Systems and methods for improving sleep disordered breathing using closed loop feedback
Through the closed-loop feedback system to monitor and adjust nerve stimulation during sleep, the problems of poor tolerance and poor efficacy of obstructive sleep apnea treatment are solved, and quality of life and health risk management are improved.
Patent Information
- Application Number
- CN202411357501.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2019-05-02
- Filing Date
- 2020-05-04
- Publication Date
- 2024-12-03
AI Technical Summary
Existing obstructive sleep apnea (OSA) treatment methods, such as poor tolerance and many side effects of continuous positive airway pressure ventilation (CPAP), surgical treatment is traumatic and poorly effective, and cannot effectively improve the quality of life and reduce the patient's quality of life. Related disease risks.
Using a closed-loop feedback system, the monitoring of physiological data, identifying sleep-related biomarkers, rules-based classification and artificial intelligence algorithms to adjust nerve or muscle stimulation to synchronous respiratory patterns. unobstructed.
Improves sleep quality, reduces daytime sleepiness and fatigue, enhances the targeted and tolerant treatment, reduces surgical trauma and side effects, and improves quality of life and health risk management.
Smart Images

Figure CN119055198A_ABST
Abstract
Description
[0001] This application is a divisional application of an application with an application date of May 4, 2020, application number 202080048485.5, and invention name “System and method for improving sleep-disordered breathing using closed-loop feedback”.
[0002] CROSS-REFERENCE TO RELATED APPLICATIONS
[0003] This application claims priority to U.S. Provisional Application No. 62 / 841,978, entitled “Systems and Methods for treating Obstructive Sleep Apnea,” filed on May 2, 2019. The entire contents of this provisional application are incorporated herein by reference for all purposes. Technical Field
[0004] The present disclosure relates generally to systems and methods for treating sleep disordered breathing (SDB), and more specifically, to systems and methods for providing neurostimulation that is delivered physiologically and adjusted based on biomarkers input into a closed-loop algorithm to treat SDB, including obstructive sleep apnea (OSA). Background Art
[0005] Upper airway sleep disorders (UASD) are conditions that occur in the upper airway that reduce sleep time and quality, causing the patient to experience symptoms such as daytime sleepiness, fatigue, and difficulty concentrating. Obstructive sleep apnea (OSA) is a type of UASD characterized by cessation of airflow due to upper airway obstruction despite simultaneous respiratory efforts. Despite the obstruction, respiratory efforts continue until the individual awakens from sleep. During sleep, the genioglossus and other muscles that keep the airway open relax, causing the mandible and tongue to move backward, which reduces the upper airway volume. The obstruction causes a decrease in blood oxygen levels, which leads to an increase in respiratory drive, and this cycle continues until the patient wakes up.
[0006] OSA is very common, affecting one in five adults in the United States. One in fifteen adults has moderate to severe OSA that requires treatment. OSA is the most common type of sleep apnea. Untreated OSA can lead to decreased quality of life indicators and increase the risk of high blood pressure, stroke, heart disease, and other diseases. Continuous positive airway pressure (CPAP) is the standard treatment for OSA. Although CPAP is non-invasive and highly effective, it is not well tolerated by patients and has a variety of side effects. It is reported that patient compliance and / or tolerance to CPAP is usually between 40% and 60%. Surgical treatment options for OSA, such as anterior tongue muscle repositioning, orthognathic bimaxillary advancement, uvula-palatopharyngoplasty, and tracheotomy, are also available. However, the surgical treatment options are often highly invasive (causing structural changes), irreversible, and have poor and / or inconsistent efficacy. Even more effective surgical options are undesirable because they often require multiple invasive and irreversible surgeries that may alter the patient's appearance (eg, maxillomandibular advancement), and / or may be socially stigmatizing (eg, tracheotomy), and are associated with widespread morbidity. Summary of the invention
[0007] The present disclosure relates generally to systems and methods for treating sleep disordered breathing (SDB), and more specifically to systems and methods for providing neural or muscle stimulation that is delivered physiologically and adjusted based on biomarkers input into a closed-loop algorithm to improve SDB, including OSA.
[0008] One aspect of the present disclosure relates to a system for providing stimulation according to a closed-loop algorithm to improve SDB. The system includes one or more sensors, a computing device, and one or more electrodes. In some cases, the sensor, computing device, and / or electrode may be part of a neurostimulation device. One or more sensors may be implanted adjacent to the anterior tongue muscle and configured to record physiological data. The computing device may include a non-transitory memory storing instructions; and a processor for accessing the non-transitory memory and analyzing, decoding, and executing instructions to at least: monitor the physiological data recorded by the sensor; identify a triggering event within the physiological data, wherein the triggering event is identified as a biomarker of a sleep-related condition; and apply rule-based classification to the triggering event to determine whether one or more parameters of the stimulation signal should be changed based on the biomarker. One or more electrodes may be implanted adjacent to a neural or muscle target site, such as the hypoglossal nerve, and configured to deliver stimulation to the target site over a period of time and change one or more parameters of the stimulation in response to a signal from the computing device.
[0009] Another aspect of the present disclosure relates to a method of providing stimulation according to a closed-loop algorithm to treat SDB including OSA. The method may include steps that can be performed by a system including a processor. The steps may include: monitoring physiological data recorded by a sensor implanted adjacent to anterior tongue muscle; identifying a triggering event within the physiological data, wherein the triggering event is identified as a biomarker of a sleep-related condition; and applying a rule-based classification by the system to the triggering event to determine, based on the biomarker, whether one or more parameters of the stimulation should be changed. BRIEF DESCRIPTION OF THE DRAWINGS
[0010] Figure 1 is a diagram illustrating an example configuration of a stimulation system according to an aspect of the present disclosure;
[0011] Figure 2 yes Figure 1 A schematic diagram of an implantable stimulator portion of a system;
[0012] Figure 3 is a block diagram of an example system that can provide neural stimulation to treat SDB according to a closed-loop algorithm, which example system can be Figure 1 part of a system;
[0013] Figure 4 yes Figure 3 A block diagram of an example of a computing device is shown;
[0014] Figure 5 is a diagram illustrating portions of an exemplary implantable stimulator; and;
[0015] Figure 6 is a process flow diagram of an example method for providing neural stimulation according to a closed-loop algorithm to treat SDB, including OSA. DETAILED DESCRIPTION
[0016] The present disclosure relates to systems and methods for improving SDB. SDB may refer to several chronic conditions in which one or more partial or complete cessation of breathing occurs throughout the night, resulting in daytime sleepiness or fatigue, which interferes with the patient's ability to work and reduces quality of life. OSA may refer to a type of SDB, a breathing disorder that occurs primarily during sleep, the consequences of which may persist throughout waking hours in the form of sleepiness. In some cases, despite continued efforts to breathe, patients with OSA may experience reduced or complete cessation of airflow due to muscle relaxation during sleep causing the soft tissue in the back of the throat to collapse and block the airway. OSA can be characterized by periodic collapse of the upper airway with apnea, hypopnea, or continuous or sustained reduction in ventilation during sleep, as well as excessive daytime sleepiness, neurocognitive deficits, and depression.
[0017] The term "modulate" may refer to causing a change in neural and / or muscle activity, chemical substances and / or metabolism, such as an increase, decrease or pattern change in neuronal and / or muscle activity. For example, modulation may refer to excitatory stimulation and / or inhibitory stimulation. "Stimulation" may refer to activation and / or inhibition of neural and / or muscle activity, and may also be referred to as "neuromuscular stimulation" or "neurostimulation". Thus, stimulation may be excitatory or inhibitory. "Patient" includes mammals, such as humans. By "improvement" or "treatment", the patient's medical disorder is better after treatment than before treatment.
[0018] As used herein, with respect to the described elements, unless otherwise specified, the terms "one" and "the" include at least one or more of the described elements, including combinations thereof. In addition, unless otherwise specified, the terms "or" and "and" refer to "and / or" and combinations thereof. The phrase "substantially" means that the shape or configuration of the described element does not need to have a mathematically accurate shape or configuration of the described element, but may have a shape or configuration that can be identified by a technician in the field as substantially or roughly having the described shape or configuration of the described element. The term "artificial intelligence" may refer to the ability of a computer to correctly interpret / execute external data. The term "learning" may refer to the ability of artificial intelligence to improve from experience. The term "machine learning" may be used interchangeably with "learning". The term "rule-based classification" may refer to any suitable classification scheme, such as a classification scheme for class prediction using IF-THEN rules. The term "titration" may refer to a technique that can continuously measure and update parameters. For example, titration may be part of learning.
[0019] 1. Overview
[0020] The present disclosure relates to improving SDB, such as OSA, through neural and / or muscle stimulation. The systems and methods described herein may provide stimulation according to a closed-loop stimulation algorithm. The closed-loop stimulation algorithm may: monitor physiological data (e.g., EMG data from anterior tongue muscle tissue such as the genioglossus), which may include characteristic signals related to breathing, but the characteristic signals may also be related to sleep position, sleep state, and / or other physiological characteristics that are critical to the treatment of OSA, recorded by sensors implanted in close proximity to anterior tongue muscles such as the genioglossus; identify trigger events within the physiological data, which are identified as biomarkers of sleep-related conditions (e.g., inspiration); and apply rule-based classification to the trigger events to determine whether one or more parameters of the stimulation should be changed based on the biomarkers. In some cases, the rule-based classification may include elements of artificial intelligence, such as undergoing a type of learning so that the stimulation becomes more appropriate for a particular patient.
[0021] As an example use, monitoring the patient's physiological or intrinsic EMG during sleep can be used to apply treatment in a closed-loop manner. EMG data from the anterior tongue muscles, particularly the genioglossus muscle, can be used to synchronize stimulation and synchronize with the patient's breathing pattern to open the patient's airway during certain phases of breathing associated with contraction of the genioglossus muscle. For example, by identifying inspiration / expiration via the EMG signal of the genioglossus muscle that innervates the tongue, the device can sense EMG activity of, for example, the diaphragm without the need for chest leads or other methods associated with other OSA neurostimulation systems. When the EMG data meets the trigger condition, a therapeutic signal, such as an electrical signal, can be delivered to the genioglossus muscle to treat the patient's occurrence of SDB in a closed-loop manner.
[0022] 2. System
[0023] In one aspect, the present disclosure may include systems that can be used to improve SDB through neuromodulation. Figure 1 , the system 10 may include an implantable stimulator 20 and an external controller 100. The controller 100 may provide electromagnetic induction to the stimulator 20 (in Figure 2 The stimulator 20 may include a power receiver 30 having an antenna 32. When the antenna 32 is positioned above the power pad 112 of the controller 100 in the electric field generated by the power transmitting antenna 112, a current may be induced in the antenna 32. The antennas 112 and 32 may also facilitate communication between the controller 100 and the stimulator 20, respectively. This power / communication link between the stimulator 20 and the controller 100 is generally provided by Figure 1 As shown by arrow 70 in FIG.
[0024] The system 10 may also include a user interface 200 in the form of a computer platform 202 running a custom application that is capable of communicating wirelessly with the controller 100, as generally indicated by arrow 204. Such communication may be accomplished, for example, using Bluetooth or WiFi radio communications. Figure 1 In the example configuration of , the computer platform 202 is a smart phone. However, the type of computer platform 202 can vary. For example, the computer platform 202 can be a doctor and / or patient platform. Each platform 202 can have a user-specific application or "app" installed thereon, i.e., a patient app or a doctor app. The patient app can allow the patient to execute certain commands necessary to control the operation of the stimulator 20, such as starting / stopping treatment, increasing / decreasing stimulation power, and selecting a stimulation program. In addition to the controls provided to the patient, the doctor app can also allow the doctor to modify stimulation settings, closed-loop and open-loop control settings, and tuning parameters such as, but not limited to, pulse settings (patterns, durations, waveforms, etc.), stimulation frequencies, amplitude settings, and electrode configurations.
[0025] As generally indicated by arrow 206, the computer platform 202 may be connected (e.g., WiFi and / or LTE) to the Internet / cloud 208, which facilitates communication 214 with a remote or cloud-based server 216. This allows data to be transferred between the server 216 and the computer platform 202 via the Internet 208. Additionally, the controller 100 itself may also be connected (e.g., WiFi) to the Internet, as shown at 210. This may also allow data to be transferred between the controller 100 and the server 216 via the Internet 208.
[0026] like Figure 1 As shown in and described above, the system 10 can be configured to provide various communication paths between system components. For example, a computer platform 202 connected to the controller 100 (see 204) and connected to the Internet 208 (see 206) can facilitate a communication path from a remote server 216 (see 214) to the stimulator 20 itself (see 70). A communication path between the server 216 and the stimulator 20 can also be established via the WiFi link 210 of the controller 100.
[0027] Additionally, recognizing that the physician may be remote from the patient, a physician communication path may be established via the Internet connection 206 of the remotely located physician platform 202. Through this connection, the remote physician platform 202 may communicate with the server 216 via the Internet connection 206. The remote physician platform 202 may also communicate with the controller 100 via the Internet connection 210 (when enabled) or through the patient controller 202.
[0028] The treatment method implemented with system 10 may involve implanting only stimulator 20, leaving controller 100 as an external component used only during the application of treatment. To facilitate this, stimulator 20 can be configured to be powered by controller 100 through electromagnetic induction. In operation, a power pad 110 operated by control unit 120 can be positioned outside the patient near stimulator 20 to position the controller's transmitting antenna 112 located in the pad close to the stimulator's receiving antenna 32. When the patient is sleeping, power pad 110 can be positioned on the sleeping surface or close enough to the sleeping surface to maintain the position of receiving antenna 32 within the target volume of the electromagnetic field generated by transmitting antenna 112.
[0029] In addition, the stimulator 20 may implement electromyography (EMG) electrodes for sensing the patient's neuromuscular response to physiological activity during sleep, as stated above. Such sensing electrodes can continuously monitor physiological intrinsic EMG signals from the anterior tongue muscle tissue. For example, the EMG sensing electrodes can be configured to detect neuromuscular responses from the genioglossus muscle, which is innervated by the HGN.
[0030] In operation, the sensed EMG response from the genioglossus muscle can allow closed-loop operation of the stimulator 20 without chest leads. To facilitate real-time, closed-loop control, a control algorithm can be implemented locally on the stimulator 20. This can be achieved, for example, by programming the control algorithm on an application-specific integrated circuit (ASIC) component of the stimulator 20, but the control algorithm can be programmed on an external control device / component of the system. When operating in real time, the stimulator 20 can record data related to the stimulation session, including, for example, stimulation settings, EMG responses, breathing, pulses, sleep states including different stages of REM and non-REM sleep, etc. After the sleep session, this recorded data can be uploaded to the user interface 200 and the server 216. In addition, the patient can be queried to use the interface 200 to record data about the patient's perceived sleep quality, which can also be uploaded to the server 216. Offline, the server 216 can execute a software application to evaluate the recorded data to determine whether the settings and control parameters can be adjusted to further optimize the stimulation treatment. The software application may, for example, include an artificial intelligence (AI) model that learns from recorded treatment sessions how certain adjustments affect the patient's treatment outcomes. In this way, through AI learning, the model can provide patient-specific optimized treatment.
[0031] refer to Figure 3, system 300 may be implemented within system 10 and / or stimulator 20 to provide stimulation according to open-loop control or closed-loop control to improve SDB. The system may include one or more sensors 302 (which may be implanted and / or external), a computing device 304 (which may be implanted and / or external, and may be part of another device similar to a controller), and one or more electrodes 306 (which may be implanted and / or external). One or more sensors may be configured to record / detect physiological data (e.g., data derived from a patient's body) over time, wherein the physiological data includes changes. Exemplary physiological data may include phasic contractions of the anterior tongue musculature, such as phasic genioglossus contractions, potential tonic activity of the anterior tongue musculature, such as tonic activity of the genioglossus, and combinations thereof. Phasic contractions of the genioglossus may indicate inspiration, particularly phasic activity within the potential tonic tension of the genioglossus. The changes in physiological data include changes in the phasic contraction of the anterior tongue muscle tissue, such as phasic genioglossus contraction, changes in the potential tonic activity of the anterior tongue muscle tissue, such as changes in the tonic activity of the genioglossus, and combinations thereof. For example, EMG signal changes may include changes in frequency, amplitude, spike rate, or other features within the EMG signal. Specifically, changes in the phasic contraction of the genioglossus may indicate changes in breathing or inspiration, and may be used as a triggering event for stimulation. Such physiological data and changes therein may be identified in the recorded EMG signal, for example, during different stages of breathing including inspiration. Therefore, one or more sensors 302 may include an EMG sensor. One or more sensors 302 may also include wireless or tethered sensors that measure, for example, body temperature, movement, breathing sounds (e.g., audio sensors), heart rate, pulse oximetry, eye movements, etc.
[0032] The computing device 304 may be configured to provide open-loop control and / or closed-loop stimulation to configure the parameters of the stimulation. In other words, with respect to closed-loop stimulation, the computing device may be configured to track the patient's breathing (e.g., each breath of the patient), and the stimulation may be applied, for example, during inspiration. However, with respect to open-loop stimulation, the stimulation may be applied without tracking specific physiological data, such as breathing or inspiration. However, even in such an "open-loop" scenario, the computing device may still adjust the stimulation and record data to act on such information. For example, one way in which the computing device may act on such information is that the computing device may configure the parameters of the stimulation to apply the stimulation in an open-loop manner, but may monitor the patient's breathing to know when to resume applying the stimulation to the breathing in a closed-loop manner to breathe, so that the system always works in a closed-loop algorithm to evaluate the data. Therefore, the adjustment of the stimulation may be based on the input of the computing device 304, which may be based on one or more trends of the physiological data recorded over time by one or more sensors 302. The treatment parameters of the system may be automatically adjusted in response to the physiological data. The physiological data may be stored over time and checked to change the treatment parameters; for example, the treatment data may be checked in real time to change the treatment parameters in real time.
[0033] One or more electrodes 306 can deliver stimulation configured according to the parameters. In some cases, the sensing component 302 and the electrode 306 can be the same structure or element. Advantageously, using a single structure or element as the sensing component 302 and the electrode 306 reduces the invasiveness of the surgical procedure associated with the implant system while also reducing the number of foreign objects introduced into the subject. In some aspects, the sensing component and the electrode are placed on the same device, such as on a neuromodulation lead.
[0034] Figure 4304 is shown in FIG. 304. The computing device 304 may include a memory 422 (e.g., non-transitory memory), a processor 424 (e.g., an integrated circuit, such as an application specific integrated circuit (ASIC)), or an ASIC that includes both a memory and a processor. For example, the memory 422 may be a computer-usable or computer-readable medium that may contain or store machine-readable instructions (e.g., a program) for use by the processor 424 or for use by the processor in conjunction with instructions or execution of a system, device, or apparatus (e.g., computing device 304). The computer-usable or computer-readable medium may be, for example, but not limited to, a random access memory (RAM)—including static or dynamic RAM, a read-only memory (ROM), a flash memory, an erasable programmable read-only memory (EPROM), a floating point memory, or a combination thereof, including combinations thereof on the same ASIC. The processor 424 may include, for example, one or more processing cores, processing units, etc. The memory 422 can store machine-readable instructions, and the processor 424 can access the memory 422 and execute the machine-readable instructions (e.g., which can include one or more programs) and cause the computing device 304 to perform operations of the monitoring component 426, the identification component 427, and / or the classification component 428. The processor 424 can interpret the physiological information from the sensor, including decoding the data, analyzing the data, recognizing patterns, etc.
[0035] The monitoring component 426 can monitor physiological data recorded by one or more sensors 302. The identification component 427 can identify triggering events (e.g., related to breathing) within the physiological data. For example, the monitoring component can monitor EMG waveform characteristics, such as spike rate, amplitude, and frequency, as well as phasic activity and tonic activity (also monitoring changes in the amplitude, frequency, or other parameters of the EMG). The identification component can identify triggering events (e.g., changes in characteristics of the EMG waveform) during such monitoring. In one example, the triggering event can be an associated change in the EMG, such as a short-term contraction of the genioglossus muscle or a longer-term change in the activity of the genioglossus muscle, the short-term contraction indicating phasic genioglossus muscle activity, and the longer-term change indicating a change in the potential tonic tension of the genioglossus muscle seen via one or more portions or repetitions of the physiological data. The triggering event can be identified as a biomarker of a condition related to sleep, such as a change in at least one parameter physiological data. In some cases, the biomarker can be inspiration. In other cases, the biomarker can be body position. In other cases, the biomarker may be a stage in the sleep cycle (e.g., awake, non-REM sleep—stage 1 light sleep, stage 2 light sleep, stage 3 deep sleep, REM sleep, etc.). In some cases, motion detection and / or other biomarkers may be used to automatically turn on treatment only after the patient has fallen asleep, and to determine the parameters of the stimulation to optimally maintain airway patency throughout the night (including adapting the stimulation based on sleep stage and position) without causing unnecessary discomfort or leading to arousal events, thereby increasing patient comfort and compliance with treatment. The stimulation may be gradually increased as the patient moves from light sleep to deep sleep or during each stimulation stage, so that the first pulse in a pulse train has a smaller amplitude and / or pulse width than the last pulse in the pulse train. In some cases, the stimulation will automatically stop when the patient wakes up and start again when the patient falls asleep again.
[0036] The wakefulness phase of the sleep cycle refers to the relaxation phase when the subject first lies down in bed or lies in bed trying to fall asleep again. Non-REM sleep has three stages and is a sleep phase without rapid eye movement. The REM phase contains REM sleep in which the eyeballs move rapidly from side to side behind tightly closed eyelids, breathing becomes faster and irregular, heart rate and blood pressure increase to near wakefulness levels, and arm and leg muscles are temporarily paralyzed.
[0037] Non-REM stage 1 refers to the transition from wakefulness to sleep (lasting several minutes). During non-REM stage 1, the subject's heartbeat, breathing, and eye movements slow, and muscles relax, with occasional twitches. Non-REM stage 2, the longest of all stages, is a period of light sleep before entering a deeper sleep, in which heartbeat and breathing slow, muscles relax further, body temperature drops, and eye movements stop. Non-REM stage 3 refers to the period of deep sleep needed to feel refreshed in the morning, in which heartbeat and breathing slow to the lowest level during sleep, muscles relax, and it may be difficult to wake up.
[0038] The sleep state may be determined, for example, based on information in the physiological data (e.g., tonic genioglossus activity as indicated on the EMG). Once the sleep state is identified, the goal is to apply the therapy in a manner that minimizes discomfort to the patient and also minimizes potential stimulation associated with arousal events. This may include reducing the amplitude of the stimulation during stage 1 and stage 2 sleep, and increasing the amplitude during stage 3 and REM. This may also include gradually increasing the therapy over a longer period of time—meaning from zero to the programmed output over a longer period of time during stage 1 and stage 2 sleep and stage 3 and REM sleep, or gradually increasing the therapy within each pulse train—for example, when applied during inspiration.
[0039] For example, if certain EMG activities are detected, such as the phase-wise changes in the EMG activities of inspiration during any stage of sleep, the system can deliver stimulation during the breathing period of inspiration. The system can, for example, use a set of specific electrodes, waveforms, pulse widths, frequencies, pulse intervals, and pulse ramp rates to apply stimulation to the hypoglossal nerve, thereby providing therapeutic airway patency during inspiration. The system can stop stimulation during the expiratory period, and can, for example, continue to monitor the physiological EMG from the genioglossus muscle during the entire inspiration and expiration period of each breath. The system can adjust the stimulation parameters and / or select electrodes for stimulation as needed to optimize the stimulation based on additional biomarkers including sleep state, body position, etc., thereby providing optimal airway patency. The closed-loop algorithm embedded in the stimulator or neuromodulation lead can continuously monitor and adjust the treatment based on physiological data and triggering events, and use rule-based classification to determine when, how, and in what time period to apply and adjust the stimulation to provide optimal airway patency during sleep.
[0040] For example, if certain EMG activity, such as tonic and phasic EMG activity, decreases or ceases during REM, the system may deliver stimulation periodically based on predetermined physician-programmed parameters, the system may rely on previously known patient-specific parameters to apply stimulation, or the system may use a default periodic stimulation that is applied throughout REM sleep. The system may also monitor the EMG through the REM period to determine when to stop using periodic stimulation and when to restart applying stimulation during each inspiratory event.
[0041] In some cases, the system may not turn on stimulation immediately when the stimulator is in the field from the transmitting coil. In this case, the system can turn on and monitor EMG signals, for example, to detect tension and phasic muscle activity to understand the sleep stage. Once the system has determined that the patient is sleeping, entering stage 1 of sleep or stage 2 of sleep, the system 10 can begin to provide treatment in a physiological manner, for example, starting to apply a small amount of stimulation using a stimulation ramp during each stimulation period so that unnecessary awakening events or discomfort are not caused during the initial stage of sleep. In this configuration, EMG may be monitored for several minutes or hours to determine the state before the system starts treatment. Many individuals with OSA also suffer from insomnia, in which the individual has difficulty falling asleep, and in this case, the negative feedback loop may cause additional anxiety in the patient, making the patient afraid that the treatment will be turned on before he falls asleep, and therefore cannot relax enough to fall asleep. This may cause the individual to turn off the treatment, or stop using the treatment over time. A "smart" system that can recognize when the patient is asleep and apply physiological treatment will improve treatment compliance and efficacy. For example, once the system recognizes Non-REM Stage 1, the system may begin recognizing Non-REM Stage 2, Non-REM Stage 3, REM sleep, and so on.
[0042] For example, the ASIC (an instance of processor 424) may be configured to control a custom algorithm that controls a treatment application. For example, the ASIC may be configured to run embedded digital logic that uses information collected by the EMG sensor to determine when, how long, and at which stimulation parameters to stimulate to provide optimal treatment to the subject, thereby controlling the volume of air that can flow through the upper airway, also known as airway patency. The embedded digital logic may sense EMG activity that may be known to the algorithm to correspond to breathing, more specifically, to inhalation and exhalation. The algorithm may decode the EMG activity to trigger stimulation of the bilateral anterior musculature and / or the hypoglossal nerve (including its distal branches), thereby, for example, opening the airway, such that treatment is associated with each breath, each inhalation, and each exhalation, for example. Therefore, treatment may be provided during each breath, particularly during inhalation, for example, by using embedded knowledge of the EMG features corresponding to breathing. The embedded logic may contain knowledge of EMG features specific to body position, jaw position, sleep state (e.g., REM, non-REM), movement, and other physiological parameters that can illuminate and optimize treatment. The algorithm can use adaptive learning to learn individual subject-specific EMG features associated with the above physiological states during sleep to provide subject-specific additional optimization. Adaptive learning can be done manually with physician input, or can be done entirely within the algorithm based on predetermined constraints and knowledge, or can be done using cloud databases and additional adaptive learning that cloud software can use to analyze data from each patient and each sleep session. The algorithm, while still based on respiratory information sensed by the EMG sensor, can also have different modes. In one mode, the algorithm can run and provide therapeutic breathing for breathing, especially during inspiration; in another mode, the algorithm can learn, look for input from the EMG and also from the user (e.g., patient, physician, etc.); in another mode, the algorithm can provide more continuous control of the airway, thereby providing periodic stimulation that can last for a period of time. In another mode, the algorithm may sense EMG information, but instead of providing therapeutic breaths to initiate breathing, it may wait until an impending collapse of the airway is identified and react by providing therapy to prevent the collapse from occurring. The EMG information may include the amplitude of the EMG, the frequency components of the EMG, spike sensing, envelope sensing, and other features that may be acquired directly from the EMG signal to control the algorithm and provide biomarkers of breathing and airway collapse. It should be understood that the algorithm may use any or all of these features throughout the sleep period and may switch between modes based on the EMG activity sensed by the EMG sensor, or the system may be hard programmed to run only one algorithm.
[0043] The system can apply treatment in a way that does not cause discomfort and / or arousal events to the patient. The system can continually adapt to the sleep stage (and / or patient needs) as the patient moves through the stages over the course of the night. For example, the greatest need for stimulation may be during deep sleep (non-REM stage 3) and REM, when discomfort and arousal are less likely, so the system can apply more stimulation because arousal and discomfort are less likely during these stages. The amount of time the patient spends in each stage of sleep can also be tracked, which is very useful for tracking results because most OSA patients typically do not enter deep sleep due to arousals.
[0044] The classification component 428 can apply rule-based classification to the triggering event to determine whether one or more stimulation parameters applied by one or more of the stimulation electrodes should be changed based on biomarkers related to sleep. As stated above, the biomarkers include respiratory stage (e.g., including inhalation within the inhalation period), sleep stage during one or more sleep cycles, and / or body position during sleep, such as indicated by EMG or other sensors or sensed activities. As stated above, the stimulation parameters include, for example, pulse width, amplitude, frequency, waveform shape, electrode position / configuration, etc. The initial rules for the rule-based classification used by the algorithm can be set for the patient and / or based on historical values of the population, historical values of the patient, and / or patient-derived values. Subsequent rules of the algorithm can be learned and / or updated and / or personalized based on an artificial intelligence learning process.
[0045] Feedback related to the stimulation (eg, after it is delivered) may be given to computing device 304. Computing device 304 may receive the feedback and may change one or more stimulation parameters.
[0046] For example, rule-based classification may employ one or more pattern recognition classifiers, each of which utilizes the extracted features or a subset of the extracted features to determine the appropriate clinical parameters. When multiple classifiers are used, an arbitration element may be utilized to provide coherent results from multiple classifiers. Each classifier is trained for multiple training patterns representing various categories of interest. The training process for a given classifier will vary with its implementation, but training generally involves statistical aggregation of training data from multiple training images into one or more parameters associated with an output category. Any of a variety of optimization techniques may be used for classification algorithms, including support vector machines, self-organizing maps, fuzzy logic systems, data fusion processes, ensemble methods, rule-based systems, or artificial neural networks. The result category may represent a specific clinical parameter of a subject. Based on the feature vector provided, the result category is selected and the confidence of the selected result may be calculated. Results below a threshold confidence value may be rejected. For example, a support vector machine (SVM) classifier may process training data (which may involve any parameter being sensed, etc.) to generate a function representing a boundary in a feature space defined by various attributes of interest. Similarly, an artificial neural network (ANN) classifier may process training data (which may be related to any parameter being sensed, etc.) to determine a set of interconnection weights corresponding to the interconnections between nodes in the associated neural network of the ANN classifier. An SVM classifier may conceptually divide boundaries in an M-dimensional feature space using a plurality of functions referred to as hyperplanes, wherein each of the M dimensions represents an associated feature of a feature vector. The boundaries define the range of feature values associated with each class. Therefore, the output category and associated confidence value may be determined based on the position of a given input feature vector in the feature space relative to the boundary. A rule-based classifier applies a set of logical rules to the extracted features to select an output category. Typically, the rules are applied sequentially, wherein the logical result of each step affects the analysis of later steps.
[0047] The ANN classifier includes multiple nodes with multiple interconnections. Values from feature vectors are provided to multiple input nodes. The input nodes each provide these input values to a layer of one or more intermediate nodes. A given intermediate node receives one or more output values from a previous node. The received values are weighted according to a series of weights established during the training of the classifier. The intermediate node converts the received values into a single output according to the transfer function at the node. For example, the intermediate node can sum the received values and assign the sum to a binary step function. The final layer of nodes provides confidence values for the output categories of the ANN, wherein each node has an associated value representing the confidence of one of the associated output categories of the classifier.
[0048] In another example, the rule-based classification may employ a regression model configured to calculate a parameter representing the likelihood that a patient exhibits a biomarker. In yet another example, the rule-based classification may employ a sensitivity analysis using the model such that the magnitude of the effect of one or more features on at least one parameter may be determined and correlated with the biomarker.
[0049] An example closed-loop control scenario involves one or more sensors 302 (implanted adjacent to anterior tongue muscles such as the genioglossus) that can detect / record physiological data over time. The physiological data may include EMG data from the musculature of the anterior airway, which may include characteristic signals related to breathing but may also be related to sleep position, sleep state, and / or other physiological characteristics that are important for the treatment of SDB. A computing device 304 may monitor the physiological data recorded by the one or more sensors 302 to identify triggering events within the physiological data. The triggering event may be identified as a biomarker of a sleep-related condition (e.g., inspiration). Rule-based classification may be applied to the triggering event to determine whether one or more parameters of stimulation (e.g., delivered to the hypoglossal nerve by one or more electrodes 306 or electrode contacts) should be changed based on the biomarker.
[0050] The voltage changes on the transmitting and receiving body and on the power receiver, and the resulting impedance changes, can be sensed to determine the position and movement of the power receiving body within the magnetic field. In this regard, the voltage and impedance changes between the two coils of the power antenna can provide additional information to the system to inform the closed-loop algorithm and inform additional improvements to the treatment. This type of position sensor can have additional uses besides treatment optimization because the position sensor can provide additional data about sleep quality over time and health-related information. In addition, the impedance data between the coils can be related to activity, which can also be used to track wakefulness and sleep cycles. These data, together with EMG data such as tonic EMG activity from the genioglossus muscle, can be used to understand and learn wakefulness and sleep during the entire period of attempted sleep (e.g., when the power receiving coil is within the inductive field volume of the transmitting coil).
[0051] Several wired or wireless input applications including smartphone applications or tablet applications, such as wireless remote controls, may also be used. These additional input applications may provide additional input to the system to adjust therapy, adjust closed loop algorithms, adjust stimulation outputs, adjust optimization, or adjust algorithm modes as needed. The input application may display EMG data to the user, allowing the user to adjust parameters that control EMG acquisition, such as input filters, trigger amplitude, frequency range, etc.
[0052] The input application can also allow automatic treatment titration. In this mode, the application can run customized software, which provides stimulation to the target part of the subject, such as a target nerve or target muscle, and monitors the EMG activity induced thereby of the muscle, such as the anterior tongue muscle of the genioglossus muscle. The EMG activity thus generated can be related to the required airway opening amount (as input into the application), and therefore can allow automatic treatment stimulation parameter setting and eliminate time-consuming parameter adjustment during sleep. Non-limiting examples of stimulation parameter settings include stimulation pulse width, amplitude, frequency, electrode position / configuration, etc. In this regard, the system can determine the treatment stimulation output and allow the subject / doctor to make fine adjustments as needed. The subject or doctor can re-run the automatic parameter adjustment application at any time, and remote monitoring can be performed through the application, so that titration and programming can be performed in the comfort of the subject's home.
[0053] Even if the response to a given set of stimulation parameters varies over time, the EMG signals induced thereby can be continuously monitored, and the stimulation parameters required to produce the tongue movement required for effective treatment can be determined, thereby effectively reducing the amount of testing required for initial programming and the need for ongoing follow-up testing. In addition, problems with treatment can be identified (e.g., stimulation set according to certain stimulation parameters does not provide the tongue movement necessary to open the airway), and warnings can be generated for the patient and / or physician (which allows for faster response and proactive management of the system).
[0054] Figure 6 The stimulator device is shown as implemented in Figure 3 The one or more sensors 302 and the one or more electrodes 306 may be included in a common electrode contact. However, the one or more sensors and the one or more electrodes need not be within a common electrode contact and may be distinct and separate.
[0055] 3. Methods
[0056] Another aspect of the present disclosure may include a method 700 for providing nerve and / or muscle stimulation according to a closed-loop algorithm to treat SDB. Figure 6 ). For example, method 700 may be performed by Figure 1-5 Portions of method 700 may be at least partially stored on a non-transitory memory and executed by a processor.
[0057] For simplicity, method 700 is shown and described as being performed sequentially; however, it should be understood and appreciated that the present disclosure is not limited by the illustrated order, as some steps may occur in a different order and / or concurrently with other steps illustrated and described herein. Furthermore, all of the illustrated aspects may not be required to implement method 700, and / or more aspects than those illustrated may be required to implement method 700. Additionally, one or more aspects of method 700 may be stored in one or more non-transitory memory devices and executed by one or more hardware processors.
[0058] At 752, physiological data recorded by one or more sensors (e.g., related to inspiration, sleep stage, and / or body position, such as indicated by EMG) can be monitored. For example, one or more sensors can be implanted adjacent to an anterior tongue muscle, such as the genioglossus, or in a plane between the genioglossus and geniohyoid muscles. At 754, a triggering event can be identified within the physiological data. The triggering event is a change in at least one parameter of the physiological data (e.g., indicative of inspiration during breathing, body position, and / or a stage in a sleep cycle, such as indicated by EMG).
[0059] At 756, rule-based classification can be applied to the triggering event to determine whether one or more parameters of the stimulation should be changed based on the biomarker represented by the triggering event. For example, a signal including configuration / setting information for the parameter can be sent to one or more electrodes located near the hypoglossal nerve. One or more stimulation parameters can be titrated and adapted based on the triggering event to optimize airway muscle tone.
[0060] Each of the disclosed aspects and embodiments of the present disclosure may be considered individually or in combination with other aspects, embodiments and variations of the present disclosure. In addition, although certain features and aspects of the embodiments of the present disclosure may only be shown in certain figures or described in other ways in certain parts of the present disclosure, such features may be incorporated into other embodiments and aspects shown in other figures or other parts of the present disclosure. Along the same line, certain features of the embodiments and aspects of the present disclosure shown in certain figures or described in other ways in certain parts of the present disclosure may be optional or may be deleted from such embodiments and aspects. In addition, when describing a scope, all points within this scope are included in the present disclosure. In addition, all references cited herein are incorporated by reference in their entirety.
Claims
1. A system, include: a sensor configured to be implanted adjacent to anterior tongue muscles of a subject; an electrode configured to be implanted adjacent to a hypoglossal nerve of the subject and configured to deliver stimulation energy to the hypoglossal nerve; and a computing device coupled to the electrode and the sensor, the computing device comprising a non-transitory computer-readable medium storing instructions that, when executed by one or more processors of the computing device, cause the one or more processors to: monitoring electromyogram (EMG) data recorded by the sensor over a period of time, wherein the EMG data describes anterior tongue muscle activity; determining the phasic activity and the tonic activity of the anterior tongue muscle based on the EMG data; determining a sleep stage of the subject during one or more portions of the time period based on at least the stressful activity; comparing the determined sleep stage for the one or more portions of the time period with a previously determined sleep stage for a previous portion of the time period, and determining, based on the previously determined sleep stage, whether to modify the determined sleep stage according to a predetermined physiological sequence of sleep stages; determining one or more parameters of the stimulation energy based on the determined sleep stage of the subject, and The electrode is caused to deliver the stimulation energy having the one or more parameters to the hypoglossal nerve.
2. The system of claim 1, wherein the one or more parameters of the stimulation energy include at least one of pulse width, amplitude, frequency, waveform shape, electrode configuration, intra-pulse interval, or pulse ramp rate.
3. A system according to claim 1, wherein the one or more parameters of the stimulation energy include amplitude, and wherein the amplitude is smaller when the sleep stage is non-REM stage 1 or non-REM stage 2 than when the sleep stage is non-REM stage 3 or REM.
4. A system according to claim 1, wherein the one or more parameters of the stimulation energy include at least one of amplitude or pulse width, and wherein when the sleep stage is non-REM stage 1 or non-REM stage 2, at least one of the amplitude or pulse width is gradually increased over a longer period of time compared to when the sleep stage is non-REM stage 3 or REM.
5. The system of claim 1, wherein the stimulation energy comprises a pulse train, and wherein a first pulse in the pulse train has at least one of a smaller amplitude or a smaller pulse width than a corresponding amplitude or pulse width of a last pulse in the pulse train.
6. The system of claim 1, wherein the one or more parameters of the stimulation energy include amplitude, and wherein the amplitude is substantially zero if the sleep stage is awake.
7. The system of claim 1, wherein the sleep stage of the subject is determined based on the tonic activity of at least the anterior tongue muscle, including determining the sleep stage to be REM if the tonic activity decreases or stops.
8. A system according to claim 7, wherein the sleep stage of the subject is determined based on the tonic activity of at least the anterior tongue muscle, including determining that the sleep stage is REM if the tonic activity is significantly reduced relative to the tonic activity of the anterior tongue muscle associated with non-REM stage 1, non-REM stage 2 and non-REM stage 3.
9. The system of claim 1, wherein determining the sleep stage of the subject comprises monitoring one or more features of the EMG data, the one or more features comprising at least one of amplitude, frequency components, spikes, or envelope.
10. The system of claim 1, wherein determining the sleep stage of the subject comprises associating subject-specific features of the EMG data with at least one sleep stage.
11. The system of claim 1, wherein the anterior tongue muscle is at least one of the genioglossus or the geniohyoid muscle.
12. The system according to claim 1, in, Upon determining that the sleep stage of the subject is entering Non-REM Stage 1 or Non-REM Stage 2, the processor causes the electrodes to provide the stimulation energy using a stimulation ramp as the subject progresses through subsequent sleep stages.
13. A system for treating a patient with reduced airway volume during sleep, the system include: an electrode configured to be implanted proximate to at least one target nerve or target muscle of the patient; a sensor configured to be implanted proximate to airway muscles associated with airway patency; as well as A computing device comprising: a non-transitory memory for storing instructions; as well as a processor configured to access the non-transitory memory and execute the instructions to at least: causing the electrode to deliver stimulation energy to the at least one target nerve or target muscle at an initial parameter, causing the sensor to acquire evoked electromyography (eEMG) data describing the airway muscle activity in response to the stimulation energy, describing, based on the eEMG data, an increase in the airway volume in response to the stimulation energy, determining a treatment parameter for the stimulation energy based on the increase in airway volume, the treatment parameter being associated with an expected increase in the airway volume in response to the stimulation energy, and The electrodes are caused to deliver the stimulation energy having the treatment parameters to at least one target nerve or target muscle while the patient is sleeping, causing the desired increase in the airway volume to improve airflow through the airway. The system of claim 13 , wherein the treatment parameters are different from the initial parameters.
15. A system according to claim 13, wherein the initial parameters and the treatment parameters each include at least one of the amplitude of the stimulation energy, the pulse width of the stimulation energy, the frequency of the stimulation energy, the waveform shape of the stimulation energy, or the position of the electrode relative to at least one target nerve or target muscle.
16. The system of claim 13, wherein the electrode comprises a plurality of electrodes, wherein the initial parameters and the treatment parameters each comprise a configuration of the plurality of electrodes.
17. A system, include: one or more sensors configured to be implanted adjacent to anterior tongue muscles of the subject and configured to record electromyography (EMG) data; one or more electrodes configured to be implanted adjacent to a hypoglossal nerve of a subject and configured to deliver stimulation energy to the hypoglossal nerve; as well as A computing device, the computing device comprising: A memory for storing instructions; A processor for accessing the memory and executing the instructions to cause the computing device to monitor the EMG data recorded by the sensor, identify the stage in the subject's sleep cycle, and adjust the stimulation energy configured to be delivered to the hypoglossal nerve through the one or more electrodes.
18. The system of claim 17, wherein optimizing the stimulation energy comprises adjusting one or more stimulation parameters and / or selecting electrodes for stimulation.
19. The system of claim 17, wherein the one or more stimulation parameters include at least one of pulse width, amplitude, frequency, waveform shape, electrode configuration, intra-pulse interval, or pulse ramp rate.
20. The system of claim 17, wherein the one or more stimulation parameters include amplitude, and wherein the amplitude is smaller when the stage of the sleep cycle is non-REM stage 1 or non-REM stage 2 than when the stage of the sleep cycle is non-REM stage 3 or REM.
21. A system according to claim 17, wherein the one or more stimulation parameters include at least one of amplitude or pulse width, and wherein when the stage of the sleep cycle is non-REM stage 1 or non-REM stage 2, at least one of the amplitude or the pulse width is gradually increased over a longer period of time compared to when the stage of the sleep cycle is non-REM stage 3 or REM.
22. The system of claim 17, wherein the one or more stimulation parameters comprise a pulse train, and wherein a first pulse in the pulse train has at least one of a smaller amplitude or a smaller pulse width than a corresponding amplitude or pulse width of a last pulse in the pulse train.
23. The system of claim 17, wherein the EMG data identifies phasic and tonic activity of anterior tongue muscles.
24. The system of claim 17, wherein the EMG data identifies the stage in a sleep cycle based on the phasic and tonic activity of the anterior tongue muscle.
25. A system according to claim 17, wherein the stage in the subject's sleep cycle is identified based on the tonic activity of the anterior tongue muscle, wherein if the tonic activity of the anterior tongue muscle decreases or increases, the stage in the sleep cycle is identified as REM.
26. The system of claim 17, wherein identifying the stage of the sleep cycle of the subject comprises monitoring one or more features of the EMG data, the one or more features comprising at least one of amplitude, frequency components, spikes, or envelope.
27. The system of claim 17, wherein identifying the stage of the sleep cycle of the subject comprises associating subject-specific features of the EMG data with at least one stage of the sleep cycle.
28. The system of claim 27, wherein the subject-specific characteristics are specific to body position, jaw position, movement, and other physiological parameters.
29. The system of claim 17, wherein the EMG data identifies changes in inspiration or exhalation of the subject.
30. The system of claim 17, wherein the anterior tongue muscle is the genioglossus muscle.
31. The system according to claim 17, in, Upon determining that the stage of the sleep cycle of the subject is entering Non-REM Stage 1 or Non-REM Stage 2, the processor causes the electrodes to provide the stimulation energy using stimulation ramping as the subject progresses through subsequent stages of the sleep cycle.
32. A system for treating a patient with reduced airway volume during sleep, the system include: one or more electrodes, at least partially implanted in the patient; a sensor configured to record evoked electromyography (eEMG) data; as well as A computing device coupled to the one or more electrodes and the sensor, the computing device comprising: a non-transitory computer-readable medium for storing instructions; and a processor configured to access the non-transitory memory and execute instructions to at least: titrating stimulation energy to be delivered to at least one target nerve or target muscle using the one or more electrodes, wherein titrating the stimulation energy comprises: delivering the stimulation energy having initial parameters to the at least one target nerve or target muscle using the one or more electrodes; obtaining, using the sensor, eEMG data describing activity of airway muscles associated with airway patency in response to the stimulation energy; describing, based on the eEMG data, an increase in the airway volume in response to the stimulation energy; and determining a treatment parameter for the stimulation energy based on the increase in airway volume, the treatment parameter being associated with an expected increase in the airway volume in response to the stimulation energy; and While the patient is sleeping, the stimulation energy having the treatment parameters is delivered to the at least one target nerve or target muscle via the one or more electrodes to produce the expected increase in the airway volume, thereby improving airflow through the airway.
33. The system of claim 32, wherein the treatment parameters are different from the initial parameters.
34. The system of claim 32, wherein the initial parameter and the treatment parameter each comprise at least one of an amplitude of the stimulation energy, a pulse width of the stimulation energy, a frequency of the stimulation energy, or a waveform shape of the stimulation energy.
35. The system of claim 32, wherein the initial parameters and the treatment parameters each include a position of the electrode relative to at least one target nerve or target muscle.
36. The system of claim 32, wherein the electrode comprises a plurality of electrodes, wherein the initial parameters and the treatment parameters each comprise a configuration of the plurality of electrodes.
37. The system of claim 32, wherein determining the treatment parameter comprises evaluating at least one of an envelope or an amplitude of the eEMG data.
38. The system of claim 32, wherein titration of the stimulation energy is controlled by a software application on a computing device.
39. The system of claim 38, wherein the computing device is remote from the patient.
40. The system of claim 38, wherein describing the airway volume increase is performed automatically by the software application.
41. The system of claim 32, further comprising re-titration of the stimulation energy after delivering the stimulation energy having the treatment parameters to at least one target nerve or target muscle while the patient is sleeping.
42. The system of claim 32, wherein the airway muscle is the anterior tongue muscle.
43. The system of claim 32, wherein the airway muscle is the genioglossus muscle.
44. The system of claim 32, wherein the target muscle is the airway muscle.
45. The system of claim 32, wherein the target nerve is the hypoglossal nerve.
46. The system of claim 32, wherein the electrode and the sensor are the same structure.
47. A system for treating a patient with reduced airway volume during sleep, the system include: one or more electrodes, at least partially implanted in the patient; A sensor configured to record evoked electromyography (eEMG) data; a computing device coupled to the one or more electrodes and the sensor, the computing device comprising: a non-transitory computer-readable medium configured to store reference data associated with an expected volume of the airway, and a processor configured to access the non-transitory memory and execute instructions to at least: titrating stimulation energy to be delivered to at least one target nerve or target muscle using the one or more electrodes, wherein titrating the stimulation energy comprises: delivering the stimulation energy having initial parameters to the at least one target nerve or target muscle using the one or more electrodes; obtaining, using the sensor, eEMG data describing activity of airway muscles associated with airway patency in response to the stimulation energy; comparing the eEMG data to the reference data; and Determining treatment parameters of the stimulation energy based on the comparison; and While the patient is sleeping, the stimulation energy having the treatment parameters is delivered to the at least one target nerve or target muscle using the one or more electrodes, causing activity of the airway muscle associated with the expected airway volume, thereby improving airflow through the airway.
48. The system of claim 47, wherein the treatment parameters are modified relative to the initial parameters.
49. A system according to claim 47, wherein the initial parameters and the treatment parameters each include at least one of the amplitude of the stimulation energy, the pulse width of the stimulation energy, the frequency of the stimulation energy, the waveform shape of the stimulation energy, or the position of the electrode relative to the at least one target nerve or target muscle.
50. The system of claim 47, wherein the electrode comprises a plurality of electrodes, wherein the initial parameters and the treatment parameters each comprise a configuration of the plurality of electrodes.
51. The system of claim 47, wherein titration of the stimulation energy is controlled by a software application on a computing device.
52. The system of claim 51, wherein the computing device is remote from the patient.
53. The system of claim 51 , wherein comparing the eEMG data to the reference data is performed automatically by the software application.
54. The system of claim 47, further comprising re-titrifying the stimulation energy after delivering the stimulation energy having the treatment parameters to the at least one target nerve or target muscle while the patient is sleeping.
Citation Information
Cited By
Electrical stimulation device and method for controlling electrical stimulation device
CN120837843A
Electrical stimulation device, method of controlling an electrical stimulation device
CN120837843B
