阅读说明：本技术 评估治疗睡眠呼吸暂停的刺激功效和用于改进osa疗法的舌肌张力感测系统 (Assessing stimulation efficacy for treating sleep apnea and tongue muscle tone sensing system for improving OSA therapy ) 是由 A·沙伊纳 R·舒尔豪瑟 J·希索恩 E·斯科特 R·哈格 于 2020-03-05 设计创作，主要内容包括：一种对植入式神经刺激器(INS)的疗法进行评估的系统和方法,所述INS包含引线和脉冲发生器,所述引线具有至少一对双极电极,所述脉冲发生器与所述双极电极电通信,所述脉冲发生器包含传感器、存储器、控制电路和遥测电路。所述系统和方法包含：外部编程器,所述外部编程器通过所述遥测电路与所述INS通信；服务器,所述服务器与所述外部编程器通信并且所述服务器上包含应用,所述应用被配置成从所述外部编程器接收来自所述INS的传感器数据并且基于接收到的传感器数据对所述INS所植入的患者的睡眠质量进行评估；以及远程计算机,所述远程计算机与所述服务器通信并且被配置成呈现对所述患者的所述睡眠质量的评估。(A system and method of assessing therapy for an Implantable Neurostimulator (INS), the INS comprising a lead having at least one pair of bipolar electrodes and a pulse generator in electrical communication with the bipolar electrodes, the pulse generator comprising a sensor, a memory, control circuitry, and telemetry circuitry. The system and method include: an external programmer in communication with the INS through the telemetry circuitry; a server in communication with the external programmer and containing an application thereon configured to receive sensor data from the INS from the external programmer and to evaluate sleep quality of a patient in which the INS is implanted based on the received sensor data; and a remote computer in communication with the server and configured to present an assessment of the sleep quality of the patient.)

1. An Implantable Neurostimulator (INS), comprising:

an electrical lead having at least one pair of bipolar electrodes formed thereon, wherein the electrical lead is configured for placement of the pair of bipolar electrodes in proximity to an extensor muscle of a patient and configured to receive an Electromyography (EMG) signal;

a pulse generator electrically connected to the electrical leads and configured to deliver electrical energy to the pair of bipolar electrodes, the pulse generator having a sensor and control circuitry therein,

wherein the sensor and the control circuit are configured to receive the EMG signal and determine a tension state of the extensor muscle in which the lead is placed.

2. The implantable neurostimulator of claim 1, wherein the control circuit is in electrical communication with a therapy delivery circuit and causes the therapy delivery circuit to deliver electrical energy to the bipolar electrode upon determining that the EMG signal is below a threshold.

3. The implantable neurostimulator of claim 1, wherein the control circuit is in electrical communication with therapy delivery circuit and causes the therapy delivery circuit to deliver electrical energy to the bipolar electrode when it is determined that the EMG signal is below a threshold and the heart rate detected by the sensor is below a threshold.

4. The implantable neurostimulator of claim 1, wherein the control circuit is in electrical communication with therapy delivery circuit and causes the therapy delivery circuit to deliver electrical energy to the bipolar electrodes when the EMG signal is determined to be below a threshold and a motion sensor determines that the INS is not moving.

5. The implantable neural stimulator of claim 1, wherein the control circuit is in electrical communication with a therapy delivery circuit and causes the therapy delivery circuit to deliver electrical energy to the bipolar electrodes when the EMG signal is determined to be below a threshold and an acoustic sensor detects a sound consistent with snoring.

6. The implantable neurostimulator of claim 1, wherein the control circuit is in electrical communication with therapy delivery circuit and causes the therapy delivery circuit to deliver electrical energy to the bipolar electrode upon determining that the EMG signal is below a threshold and a temperature sensor detects a body temperature consistent with sleep.

7. The implantable neurostimulator of claim 1, wherein the control circuit is in electrical communication with therapy delivery circuit and causes the therapy delivery circuit to deliver electrical energy to the bipolar electrodes when the EMG signal is determined to be below a threshold and a respiration rate consistent with sleep is detected by a respiration rate sensor.

8. A system, comprising:

an Implantable Neurostimulator (INS) comprising a lead having at least a pair of bipolar electrodes and a pulse generator in electrical communication with the bipolar electrodes, the pulse generator comprising a sensor, a memory, a control circuit, and a telemetry circuit;

an external programmer in communication with the INS through the telemetry circuitry;

a server in communication with the external programmer and containing an application thereon configured to receive sensor data from the INS from the external programmer and to evaluate sleep quality of a patient in which the INS is implanted based on the received sensor data; and

a remote computer in communication with the server and configured to present an assessment of the sleep quality of the patient.

9. The system of claim 8, further comprising a user interface presented on the external programmer and configured to receive various self-reported data entered by the patient.

10. The system of claim 9, wherein the application is further configured to assess the sleep quality of the patient in which the INS is implanted based on the received sensor data and the self-reported data.

11. The system of claim 10, wherein the assessment of the sleep quality is presented in the form of a sleep score.

12. The system of claim 8, wherein the application is configured to evaluate the sleep quality of a patient in which the INS is implanted and determine a set of recommended updated stimulation parameters for the INS based on the received sensor data and self-reported data entered through a user interface on the external programmer.

13. The system of claim 12, wherein the received sensor data includes one or more of extensor muscle tension status, heart rate, blood pressure, blood oxygen saturation, patient temperature, arousal, wakefulness, and electromyography data.

14. The system of claim 12, wherein the updated stimulation parameters are available for review, acceptance, modification, or rejection on the remote computer.

15. The system of claim 14, wherein the updated stimulation parameters are transmitted to the external programmer upon acceptance or modification of the updated stimulation parameters.

16. The system of claim 15, wherein the external programmer transmits the updated stimulation parameters to the INS.

17. A method of providing feedback to an Implantable Neurostimulator (INS), the method comprising:

receiving sensor data from an INS having at least one lead implanted in an extensor muscle of a patient;

receiving self-reported data entered through a user interface;

analyzing the sensor data and the self-reported data to determine a sleep score;

recording the sleep score; and

presenting the sleep score for analysis.

18. The method of claim 17, further comprising providing recommendations for updating stimulation parameters of the INS.

19. The method of claim 18, further comprising updating the INS with updated stimulation parameters to apply stimulation to the patient.

20. The method of claim 19, further comprising analyzing the self-reported data and the sensor data with artificial intelligence to determine whether recovery to the stimulation parameters of the INS prior to the update is required.

Technical Field

The present disclosure relates to a medical device system and method for therapeutic electrical stimulation of the hypoglossal nerve for treatment of obstructive sleep apnea. More specifically, the present disclosure relates to methods of measuring tongue muscle tone and assessing the efficacy of stimulation therapy.

Background

Implantable medical devices capable of delivering electrical stimulation pulses have been proposed and are available for treating various medical conditions, such as cardiac arrhythmias and chronic pain, for example. Obstructive Sleep Apnea (OSA), which encompasses apnea and hypopnea, is a serious condition in which breathing irregularly and repeatedly stops and starts during sleep, resulting in sleep disruption and a reduction in blood oxygen levels. OSA is caused by a complete or partial pharyngeal trapping during sleep. In particular, muscles in the mouth and throat of a patient may relax intermittently, thereby blocking the airway while sleeping. Airflow into the upper airway may be blocked by the tongue or soft palate, which moves to the back of the throat and covers less than the normal airway. When one attempts to breathe in the event of an airway obstruction, the loss of air flow can also result in abnormal transthoracic pressure. Lack of sufficient levels of oxygen during sleep can lead to increased heart rhythm abnormalities, heart attacks, heart failure, hypertension, stroke, memory problems, and accidents. In addition, sleep deficits occur when a person is awakened during an apnea event. Implantable medical devices capable of delivering electrical stimulation pulses have been proposed for treating OSA by electrically stimulating muscles surrounding the airway that may block the airway during sleep.

Disclosure of Invention

One aspect of the present disclosure relates to an Implantable Neurostimulator (INS), the INS comprising: an electrical lead having at least one pair of bipolar electrodes formed thereon, wherein the electrical lead is configured for placement of the pair of bipolar electrodes in proximity to an extensor muscle of a patient and configured to receive an Electromyography (EMG) signal; a pulse generator electrically connected to the electrical lead and configured to deliver electrical energy to the pair of bipolar electrodes, the pulse generator having a sensor and a control circuit therein, wherein the sensor and the control circuit are configured to receive the EMG signal and determine a tension state of the extensor tongue muscle in which the lead is placed. Other embodiments of this aspect include corresponding computer systems, apparatus, and computer programs recorded on one or more computer storage devices, each configured to perform the actions of the methods and systems described herein.

Embodiments of this aspect of the disclosure may include one or more of the following features. In the implantable neurostimulator, the control circuit is in electrical communication with a therapy delivery circuit and causes the therapy delivery circuit to deliver electrical energy to the bipolar electrode upon determining that the EMG signal is below a threshold. In the implantable neurostimulator, the control circuit is in electrical communication with a therapy delivery circuit and causes the therapy delivery circuit to deliver electrical energy to the bipolar electrode when it is determined that the EMG signal is below a threshold and the heart rate detected by the sensor is below a threshold. In the implantable neurostimulator, the control circuit is in electrical communication with the therapy delivery circuit and causes the therapy delivery circuit to deliver electrical energy to the bipolar electrode when the EMG signal is determined to be below a threshold and a motion sensor determines that the INS is not moving. In the implantable neurostimulator, the control circuit is in electrical communication with a therapy delivery circuit and causes the therapy delivery circuit to deliver electrical energy to the bipolar electrodes when the EMG signal is determined to be below a threshold and the acoustic sensor detects a sound consistent with snoring. In the implantable neurostimulator, the control circuit is in electrical communication with the therapy delivery circuit and causes the therapy delivery circuit to deliver electrical energy to the bipolar electrode upon determining that the EMG signal is below a threshold and the temperature sensor detects a body temperature consistent with sleep. In the implantable neurostimulator, the control circuit is in electrical communication with a therapy delivery circuit and causes the therapy delivery circuit to deliver electrical energy to the bipolar electrode when it is determined that the EMG signal is below a threshold and a respiration rate sensor detects a respiration rate consistent with sleep.

Additional aspects of the present disclosure relate to a system comprising: an Implantable Neurostimulator (INS) comprising a lead having at least a pair of bipolar electrodes and a pulse generator in electrical communication with the bipolar electrodes, the pulse generator comprising a sensor, a memory, a control circuit, and a telemetry circuit; an external programmer in communication with the INS through the telemetry circuitry; a server in communication with the external programmer and containing an application thereon configured to receive sensor data from the INS from the external programmer and to evaluate sleep quality of a patient in which the INS is implanted based on the received sensor data; and a remote computer in communication with the server and configured to present an assessment of the sleep quality of the patient. Other embodiments of this aspect include corresponding computer systems, apparatus, and computer programs recorded on one or more computer storage devices, each configured to perform the actions of the methods and systems described herein.

Embodiments of this aspect of the disclosure may include one or more of the following features. The system further includes a user interface presented on the external programmer and configured to receive various self-reported data entered by the patient. In the system, the application is further configured to assess the sleep quality of the patient in which the INS is implanted based on the received sensor data and the self-reported data. In the system, the assessment of the sleep quality is presented in the form of a sleep score. In the system, the application is configured to evaluate the sleep quality of a patient in which the INS is implanted based on the received sensor data and the self-reported data entered through a user interface on the external programmer and determine a set of suggested updated stimulation parameters for the INS. In the system, the received sensor data includes one or more of extensor muscle tension status, heart rate, blood pressure, blood oxygen saturation, patient temperature, arousal, wakefulness, and electromyography data. In the system, the updated stimulation parameters are available for review, acceptance, modification, or rejection on the remote computer. In the system, upon accepting or modifying the updated stimulation parameters, the updated stimulation parameters are transmitted to the system of the external programmer. In the system, the external programmer transmits the updated stimulation parameters to the INS.

Still further aspects of the present disclosure relate to a method of providing feedback to an Implantable Neurostimulator (INS), the method comprising receiving sensor data from an INS having at least one lead implanted in an extensor muscle of a patient. The method also includes receiving self-reported data entered through a user interface. The method also includes analyzing the sensor data and the self-reported data to determine a sleep score. The method also includes recording the sleep score. The method also includes presenting the sleep score for analysis. Other embodiments of this aspect include corresponding computer systems, apparatus, and computer programs recorded on one or more computer storage devices, each configured to perform the actions of the methods and systems described herein.

Embodiments of this aspect of the disclosure may include one or more of the following features. The method further includes providing a recommendation for updating the INS' stimulation parameters. The method further includes transmitting the updated stimulation parameters to an external programmer associated with the INS and updating the INS. The method further includes analyzing the self-reported data and the sensor with artificial intelligence to determine whether a recovery to the stimulation parameters of the INS prior to the update is required. Implementations of the described technology may include hardware, methods or processes, or computer software on a computer-accessible medium including software installed on a system that in operation causes the system to perform actions, firmware, hardware, or a combination thereof. One or more computer programs may be configured to perform particular operations or actions by virtue of containing instructions that, when executed by a data processing apparatus, cause the apparatus to perform the actions.

Drawings

Fig. 1 is a conceptual diagram of an Implantable Neurostimulator (INS) for delivering OSA therapy;

FIG. 2 is a conceptual diagram of a pulse generator included in the INS of FIG. 1;

fig. 3 is a diagram of a distal end portion of a lead of fig. 1 deployed for delivering an INS for OSA therapy, according to an aspect of the present disclosure;

fig. 4 is a diagram of a distal end portion of a dual-lead INS deployed for delivering OSA therapy, according to a further aspect of the present disclosure;

fig. 5 is a timing diagram illustrating a method performed by the system of fig. 1 for delivering selective stimulation to the extensor muscles for promoting upper airway patency during sleep, according to one example; and is

Fig. 6 is a timing diagram of a method for delivering OSA therapy by the system of fig. 1, according to another example.

FIG. 7A depicts an electromyography plot of the tongue versus pharyngeal pressure during breathing;

FIG. 7B depicts the change in electromyographic activity from awake to sleep in both a normal subject and those with OSA;

FIG. 8 depicts electromyography of the tongue in different sleep and awake states;

FIG. 9 depicts the steps of signal processing required to convert a raw electromyography signal into a usable data stream in accordance with the present disclosure;

FIG. 10 depicts a simplified system for collecting, transmitting and analyzing data originating from or destined for an INS;

FIG. 11 depicts a flow diagram for collecting, transmitting and analyzing data originating from or destined for an INS.

Detailed Description

An Implantable Neurostimulator (INS) for delivering electrical stimulation to the lingual muscle of the tongue, particularly the extensor tongue muscle, for treating OSA is described herein. Electrical stimulation is delivered to leave the patient's tongue in a protruding state during sleep to avoid or reduce upper airway obstruction. As used herein, the term "protruding state" with respect to the tongue refers to a position that moves forward and/or downward as compared to an unstimulated or relaxed position. Those skilled in the art will recognize that in the protruding state it is not necessary for the tongue to come out of the patient's mouth, and in fact it is preferred that the tongue not extend out of the patient's mouth, but merely advance forward to the point where the airway obstruction is reduced or eliminated. A protrusive state is a state associated with recruitment of the anterior process muscle of the tongue (sometimes also referred to as the "protrusive" muscle of the tongue), which includes the genioglossus muscle and the geniohyoid muscle. The protruding state may be opposite a retracted and/or elevated position associated with the recruitment of retracting and elevating retractor muscles of the tongue, e.g., the glossogyne and hyoglossus muscles. Electrical stimulation is delivered to move the tongue and maintain the protrusive state to prevent trapping, opening, or widening of the patient's upper airway to facilitate unrestricted or at least reduced airflow during breathing.

Current INS systems must be manually turned on and off by the patient when he is ready to sleep and wake up. As can be appreciated, manual switching is not always a desirable feature of implantable devices associated with sleep. According to one aspect of the disclosure, turning on an INS is only required when there is a loss of tongue muscle tone (i.e., the prominent muscle is not naturally stimulated sufficiently). Loss of tongue muscle tone increases the susceptibility of a patient to experience OSA events. Accordingly, one aspect of the present disclosure relates to systems and methods for assessing muscle tone of a projecting muscle and applying a desired therapy based on determining a tension state of a patient in need of the therapy. The result is a therapy system that will be more "natural" and convenient to the patient and improve therapy compliance.

Additional aspects of the present disclosure relate to systems and methods for developing patient feedback sleep scores for counseling, stimulus modification, and healthcare compensation support using sensed extensor muscle tension states in conjunction with various self-reported and detected patient data.

Fig. 1 is a conceptual diagram of an Implantable Neurostimulator (INS) for delivering OSA therapy. The INS system 10 includes at least one electrical lead 20 and a pulse generator 12. The pulse generator 12 includes a housing 15 that encloses circuitry including control circuitry, therapy delivery circuitry, optional sensors, a battery, and telemetry circuitry, as described below in connection with fig. 2. The connector assembly 17 is hermetically sealed to the housing 15 and includes one or more connector apertures for receiving at least one medical electrical lead for delivering OSA therapy and, in some instances, for sensing a physiological condition such as an Electromyography (EMG) signal. As depicted in fig. 1, the pulse generator 12 is implanted in the neck of the patient 8. The present disclosure is not so limited and the pulse generator 12 may be located in other locations, such as the chest area or other areas known to those skilled in the art.

The lead 20 includes a flexible, elongate lead body 22 extending from a lead proximal end 24 to a lead distal end 26. At least two electrodes 30 are carried along the lead distal end portion adjacent the lead distal end 26 configured for insertion within the extensor muscles 42a, 42b, and 46 of the patient's tongue 40. The electrodes 30 are configured for implantation within soft tissue, such as muscle tissue near the medial branch of one or both hypoglossal nerves (HGNs) innervating the extensor muscles of the tongue. The electrodes may be placed about 5mm (e.g., 2mm to 8mm) from the stem of the HGN. As such, the electrodes 30 may be referred to herein as "intramuscular electrodes," as opposed to electrodes placed on or along a nerve trunk or branch, such as cuff electrodes, for directly stimulating the nerve trunk or branch. Lead 20 may be referred to herein as an "intramuscular lead" because the lead distal end and electrode 30 are configured for advancement through soft tissue, which may contain extensor tissue, to anchor electrode 30 near the HGN branch innervating extensor muscles 42a, 42b, and 46. However, the term "intramuscularly" with respect to electrode 30 and lead 20 is not intended to be limiting, as electrode 30 may be implanted in connective or other soft tissue near the medial HGN and its branches. One or more electrodes 30 may be placed in the extensor muscles 42a, 42b and 46 containing the points of movement where each nerve axon terminates in a region of the muscle (also referred to as the nerve-muscle junction). The motion points are not located at one location, but rather spread out in the extensor muscles. Lead 20 may be implanted such that one or more electrodes 30 may be generally located in the area of the motion point (e.g., such that the motion point is within 301mm to 10mm from one or more electrodes).

The extensor tongue muscle is activated by electrical stimulation pulses generated by pulse generator 12 and delivered through intramuscular electrodes 30 to move tongue 40 forward to promote a reduction in obstruction or narrowing of upper airway 6 during sleep. As used herein, the term "activation" with respect to electrical stimulation of the extensor muscles refers to electrical stimulation that causes depolarization or action potentials of cells of the nerve innervating the extensor muscles and motor points (e.g., the hypoglossal nerve) and subsequent depolarization and mechanical contraction of the extensor muscle cells. In some cases, the muscle may be directly activated by the electrical stimulation pulse. The extensor muscles that may be activated by means of stimulation by intramuscular electrode 30 may comprise at least one or both of right and/or left genioglossus muscle (GG)42 and/or right and/or left genioglossus muscle (GH)46, said GG comprising oblique compartment (GGo)42a and horizontal compartment (GGh)42b (collectively referred to as GG 42). The GG and GH muscles are innervated by the medial branch of the HGN (also known as the XII cranial nerve), while the hyoglossus and glossogyne muscles, which cause the tongue to retract and elevate, are innervated by the lateral branch of the HGN.

The plurality of distal electrodes 30 may be used to deliver bilateral or unilateral stimulation to GG 42 and/or GH 46 muscles through the medial branch of the HGN or a medial branch thereof (also referred to herein as a "medial HGN"). The distal electrode 30 may be switchably coupled to the output circuitry of the pulse generator 12 to enable delivery of electrical stimulation pulses in a manner that selectively activates the right and left extensor muscles in a cyclical or alternating pattern while maintaining upper airway patency to avoid muscle fatigue. Additionally or alternatively, electrical stimulation may be delivered to selectively activate GG 42 and/or GH 46 muscles or portions thereof during unilateral stimulation of the left or right extensor muscles.

The lead proximal end 24 includes a connector (not shown in fig. 1) that can be coupled to the connector assembly 17 of the pulse generator 12 to provide an electrical connection between circuitry enclosed by the housing 15 of the pulse generator 12, for example, including therapy delivery circuitry and control circuitry as described below in connection with fig. 2. The lead body 22 encloses electrical conductors extending from each of the distal electrodes 30 to a proximal connector at the proximal end 24 to provide an electrical connection between the output circuitry of the pulse generator 12 and the electrodes 30.

Although shown in fig. 1 as being separate from and extending from the pulse generator 12, the leads 20 may be integrated into a portion of the pulse generator 12 and may be merely an exposed surface of the pulse generator 12. In this embodiment, the pulse generator would be implanted near the patient's tongue muscle under the chin. In contrast, the embodiment shown in fig. 1 allows for more flexibility in placing the pulse generator in the neck or chest region of the patient.

Fig. 2 is a schematic diagram of the pulse generator 12. The pulse generator 12 includes control circuitry 80, memory 82, therapy delivery circuitry 84, sensors 86, telemetry circuitry 88, and a power supply 90. The power supply 90 may contain one or more rechargeable or non-rechargeable batteries for supplying current to each of the control circuitry 80, memory 82, therapy delivery circuitry 84, sensors 86, and telemetry circuitry 88. Although only the power supply 90 is shown in communication with the control circuit 80 for clarity, it should be understood that the power supply 90 provides the required power to each of the circuits and components of the pulse generator 12 as needed. For example, the power supply 90 provides power to the therapy delivery circuit 84 for generating electrical stimulation pulses.

The sensors 86 may include one or more individual sensors for monitoring the condition of the patient. These sensors may include one or more accelerometers, Inertial Measurement Units (IMUs), fiber bragg gratings (e.g., shape sensors), optical sensors, acoustic sensors, pulse oximeters, and the like, without departing from the scope of the present disclosure and as will be described in more detail below. In one aspect of the present disclosure, the sensors 86 are configured as, among other things, patient posture sensors. When included, the sensors 86 may store patient posture data in the memory 82 in accordance with the detected posture state of the patient, and may present the patient posture data on a display of the external programmer 50, such as that generally described in U.S. patent No. 9,662,045 (Skelton et al), which is incorporated by reference in its entirety.

Additionally or alternatively, the sensor 86 may detect signals related to movement of the patient's tongue into and out of the protruding state. This signal may be used to detect sufficient protrusion and/or fatigue of the stimulated muscle for controlling the duty cycle, pulse amplitude, and/or electrode vector stimulating the electrical stimulation therapy delivered by therapy delivery circuit 84.

The functional blocks shown in fig. 2 represent functionality included in the pulse generator 12 configured to deliver OSA therapy, and may include any discrete and/or integrated electronic circuit components implementing analog and/or digital circuitry capable of producing the functionality attributed to the pulse generator herein. Various components may include an Application Specific Integrated Circuit (ASIC), an electronic circuit, a processor (shared, dedicated, or group) and memory that execute one or more software or firmware programs, a combinational logic circuit, a state machine, or other suitable components or combinations of components that provide the described functionality. Given the disclosure herein, it is within the ability of one skilled in the art to provide software, hardware, and/or firmware to implement the described functionality in the context of any modern medical device system.

The control circuit 80 communicates with the memory 82, therapy delivery circuit 84, telemetry circuit 88, and sensors 86 (when included), for example, over a data bus, to control OSA therapy delivery and other pulse generator functions. As disclosed herein, the control circuitry 80 may communicate control signals to the therapy delivery circuitry 84 to cause the therapy delivery circuitry 84 to deliver electrical stimulation pulses through the electrodes 30 according to a therapy regime that may include selective stimulation patterns of right and left portions of the GG and GH muscles and/or proximal and distal portions of the GG and GH muscles. Control circuit 80 may be further configured to communicate therapy control signals to therapy delivery circuit 84 that include stimulation pulse amplitude, stimulation pulse width, stimulation pulse number, and frequency of stimulation pulse trains.

The memory 82 may store instructions for execution by a processor included in the control circuit 80, stimulation control parameters, and other device-related or patient-related data. The control circuit 80 can retrieve the therapy delivery control parameters and the therapy delivery protocol from the memory 82 to enable the control circuit 80 to pass control signals to the therapy delivery circuit 84 for controlling OSA therapy. The memory 82 may store historical data related to therapy delivery for retrieval by a user through the telemetry circuitry 88. The therapy delivery data or information stored in memory 82 may include therapy control parameters for delivering stimulation pulses as well as delivered therapy regimen, hours of therapy delivery, and the like. Patient-related data, such as data received from the sensor 86 signals, may be stored in the memory 82 for retrieval by the user.

Therapy delivery circuit 84 may include a charging circuit 92, an output circuit 94, and a switching circuit 96. Charging circuit 92 may include one or more holding capacitors that are charged using, for example, a multiple of the battery voltage of power supply 90. The holding capacitors are switchably connected to an output circuit 94, which may include one or more output capacitors coupled to selected bipolar electrode pairs by switching circuits 96. The hold capacitor is charged to the programmed pacing pulse voltage amplitude by the charging circuit 92 and discharged across the output capacitor for the programmed pulse width. The charging circuit 92 may include a capacitor charge pump or amplifier for a charge source to enable rapid recharging of the holding capacitor included in the charging circuit 92. Therapy delivery circuit 84 reacts to the control signal from control circuit 80 to generate and deliver a pulse train to produce a sustained tonic contraction of the GG and/or GH muscles, or portions thereof, to move the tongue forward and avoid upper airway obstruction.

The output circuit 94 may be selectively coupled to the bipolar electrode pairs 30a-30d by a switching circuit 96. Switching circuit 96 may include one or more switches that are activated by timing signals received from control circuit 80. Electrodes 30a-30d may be selectively coupled to output circuit 94 in a time-varying manner to deliver stimulation to different portions of the extensor muscle at different times to avoid fatigue without the need to completely suppress stimulation. The switching circuitry 96 may include a switch array, switch matrix, multiplexer, or any other type of switching device suitable for selectively coupling the therapy delivery circuit 84 to bipolar electrode pairs selected from the electrodes 30. The bipolar electrode pairs may be selected one at a time, or two or more at a time, to allow for overlapping stimulation of two or more different portions of the extensor muscle. The number of overlapping stimulations of two portions of the extensor tongue (e.g., left and right or proximal and distal) may maintain the anterior positioning of the tongue and allow for the ramping up and down of the electrical stimulation delivered to the two different portions of the extensor tongue.

Telemetry circuitry 88 is optional, but may be included to enable bi-directional communication with external programmer 50. A user, such as the Patient 8, may manually adjust therapy control parameter settings as described, for example, in Medtronic Patient Programmer Model 37642, which is incorporated by reference in its entirety. The patient may make limited programming changes, such as small changes in stimulation pulse amplitude and pulse width. The patient may turn therapy on and off or set a timer to turn therapy on or off using the external programmer 50 in wireless telemetry communication with the telemetry circuitry 88.

In other instances, a user, such as a clinician, may interact with a user interface of the external programmer 50 to program the pulse generator 12 according to a desired OSA therapy protocol. For example, a Physician may program the pulse generator 12 to deliver electrical stimulation using a Physician Programmer Model (physical Programmer Model)8840 available from Medtronic, inc.

Programming of pulse generator 12 may generally refer to generating and communicating commands, programs, or other information for controlling the operation of pulse generator 12. For example, the external programmer 50 may transmit programs, parameter adjustments, program selections, group selections, or other information to control the operation of the pulse generator 12, e.g., via wireless telemetry. As one example, external programmer 50 may transmit parameter adjustments to support therapy changes. As another example, a user may select a program or group of programs. The procedure may be characterized by a pattern of electrode selection of electrode combinations, electrode polarities, voltage or current amplitudes, pulse widths, pulse rates, therapy durations, and/or patterns for delivering alternating portions of the extensor muscle being stimulated. A group may be characterized by delivering multiple programs simultaneously or in an interleaved or rotating fashion. These programs may adjust output parameters or turn therapy on or off at different time intervals.

The external programmer 50 may present patient-related and/or device-related data retrieved from the memory 82 via the telemetry circuitry 88. Additionally or alternatively, external programmer 50 may present sleep sounds or movement data stored in memory 82 as determined from signals from sensors 86. Further, the time period for which the patient is lying down may be acquired based on patient posture detection using sensor 86, and a history of this data may be stored in memory 82 and retrieved and displayed by external programmer 50.

Fig. 3 depicts a single intramuscular lead 20 inserted into the patient's tongue 40. Lead 20 may include two or more electrodes, and in the example shown, lead 20 includes four electrodes 30a, 30b, 30c, and 30d (collectively "electrodes 30") spaced longitudinally along lead body 22. The lead body 22 is a flexible lead body that can define one or more lumens in which insulated electrical conductors extend to the respective electrodes 30a-30 d. Distal-most electrode 30a may be adjacent to or near the lead distal end 26. Each of the electrodes 30b, 30c, and 30d is spaced proximally from a respective adjacent electrode 30a, 30b, and 30c by a respective inter-electrode distance 34, 35, and 36.

Each electrode 30a-30d is shown having an equivalent electrode length 31. However, in other examples, the electrodes 30a-30d may have different electrode lengths 31 from one another in order to optimize placement of the electrodes 30 or the resulting stimulation electric field relative to the target stimulation site corresponding to the motion points of the HGN or the left and right portions of its branches and/or the GG and GH muscles. The inter-electrode spacing between electrodes 30a, 30b, 30c and 30d is shown in fig. 3 as being approximately equal, however, they may also be different from one another to optimize placement of electrodes 30 relative to a target stimulation site or placement of the resulting stimulation electric field relative to a target stimulation site corresponding to the branches of the left and right hypoglossal nerves or hypoglossal nerves and/or the motor points of the extensor muscles 42a, 42 or 46.

In some examples, electrodes 30a and 30b form an anode and cathode pair for delivering bipolar stimulation in a portion of the extensor muscle, e.g., a proximal portion or a distal portion of the left or right GG and/or GH muscle or GG and/or GH muscles. Electrodes 30c and 30d may form a second anode and cathode pair for delivering bipolar stimulation in a different portion of the extensor muscle (e.g., the other of the left or right portions or the other of the proximal or distal portions). Thus, the inter-electrode spacing 35 between the two bipolar pairs 30a-30b and 30c-30d may be different than the inter-electrode spacing 34 and 36 between the anode and cathode within each bipolar pair 30a-30b and 30c-30 d.

In one example, the total distance covered by the electrodes 30a-30d along the lead body 22 can be, for example, about 20 millimeters, 25 millimeters, or 30 millimeters. In one example, the total distance is between 20 millimeters and 22 millimeters. The inter-electrode spacing between the proximal electrode pair 30c-30d and the distal electrode pair 30a-30b, respectively, may be between 2mm and 6mm, including all integer values therebetween. The inter-electrode spacing separating the distal and proximal pairs 30a-30b and 30c-30d may be the same or different from each other and may be the same or different from the spacing between the individual electrodes of any such pair.

In the example shown, each of the electrodes 30a-30d is shown as a circumferential ring electrode that may be consistent in diameter with the lead body 22. In other examples, the electrode 30 may comprise other types of electrodes, such as tip electrodes, spiral electrodes, coil electrodes, segmented electrodes, button electrodes, for example. For example, distal-most electrode 30a may be provided as a tip electrode at lead distal end 26, while the remaining three electrodes 30b, 30c, and 30d are ring electrodes. When electrode 30a is positioned at distal end 26, electrode 30a may be a helical electrode configured to be screwed into muscle tissue at the implantation site to otherwise serve as a fixation member for anchoring lead 20 at the targeted therapy delivery site. In other examples, one or more of electrodes 30a-d may be a hooked or barbed electrode for providing active fixation of lead 20 at a therapy delivery site.

The lead 20 may include one or more fixation members 32 for minimizing the likelihood of lead migration. In the example shown, the fixation member 32 includes multiple sets of tines that engage the surrounding tissue when the lead 20 is positioned at the target therapy delivery site. The tines of the fixation member 32 can extend radially and proximally at an angle relative to the longitudinal axis of the lead body 22 to prevent or reduce retraction of the lead body 22 in the direction of the proximal end. The tines of the fixation member 32 may be trapped against the lead body 22 when the lead 20 is held within the boundaries of a lead delivery tool (e.g., a needle or introducer) used to deploy the lead 20 at a target implantation site. After removal of the lead delivery tool, the tines of fixation member 32 can be extended to a normally extended position to engage the surrounding tissue and resist proximal and lateral migration of lead body 22. In other examples, fixation member 32 can include one or more hooks, barbs, spirals, or other fixation mechanisms extending from one or more longitudinal locations along lead body 22 and/or lead distal end 26. The fixation member 32 may partially or entirely engage GG, GH muscle, and/or other muscles under the tongue and/or other soft tissue of the neck, such as adipose tissue and connective tissue, as the proximal end of the lead body 20 is tunneled into the implant pocket of the pulse generator 12. In other examples, fixation member 32 may include one or more fixation mechanisms located at other positions than that shown in fig. 3, including being located at or near distal end 26, between electrodes 30, or otherwise further or closer than that shown. The implant pocket of the pulse generator 12 may be located in the chest along the patient's neck 8 (see fig. 1) or at another location as deemed appropriate by the surgeon performing the implantation. Thus, the length of the elongate lead body 22 from the distal end 26 to the lead proximal end 24 (fig. 1) can be selected to extend from the target therapy delivery site in the extensor muscle to the location along the patient's neck where the pulse generator 12 is implanted. By way of example, this length may be up to 10cm or up to 20cm, but typically may be 25cm or less, although longer or shorter lead body lengths may be used, depending on the anatomy and size of the individual patient.

Fig. 2 is a conceptual diagram 120 of a lead 20 deployed for delivering OSA therapy according to another example. In this example, the lead 20 carrying the electrode 30 is advanced approximately along or parallel to the midline 102 of the tongue 40. In the illustrated example, the lead body 22 is shown approximately centered along the centerline 102, however in other examples, the lead body 22 may be laterally offset from the centerline 102 in a left or right direction, but generally inboard of the left HGN 104L and the right HGN 104R. The distal end 26 of the lead 20 may be inserted into the body of the tongue 40 below, for example, in the musculature below the floor of the oral cavity at a percutaneous insertion point along the submandibular triangle. The distal end 26 is advanced to position the electrode 30 inside the left HGN 104L and the right HGN 104R, e.g., approximately in the middle of the genioglossus eminence (chin). The electric field generated by the stimulation pulses delivered between any bipolar electrode pair selected from electrodes 30 may encompass a portion of both the left target region 106L and the right target region 106R to produce bilateral stimulation of the HGNs 104L and 104R and thus bilateral recruitment of the extensor muscles. Bilateral recruitment of the extensor muscles may provide a larger airway opening than unilateral stimulation typically performed along the HGN using nerve cuff electrodes. For example, electrical stimulation pulses delivered using electrodes 30a and 30b may generate an electric field 122 (shown conceptually) encompassing a portion of both left side target region 106L and right side target region 106R. The electrical stimulation pulses delivered using electrodes 30c and 30d may generate an electric field 124 (shown conceptually) that encompasses a portion of both left target region 106L and right target region 106R. The portion of the left target region 106L and the right target region 106R encompassed by the electric field 122 is a posterior portion with respect to the portion of the left target region 106L and the right target region 106R encompassed by the electric field 124.

In some examples, the electrical stimulation is delivered by the pulse generator 12 by sequentially selecting different pairs of electrodes from the available electrodes 30 to sequentially recruit different bilateral anterior portions and bilateral posterior portions of the HGNs 104L and 104R. This electrode selection may result in the recruitment of different anterior and posterior portions of the extensor tongue. The sequential selection of different electrode pairs may or may not overlap. The electrical stimulation is delivered over an extended period of time, independent of the timing of the breathing cycle, covering multiple breathing cycles, to maintain the tongue 40 in a prominent state from the beginning of the period to the end of the period. The electrodes 30 may be selected to be in the form of bipolar pairs, including the most distal pair 30a and 30b, the most distal pair 30a and 30d, the most inner pair 30b and 30c, the most proximal pair 30c and 30d, or alternating electrodes, such as 30a and 30c or 30b and 30d, along the lead body 22. The sequential selection of two or more different electrode pairs allows for the sequential recruitment of different portions of the extensor muscle to reduce the likelihood of fatigue.

In some examples, electrical stimulation delivered using electrode pairs (e.g., 30a and 30b) implanted relatively farther along the distal lead portion 28 and relatively anterior along the tongue 40 may recruit anterior muscle fibers, e.g., within the GG muscles. Electrical stimulation delivered using electrode pairs (e.g., 30c and 30d) implanted relatively closer along the distal lead portion 28 and relatively posterior along the tongue 40 may recruit anterior muscle fibers, e.g., within the GH muscle. The sequential selection of electrodes 30 for delivering electrical stimulation pulses allows for the sequential recruitment of anterior and posterior portions of the extensor tongue muscle in an overlapping or non-overlapping pattern to maintain the tongue in a prominent state throughout an extended period of time while reducing or avoiding muscle fatigue.

Fig. 4 is a conceptual diagram of a distal portion of a dual lead system for delivering OSA therapy. In this example, one lead 20 is advanced approximately parallel to the midline 102 and offset to the left of the midline 102 by, for example, 5-8 millimeters to position the distal portion 28 and electrodes 30 in or adjacent to the left target area 106L. The second lead 220 is advanced approximately parallel to the midline 102 but laterally offset to the right of the midline 102 to position the distal portion 228 and the electrode 230 in or adjacent the right target region 106R. The lead 20 may be inserted from a left lateral or posterior approach to the body of the tongue 40, and the lead 230 may be inserted from a right lateral or posterior approach to the body of the tongue 40. In other examples, both leads 20 and 220 may be inserted from only a left or only a right approach, with one lead transverse to the midline 102 to position the electrode 30 or 230 from the proximal side along the opposite side of the midline 102. Lead 20 and/or lead 220 may be advanced at an oblique angle relative to midline 102, but may not cross midline 102. In other examples, one or both leads 20 and 220 may approach and cross midline 102 at an oblique angle such that one or both of distal portions 28 and 228 extend in or adjacent to both right and left target areas 106L and 106R, similar to the orientation in fig. 6.

In the example shown, relatively more local control over recruitment of left, right, anterior and posterior portions of the extensor tongue can be achieved by selecting different pairs of electrodes from electrodes 30a through 30d and 230a through 230 d. For example, any combination of electrodes 30a-30d may be selected to deliver electrical stimulation pulses to the left portion of the extensor muscle. More distal electrodes 30a and 30b may be selected to stimulate more anterior portions of the left extensor muscle (corresponding to electric field 144) and more proximal electrodes 30c and 30d may be selected to stimulate more posterior portions of the left extensor muscle (corresponding to electric field 142). Any combination of electrodes 230a through 230d may be selected to deliver electrical stimulation pulses to the right portion of the extensor muscle. More distal electrodes 230a and 230b may be selected to stimulate more anterior portions of the right extensor muscle (corresponding to electric field 154), and more proximal electrodes 230c and 230d may be selected to stimulate more posterior portions of the right extensor muscle (corresponding to electric field 152).

Switching circuit 96 may be configured to select an electrode pair comprising one electrode located on one of leads 20 or 220 and another electrode located on the other lead 20 or 220 to generate an electric field (not shown) encompassing a portion of both left side target region 106L and right side target region 106R for bilateral stimulation. Any combination of available electrodes 30a-30d and electrodes 230a-230 d may be selected to select in a repeating, sequential pattern to sequentially recruit two or more dipole pairs of different portions of the two target regions 106L and 106R. The sequential selection of electrode pairs may be overlapping or non-overlapping, but the electrical stimulation pulses are delivered uninterrupted throughout an extended period of time at one or more selected frequencies to maintain the tongue 40 in a protruding state from the beginning of the period to the end of the period, which encompasses multiple respiratory cycles.

In the example of fig. 4, which includes two leads, two electrode pairs may be selected simultaneously and sequentially with one or more other electrode pairs. For example, electrodes 30a and 30b may be selected as one bipolar pair and electrodes 230c and 230d may be selected as a second bipolar pair for simultaneously stimulating a left anterior portion of target region 106L and a right posterior portion of target region 106R. Electrodes 30c and 30d may be selected as the next bipolar pair from lead 20, while electrodes 230a and 230b are selected as the next bipolar pair from lead 220. In this way, electrical stimulation may be delivered bilaterally, alternating between the posterior and anterior regions of each side. The anterior left (30a and 30b) and posterior right (230c and 230d) bipolar pairs may be selected first, and then the posterior left (30c and 30d) and anterior right (230a and 230b) bipolar pairs may be selected in a repeating alternating manner to continuously maintain the tongue 40 in the prominent state for an extended period of time encompassing multiple respiratory cycles. In other examples, both the front pairs (30a-30b and 230a-230b) may be selected simultaneously first, and then the back pairs (30c-30d and 230c-230d) may be selected simultaneously sequentially after the front pairs. In this way, sustained bilateral stimulation may be achieved while sequentially alternating between posterior and anterior portions to avoid or reduce fatigue. In contrast to other OSA therapy systems that rely on sensors for sensing the inspiratory phase of breathing to coordinate therapy with inhalation, the intramuscular electrodes 30 positioned to stimulate different portions of the extensor tongue muscle need not be synchronized with the breathing cycle. The alternation of the stimulation site within the extensor allows different parts of the muscle to rest while other parts are activated to avoid the tongue from being trapped against the upper airway, while also avoiding muscle fatigue.

It should be understood that more or fewer than the four electrodes shown in the examples presented herein may be included along the distal portion of the lead used in conjunction with the OSA therapy techniques disclosed herein. A lead carrying multiple electrodes for delivering OSA therapy may contain 2,3, 5, 6, or other selected number of electrodes. When the leads include only two electrodes, a second lead having at least one electrode may be included to provide at least two different bipolar electrode pairs for sequential stimulation of different portions of the right and/or left medial HGN. Further, while the selected electrode pair is generally referred to herein as a "bipolar pair" comprising one cathode and one return anode (return anode), it should be recognized that three or more electrodes may be selected at a time to provide a desired electric field or stimulation vector for recruiting a desired portion of the extensor muscle. Thus, the cathode of a bipolar "pair" may comprise one or more electrodes simultaneously selected from the available electrodes and/or the anode of a bipolar "pair" may comprise one or more electrodes simultaneously selected from the available electrodes.

FIG. 5: a timing diagram of a method performed by pulse generator 12 for delivering selective stimulation to the extensor digitorum anteriorum during sleep to promote upper airway patency according to one example is presented. The electrical stimulation is delivered over a therapy time period 401 having a start time 403 and an end time (not shown). The electrical stimulation pulses delivered when the pulse generator sequentially selects the first bipolar electrode pair 402 and the second bipolar electrode pair 412 in an alternating, repeating manner are shown. The first bipolar electrode pair 402 and the second bipolar electrode pair 412 may correspond to any two different electrode pairs described above in connection with the example of fig. 3-4.

The first electrical pulse train 406 is shown to begin at the onset 403 or therapy period 401. A first electrical pulse train 406 is delivered using the bipolar electrode pair 402 during the duty cycle time interval 404. The first electrical pulse train 406 has a pulse amplitude 405 and a pulse frequency, e.g., 20Hz to 50Hz, defined by an inter-pulse interval 407. A first electrical pulse train 406, also referred to as "pulse train" 406, may have a ramp-up portion 408 during which the pulse amplitude gradually increases from a starting voltage amplitude up to a pulse voltage amplitude 405. In other examples, the pulse width may be gradually increased. In this manner, the delivered pulse energy is gradually increased to facilitate a gentle transition from the relaxed, non-stimulated state to the protruding state of the tongue.

The electrical pulse train 406 may include a ramp-off portion 410 during which the pulse amplitude (and/or pulse width) is decremented from the pulse voltage amplitude 405 to an end amplitude at the expiration of the duty cycle time interval 404. In other examples, the burst 406 may include a ramp-up portion 408 and a non-ramp-down portion 410. In this case, the last pulse of the pulse train 406 delivered at the expiration of the duty cycle time interval 404 may be delivered at the full pulse voltage amplitude 405. Delivery of electrical stimulation through the bipolar electrode pair 402 is terminated upon expiration of the duty cycle time interval 404.

In the example shown, the second electrode pair 412 is selected when the duty cycle time interval 404 expires. The second electrode pair 412 may be selected such that delivery of the electrical stimulation pulse train 416 begins a ramp-up portion 418 that occurs simultaneously with the ramp-down portion 410 of the pulse train 406. In other examples, the ramp-up portion 418 of the burst 416 may begin when the first duty cycle time interval 404 expires. When the burst 406 does not contain the ramp down portion 410, the burst 416 may begin such that the ramp up portion 418 ends only before, only after, or both before and at the same time the duty cycle time interval 404 expires. The second pulse train 416 has a duty cycle time interval 414 of duration and may end in an optional ramp down portion 420, which may overlap with a ramp up portion of a next pulse train delivered using the first electrode pair 402.

In this example, pulse trains 406 and 416 are shown to be equivalent in terms of amplitudes 405 and 415, pulse width, pulse frequency (and inter-pulse interval 407), and duty cycle time intervals 404 and 414. However, it is contemplated that each of the stimulation control parameters used to control the delivery of the sequential pulse trains 406 and 416 may be controlled individually and set to different values as needed to achieve the desired sustained protrusion of the tongue 40 while avoiding or minimizing fatigue.

Sequential pulse trains 406 and 416 are delivered using two different electrode pairs 402 and 412, such that different portions of the extensor muscle are recruited by the pulse trains 406 and 416, allowing one portion to rest while the other portion is stimulated. However, the pulse trains 404 and 406 occur in a sequential overlapping or non-overlapping manner such that the electrical pulses are delivered at one or more selected frequencies for the entire duration of the therapy session 401 to maintain the tongue in the protruding state 401 for the entire duration. It should be understood that the relative downward and/or forward positioning of the protruding tongue may move or change as different electrode pairs are selected, but the tongue remains in the protruding state throughout the therapy session 401.

At times, the pulse trains 404 and 406 may overlap to simultaneously recruit the left GG muscle and/or the right GH muscle to produce a relatively greater force (as compared to recruiting a unilateral) to pull the tongue forward to open the blocked upper airway. In some cases, overlapping pulse trains 404 and 406 may cause temporary fatigue of the extensor muscles along the left or right side, but temporary fatigue may improve therapy effectiveness to ensure that the upper airway is open during an apneic event. Recovery from fatigue will occur during the duty cycle and the end of the apnea event. Depending on the fatigue characteristics of individual patients, the duty cycle length may vary from patient to patient depending on the fatigue properties of the individual patient. The control circuit 80 may control the duty cycle on time in a manner that minimizes or avoids fatigue and subsequent fatigue in a closed loop system using signals from the sensor 86, such as the motor sensor signal and/or the extensor muscle contraction force associated with the Electromyography (EMG) signal.

Fig. 6 is a timing diagram 500 of a method for delivering OSA therapy by pulse generator 12 according to another example. In this example, the therapy delivery period 501 begins at 503 with a ramp-up interval 506 delivered using the first bipolar electrode pair 502. The ramp-up interval 506 is followed by a duty cycle time interval 504. Upon expiration of the duty cycle time interval 504, a second bipolar electrode pair 512 is selected for delivery of electrical stimulation pulses within a second duty cycle time interval 514. The third duty cycle time interval 524 begins when the second duty cycle time interval 514 expires and a stimulation pulse is delivered by selecting a third bipolar electrode pair 522 that is different from the first two pairs 502 and 512. A fourth bipolar pair 532 is selected upon expiration of the third duty cycle time interval 524 and is used to deliver stimulation pulses within a fourth duty cycle time interval 534. Upon expiration of fourth duty cycle time interval 534, the sequence repeats again beginning with duty cycle time interval 504.

In this example, four different dipole pairs are sequentially selected. The four different bipolar electrode pairs may differ in the polarity of at least one electrode and/or another bipolar electrode pair. For example, when a single quadrupole lead 20 is used, the four dipole pairs may comprise 30a-30b, 30b-30c, 30c-30d, and 30a-30 d. The portions of the extensor muscles recruited by the four different pairs may not be mutually exclusive, as the electric fields of the four different pairs may stimulate some of the same nerve fibers. Four different portions of the extensor tongue muscle may be recruited, which may include overlapping portions. The relatively long recovery phases 540, 542, 544, and 546 between the respective duty cycle intervals allow each different portion of the extensor muscle to recover before the next duty cycle. When the recruited muscle portions overlap between the selected electrode pairs, bipolar electrode pairs may be sequentially selected that avoid continuously stimulating the overlapping recruited muscle portions. All of the recruited muscle portions are allowed to recover during at least a portion of each respective recovery phase 540, 54, 544 and/or 546. For example, if bipolar electrode pair 502 and bipolar electrode pair 522 recruit an overlapping portion of the extensor muscle, the recruited portion may still recover during second duty cycle time interval 514 and during fourth duty cycle time interval 534.

The duration of each duty cycle time interval 504, 514, 524, and 534 may be the same or different from each other, resulting in the same or different overall duty cycles. For example, when four bipolar electrode pairs are selected in sequence, the stimulation delivery for each individual pair may be a 25% duty cycle. In other examples, combinations of different duty cycles may be selected, e.g., 30%, 10%, 40%, and 20%, to promote continued tongue protrusion and adequate airway opening while minimizing or avoiding fatigue. The choice of duty cycle for a given electrode pair choice may depend on the particular muscle or muscle portion being recruited and the associated response (location) of the tongue to the stimulation.

The stimulation control parameters used to deliver electrical pulses using each of the different bipolar electrode pairs 502, 512, 522, and 532 during each of the duty cycle time intervals 504, 514, 524, and 534 may be the same or different. As shown, different pulse voltage amplitudes and different inter-pulse intervals may be used, as well as the resulting pulse train frequency. The pulse amplitude, pulse width, pulse frequency, pulse shape, or other pulse control parameters may be controlled according to the settings selected for each bipolar electrode pair.

In the example shown, a ramp-up portion 506 of the stimulation protocol is shown as being initiated at therapy delivery time period 501. Once the stimulation ramp is to position the tongue in the prominent position, no other subsequent duty cycle time intervals 504 (other than the first), 514, 524, and 534 may contain the ramp portion or may continue through the ramp portion. In other examples, the ramp-up portion may precede (or be included in) each duty cycle time interval (or be included in a duty cycle time interval as shown in fig. 5) and may overlap the preceding duty cycle time interval. The ramp down portion is not shown in the example of fig. 6. In other examples, the ramp up portion may follow each duty cycle time interval 504, 514, 524, and 534 or may be included in each duty cycle time interval and may overlap with the firing of the next duty cycle time interval, as shown in fig. 5. In some examples, only the last duty cycle time interval (not shown in fig. 6) may contain or be followed by a ramp down portion to gently allow the tongue to return to a relaxed position at the end of therapy delivery period 501.

After implantation and calibration by a surgeon or other caregiver as depicted in fig. 3 and 4, the INS 10 is ready for use. In accordance with one aspect of the present disclosure, the INS system 10 is manually turned on by the patient as part of its pre-sleep routine. This may be the function of external programmer 50 or another similar device that may communicate with pulse generator 12 via telemetry circuitry 88. The delay period may be programmed into the software or firmware employed by the control circuit 80. The delay period allows the patient to have a period of falling asleep before the therapy begins. The time period may be established for the patient based on a variety of factors, including the average time to sleep observed during, for example, a sleep study, and may be adjusted by the patient via external programmer 50. Without a delay period, the patient will immediately begin to experience the effect of stimulating the extensor muscles, which, while not dangerous or painful, can be observed and may be considered annoying to experience while awake.

As can be appreciated, manual switching as described herein is not always a desirable feature of implantable devices associated with sleep. In further aspects of the disclosure, OSA therapy can be started and stopped at predetermined times of the day. The control circuit 80 may contain a clock for scheduling the time at which OSA therapy is started and stopped by the therapy delivery circuit 84. However, many patients are not as strict with respect to their scheduling as would be desirable to make the scheduling most effective. Further, a patient may find himself in a social gathering or other business at a time when he is typically scheduled to sleep. Additionally or alternatively, the patient may find himself in a motor vehicle, airplane, or train with an unscheduled nap and have no opportunity to initiate or schedule therapy. Since OSA is often co-morbid with heart-related diseases, any situation that experiences OSA may have a complex factor that affects the patient's heart. Therefore, it is desirable to perceive the sleep condition and initiate therapy. One aspect of the present disclosure relates to a mechanism for initiating therapy based on a detected state of extensor muscle tone.

In accordance with the present disclosure and as noted, the electrodes 30, alone or in combination with the sensor 86, may be configured to detect an Electromyography (EMG) signal. Electromyography is a technique for evaluating and recording the electrical activity produced by skeletal muscles. Electromyography detects the electrical potential produced by muscle cells when the cells are electrically or neuro-activated. Fig. 7A depicts in the top graph EMG signals observed in the patient's genioglossus muscle (GG)42 during normal breathing. The lower graph in fig. 7A depicts pharyngeal pressure during a time period that is the same as the EMG signal in the upper graph. As seen in fig. 7A, during respiration as pharyngeal pressure decreases, the EMG signal increases significantly, consistent with inhalation. Those skilled in the art will recognize that this increase in EMG signal during breathing is indicative of stimulation of the muscles of the tongue, such as the genioglossus muscle (GG)42, ensuring that the airway does not close or collapse, thereby reducing the ability of the subject to gasp. That is, during periods of high EMG signal, the extensor muscle tone state contracts. During the time period when low EMG signals are present, the tension state of the extensor muscles is relaxed.

Fig. 7B depicts a comparison of observed EMG signals observed in the extensor muscles of two groups of subjects. For all subjects, EMG signals may decline when the subject is in a sleep state as compared to a wake state. However, it is evident to the present disclosure that the EMG signal of subjects experiencing OSA events is significantly lower. This decreased EMG signal is evidence of a decreased tension state of the extensor muscles in subjects experiencing OSA. Fig. 8 depicts similar data for comparing EMG signals of a subject during REM, non-REM, quiet wakefulness and active wakefulness. This data confirms that the top line of fig. 7B, i.e., the EMG signal, decreased while asleep, and as indicated in fig. 7B, the decrease was more pronounced in subjects experiencing OSA events.

According to one aspect of the present disclosure, electrodes may be employed to detect the potential of the muscle when no stimulation pulse is delivered through the electrodes 30a-30 d. That is, the electrodes 30a-30d may detect EMG signals applied to the extensor muscles by the patient's nervous system. These signals may be communicated to the control circuit 80 for monitoring and applying rules in the software or firmware stored therein. In other examples, dedicated EMG sensing electrodes may be carried by the housing 15 and/or lead body 22 and coupled to the sensor 86 for EMG signal monitoring. Monitoring of the EMG signal by the control circuit 80 allows detection of a low tension state of the GG and/or GH muscles. Referring to fig. 7A, a low tension state (i.e., low incidence of EMG signals) indicates the likelihood that the patient is asleep and susceptibility to upper airway collapse. The detection of a low tension state of the extensor muscles may be used alone or in combination with other sensor data, such as the detection of a patient posture indicating that the patient is in a recumbent position or the detection of a sleep-consistent heart rate to initiate therapy and prevent the onset of OSA events. Thus, the control circuit 80 may use the EMG signals to detect a sleep state and/or a low tension state of the extensor muscles for controlling the therapy delivery circuit 84 to deliver stimulation pulses to protrude the patient's tongue. As will be appreciated, in the detection of EMG signals, the control circuit 80, or intervening hardware, may employ various band-pass filtering, rectification and normalization to produce a usable signal that provides a clear indication of the state of the extensor muscles. An example of this processing of EMG signals is depicted in fig. 9.

EMG monitoring may further be used to monitor fatigue of stimulated GG and/or GH muscles. If muscle fatigue is detected, the control circuitry 80 may vary to control the duty cycle of the electrical stimulation pulse train delivered by the therapy delivery circuitry 84 to minimize or avoid fatigue and/or allow for sufficient fatigue recovery time between duty cycle on times. In this manner, the sensor 86 may be configured to generate a signal related to the extensor tension state for use by the control circuit 80 for detecting a low tension state that is predictive of upper airway obstruction, detecting extensor fatigue, and/or detecting an extensor tension state of the tongue 40. Therapy delivery circuit 84 may be configured to react to detection of the extensor tongue tension state by control circuit 80 by adjusting one or more control parameters used to control the delivery of stimulation pulses.

As indicated above, EMG monitoring may not be the only signal that the pulse generator 12, and in particular the control circuit 80, employs in determining the level of wakefulness. As an example, the sensor 86 may include an accelerometer that may provide an indication of the patient's motion. Further, where the accelerometer 86 is a three-axis accelerometer, the posture of the patient may be determined. Additionally, an accelerometer may be employed to detect snoring sounds and body movements of the patient. Still further, a temperature sensor in which the circadian temperature of the patient is measured and stored in a memory as the sleep temperature may be employed. The sensor 86 may also be one or more accelerometers for detecting the patient's heart rate. In another example, the sensor 86 may be an accelerometer for detecting the rate of respiration or the volume of airflow into and out of the patient. The volume of airflow may be determined by placing an accelerometer at a point on the patient's chest and comparing the accelerometer travel to previously observed lung volume data associated with sensor 86 movement data. The respiration rate can be determined by monitoring only the change in the direction of the accelerometer.

Still further, sensor 86 may be an implantable pulse oximeter that may be used to measure the blood oxygen saturation level. In one example, a pulse oximeter is a cuff that is placed substantially around a blood vessel and measures blood oxygen levels using light sources known in the art. As described herein, the sensor 86 may be one or more of the various types of sensors described herein.

The sensor 86 may be an ECG sensor. An ECG is a recording of the electrical activity of the heart over a period of time. While ECGs typically employ skin-placed sensors, an effective ECG may be employed in an implantable device in which at least two electrodes separated by a distance (e.g., at least about 35mm) are employed to detect electrical changes caused by depolarization and repolarization of the heart during each cardiac cycle.

Additional aspects of the disclosure are described in conjunction with fig. 10 and 11, in which a simplified diagram of an INS system is depicted and a method of system operation is described. The system 600 includes the INS device 10, the external programmer 50, a server 602 in communication with the external programmer, and a remote computer 604 in communication with the server 602. Prior to implanting the INS 10, the patient typically performs one or more analyses of the patient's assessment (step 702) along with their physician. During this analysis, various self-reported issues may be identified, including daytime sleepiness, snoring interruptions, panting, comorbidities, and the like. Data relating to these issues may be stored on server 602 as part of a patient Electronic Medical Record (EMR) or as part of a file for treatment and remediation of a particular OSA. Discussion with the healthcare provider may lead to an initial diagnosis of OSA. This initial diagnosis is typically confirmed by using one or more sleep studies performed on the patient. During a sleep study, various physiological data as well as some self-reported data are collected. Such as heart rate, blood oxygen saturation level, temperature, electroencephalogram (EEG), Electrocardiogram (ECG), total sleep, sleep quality, sleep efficiency, sleep stage, number of awakenings (less than 15 seconds), number of wakefulness (greater than 15 seconds), apnea hypopnea index, etc. This data may be recorded by the remote computer 604, directly or through additional hardware, and stored on the remote server 602 (step 704).

These collected data from the sleep study, along with data collected by the healthcare provider, may be used to generate a set of initial stimulation parameters (e.g., pulse width, frequency, amplitude, electrode pairing, etc.) for the INS 10 (step 706). This may be based in part on larger population studies to identify some aspects of more comprehensive therapy parameters. The initial stimulation parameters may be set at the remote computer 604 or directly at the external programmer 50, and in either case the server 602 (e.g., cloud computer storage) may be saved for access by either device. And the external programmer 50 may be employed to install the initial stimulation parameters in the INS 10 (step 708). Typically, the patient is allowed to utilize the INS 10 for a period of time, and subsequent sleep studies may be performed. According to this second sleep study, the initial stimulation parameters may be changed, or additional surgical procedures may be recommended to those who do not respond to the stimulation therapy. Additional sleep studies may be required on a regular basis to adjust the stimulation parameter settings in an effort to improve the therapy of the individual patient.

In accordance with the present disclosure, the data collected from sensors 86 may be combined with various self-reported data that the user may enter through a user interface on external programmer 50 and used to replace at least a second sleep study (and possibly also the first). External programmer 50 or another device in communication with server 602 presents a user interface to the patient. The user interface may be presented to the user on a regular basis, including daily, weekly, biweekly, or monthly. According to daily embodiments, the user interface may request that the patient enter various self-reported data. This may include night alcohol intake, smoking, stress, time the patient went to bed, the patient's perception of last night sleep quality, fatigue, discomfort, extensor muscle pain or soreness that may be caused by irritation, etc. In addition, data from other instruments may also be reported. For example, a patient suffering from OSA may also suffer from hypertension, and may be in a regimen where their blood pressure is periodically tested. This blood pressure data may be self-reported through the user interface. Similarly, if the patient is diabetic, it may need to test their blood glucose levels both before and after sleep. These data may also be self-reported through the user interface. In addition, the patient may be asked to answer an inquiry from the Epsworth Sleepiness Scale (ESS). In one embodiment, ESS queries may be requested from the patient at different intervals than other data. In this way, ESS can be used as a gauge of the effectiveness of therapy.

As noted above, the sensor 86 may provide a variety of data depending on how it is configured. As one example of using EMG data, sleep start and end times may be determined. Using one or more accelerometers and various band-pass filtered locations, activity (wake vs. wake), sleep stages, respiration rate, and heart rate can be collected. This data may be reported to the control circuit 80 and stored at least temporarily in the memory 82. The external programmer 50 may be set to automatically interface with the INS 10 daily or at another periodic interval. External programmer 50 may then download the sensor data via telemetry circuitry and transmit the sensor data from the INS and self-reported data entered via the user interface to server 602 (step 710).

The server 602 may have one or more software applications embodied thereon. One of these applications may review data received from external programmer 50 and assign a value to each data point received. These values may be analyzed and a sleep score determined based on the received sensor data and self-reported data (step 712). The sleep score provides an overall assessment of the sleep of the patient that can be assessed by both the patient and the healthcare provider. As will be apparent, some data points may be more important to assess the overall sleep of the patient, and thus some form of scaling of the values may be required. The application will also be able to label any relevant data that is significant for poor sleep scores. For example, if the patient reported drinking several alcoholic beverages in the evening before recording the data leading to poor sleep scores, this may be a highly relevant factor and indicate that the day's sleep score is not an accurate indicator of the effectiveness of the current stimulation parameters.

Regardless, the sleep score may then be recorded as part of the patient's sleep record and reported to the healthcare provider via the remote computer 604 (step 714). This data may be reviewed in a variety of ways for providing an assessment of current stimulation parameters to a medical provider. For example, the healthcare provider may review daily results, an average over a period of time, a graphical representation of sleep scores, or percentages or rates of change (if any) compared to a previous reporting period of time. By periodically assessing the effectiveness of the parameters and comparing the effectiveness to otherwise self-reported data, a multi-tiered analysis can be performed. In one example, if the patient's data indicates that the therapy is effective in achieving high quality sleep with few OSA events, but the patient expresses a sensation of extensor muscle soreness or fatigue, the stimulation parameters can be varied to increase the frequency of switching between bipolar pairs and extend the interval between any set of bipolar pairs that are stimulated. Alternatively, the amplitude of the signal may be reduced. Further, after additional sampling of the data collected by the sensors 86 and self-reported by the patient via the user interface, a second data may be attempted if a first of these data is invalid. In this way, the healthcare provider can continue in a manner that gradually changes stimulation parameters, makes adjustments, and observes the results of those adjustments while not just considering self-reported data. The collection of this data and the reporting of sleep scores (step 710-714) may be iteratively repeated before proceeding to the next step. Those skilled in the art will recognize that the remote computer may in fact be an external programmer 50 configured for use by a physician or healthcare provider.

In further aspects of the disclosure, the server 602 may collect or communicate with one or more additional servers that receive similar data from other patients. All of the collected data may then be analyzed by one or more neural networks to evaluate the combined data and identify patterns within the data to provide a global assessment of stimulation parameters and effectiveness of stimulation patterns when applied across a wide array of patients. Some of these patients will develop similar complications, while others will not. Through additional evaluation of the data, the neural network can find similar patient groups and provide refined initial stimulation parameters for the subgroups based on these similarities (e.g., age, demographics, weight, heart disease, blood pressure, etc.). Neural networks may also be used to evaluate individual patients to provide personalized guidance regarding updating stimulation parameters. In a similar manner, the server may contain one or more applications that employ fuzzy logic to analyze data from individuals and from a broader patient population to provide recommendations for updating stimulation parameters (step 716). In both the use of neural networks and fuzzy logic, an application on server 602 may present the healthcare provider with the option of rejecting suggested updated stimulation parameters (step 718) or accepting or modifying suggested updated stimulation parameters (step 720). As will be appreciated, the medical provider may forego using neural networks or fuzzy logic and update or modify the stimulation parameters. Once the updated stimulation parameters are accepted/modified by the healthcare provider, the updated stimulation parameters are transmitted to external programmer 50 (step 722). Once received at the external programmer 50, the patient may again have the option of accepting the updated stimulation parameters (step 724) or rejecting the updated stimulation parameters (step 726). If the patient accepts, the external programmer 50 may update the stimulation parameters on the INS (step 728). If the updated stimulation parameters are rejected at steps 718 or 726, the method simply returns to step 710. Similarly, after updating the stimulation parameters on the INS 10, the process similarly returns to step 710.

These updated stimulation parameters may be stored on server 602 until the next communication with remote programmer 50, at which point the improved stimulation parameters may be downloaded into external programmer 50. During the next collection of data from the INS 10, the external programmer 50 may then download updated stimulation parameters to the INS 10. In this way, the stimulation parameters of the INS are updated and the sleep score of the patient is also improved. As will be expected, the user interface on the external programmer 50 will indicate to the patient that new stimulation parameters are ready to be installed on the INS, and steps 726 and 724 may be performed at this time.

A further aspect of the present disclosure is the presence of Artificial Intelligence (AI) in the external programmer 50. Authorization for AI may be limited for patient safety purposes to limit the number of consecutive nights that result in poor sleep scores. In one aspect of the disclosure, after updating the stimulation parameters and receiving data from the next night's sleep, the AI may analyze the data from the sensors 86 and the self-reported data and immediately evaluate in terms of sleep scores (step 730). If the sleep score is not good, the user interface may enable the patient to revert to the previous stimulation parameters until the patient can dock with their medical provider with respect to sleep on the previous bad night (step 732). Of course, AI may require more than a single night of data to identify problems or require enough data to be of concern to the patient. Further, AI can be enabled to transmit a request for intervention directly to a medical provider through the server 602 and the remote computer 604.

In this way, true actionable feedback on the effectiveness of the stimulation parameters is provided to both the medical provider and the patient. Continuous care and assessment of patient experiences with the INS device is enabled so that adverse outcomes from therapy may be corrected and behavioral modifications may be suggested to the patient based on the patient's self-reported data.

Those skilled in the art will appreciate that one or more of the calculations, evaluations, and user interfaces described above with respect to the server 602 and remote computer 604 may also be performed directly at the external programmer 50. In some embodiments, this may provide near-instantaneous feedback to the patient regarding previous nights sleeping through a user interface on the external programmer. In other embodiments, where external programmer 50 is of the type commonly used by physicians during a visit, the capabilities and functionality of external programmer 50 may be more robust and potentially may even eliminate or at least reduce the use of server 602 and remote computer 604. As a further example, in applications employing AI, the AI may be trained to perform all evaluations and analyses of the server and provide recommendations to the patient or care provider for modifying stimulation parameters, allowing a deeper understanding of therapy, efficacy, and evaluation of possible changes without having to access data stored on the server 602 or the application running thereon. And in yet further examples, the AI on the external programmer 50 may evaluate the data received from the INS 10 and autonomously adjust or present the stimulation parameters to the patient for acceptance. As will be appreciated, these updates may be limited to prevent large changes in stimulation parameters without intervention from a healthcare provider.

It should be understood that, depending on the example, certain acts or events of any of the methods described herein can be performed in a different order, can be added, merged, or left out altogether (e.g., not all described acts or events are necessary for the practice of the methods). Further, in some instances, actions or events may be performed concurrently, e.g., through multi-threaded processing, interrupt processing, or multiple processors rather than sequentially. Additionally, although certain aspects of the disclosure are described as being performed by a single module or unit for clarity, it should be understood that the techniques of the disclosure may be performed by a combination of units or modules associated with, for example, a medical device.

In one or more examples, the functions described may be implemented in hardware, software, firmware, or any combination thereof. If implemented in software, the functions may be stored on a computer-readable medium in the form of one or more instructions or code and may be executed by a hardware-based processing unit. The computer readable medium may include a computer readable storage medium, which corresponds to a tangible medium, such as a data storage medium (e.g., RAM, ROM, EEPROM, flash memory, or any other medium that can be used to store desired program code in the form of instructions or data structures and that can be accessed by a computer).

The instructions may be executed by one or more processors, such as one or more Digital Signal Processors (DSPs), general purpose microprocessors, Application Specific Integrated Circuits (ASICs), field programmable logic arrays (FPGAs), or other equivalent integrated or discrete logic circuitry. Thus, as used herein, the term "processor" may refer to any of the foregoing structure or any other structure suitable for implementation of the techniques described herein. Furthermore, the techniques may be fully implemented in one or more circuits or logic elements.

Accordingly, an implantable medical device system has been presented in the foregoing description with reference to specific examples. It should be understood that the various aspects disclosed herein may be combined in different combinations than the specific combinations presented in the figures. It should be appreciated that various modifications to the reference examples may be made without departing from the scope of the disclosure and the appended claims.