical stimulation energy can be precisely applied to the target muscle and the medial branch of the hypoglossal nerve. The two circular main electrodes are covered with a medical conductive gel layer, which reduces skin contact impedance, makes the current distribution more uniform, and reduces local stinging sensation. The electrical stimulation output uses biphasic balanced square wave pulses, which can avoid the accumulation of skin charge and electrolytic damage caused by unidirectional current, improving the safety of long-term wear.

[0033] The synchronization control subsystem includes a millisecond-level signal processing program. The program generates a trigger command within a 0-50ms time window after receiving the inspiratory phase start signal from the respiratory sensing subsystem, controlling the electrical stimulation subsystem to output pulses, with each pulse lasting 1-2 seconds. The synchronization control subsystem also stops the electrical stimulation subsystem from outputting after receiving the expiratory phase signal from the respiratory sensing subsystem. Furthermore, the synchronization control subsystem limits the output current intensity to below the critical current value that triggers muscle tetany based on the amplitude of the electromyographic signal fed back by the AI ​​intelligent adjustment subsystem.

[0034] In practical implementation, the synchronization control subsystem employs an embedded microcontroller with a real-time operating system to ensure low-latency signal processing. The system uses a waveform inflection point recognition algorithm to locate the inspiratory phase start point of the respiratory waveform in real time. Within 0-50ms of identifying the start point, a stimulation trigger command is generated, with each stimulation lasting 1-2 seconds, fully covering the core period of the inspiratory phase. When the expiratory phase start signal is detected, a rest command is immediately generated to stop pulse output, achieving a precise synchronization mechanism of "inspiratory stimulation, expiratory rest," enabling the genioglossus muscle contraction to coordinate with inspiratory movement and maximizing the upper airway expansion effect.

[0035] Meanwhile, the synchronous control subsystem receives real-time electromyographic amplitude data of the genioglossus muscle from the AI ​​intelligent adjustment subsystem. Based on the preset critical current value for muscle tetanic contraction, it sets a dynamic upper limit for the output intensity of electrical stimulation to ensure that the stimulation intensity is always below the sleep-wake threshold, thus avoiding stimulation that could cause the patient to wake up from sleep or experience soreness or discomfort in the jaw muscles.

[0036] The AI ​​intelligent regulation subsystem includes an EMG electromyography sensor and a deep learning model processing unit. The EMG electromyography sensor is used to collect electromyographic signals of the genioglossus muscle in real time. The deep learning model processing unit integrates respiratory signals, electromyographic signals and sleep stage data to adjust the frequency, pulse width and intensity of electrical stimulation.

[0037] In practice, the EMG (electromyography) sensor is integrated around the electrodes of the electrical stimulation patch, enabling real-time acquisition of electromyographic activity signals of the genioglossus muscle, quantifying the intensity of muscle contraction and resting muscle tone. The deep learning model processing unit employs a lightweight embedded deployment, allowing inference operations to be performed locally on the device. Sleep stage data is derived from accelerometer motion characteristics and heart rate changes, distinguishing between different sleep states such as deep sleep, light sleep, and wakefulness.

[0038] During treatment, the model can dynamically adjust the electrical stimulation parameters based on the patient's real-time status: when an increased risk of sleep apnea or decreased genioglossus muscle tone is detected, the stimulation parameters are appropriately increased to enhance muscle contraction; when the patient is in a light sleep phase, has a strong electromyographic response, or exhibits increased body movement, the stimulation parameters are decreased to reduce sleep disturbance. The system also retains a manual adjustment channel, allowing patients or medical staff to manually set the upper and lower limits of the parameters via a companion terminal, balancing automated adjustment with personalized customization needs.

[0039] The multimodal comfort subsystem is a microcapsule fragrance module, which is embedded in a flexible gel patch. The microcapsule fragrance module is specifically composed of natural essential oil microcapsules, which are released in response to the patient's body temperature, with a release cycle of 7-10 days.

[0040] In practice, the microcapsule fragrance module is embedded in the reserved area of ​​the flexible gel patch, without occupying space for electrode placement. The module encapsulates thermosensitive natural essential oil microcapsules, which can be made of plant essential oils with different scents such as jasmine, peppermint, and lavender. The wall material of the microcapsules is made of thermosensitive polymer material, which can slowly rupture in response to body surface temperature, releasing the essential oil components at a uniform rate. This helps to soothe nerves, relieve pre-sleep anxiety, and further improve sleep comfort and treatment adherence.

[0041] The effective release cycle of a single fragrance module is 7-10 days, and the release intensity can be adjusted by the opening of the vents on the surface of the patch. The fragrance module adopts a replaceable design and can be replaced individually after the use cycle, without the need to replace the entire electrical stimulation patch, thus reducing long-term use costs.

[0042] The system is equipped with a mode switching unit that switches between OSA treatment mode and swallowing function improvement mode. In swallowing function improvement mode, the electrical stimulation subsystem outputs biphasic square wave pulses with a frequency of 30-60Hz, a pulse width of 300-500μs, and a current intensity of 15-40mA, and the stimulation timing is synchronized with the swallowing action.

[0043] In practice, the mode switching unit can switch modes via physical buttons on the device or a mobile app, adapting to both nighttime sleep therapy and daytime rehabilitation training scenarios.

[0044] OSA treatment mode is the default nighttime sleep mode, which uses respiratory-synchronized stimulation logic. The pulse parameters range from 20-50Hz frequency, 200-300μs pulse width, and 10-30mA current intensity. Stimulation triggered by the inspiratory phase maintains the tone of the genioglossus muscle and prevents the tongue root from falling back and the airway from collapsing.

[0045] The swallowing function improvement mode is mainly used for swallowing rehabilitation training in the daytime waking state. In this mode, the intensity of stimulation parameters is increased to match the training needs of the swallowing muscle group. The stimulation sequence can be triggered by manual button or synchronized with swallowing action detection, which strengthens the contraction ability of the genioglossus muscle and related swallowing muscle groups, improves the sensitivity of swallowing reflex, and is suitable for OSA patients with swallowing dysfunction to carry out daytime rehabilitation.

[0046] The wireless power supply system includes a portable power supply unit and a wireless receiving coil; the portable power supply unit is worn on the waist and has a built-in lithium battery; the wireless receiving coil is integrated on a flexible gel patch; the wireless power supply system adopts magnetic resonance wireless power supply technology, with a transmission distance of ≤5cm and a transmission efficiency of ≥80%; the portable power supply unit supports continuous treatment time of ≥12 hours on a single charge.

[0047] In practice, the portable power supply unit is a miniaturized wearable device that can be worn on the patient's waist via a clip. It has a built-in high-capacity rechargeable lithium battery and a wireless transmitting coil, and integrates power management and low-power control circuitry. The wireless receiving coil is integrated inside a flexible gel patch, and achieves contactless energy transmission with the power supply unit's transmitting coil through magnetic resonance coupling. The effective transmission distance is ≤5cm, and the energy transmission efficiency is ≥80%. It can maintain a stable power supply even when the patient turns over at night or makes slight changes in position, completely eliminating the constraints of wired connections and improving sleep freedom.

[0048] The system has a multi-level low-power management mechanism, with extremely low power consumption in standby and hibernation states; the portable power supply host can support continuous treatment time of ≥12 hours after a single full charge, which can fully cover more than 8 hours of sleep treatment overnight, while reserving power reserve for daytime swallowing function training.

[0049] The data management subsystem includes a mobile app and a cloud server; treatment data is transmitted to the mobile app via Bluetooth and synchronized to the cloud server via Wi-Fi; the mobile app is used to display real-time treatment data and generate efficacy reports; the cloud server is used to store data and receive remote parameter adjustment commands.

[0050] In practice, real-time treatment data such as respiratory waveforms, stimulation parameters, electromyographic data, and sleep status are transmitted to a mobile app via a Bluetooth Low Energy link. The app can intuitively display information such as real-time respiratory curves, stimulation sequence, and treatment duration, and can generate efficacy analysis reports on a daily, weekly, and monthly basis, and statistically analyze core indicators such as changes in the apnea-hypopnea index (AHI), number of awakenings, and treatment compliance.

[0051] Data collected by the mobile app can be synchronously uploaded to the cloud server via Wi-Fi for long-term storage. The cloud platform supports access from multiple terminals, including medical staff and patients. Medical staff can view patients' historical treatment data through the cloud backend and remotely adjust the range of stimulation parameters based on the recovery status of the treatment effect. Parameter adjustment instructions can be sent to the patient's device through the cloud, realizing remote treatment management and individualized plan optimization, and building a complete treatment closed loop.

[0052] System working principle In nighttime sleep therapy scenarios, the respiratory sensing subsystem collects real-time chest and abdominal breathing and airflow signals from the patient, and identifies the respiratory phase through a multi-sensor fusion algorithm; the synchronization control subsystem triggers electrical stimulation output at the start of the inspiratory phase in milliseconds, and automatically stops during the expiratory phase, achieving precise synchronization between breathing and stimulation; the AI ​​intelligent adjustment subsystem dynamically optimizes stimulation parameters by combining multi-dimensional data from electromyography, respiration, and sleep, ensuring treatment effectiveness while reducing sleep disturbances; the multimodal comfort subsystem enhances the wearing experience through aromatherapy, the wireless power supply system provides unrestricted power overnight, and the data management subsystem completes data storage, analysis, and remote management, ultimately achieving non-invasive, precise, comfortable, and personalized OSA treatment, and can switch modes to also improve swallowing function.

[0053] Example 1: Basic OSA Treatment System This embodiment employs a dual-sensor breathing band integrating an accelerometer and an impedance sensor, paired with a 4cm×4cm flexible gel patch under the chin. It features a basic AI parameter adjustment algorithm, a jasmine essential oil fragrance module, and a wireless power supply solution for the portable waist-mounted unit. In a 4-week clinical trial involving 20 patients with moderate to severe obstructive sleep apnea syndrome, the patients' nocturnal apnea-hypopnea index (AHI) decreased by an average of 45% compared to before treatment. The sleep awakening rate during treatment was 3.2%, and the subjective comfort score for patch wearing was 9.2 out of 10. No adverse reactions such as skin allergies or local muscle soreness were observed in any of the subjects.

[0054] Example 2: Dual-Mode Integrated Treatment System This embodiment employs a three-sensor fusion respiratory sensing strip, equipped with an electrical stimulation patch supporting dual-mode switching between OSA treatment and swallowing function improvement, and features a full-function AI adjustment model and a replaceable fragrance module, powered by a built-in high-capacity lithium battery. A 6-week treatment observation was conducted on 30 elderly OSA patients with dysphagia. Results showed that the patients' average AHI decreased by 52% compared to before treatment, and their sluggish swallowing reflex score improved by 60%. The device can support 12 hours of continuous operation on a single full charge, simultaneously meeting the power needs of nighttime sleep therapy and daytime swallowing training.

[0055] Example 3: Cloud-based managed treatment system This embodiment, based on the hardware of Embodiment 2, adds sleep monitoring modules for blood oxygen, heart rate, and body movement, and connects to a cloud-based data management platform, supporting remote data viewing and parameter adjustment by medical staff. A controlled application involving remote management of 50 OSA patients showed that under the remote dynamic parameter adjustment mode, the compatibility of the treatment plan with the individual patient's condition improved by 70%, and the long-term efficacy stability was improved by 40% compared to the traditional offline single-parameter adjustment method.

[0056] The above description is merely a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural transformations made using the contents of the present invention's specification and drawings under the inventive concept of the present invention, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A respiratory-synchronized dual-mode non-invasive intelligent electrical stimulation therapy system, characterized in that, It includes a respiratory sensing subsystem, an electrical stimulation subsystem, a synchronization control subsystem, an AI intelligent adjustment subsystem, a multimodal comfort subsystem, a wireless power supply system, and a data management subsystem; The respiratory sensing subsystem is used to collect the patient's respiratory signals; The electrical stimulation subsystem is used to output biphasic square wave electrical stimulation pulses; The synchronization control subsystem is electrically connected to the respiratory sensing subsystem and the electrical stimulation subsystem, respectively, and is used to receive respiratory signals and synchronously control the pulse output and stop of the electrical stimulation subsystem. The AI ​​intelligent adjustment subsystem is electrically connected to the synchronization control subsystem and the electrical stimulation subsystem, respectively, and is used to collect electromyographic signals and adjust the output parameters of the electrical stimulation subsystem. The multimodal comfort subsystem is integrated into the electrical stimulation subsystem to improve wearing comfort. The wireless power supply system is connected to the power supply of each subsystem to provide working power to each module of the system. The data management subsystem is communicatively connected to the respiratory sensing subsystem and the AI ​​intelligent adjustment subsystem, respectively, and is used for the transmission, storage and management of treatment data.

2. The respiratory-synchronized dual-mode non-invasive intelligent electrical stimulation therapy system according to claim 1, characterized in that, The respiratory sensing subsystem includes a chest / abdomen fitted respiratory sensing band, on which an accelerometer, an impedance sensor, and an airflow sensor are integrated. The accelerometer, impedance sensor, and airflow sensor are all electrically connected to the signal processing unit of the respiratory sensing subsystem. The signal processing unit identifies the inspiratory phase, expiratory phase, and abnormal respiratory events through a multi-sensor fusion algorithm, and outputs the identified respiratory phase signals to the synchronization control subsystem.

3. The respiratory-synchronized dual-mode non-invasive intelligent electrical stimulation therapy system according to claim 1, characterized in that, The electrical stimulation subsystem includes a flexible gel patch and a dual-electrode array; the dual-electrode array is disposed on the surface of the flexible gel patch, the dual-electrode array is electrically connected to a pulse generation circuit, and the control terminal of the pulse generation circuit is electrically connected to a synchronization control subsystem for outputting biphasic square wave pulses to the dual-electrode array according to control commands.

4. The respiratory-synchronized dual-mode non-invasive intelligent electrical stimulation therapy system according to claim 3, characterized in that, The parameters of the biphasic square wave pulse output by the pulse generation circuit are: frequency 20-50Hz, pulse width 200-300μs, and current intensity 10-30mA.

5. The respiratory-synchronized dual-mode non-invasive intelligent electrical stimulation therapy system according to claim 1, characterized in that, The synchronization control subsystem has a built-in millisecond-level signal processing program. After receiving the inspiratory phase start signal from the respiratory sensing subsystem, the synchronization control subsystem generates a trigger command within a 0-50ms time window and sends it to the electrical stimulation subsystem, controlling the duration of a single pulse output to be 1-2s. After receiving the expiratory phase start signal from the respiratory sensing subsystem, the synchronization control subsystem generates a stop command and sends it to the electrical stimulation subsystem. The synchronization control subsystem also receives electromyographic signals fed back from the AI ​​intelligent adjustment subsystem, which is used to limit the output current intensity of the electrical stimulation to a range below the critical current value that triggers muscle tetanic contraction.

6. The respiratory-synchronized dual-mode non-invasive intelligent electrical stimulation therapy system according to claim 3, characterized in that, The AI ​​intelligent adjustment subsystem includes an EMG electromyography sensor and a deep learning model processing unit. The EMG electromyography sensor is mounted on a flexible gel patch of the electrical stimulation subsystem to collect electromyographic signals of the genioglossus muscle in real time and transmit them to the deep learning model processing unit. The deep learning model processing unit is electrically connected to the synchronization control subsystem and the electrical stimulation subsystem, respectively, and is used to fuse respiratory signals, electromyographic signals and sleep stage data to dynamically adjust the frequency, pulse width and intensity of electrical stimulation.

7. The respiratory-synchronized dual-mode non-invasive intelligent electrical stimulation therapy system according to claim 6, characterized in that, The multimodal comfort subsystem is a microcapsule fragrance module; the microcapsule fragrance module is embedded inside the flexible gel patch, and the microcapsule fragrance module uses temperature-sensitive natural essential oil microcapsules to release soothing ingredients in response to human body surface temperature.

8. The respiratory-synchronized dual-mode non-invasive intelligent electrical stimulation therapy system according to claim 1, characterized in that, The system also includes a mode switching unit, the output of which is electrically connected to the control terminal of the electrical stimulation subsystem, for switching between OSA treatment mode and swallowing function improvement mode; in swallowing function improvement mode, the electrical stimulation subsystem outputs biphasic square wave pulses with a frequency of 30-60Hz, a pulse width of 300-500μs, and a current intensity of 15-40mA, and the stimulation timing is synchronized with the swallowing action.

9. The respiratory-synchronized dual-mode non-invasive intelligent electrical stimulation therapy system according to claim 3, characterized in that, The wireless power supply system includes a portable power supply host and a wireless receiving coil; the portable power supply host has a built-in lithium battery and a wireless transmitting circuit, and the wireless receiving coil is integrated on the flexible gel patch of the electrical stimulation subsystem; the portable power supply host transmits electrical energy to the wireless receiving coil through magnetic resonance coupling to power the various functional modules on the patch side.

10. The respiratory-synchronized dual-mode non-invasive intelligent electrical stimulation therapy system according to claim 1, characterized in that, The data management subsystem includes a mobile APP and a cloud server. The mobile APP communicates with the respiratory sensing subsystem and the AI ​​intelligent adjustment subsystem via Bluetooth to display real-time treatment data and generate efficacy reports. The cloud server communicates with the mobile APP via a wireless network to store treatment data and issue remote parameter adjustment commands.

Images