An electromyographic feedback-based electrical stimulation closed-loop control system, method and storage medium

Abstract

The application discloses an electromyographic feedback-based electric stimulation closed-loop control system, method and storage medium, which is used for neuromuscular electric stimulation treatment of sleep apnea. The system comprises an electrode module, a control circuit module and a control software module; the electrode module is used for electric stimulation output and electromyographic signal acquisition; the control circuit module is used for generating an electric stimulation signal, collecting an electromyographic signal and controlling the electrode module to switch between a stimulation mode and an acquisition mode; and the control software module is used for coordinating the stimulation and acquisition processes in a time-sharing manner and adjusting stimulation parameters based on electromyographic feedback. By setting a first isolation circuit and a second isolation circuit, patient end safety isolation and time domain isolation of the stimulation signal and the acquisition channel are realized, and signal coupling interference is reduced. The application realizes integrated integration of stimulation and electromyographic acquisition on a single miniaturized platform, improves the stability and anti-interference ability of electromyographic signal acquisition, and enhances the real-time performance and reliability of closed-loop control.

Description

Technical Field

[0001] This invention belongs to the field of biomedical engineering technology, specifically relating to an electromyographic feedback-based closed-loop control system, method, and storage medium for electrical stimulation. Background Technology

[0002] Obstructive sleep apnea (OSA) is a common sleep disorder characterized by recurrent upper airway collapse during sleep, leading to intermittent nocturnal hypoxia and disrupted sleep structure. OSA not only severely impacts patients' quality of life but is also an independent risk factor for various cardiovascular and cerebrovascular diseases, significantly increasing the risk of serious complications such as stroke and heart failure. Furthermore, studies show that OSA patients have a significantly higher incidence of motor vehicle accidents due to daytime sleepiness and cognitive decline than the general population, resulting in a heavy social disease burden.

[0003] Currently, the first-line standard treatment for OSA is non-invasive positive pressure ventilation (CPAP). However, this therapy suffers from poor long-term adherence and intolerance in some patients, and there are currently no widely applicable and effective drug treatments. Therefore, developing new and feasible treatment technologies and drugs has become a research hotspot in this field. Targeting upper airway muscle insufficiency, a key pathological mechanism of OSA, implantable hypoglossal nerve electrical stimulation technology increases the activity of upper airway dilators by stimulating the hypoglossal nerve during sleep, thereby maintaining airway patency and treating OSA. However, this technology still has limitations in its widespread clinical application, including significant surgical trauma, easy muscle fatigue, strict patient selection criteria, and significant individual differences in treatment response.

[0004] Given the crucial role of upper airway muscles in OSA and the limitations of existing monitoring technologies, non-invasive submental electrical stimulation has attracted attention as a potential therapy. Its advantage lies in its ability to stimulate muscles in a non-invasive manner, thereby lowering the treatment threshold.

[0005] In the prior art, CN114470520A discloses an electromyography (EMG) feedback electrical stimulation control system. This system acquires EMG signals through electrodes and stimulates muscles through an electrical stimulation pulse module. Simultaneously, a signal preprocessing module processes the composite signal containing the electrical stimulation pulse and the EMG signal to eliminate the influence of the electrical stimulation pulse on the EMG signal acquisition. Furthermore, the system coordinates the operation of the electrical stimulation module, sampling module, and signal processing module through a system control module to achieve a certain degree of EMG feedback control.

[0006] However, the aforementioned existing technologies still have the following technical problems: 1. Such systems typically use a multi-module cascade approach for signal acquisition and processing. They rely on pulse sampling and signal preprocessing modules to post-process and eliminate stimulation artifacts. When the amplitude of the electrical stimulation signal is high or the electrode contact state changes, the electromyographic signal may still be submerged by the stimulation pulse, resulting in a decrease in the signal-to-noise ratio of the electromyographic signal and affecting the stability and reliability of the electromyographic signal acquisition.

[0007] 2. When used in small spaces such as under the chin, the electrical stimulation and electromyography acquisition functions usually rely on separate circuits or multi-module collaboration, resulting in a relatively complex system structure and high requirements for synchronous control between modules. Under dynamic electrode-skin contact impedance changes, it is difficult to compensate for signal quality in real time, thereby reducing the overall stability of the system.

[0008] 3. Existing technologies mostly focus on post-processing and filtering of electromyographic signals, lacking a more efficient time-domain coordination mechanism between electrical stimulation output and electromyographic acquisition. This results in a limited effective sampling window for electromyographic signals in scenarios with rapid switching between stimulation and acquisition or high-frequency stimulation, further affecting the real-time performance and accuracy of electromyographic feedback control.

[0009] Therefore, existing technologies still have shortcomings in terms of anti-interference ability of electromyographic signals, signal stability under complex electrical stimulation environments, and collaborative control capabilities of small-area multifunctional electrode systems. Summary of the Invention

[0010] To address the core challenges of signal interference and impedance mismatch mentioned above, this invention creatively proposes a dual-modal electrode integration scheme and a collaborative control algorithm. This scheme is designed collaboratively at both the hardware architecture and software algorithm levels: On the hardware side, a time-division multiplexing electrode array structure and intelligent control circuit unit are employed, integrating a high-speed analog switch to achieve rapid switching between stimulation and acquisition modes, physically constructing a temporal isolation barrier to effectively isolate the direct coupling of stimulation pulses to the acquisition channel; on the algorithm side, advanced artifact elimination and dynamic impedance compensation technologies are developed to ensure rapid recovery of high-quality electromyographic signal acquisition after high-intensity stimulation. Through this integrated hardware and software design, reliable integration of stimulation and acquisition functions is ultimately achieved on a single miniaturized platform, clearing key technical obstacles for building a truly effective closed-loop treatment system.

[0011] The purpose of this invention is to overcome the shortcomings of the prior art by providing a closed-loop control system, method and storage medium for electromyographic feedback-based electrical stimulation.

[0012] The objective of this invention can be achieved through the following technical solutions: The present invention provides an electromyographic feedback-based closed-loop control system for treating sleep apnea, comprising: an electrode module, a control circuit module, and a control software module; The electrode module is used to output electrical stimulation signals and acquire electromyographic signals. The control circuit module is connected to the electrode module and the control software module respectively, and is used to generate electrical stimulation signals, acquire electromyographic signals, and control the electrode module to work in stimulation output mode or electromyographic acquisition mode. The control software module is used to control the control circuit module to output electrical stimulation and acquire electromyographic signals in a time-sharing manner, and to control the output parameters of the electrical stimulation signal according to the acquired electromyographic signals, forming a closed-loop control based on electromyographic feedback.

[0013] Furthermore, the electrode module includes multiple electrode contacts and a base layer. The multiple electrode contacts are disposed on the base layer and in contact with the human body for transmitting the electrical stimulation signal and acquiring the electromyographic signal. Under the control of the control circuit module, the multiple electrode contacts selectively operate in either a stimulation output mode or an electromyographic acquisition mode.

[0014] Furthermore, the electrode contacts include metal electrode contacts and / or conductive gel electrode contacts, and the base layer is a flexible base layer; the electrode module also includes a shielding layer and / or a biocompatible bonding layer disposed on the base layer, which are used to reduce the influence of external interference signals on electromyography signal acquisition and improve the bonding stability between the electrode module and the human body.

[0015] Furthermore, the control circuit module includes a control unit, an electrical stimulation signal generation unit, an electrophysiological signal acquisition unit, and an isolation circuit.

[0016] Furthermore, the electrical stimulation signal generation unit includes a waveform generation circuit and an output signal amplitude control circuit. The waveform generation circuit is used to generate an electrical stimulation waveform, and the output signal amplitude control circuit is used to adjust the output amplitude of the electrical stimulation signal. The electrophysiological signal acquisition unit is connected to the electrode module and is used to receive electromyographic signals; The isolation circuit includes a first isolation circuit and a second isolation circuit.

[0017] Furthermore, the first isolation circuit is disposed between the patient end and the control circuit power supply to achieve electrical isolation; the first isolation circuit includes an isolation transformer and / or an isolation power supply. The second isolation circuit is disposed between the electrical stimulation signal generation unit and the electrophysiological signal acquisition unit, and is used to isolate the electrophysiological signal acquisition unit when outputting the electrical stimulation signal; the second isolation circuit includes a switching circuit and / or a filtering circuit, the switching circuit includes an analog switch or a relay, and the filtering circuit is used to filter out interference components in the electrical stimulation signal.

[0018] Furthermore, the control software module is used to perform the following processes: The electrical stimulation signal generation unit is controlled to output an electrical stimulation signal during the first time period T1 and stop outputting the electrical stimulation signal during the second time period T2. At the same time, the electrophysiological signal acquisition unit in the control circuit module is controlled to acquire electromyographic signals during the second time period T2. The acquired electromyographic signals were preprocessed; Within each sampling period, electromyographic sampling data within a preset time T3 after the cessation of electrical stimulation are discarded to eliminate interference from electrical stimulation artifacts, wherein T3 satisfies... ; Feature extraction is performed on the preprocessed electromyographic signal to obtain electromyographic feature parameters, wherein the electromyographic feature parameters include the root mean square value and / or the mean absolute value of the electromyographic signal. Calculate the muscle activation index based on the electromyographic characteristic parameters. And determine the physiological state of the target muscle based on a threshold, where: in, , These are the weighting coefficients; This refers to the root mean square value among the electromyographic characteristic parameters. This represents the average absolute value among the electromyographic characteristic parameters; when Below the preset threshold At that time, it is determined that the target muscle is in a relaxed state or the risk of airway collapse is increased; Based on the physiological state, a control strategy signal is generated, and the output parameters of the electrical stimulation signal generation unit in the control circuit module are adjusted. These output parameters include stimulation amplitude A, stimulation duration T1, and / or acquisition time T2. When… At the same time, increasing the stimulation amplitude A or increasing T1 can enhance the activation of the target muscle.

[0019] Furthermore, the preprocessing includes denoising filtering and artifact suppression, wherein the denoising filtering includes low-pass filtering, band-pass filtering, notch filtering, or adaptive filtering.

[0020] Another aspect of the present invention provides a closed-loop control method for electrical stimulation based on electromyographic feedback, comprising the following steps: A time-division control cycle for electrical stimulation and electromyography (EMG) acquisition is established, and a single control cycle is divided into an electrical stimulation output phase T1 and an EMG acquisition phase T2. During the electrical stimulation output phase T1, the electrical stimulation signal generation unit is controlled to output an electrical stimulation signal to the electrode module. During the EMG acquisition phase T2, the electrophysiological signal acquisition unit is controlled to acquire the EMG signal of the target muscle through the electrode module. Within each control cycle, the electrode module is switched in working mode by the control circuit module, so that it alternates between electrical stimulation output mode and electromyography acquisition mode, so as to achieve temporal isolation between stimulation signal and electromyography signal. The acquired electromyographic signals are preprocessed, including noise reduction filtering and artifact suppression processing, wherein the noise reduction filtering is used to suppress environmental noise and electrical interference signals. During the electromyography (EMG) acquisition phase T2, data from the pre-defined artifact time T3 following the end of stimulation are discarded from the acquired EMG signals to obtain valid EMG signals. ; Feature extraction is performed on the effective electromyography (EMG) signals to obtain EMG feature parameters; Calculate the muscle activation index based on the electromyographic characteristic parameters. ; The muscle activation index With preset threshold When comparing, At that time, it is determined that the target muscle is in a relaxed state or there is a risk of airway collapse; A control feedback signal is generated based on the determination result, and the control feedback signal is fed back to the electrical stimulation signal generation unit to dynamically adjust the electrical stimulation parameters. The electrical stimulation parameters include stimulation amplitude A and / or stimulation duration T1 and / or electromyography acquisition time T2, wherein when... At the same time, increasing the stimulation amplitude A and / or prolonging T1 enhances the electrical stimulation response of the target muscle.

[0021] In another aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described closed-loop control method for electrical stimulation based on electromyographic feedback.

[0022] Compared with the prior art, the present invention has the following advantages: (1) Existing technologies in electromyography (EMG) feedback electrical stimulation systems typically employ a separate or cascaded structure for electrical stimulation and EMG acquisition. When applied in small spaces such as the submental region, signal coupling interference between electrical stimulation and EMG acquisition is common, especially when the amplitude of the electrical stimulation pulse is high. EMG signals are easily overwhelmed by stimulation artifacts, leading to a decrease in the signal-to-noise ratio of the acquisition channel and affecting the effective extraction and stability of EMG signals. This invention sets up an electrode module and controls it to switch between stimulation output mode and EMG acquisition mode by a control circuit module. Combined with a second isolation circuit, it achieves isolation between the electrical stimulation signal generation unit and the electrophysiological signal acquisition unit, enabling the electrical stimulation and EMG acquisition to be separated and controlled in the time domain. This forms a time-domain isolation mechanism at the physical level, effectively reducing the direct coupling effect of electrical stimulation pulses on the EMG signal acquisition channel, improving the signal-to-noise ratio and stability of EMG signal acquisition, and achieving the technical effect of stably acquiring effective EMG signals even under high-intensity electrical stimulation conditions.

[0023] (2) Existing technologies typically rely on fixed procedures or sequential processing of multiple modules during the coordinated process of electrical stimulation and electromyography (EMG) acquisition. They lack a dynamic feedback control mechanism based on EMG status, making it difficult to adjust stimulation parameters in a timely manner according to changes in muscle physiological state. This results in the inability to intervene in a timely manner during the muscle relaxation phase before respiratory events occur, thereby reducing the real-time nature and specificity of treatment. This invention extracts features from the acquired EMG signals through a control software module and calculates the muscle activation state index S based on EMG feature parameters. When S is lower than a preset threshold, the amplitude, duration, or acquisition cycle parameters of the electrical stimulation signal are dynamically adjusted, thereby achieving closed-loop control based on EMG feedback. This enables the system to intervene in the early stages when the risk of muscle relaxation or airway collapse increases, improving the predictive ability and control response speed for sleep apnea events.

[0024] (3) In the existing technology, during the acquisition of electromyographic signals, strong transient artifacts still exist after the electrical stimulation ends. This usually relies on complex back-end filtering algorithms for noise reduction. However, when stimulation is frequent or the electrode contact state changes, the duration of artifacts is difficult to accurately define, leading to contamination or loss of effective electromyographic signals and affecting the accuracy of subsequent feature analysis. This invention controls the software module to remove sampled data within a preset time T3 after each electrical stimulation stops. Combined with filtering and adaptive signal processing methods, it preprocesses the electromyographic signals, thereby achieving artifact suppression and effective data screening during the acquisition stage. This reduces reliance on complex post-processing algorithms and improves the reliability and real-time performance of electromyographic signal extraction.

[0025] (4) In the prior art, when applying electrical stimulation and electromyography (EMG) acquisition systems in small areas such as the subchin, the electrode-skin interface impedance is prone to dynamic changes due to the limited electrode contact area and variations in sweating, displacement, or pressure. This leads to unstable stimulation current output or large fluctuations in EMG signal amplitude, affecting the overall stability of the system. The present invention achieves regulation and safe isolation of the electrical stimulation output amplitude through the output signal amplitude control circuit in the control circuit module, combined with the first isolation circuit and the electrical isolation structure between the patient end and the circuit. This ensures the stability of the stimulation output even under conditions of electrode impedance changes, reduces stimulation inconsistency caused by impedance mismatch, and improves the robustness of the system in dynamic physiological environments.

[0026] (5) Existing technologies typically employ multi-module discrete structures or complex cascaded circuit designs when integrating stimulation and acquisition functions, resulting in large system size and low integration, which is not conducive to long-term wear and application in small anatomical areas such as the submental region. This also increases the complexity of system wiring and signal transmission loss. This invention, through the integrated collaborative design of electrode modules, control circuit modules, and control software modules, and by adopting a time-division multiplexing electrode structure and isolation circuit integration scheme, enables the stimulation and acquisition functions to operate collaboratively on a single miniaturized platform. This significantly reduces system structural complexity, improves integration and wearability, and enhances its practicality and stability in long-term sleep apnea monitoring and treatment scenarios. Attached Figure Description

[0027] Figure 1 This is a block diagram of the electrical stimulation closed-loop control system according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the electrode module structure according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the control circuit module structure according to an embodiment of the present invention; Figure 4 This is a flowchart illustrating the control software module of an embodiment of the present invention. Detailed Implementation

[0028] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0029] Example 1: like Figure 1 As shown, this embodiment provides an electromyography feedback-based closed-loop control system for treating sleep apnea, including: an electrode module, a control circuit module, and a control software module; like Figure 2 As shown, in one specific embodiment, the electrode module is used to output electrical stimulation signals and acquire electromyographic (EMG) signals. The electrode module is electrically connected to the control circuit module, used to transmit the electrical stimulation signals generated by the control circuit module to the target stimulation area, and to transmit the EMG signals generated by the target muscle to the control circuit module for subsequent processing. The target muscle can be the submental muscles, the genioglossus muscle, or other muscle groups related to upper airway opening. Since the amplitude of EMG signals is usually low during sleep, the electrode module simultaneously performs stimulation output and weak bioelectric signal acquisition functions to improve system integration and reduce the wearing area.

[0030] The electrode module and the control circuit module are connected via wires, circuit board traces, or flexible circuitry. When flexible circuitry or chip-based traces are used, the electrode module and control circuit module are integrated into a single structure, thereby reducing the number of connection nodes, minimizing the impact of impedance changes at the connection points on the stability of EMG acquisition, reducing the overall system size, and improving long-term wearing comfort.

[0031] The electrode module comprises multiple electrode contacts and a base layer. The electrode contacts are disposed on the surface of the base layer and in contact with human skin. The electrode contacts are used to transmit electrical stimulation signals and acquire electromyographic (EMG) signals. The electrode contacts are made of materials with good conductivity and biocompatibility, such as stainless steel, silver, silver chloride, platinum-iridium alloy, or conductive hydrogel materials. Among these, Ag / AgCl electrodes can reduce polarization effects and improve the stability of low-frequency EMG signal acquisition; Pt / Ir alloy materials have high corrosion resistance, which can improve service life under repetitive stimulation conditions.

[0032] The base layer is made of flexible materials, such as non-woven fabric, flexible polymer film, or leather, to adapt to the curved structure of the subchinal region and reduce electrode displacement caused by human movement. The base layer has a certain degree of flexibility and mechanical cushioning, which can reduce the impact of turning over or muscle movement during sleep on the electrode contact state and improve the stability of the electrode-skin interface.

[0033] The DC impedance of the electrode contacts is controlled below 10KΩ. Lower contact impedance reduces energy loss during stimulation output and lowers thermal noise and power frequency interference during EMG acquisition, thereby improving the signal-to-noise ratio of the EMG signal. When the electrode impedance is too high, the stimulation current is prone to fluctuation, and the acquisition channel is easily affected by environmental noise. Therefore, limiting the electrode impedance helps improve the stability of the closed-loop control system.

[0034] The electrode module also features a biocompatible bonding layer made of conductive gel, hydrogel, or medical pressure-sensitive adhesive material. This layer enhances the adhesion between the electrode module and the skin, reducing contact impedance fluctuations caused by air gaps. The conductive gel forms a stable ion-conducting pathway between the electrode and the skin, thereby improving the quality of weak electromyographic signal acquisition.

[0035] To reduce the impact of external electromagnetic environment on electromyography (EMG) acquisition, the electrode module is also equipped with a shielding layer, which is located inside the base layer or on the side away from the skin. The shielding layer is made of graphene conductive film, metal fabric, or conductive coating material to shield against power frequency interference and high-frequency electromagnetic interference in the environment. Since the subchinar region is close to respiratory monitoring equipment and wireless communication equipment, external electromagnetic interference can easily couple into the acquisition channel. The shielding layer can improve the stability of EMG acquisition signals.

[0036] In one embodiment, the electrode module includes 2N electrode contacts, where N is an integer greater than or equal to 1. The control circuit module configures the 2N electrode contacts as either N pairs of stimulation output electrodes or N pairs of electromyography (EMG) acquisition electrodes according to the operating state. During different operating phases, the same electrode contacts can switch between stimulation output and EMG acquisition, thereby reducing the number of electrodes and the system size.

[0037] In another embodiment, the electrode module includes N functional electrode contacts and M reference electrode contacts, where M is less than or equal to N. The functional electrode contacts are used to output stimulation signals or acquire electromyographic signals, while the reference electrode contacts are used to establish reference potentials or suppress common-mode interference. The reference electrode contacts can reduce the impact of human body common-mode voltage on the acquisition channel, improving differential acquisition accuracy.

[0038] In another embodiment, the electrode module includes a first electrode contact group for stimulation output and a second electrode contact group for electromyography (EMG) acquisition, the first and second electrode contact groups being independently configured. This independent electrode structure reduces the probability of stimulation signals directly coupling to the acquisition channel, thereby mitigating the impact of stimulation artifacts on EMG acquisition and improving the stability of EMG detection during stimulation.

[0039] like Figure 3As shown, in one specific embodiment, the control circuit module includes a control unit, an electrical stimulation signal generation unit, an electrophysiological signal acquisition unit, and an isolation circuit. The control unit coordinates the working timing of each functional module and controls the output of electrical stimulation signals, acquisition of electromyographic signals, and switching of electrode working modes according to the control instructions output by the control software module. The control unit can be implemented using a microcontroller, digital signal processor (DSP), field-programmable array (FPGA), or other programmable control chips. The programmable control structure allows for dynamic adjustment of stimulation parameters based on the electromyographic state of different patients, improving the flexibility and adaptability of closed-loop control.

[0040] The electrical stimulation signal generation unit is used to generate electrical stimulation signals that act on the target muscle. The unit includes a waveform generation circuit, an output signal amplitude control circuit, and an isolation circuit. The waveform generation circuit generates a preset stimulation waveform, which may include a square wave, a biphasic pulse wave, a sine wave, a triangle wave, or a combination of these waveforms. When using digital waveform synthesis, the waveform generation circuit outputs the target waveform data through a digital-to-analog converter; when using filtering, it obtains the target frequency band waveform by filtering the input square wave. The biphasic pulse wave can reduce the charge accumulation effect, thereby reducing the irritation and damage to the skin and muscle tissue under long-term stimulation conditions.

[0041] The output signal amplitude control circuit adjusts the amplitude of the stimulation signal output to suit the muscle activation thresholds of different patients. Based on parameters output by the control software module, the circuit adjusts the stimulation current or voltage and can dynamically adjust it in conjunction with impedance detection results. The impedance detection circuit corrects the stimulation output parameters in real time by detecting changes in impedance at the electrode-skin interface. When the electrode contact impedance increases, the output signal amplitude control circuit automatically increases the output voltage or adjusts the constant current output range to maintain a stable stimulation current; when the contact impedance decreases, it reduces the output amplitude to avoid excessive stimulation that could cause patient discomfort.

[0042] In one embodiment, the output signal amplitude control circuit employs a constant current source structure to maintain a constant output stimulation current, thereby reducing the impact of electrode impedance variations on the stimulation effect. In another embodiment, the output signal amplitude control circuit employs a constant voltage source structure to achieve stimulation control by stabilizing the output voltage. The constant current output method is beneficial for improving the consistency of stimulation effects under different skin conditions.

[0043] The isolation circuit includes a first isolation circuit and a second isolation circuit. The first isolation circuit is located between the patient end and the power supply end of the control circuit to achieve electrical isolation between the patient and the external power supply, preventing dangerous voltage from being transmitted to the human body in case of power failure. The first isolation circuit is implemented using an isolation transformer, an isolation power supply module, or an opto-isolation structure. When an isolation transformer is used, voltage adjustment is achieved by setting the turns ratio of the primary and secondary coils to meet different stimulation output requirements.

[0044] The second isolation circuit is located between the electrical stimulation signal generation unit and the electrophysiological signal acquisition unit to reduce the impact of the stimulation signal on the electromyography acquisition channel. Since the amplitude of the stimulation pulse is much higher than the amplitude of the electromyography signal, if it is directly coupled to the acquisition circuit, it can easily lead to saturation or even damage of the acquisition channel. Therefore, the second isolation circuit is used to isolate the stimulation and acquisition.

[0045] In one implementation, the second isolation circuit is implemented using a switching circuit. The switching circuit includes an analog switch, a relay, or a semiconductor switching device. During the electrical stimulation output phase, the control unit controls the switching circuit to disconnect the acquisition path, preventing the stimulation signal from entering the electrophysiological signal acquisition unit; after stimulation ends, it controls the switching circuit to reopen, allowing the electromyographic signal to enter the acquisition channel. This high-speed switching method creates temporal isolation between stimulation and acquisition, thereby reducing the impact of stimulation artifacts on electromyographic acquisition.

[0046] In another embodiment, the second isolation circuit is implemented using a filter circuit. The filter circuit attenuates the high-energy frequency bands in the stimulation pulse based on the difference in frequency range between the electrical stimulation signal and the electromyographic signal. The filter circuit includes a low-pass filter, a band-pass filter, a notch filter, or a combination of filter structures. The input impedance of the filter circuit is set to be more than 10 times the impedance at the patient's end to reduce the influence of the acquisition channel on the stimulation output waveform.

[0047] To improve the input impedance of the filter circuit, a voltage follower circuit is incorporated at its input. This voltage follower circuit utilizes a high-input-impedance operational amplifier to reduce the load effect of the preceding signal. To prevent excessively high common-mode voltage from the stimulus pulse from damaging subsequent components, a high-resistance resistor network or a voltage limiting protection circuit is installed at the input of the follower circuit. This voltage limiting protection circuit includes transient suppression diodes, clamping diodes, or Zener devices to limit the peak voltage at the input.

[0048] The electrophysiological signal acquisition unit is used to receive and process electromyographic (EMG) signals. The unit includes a preamplifier circuit, a filter circuit, and an analog-to-digital converter circuit. The preamplifier circuit uses a high common-mode rejection ratio (CMRR) instrumentation amplifier to amplify weak EMG signals and suppress power frequency common-mode interference. The analog-to-digital converter circuit converts analog EMG signals into digital signals for subsequent analysis and processing by the control software module.

[0049] In one embodiment, the electrophysiological signal acquisition unit is connected to the electrode module using independent electrode contacts, with the stimulation output electrode and the electromyography (EMG) acquisition electrode being set independently of each other. This independent electrode structure reduces the probability of stimulation signals directly coupling into the acquisition channel, thereby improving the stability of EMG acquisition. In this structure, a first isolation circuit and a second isolation circuit are also provided between the acquisition channel and the patient end to ensure human safety and the anti-interference capability of the acquisition channel.

[0050] like Figure 4 As shown, in one specific embodiment, the control software module is used to complete system workflow control, electrical stimulation parameter control, electrophysiological signal sampling control, and electromyographic signal processing. The control software module runs in a microcontroller, digital signal processor (DSP), or embedded processing platform and communicates with the control circuit module to realize stimulation output, signal acquisition, and closed-loop feedback control.

[0051] The control software module employs a time-sharing control method to coordinate the electrical stimulation output and electromyography (EMG) signal acquisition process. Within one control cycle, the electrical stimulation signal generation unit outputs an electrical stimulation signal during the first time period T1, then stops outputting, and acquires EMG signals during the second time period T2. This process then repeats into the next stimulation cycle, forming a periodic alternating working mode. T1 represents the stimulation duration, and T2 represents the EMG acquisition time. Both can be fixed parameters or dynamically adjusted based on EMG feedback results, respiratory status, stimulation response level, or user-defined parameters.

[0052] The reason for adopting a time-sharing working mode is that the amplitude of electrical stimulation pulses is usually much higher than that of electromyographic (EMG) signals. If stimulation and acquisition are performed simultaneously, the stimulation pulses can easily cause the acquisition circuit to saturate, thus affecting the accuracy of EMG signal detection. By separating the stimulation output phase and the EMG acquisition phase in time, the impact of stimulation artifacts on the acquisition channel can be effectively reduced, and the effective sampling rate of EMG signals can be improved.

[0053] When the second isolation circuit adopts a switching circuit structure, the control software module synchronously controls the switching circuit state. During the electrical stimulation output phase, the control software module controls the switching circuit to disconnect, closing the acquisition path between the electrophysiological signal acquisition unit and the electrode module, preventing high-amplitude stimulation signals from entering the acquisition circuit. During the electromyography (EMG) acquisition phase, the control software module controls the switching circuit to conduct, allowing EMG signals to enter the electrophysiological signal acquisition unit. Through the synchronous control of the switching state and the stimulation output state, temporal isolation between stimulation and acquisition can be achieved, improving the anti-interference capability of the acquisition channel.

[0054] When the second isolation circuit employs a filter circuit structure, the stimulation output and electromyography (EMG) acquisition can operate synchronously. The stimulation signal generation unit and the electrophysiological signal acquisition unit can operate continuously and simultaneously, or they can overlap for certain time periods depending on different operational needs. Because the filter circuit can attenuate the high-energy frequency components in the stimulation pulse, EMG signals can still be acquired under synchronous operation conditions.

[0055] The control software module processes the acquired electromyographic (EMG) signals to determine the physiological state of the target muscle and generate feedback control signals. The EMG signal processing includes signal filtering, artifact removal, feature extraction, and state recognition.

[0056] Electromyography (EMG) signal filtering is used to reduce the impact of environmental noise, power supply frequency interference, and random high-frequency interference on EMG acquisition results. Filtering methods include low-pass filtering, band-pass filtering, notch filtering, adaptive filtering, wavelet transform filtering, or pattern recognition-based filtering methods. Band-pass filtering is used to retain effective components within the main frequency range of EMG signals, while notch filtering is used to suppress 50Hz or 60Hz power supply frequency interference. Wavelet transform methods can perform multi-scale decomposition of non-stationary EMG signals, thereby improving the extraction capability of weak EMG signals.

[0057] Since residual stimulation artifacts may remain after stimulation ends, the control software module suppresses these artifacts during the EMG acquisition phase. At the beginning of each acquisition cycle, sampling data within the first T3 time period after stimulation cessation is skipped to reduce the impact of residual stimulation charge on the EMG acquisition results. T3 satisfies: Where T3 represents the stimulus artifact inhibition time, and T2 represents the electromyography acquisition time.

[0058] T3 can be set to a fixed value or dynamically adjusted according to the stimulation parameters. For example, when the stimulation amplitude or stimulation frequency increases, the artifact decay time of the stimulation is correspondingly prolonged, so the control software module increases T3; when a significant stimulation frequency component is detected in the frequency domain of the electromyographic signal, the T3 value is also increased, thereby improving the accuracy of effective electromyographic signal acquisition.

[0059] After removing artifacts, the control software module further employs time-domain filtering, frequency-domain filtering, or adaptive filtering methods to eliminate residual stimulus interference. Among these, the adaptive filtering algorithm dynamically updates the filtering parameters based on the stimulus signal characteristics, thereby improving artifact suppression capabilities under different stimulus conditions.

[0060] The control software module extracts features from the processed electromyographic signals to obtain characteristic parameters that reflect the activation state of the target muscle. These characteristic parameters include root mean square (RMS), mean absolute value (MAV), integrated electromyographic value (IEMG), spectral energy, or median frequency (MF).

[0061] The formula for calculating the root mean square (RMS) value is as follows: in, Indicates the first The amplitude of electromyographic signals at each sampling point Indicates the number of sampling points. Used to characterize the energy level of electromyographic signals The higher the value, the higher the degree of activation of the target muscle.

[0062] The formula for calculating the Mean Absolute Value (MAV) is as follows: in, It can reflect the overall activity level of electromyographic signals.

[0063] The control software module calculates the muscle activation state index S based on the extracted electromyographic feature parameters: in, , Here, RMS represents the root mean square value, and MAV represents the absolute mean value. The weighting coefficients are set based on different patient characteristics or training data.

[0064] When the muscle activation index S is lower than the preset threshold When the target muscle is in a relaxed state or there is a risk of airway collapse, the control software module determines that the target muscle is in a relaxed state or that there is a risk of airway collapse, and generates a feedback control signal to increase the electrical stimulation output intensity or prolong the stimulation duration T1 to enhance the activation level of the target muscle; when S is higher than the threshold When necessary, the intensity of the stimulus output is reduced to minimize unnecessary stimulation, improve wearing comfort, and reduce the risk of muscle fatigue.

[0065] By coordinating the stimulation output and electromyography (EMG) acquisition, and combining EMG characteristic analysis to achieve dynamic feedback regulation, the real-time response capability and control stability during sleep apnea treatment can be improved.

[0066] Example 2: This embodiment provides a closed-loop genioglossus electromyography (EMG) device based on a switch-isolated circuit for the adjunctive treatment of sleep apnea. The device includes an electrode module, a control circuit module, and a control software module. The electrode module is positioned in the submental region to acquire EMG signals from the genioglossus muscle and output electrical stimulation signals. The electrode module employs a dual-modal electrode structure, comprising four electrode contacts. Each electrode contact, under the control of the control circuit module, switches between stimulation output mode and EMG acquisition mode in a time-division manner, thereby achieving multiplexing of stimulation and acquisition functions within a limited submental region. The electrode base layer is made of a flexible material to improve fit with the chin skin and wearing comfort; a metal fabric shielding layer is installed inside the base layer to reduce interference from environmental electromagnetic noise on the EMG acquisition signal and improve the signal-to-noise ratio of the EMG signal.

[0067] A safety isolation structure is set between the electrode module and the control circuit module, with an isolation withstand voltage greater than 1500V. This is used to achieve electrical isolation between the patient end and the system power supply, preventing high voltage from being conducted to the human body when the control circuit malfunctions, and improving the safety of system use.

[0068] The control circuit module includes a stimulation signal generation unit, a signal acquisition unit, a switching isolation circuit, an impedance detection unit, and a main control unit. The stimulation signal generation unit generates an electrical stimulation signal; the output waveform can be a sine wave, a pulse wave, or a modulated combination waveform, with the stimulation frequency range set to 2kHz–20kHz. The stimulation output terminal employs a constant current source structure to reduce the impact of skin impedance changes on stimulation intensity and improve stimulation stability. The impedance detection unit detects the contact impedance between the electrode and the skin in real time and feeds the detection results back to the main control unit. The main control unit compensates and adjusts the stimulation output parameters based on impedance changes to maintain a stable stimulation current output.

[0069] The signal acquisition unit is used to acquire electromyographic (EMG) signals from the genioglossus muscle. The acquisition path includes a preamplifier circuit, a bandpass filter circuit, and a notch filter circuit. The bandpass filter range is preferably 10Hz to 500Hz to preserve the effective frequency band of EMG; the notch filter is used to suppress 50Hz or 60Hz power frequency interference. To prevent stimulation pulses from entering the acquisition path, the switch isolation circuit is constructed using analog switching devices, and its on / off state is controlled by the main control unit.

[0070] The system operates in an alternating cycle of stimulation phase T1 and acquisition phase T2. During stimulation phase T1, the stimulation signal generation unit outputs an electrical stimulation signal, while the main control unit controls the isolation circuit to disconnect the acquisition path, preventing high-amplitude stimulation pulses from entering the acquisition front end and causing amplifier saturation or device damage. During acquisition phase T2, the stimulation signal output stops, and the main control unit controls the isolation circuit to turn on, allowing the electromyography (EMG) signal to be input to the acquisition unit. To further reduce the impact of stimulation artifacts, no effective EMG data is read during the first T3 time period after the start of acquisition phase T2, allowing residual stimulation artifacts to decay. The parameters T1, T2, and T3 can all be set by the user through the operating interface, with a setting range of 1ms to 5s.

[0071] Control software module: used for signal processing and closed-loop regulation. The software executes a time-segmented cycle of T1 (stimulation) and T2 (acquisition); during the T2 time segment, it acquires electromyographic signals of the chin and lingua muscle, which are then filtered, artifact eliminated, and feature extracted; when the signal amplitude is below the threshold (30% of the awake baseline), muscle relaxation is determined, indicating a risk of upper airway collapse; the system will gradually increase the stimulation current until the electromyographic level recovers.

[0072] This embodiment effectively avoids stimulation artifacts entering the acquisition circuit, ensuring the signal-to-noise ratio of the acquisition; the hardware circuit has strong safety, preventing damage to the front-end acquisition and amplification circuit due to overvoltage; it enables early monitoring and intervention of the genioglossus muscle, forming a practical closed-loop treatment mechanism. This embodiment effectively reduces the impact of stimulation artifacts on electromyography (EMG) acquisition and improves the quality of EMG signal acquisition by using time-division multiplexing of the acquisition and stimulation pathways, combined with a switch isolation structure; at the same time, it improves the stability of stimulation output and treatment safety by combining impedance detection and dynamic stimulation adjustment mechanisms, enabling real-time monitoring and early intervention of the genioglossus muscle state.

[0073] Example 3: This embodiment provides a closed-loop genioglossus electromyography (EMG) device based on a filter isolation circuit for non-invasive auxiliary treatment of sleep apnea. The device includes an electrode module, a control circuit module, and a control software module. Unlike implementations using switch isolation, this embodiment achieves frequency domain isolation between the electrical stimulation signal and the EMG acquisition signal through a high input impedance filter isolation structure, enabling synchronous stimulation and acquisition, thereby improving the system's real-time performance and continuous monitoring capabilities.

[0074] The electrode module is positioned in the submental region to acquire electromyographic (EMG) signals from the genioglossus muscle and output electrical stimulation signals. The electrode module employs a dual-modal, four-contact structure, with all four contacts simultaneously handling both EMG acquisition and stimulation output. The electrode contacts are made of conductive metal and / or conductive gel materials, and a flexible conductive gel patch is applied to the electrode surface to reduce skin contact impedance. The overall electrode impedance is preferably less than 10kΩ to reduce contact noise and improve the stability of the stimulation current transmission. The electrode base layer is made of a flexible material to adapt to the curved shape of the submental region and improve wearing comfort. A shielding layer structure is incorporated within the base layer; this shielding layer can be made of metal fabric, a conductive film, or a graphene conductive layer to reduce the impact of external electromagnetic interference on weak EMG signals.

[0075] A safety isolation structure is set between the electrode module and the control circuit module, with an isolation withstand voltage greater than 1500V, to achieve electrical isolation between the patient end and the system power supply, prevent abnormal voltage from acting on the human body, and improve the safety of the device.

[0076] The control circuit module includes a stimulation signal generation unit, a signal acquisition unit, a high input impedance filter circuit, an impedance detection unit, and a main control unit. The stimulation signal generation unit outputs a mid-frequency electrical stimulation signal, preferably in the range of 5kHz to 20kHz, with a preferred stimulation current range of 0mA to 100mA. The stimulation output terminal employs a constant current source structure to reduce the impact of skin impedance changes on the stimulation current and improve stimulation stability.

[0077] The signal acquisition unit is used to acquire electromyographic (EMG) signals from the surface of the genioglossus muscle. The acquisition path includes a preamplifier circuit, a buffer circuit, and a filter circuit. To reduce the load of the acquisition circuit on the human side signal, a high input impedance buffer is set at the input of the filter circuit. The input impedance of the buffer is preferably more than 10 times that of the patient side impedance to improve the accuracy of EMG signal acquisition and reduce signal attenuation. The high input impedance filter circuit adopts a bandpass filter structure, with a preferred filtering frequency band of 10Hz to 500Hz, to retain the effective frequency components of the genioglossus muscle EMG while filtering out mid-to-high frequency interference components and harmonic signals generated by the stimulation pulse. Since the stimulation signal frequency and the EMG signal frequency are in different frequency bands, the filtering isolation structure can realize the synchronous operation of the stimulation path and the acquisition path without the need for time-segmented switching between the stimulation and acquisition processes.

[0078] The impedance detection unit is used to detect the contact impedance between the electrode and the skin in real time and feeds the impedance detection results back to the main control unit. The main control unit automatically adjusts the stimulation output parameters according to the impedance changes to compensate for impedance fluctuations caused by skin sweating, displacement, or changes in adhesion, thereby improving stimulation stability.

[0079] Control software module: Implements a synchronous acquisition + control closed-loop strategy. The software continuously acquires filtered electromyographic (EMG) signals; the signals undergo preprocessing and feature extraction, and amplitude indices are calculated; when the EMG signal falls below a set threshold, the stimulation amplitude or frequency is automatically increased; the algorithm can be combined with an adaptive filter to further eliminate residual stimulation artifacts.

[0080] In this embodiment, stimulation and acquisition can be performed simultaneously, ensuring uninterrupted signal transmission and improving real-time performance; filtering and isolation enhance signal effectiveness and avoid circuit risks caused by large pulses; the closed-loop mechanism continuously adjusts stimulation based on electromyography, improving individualized treatment effects.

[0081] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0082] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A closed-loop control system for electrical stimulation based on electromyographic feedback, used to treat sleep apnea, characterized in that, include: Electrode module, control circuit module, and control software module; The electrode module is used to output electrical stimulation signals and acquire electromyographic signals. The control circuit module is connected to the electrode module and the control software module respectively, and is used to generate electrical stimulation signals, acquire electromyographic signals, and control the electrode module to work in stimulation output mode or electromyographic acquisition mode. The control software module is used to control the control circuit module to output electrical stimulation and acquire electromyographic signals in a time-sharing manner, and to control the output parameters of the electrical stimulation signal according to the acquired electromyographic signals, forming a closed-loop control based on electromyographic feedback.

2. The closed-loop control system for electrical stimulation based on electromyographic feedback according to claim 1, characterized in that, The electrode module includes multiple electrode contacts and a base layer. The multiple electrode contacts are disposed on the base layer and in contact with the human body for transmitting the electrical stimulation signal and acquiring the electromyographic signal. Under the control of the control circuit module, the multiple electrode contacts selectively operate in either the stimulation output mode or the electromyographic acquisition mode.

3. The closed-loop control system for electrical stimulation based on electromyographic feedback according to claim 2, characterized in that, The electrode contacts include metal electrode contacts and / or conductive gel electrode contacts, and the base layer is a flexible base layer; the electrode module also includes a shielding layer and / or a biocompatible bonding layer disposed on the base layer, which are used to reduce the influence of external interference signals on electromyography signal acquisition and improve the bonding stability between the electrode module and the human body.

4. The closed-loop control system for electrical stimulation based on electromyographic feedback according to claim 1, characterized in that, The control circuit module includes a control unit, an electrostimulation signal generation unit, an electrophysiological signal acquisition unit, and an isolation circuit.

5. The closed-loop control system for electrical stimulation based on electromyographic feedback according to claim 4, characterized in that, The electrical stimulation signal generation unit includes a waveform generation circuit and an output signal amplitude control circuit. The waveform generation circuit is used to generate an electrical stimulation waveform, and the output signal amplitude control circuit is used to adjust the output amplitude of the electrical stimulation signal. The electrophysiological signal acquisition unit is connected to the electrode module and is used to receive electromyographic signals; The isolation circuit includes a first isolation circuit and a second isolation circuit.

6. The closed-loop control system for electrical stimulation based on electromyographic feedback according to claim 5, characterized in that, The first isolation circuit is disposed between the patient end and the control circuit power supply to achieve electrical isolation; the first isolation circuit includes an isolation transformer and / or an isolation power supply. The second isolation circuit is disposed between the electrical stimulation signal generation unit and the electrophysiological signal acquisition unit, and is used to isolate the electrophysiological signal acquisition unit when outputting the electrical stimulation signal; the second isolation circuit includes a switching circuit and / or a filtering circuit, the switching circuit includes an analog switch or a relay, and the filtering circuit is used to filter out interference components in the electrical stimulation signal.

7. The closed-loop control system for electrical stimulation based on electromyographic feedback according to claim 1, characterized in that, The control software module is used to execute the following processes: The electrical stimulation signal generation unit is controlled to output an electrical stimulation signal during the first time period T1 and stop outputting the electrical stimulation signal during the second time period T2. At the same time, the electrophysiological signal acquisition unit in the control circuit module is controlled to acquire electromyographic signals during the second time period T2. The acquired electromyographic signals were preprocessed; Within each sampling period, electromyographic sampling data within a preset time T3 after the cessation of electrical stimulation are discarded to eliminate interference from electrical stimulation artifacts, wherein T3 satisfies... ; Feature extraction is performed on the preprocessed electromyographic signal to obtain electromyographic feature parameters, wherein the electromyographic feature parameters include the root mean square value and / or the mean absolute value of the electromyographic signal. Calculate the muscle activation index based on the electromyographic characteristic parameters. And determine the physiological state of the target muscle based on a threshold, where: in, , These are the weighting coefficients; This refers to the root mean square value among the electromyographic characteristic parameters. This represents the average absolute value among the electromyographic characteristic parameters; when Below the preset threshold At that time, it is determined that the target muscle is in a relaxed state or the risk of airway collapse is increased; Based on the physiological state, a control strategy signal is generated, and the output parameters of the electrical stimulation signal generation unit in the control circuit module are adjusted. These output parameters include stimulation amplitude A, stimulation duration T1, and / or acquisition time T2. When… At the same time, increasing the stimulation amplitude A or increasing T1 can enhance the activation of the target muscle.

8. The closed-loop control system for electrical stimulation based on electromyographic feedback according to claim 7, characterized in that, The preprocessing includes denoising filtering and artifact suppression, wherein the denoising filtering includes low-pass filtering, band-pass filtering, notch filtering, or adaptive filtering.

9. A control method based on the electrical stimulation closed-loop control system according to any one of claims 1 to 8, characterized in that, Includes the following steps: A time-division control cycle for electrical stimulation and electromyography (EMG) acquisition is established, and a single control cycle is divided into an electrical stimulation output phase T1 and an EMG acquisition phase T2. During the electrical stimulation output phase T1, the electrical stimulation signal generation unit is controlled to output an electrical stimulation signal to the electrode module. During the EMG acquisition phase T2, the electrophysiological signal acquisition unit is controlled to acquire the EMG signal of the target muscle through the electrode module. Within each control cycle, the electrode module is switched in working mode by the control circuit module, so that it alternates between electrical stimulation output mode and electromyography acquisition mode, so as to achieve temporal isolation between stimulation signal and electromyography signal. The acquired electromyographic signals are preprocessed, including noise reduction filtering and artifact suppression processing, wherein the noise reduction filtering is used to suppress environmental noise and electrical interference signals. During the electromyography (EMG) acquisition phase T2, data from the pre-defined artifact time T3 following the end of stimulation are discarded from the acquired EMG signals to obtain valid EMG signals. ; Feature extraction is performed on the effective electromyography (EMG) signals to obtain EMG feature parameters; Calculate the muscle activation index based on the electromyographic characteristic parameters. ; The muscle activation index With preset threshold When comparing, At that time, it is determined that the target muscle is in a relaxed state or there is a risk of airway collapse; A control feedback signal is generated based on the determination result, and the control feedback signal is fed back to the electrical stimulation signal generation unit to dynamically adjust the electrical stimulation parameters. The electrical stimulation parameters include stimulation amplitude A and / or stimulation duration T1 and / or electromyography acquisition time T2, wherein when... At the same time, increasing the stimulation amplitude A and / or prolonging T1 enhances the electrical stimulation response of the target muscle.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the control method as described in claim 9.

Images