阅读说明：本技术 基于传感器的膈神经刺激检测 (Sensor-based phrenic nerve stimulation detection ) 是由 N·A·格伦兹 K·J·克莱克纳 张绪省 于 2018-12-07 设计创作，主要内容包括：一种用于在心脏医疗设备中或使用心脏医疗设备检测膈神经刺激(PNS)的方法和设备。对患者的隔膜的收缩敏感的测试信号可被感知,并且测试信号在预先确定的心脏信号之前的第一窗口和测试信号在预先确定的心脏信号之后的第二窗口中的每一个内的测试信号的信号伪迹可被确定。例如,可使用测试信号(其可以是心音信号)评估PNS搏动标准。(A method and apparatus for detecting Phrenic Nerve Stimulation (PNS) in or using a cardiac medical device. A test signal sensitive to contraction of a diaphragm of the patient may be sensed, and signal artifacts of the test signal within each of a first window of the test signal before the predetermined cardiac signal and a second window of the test signal after the predetermined cardiac signal may be determined. For example, the PNS beat criteria may be evaluated using a test signal (which may be a heart sound signal).)

1. A method of detecting Phrenic Nerve Stimulation (PNS) in a cardiac medical device, the method comprising:

sensing a test signal that is sensitive to contraction of a patient's diaphragm;

determining a signal artifact of the test signal within each of a first window of the test signal preceding a predetermined cardiac signal and a second window of the test signal following the predetermined cardiac signal;

determining whether PNS beat criteria have been met in response to signal artifacts of the test signal within the first window and the second window;

in response to the PNS beat criteria being met, determining whether PNS episode criteria have been met; and

detecting a PNS episode in response to meeting the PNS episode criteria.

2. The method of claim 1, wherein the test signal comprises one of an acoustic signal and an accelerometer signal.

3. The method of any of claims 1-2, wherein the predetermined cardiac signal comprises at least one of a ventricular pacing (Vp) beat, an atrial sensing (As) beat, and an atrial pacing (Ap) beat.

4. The method of any one of claims 1-3, wherein the test signal comprises a heart sound signal, and wherein determining whether the PNS beat criteria has been met comprises:

determining whether a maximum value of an absolute value (| FHS |) of the heart sound signal within the second window is satisfied;

determining whether a sum of the | FHS | within the second window is satisfied; and

determining that a PNS criterion has been met in response to both the maximum value of the | FHS | within the second window and a sum of the | FHS | within the second window being met.

5. The method of claim 4, further comprising determining a signal artifact of the test signal within a third window of the test signal after the predetermined cardiac signal and different from the second window, wherein determining whether the maximum value of the | FHS | within the second window is satisfied comprises:

determining whether the maximum value of the | FHS | within the second window is greater than the sum of three times the mean of the | FHS | within the first window and two times the standard deviation of the | FHS | within the first window; and

determining whether the maximum value of the | FHS | within the second window is greater than a variable PNS max threshold, wherein the variable PNS max threshold is a function of a range of the heart sound signal (FHS) in the third window.

6. The method of claim 5, wherein determining whether the sum of | FHS | within the second window is satisfied comprises:

determining whether a sum of said | FHS | within said second window is greater than a sum of said | FHS | within said first window and a variable PNS sum threshold, wherein said variable PNS beat sum threshold is a function of a range of said heart sound signal FHS in said third window; and is

Determining whether a sum of the | FHS | within the second window is greater than a multiple of a sum of the | FHS | within the first window.

7. The method of any one of claims 1-6, further comprising:

detecting signal artifacts defining the test signal within a third window of the test signal subsequent to the predetermined cardiac signal and different from the second window; and

determining whether a noise criterion has been met in response to signal artifacts of the test signal within the third window.

8. The method of claim 7, wherein determining whether a noise criterion has been met comprises:

determining whether a sum of the absolute values (| FHS |) of the heart sound signals within the third window is greater than a noise sum threshold;

determining whether a range of the heart sound signal (FHS) within the third window is less than a noise range threshold; and

identifying the beat as a noise beat in response to not only the sum of the FHS within the third window being greater than the noise sum threshold but also the range of the FHS within the third window being less than the noise threshold range.

9. The method of claim 8, further comprising:

in response to at least one of the sum of the FHS in the third window being not greater than the noise sum threshold and the range of the FHS in the third window being not less than the noise range threshold, determining whether the product of the sum of the FHS in the third window and the range of the FHS in the third window is greater than a combination threshold; and

identifying the predetermined cardiac signal as not being a noise beat in response to the product of the sum of the FHS in the third window and the range of the FHS not being greater than the combined threshold.

10. The method of any of claims 1-8, wherein determining whether PNS episode criteria have been met comprises one of:

determining whether a predetermined number of consecutive beats meet the PNS criteria; and

determining whether in a plurality of beats, each predetermined order of beats satisfies the PNS criteria.

11. The method of any one of claims 1-10, further comprising, in response to detecting the PNS episode, performing one of:

storing a determination that the PNS episode was detected,

an alert is generated that is sent to the user,

adjusting pacing vectors, an

Delivery of pacing therapy is suspended.

12. A cardiac medical device, comprising:

a first sensor for sensing a test signal, the test signal being sensitive to contraction of a patient's diaphragm;

a second sensor for sensing a predetermined cardiac signal; and

a processor operably coupled to the first sensor and the second sensor and configured to:

determining a signal artifact of the test signal within each of a first window of the test signal preceding the predetermined cardiac signal and a second window of the test signal following the predetermined cardiac signal,

determining whether PNS beat criteria have been met in response to signal artifacts of the test signal within the first window and the second window,

in response to the PNS beat criteria being met, determining whether PNS episode criteria have been met, and

detecting a PNS episode in response to meeting the PNS episode criteria.

13. The apparatus of claim 12, wherein the test signal comprises one of an acoustic signal and an accelerometer signal.

14. The apparatus of any of claims 12-13, wherein the predetermined cardiac signal comprises at least one of a ventricular pacing (Vp) beat, an atrial sensing (As) beat, and an atrial pacing (Ap) beat.

15. The device of any one of claims 12-14, wherein the test signal comprises a heart sound signal, wherein to determine whether a PNS beat criterion has been met, the processor is further configured to:

determining whether a maximum value of the absolute value (| FHS |) of the heart sound signal within the second window is satisfied,

determining whether a sum of the | FHS | within the second window is satisfied, and

determining that a PNS criterion has been met in response to both the maximum value of the | FHS | within the second window and a sum of the | FHS | within the second window being met.

16. The device of claim 15, wherein the processor is further configured to:

determining signal artifacts of the test signal within a third window of the test signal subsequent to the predetermined cardiac signal and different from the second window,

determining whether the maximum value of the | FHS | within the second window is greater than a sum of three times a mean value of the | FHS | within the first window and two times a standard deviation of the | FHS | within the first window, and

determining whether the maximum value of the | FHS | within the second window is greater than a variable PNS maximum threshold, wherein the variable maximum threshold is a function of a range of the heart sound signal (FHS) in the third window.

17. The device of claim 16, wherein to determine whether the sum of the | FHS | within the second window is satisfied, the processor is further configured to:

determining whether a sum of the | FHS | within the second window is greater than a sum of the | FHS | within the first window and a variable PNS sum threshold, wherein the variable PNS beat sum threshold is a function of a range of the heart sound signal FHS in the third window, and

determining whether a sum of the | FHS | within the second window is greater than a multiple of a sum of the | FHS | within the first window.

18. The device of any one of claims 12-17, wherein the processor is further configured to:

determining signal artifacts of the test signal within a third window of the test signal after the predetermined cardiac signal and different from the second window, and

determining whether a noise criterion has been met in response to signal artifacts of the test signal within the third window.

19. The device of claim 18, wherein to determine whether a noise criterion has been met, the processor is further configured to:

determining whether a sum of the absolute values (| FHS |) of the heart sound signals within the third window is greater than a noise sum threshold,

determining whether the range of the heart sound signal (FHS) within the third window is less than a noise range threshold, and

identifying the predetermined cardiac signal as a noise beat in response to not only the sum of the FHS within the third window being greater than the noise sum threshold but also the range of the FHS within the third window being less than the noise threshold range.

20. The device of claim 19, wherein the processor is further configured to:

in response to at least one of the sum of the FHS in the third window not being greater than the noise sum threshold and the range of the FHS in the third window not being less than the noise range threshold, determining whether a product of the sum of the FHS in the third window and the range of the FHS in the third window is greater than a combination threshold, and

identifying the predetermined cardiac signal as not being a noise beat in response to the product of the sum of the FHS in the third window and the range of the FHS not being greater than the combined threshold.

21. The device of any of claims 12-20, wherein the processor is further configured to determine one of:

whether a predetermined number of consecutive beats meet the PNS criteria, an

Whether each predetermined order beat of a plurality of beats meets the PNS criteria.

22. The device of any of claims 12-21, wherein the processor is configured to perform one of:

storing a determination that the PNS episode was detected,

an alert is generated that is sent to the user,

adjusting pacing vectors, an

Delivering, by the device, pacing therapy is suspended in response to detecting the PNS episode.

23. A non-transitory computer readable medium storing instructions that cause a cardiac medical device to perform the method of any of claims 1-11.

Disclosure of Invention

In general, the present disclosure relates to detecting Phrenic Nerve Stimulation (PNS) in a cardiac medical device using a heart sound sensor. In some examples, pacing-induced phrenic nerve stimulation is detected using the techniques described herein. In some examples, asymptomatic phrenic nerve stimulation is detected. In some examples, intentional (e.g., therapeutic) PNS is detected using the techniques described herein, e.g., for assessing whether PNS has been achieved or assessing the efficacy of PNS.

In one example, the present disclosure is directed to a method for detecting Phrenic Nerve Stimulation (PNS) in a cardiac medical device, the method comprising: sensing a test signal, the test signal being sensitive to contraction of a patient's diaphragm; determining signal artifacts of the test signal within each of a first window of the test signal preceding the predetermined cardiac signal and a second window of the test signal following the predetermined cardiac signal; determining whether a PNS beat criterion has been met in response to signal artifacts of the test signal within the first window and the second window; in response to the PNS beat criteria being met, determining whether PNS episode criteria have been met; and detecting a PNS episode in response to satisfying the PNS episode criteria.

In another example, the present disclosure is directed to a cardiac medical device comprising: a first sensor configured to sense a test signal, the test signal being sensitive to contraction of a patient's diaphragm; a second sensor, the second sensor being directed to a predetermined cardiac signal; and a processor configured to, within each of a first window of the test signal preceding the predetermined cardiac signal and a second window of the test signal following the predetermined cardiac signal, determine a signal artifact of the test signal, determine whether a PNS beat criterion has been met in response to the signal artifact of the test signal within the first window and the second window being met, determine whether a PNS seizure criterion has been met in response to the PNS beat criterion being met, and detect.

In another example, the present disclosure is directed to a non-transitory computer-readable medium storing instructions that cause a cardiac medical device to perform a method comprising: sensing a test signal, the test signal being sensitive to contraction of a patient's diaphragm; determining signal artifacts of the test signal within each of a first window of the test signal preceding the predetermined cardiac signal and a second window of the test signal following the predetermined cardiac signal; determining whether a PNS beat criterion has been met in response to signal artifacts of the test signal within the first window and the second window; in response to the PNS beat criteria being met, determining whether PNS episode criteria have been met; and detecting a PNS episode in response to satisfying the PNS episode criteria.

These and other aspects of the disclosure will become apparent from the following detailed description. In no event, however, should the above summaries be construed as limitations on the claimed subject matter, which subject matter is defined solely by the attached claims, as may be amended during prosecution.

Drawings

Reference is made throughout the specification to the accompanying drawings, wherein like reference numerals designate like elements, and wherein:

fig. 1 is a conceptual diagram illustrating an example system of detecting phrenic nerve stimulation consistent with examples of the present disclosure.

Fig. 2 is a conceptual diagram illustrating an Implantable Medical Device (IMD) and a lead of the system shown in fig. 1 in more detail.

Fig. 3 is a block diagram illustrating an example configuration of the IMD of fig. 1.

Fig. 4 is a conceptual diagram illustrating an example system for delivering phrenic nerve stimulation consistent with examples of the present disclosure.

Fig. 5 is a block diagram illustrating an example computer system including one or more computing devices and an external device (such as a server) coupled to the IMD and programmer shown in fig. 1 via a network.

Figure 6 is a flow chart of a method for determining the presence of phrenic nerve stimulation in a medical device according to one example of the present disclosure.

Figure 7 is a flowchart of a method for determining the presence of phrenic nerve stimulation in a medical device according to one example of the present disclosure.

Fig. 8 is a graphical representation of determining acoustic artifacts of a heart sound signal for determining the presence of phrenic nerve stimulation in a medical device according to an example of the present disclosure.

Figure 9 is a flowchart of a method for determining the presence of phrenic nerve stimulation in a medical device according to one example of the present disclosure.

Figure 10 is a flowchart of a method for determining the presence of phrenic nerve stimulation in a medical device according to one example of the present disclosure.

Detailed Description

The techniques described in this disclosure may allow a medical device to automatically detect the presence of phrenic nerve stimulation. In some examples, phrenic nerve stimulation is an unintended side effect of the electrical stimulation applied to the patient's heart. In other examples, the detected phrenic nerve stimulation may be purposeful. For example, phrenic nerve stimulation may be used to treat nervous system disorders that affect mechanical ventilation. In various examples, detection of phrenic nerve stimulation occurs in response to an activation event. The activation event may be a change in electrical stimulation applied to the heart of the patient. In other examples, the activation event may be a posture or activity level of the patient detected by a posture or activity sensor. For example, an activation event may be an indication that the patient is lying down. In other examples, the activation event may be detection of low activity by the patient. In some examples, an activation event may be the detection of a combination of several factors. For example, an activation event may be based on the occurrence of a particular gesture or activity in conjunction with a change in electrical stimulation. In other examples, activation events may be based on time. For example, an activation event may be the passage of a predetermined amount of time from a previous phrenic nerve detection sequence. In still other examples, activation events may be based on the occurrence of a particular time of day. In another example, detection of phrenic nerve stimulation may be initiated by an input at a user interface of the device external to the medical device, such as, for example, during implantation of the medical device, during post-implantation, or during a consulting room follow-up visit by the patient remotely or to the consulting room. In another example, the medical device may continuously and fluidly analyze heart sound signals to detect the phrenic nerve without using activation events.

Heart sounds are associated with mechanical vibrations from the activity of the patient's heart and the flow of blood through the heart. The heart sounds recur with each cardiac cycle and are separated and classified according to the activity associated with the vibrations. As used herein, the term "heart sound" refers to a characteristic of a heart sound signal, such as S1, S2, S3, or S4 heart sounds. There may be multiple heart sounds for any given cardiac cycle or beat, e.g., each of the S1, S2, S3, and S4 heart sounds. The first heart sound (S1) is a vibrational sound emitted by the heart during mitral valve tensioning. The second heart sound (S2) is related to the closure of the aortic and pulmonary valves. The third heart sound (S3) and the fourth heart sound (S4) are associated with left ventricular filling during diastole. The heart sound sensor generates an electrical signal indicative of mechanical activity of the patient's heart. Examples of heart sound sensors include accelerometers or microphones. Means for measuring heart sounds can be found in U.S. patent No. 7,115,096 entitled "Method and Apparatus for Monitoring Diastolic Hemodynamics" filed on 30.12.2002 by Seijko et al, which is hereby incorporated by reference in its entirety. The evoked response parameters may include at least one of a measured amplitude of a heart sound associated with the evoked response, a time of occurrence of a heart sound associated with the evoked response, and a power of a heart sound associated with the evoked response.

In some examples, the medical device may classify the heartbeat or cardiac cycle as normal or abnormal based on a classification of one or more heart sounds detected during the heartbeat or cardiac cycle. In such examples, the medical device may confirm that the cardiac rhythm is treatable when the one or more heartbeats are classified as abnormal, or may withhold treatment when the one or more heartbeats are classified as normal. In other examples, the heart sound signals may include signals representing other acoustic events, including, for example, diaphragm movement in response to phrenic nerve stimulation.

Both symptomatic and asymptomatic phrenic nerve stimulation by pacing can lead to patient's appearance of unpleasant symptoms as well as reduced hemodynamic performance. In various examples consistent with the present disclosure, phrenic nerve stimulation may be detected, and may be avoided in the future in response to the detection.

When pacing is provided by a left ventricular lead (such as a left ventricular quadrupolar lead), particular attention should be paid to pacing-induced phrenic nerve stimulation. This is because the left ventricular lead may position one or more electrodes in close proximity to the left phrenic nerve. A physician may desire to program an IMD to provide cardiac resynchronization therapy (including left ventricular pacing) that provides near normal cardiac function while avoiding capture of one or more phrenic nerves using applied pulses.

In some examples, the disclosure relates to detecting pacing-induced PNS using a PNS test signal, such as, for example, a heart sound signal or an accelerometer signal, and in response, reprogramming the IMD to provide CRT in a manner that does not capture the phrenic nerve. In some examples, reprogramming the IMD includes changing one or more pacing vectors to avoid phrenic nerve stimulation. In some examples, reprogramming the IMD includes modifying various pacing parameters (such as pacing strength) with or without changing the pacing vector to avoid phrenic nerve stimulation. In some examples, an attempt is made to modify the pulse intensity first, and if phrenic nerve stimulation cannot be avoided without compromising cardiac capture, an attempt is made to modify the pacing vector. A determination of a new pulse strength or pacing vector may be made based on information extracted from the PNS test signal. This is possible because sensors such as activity/posture sensors or heart sound sensors are able to detect diaphragm muscle movements in the form of activity artifacts or sound artifacts, respectively, caused by both symptomatic and asymptomatic PNS.

As described in more detail below with respect to various figures, a PNS test signal (such as an accelerometer signal or an acoustic signal) may be used to detect both symptomatic and asymptomatic PNS. In various examples, the IMD does not continuously monitor the PNS test signal for PNS. Alternatively, the detection sequence may be initiated at a given time of day, for example. This allows the IMD to conserve battery power and perform other functions at other times using the same sensors and processors. In some examples, PNS detection is initiated when PNS is most likely detected. For example, the PNS detection sequence may be initiated when the patient is lying on his or her left side. In instances where a left ventricular lead is used to deliver cardiac pacing, phrenic nerve stimulation may occur when the patient is lying on his or her left side, but not when the patient is in other positions. In other examples, the PNS detection sequence may be initiated by a clinician at a clinic using a monitoring device.

For pacing induced PNS, the IMD or another device in communication with the IMD may determine signal artifacts of the sensed PNS test signal, such As, for example, a time delay of 150ms, just before or just after the predetermined timing signal, or just before or just after the predetermined cardiac signals, such As ventricular pacing (Vp) beats, and atrial sensing (As) beats, or atrial pacing (Ap) beats. The device determines whether the PNS beat criterion has been met based on signal artifacts determined before or after a predetermined timing signal or based on signal artifacts determined before or after a predetermined cardiac signal. If the PNS beat criteria are not met, then the beat is determined not to be a PNS beat, and the process resumes for the next beat. On the other hand, if the PNS beat criteria are met, the IMD determines that a PNS episode is occurring. For example, as described below, PNS detection of the present disclosure may assess the presence or absence of PNS on a beat-by-beat basis by detecting acoustic artifacts of heart sound signals occurring after Left Ventricular (LV) pacing (Vp). Although the determination of signal artifacts is described below with reference to signal artifacts occurring before and after a predetermined cardiac signal, it can be appreciated that the determination of signal artifacts may be performed with reference to a timing signal, such as, for example, a time delay of 150 ms.

The analyzed heart sound signal data may be digitized by a 16-bit ADC sampling ± 64mV at 256Hz and band-pass filtered. The absolute value of the filtered heart sound signal (| FHS |) is determined and a predetermined number of beats is evaluated for PNS at a voltage and/or polarity that can be selected by the user. In one example, to ensure that the evaluation is completed in less than 3 minutes (176 s cycle length 11x 16x 1000 ms) for 16 vectors at an estimated heart rate of 60bpm, up to 11 beats are evaluated for PNS with a voltage output selectable by the user. For each PNS detected beat, a PNS window including a pre-Vp window and a post-Vp window, and a noise window of the heart sound signal may be determined for use in assessing and detecting the presence or absence of PNS. In one example, the pre-Vp window may include 25 samples before and up to the Vp beat (98ms), the post-Vp window may occur 7 to 21 samples (27 to 82ms) after the Vp beat for the left device implant case, or 20 to 32 samples (78 to 125ms) after the Vp beat for the right device implant case, and the noise window may occur 7 to 80 samples (27 to 313ms) after the Vp beat.

Signal characteristics of the heart sound signal, such as the maximum, minimum, range, average, sum, and absolute difference of the signal, are calculated within each window. The absolute difference is similar to the Standard Deviation (SD) and is calculated by subtracting the mean from the signal, adding the absolute value of the resulting time series and dividing by the length of the signal. The heart sound signal features within the noise window are first evaluated for each beat to detect or determine whether the beat is associated with noise, and if the beat is not detected noise, the heart sound signal features within the PNS window are evaluated for the beat to detect or determine whether the beat is associated with the presence of PNS. If the beat is neither a noise beat nor a PNS beat, the beat is classified as a non-PNS beat and the process continues for the next beat identified by the heart sound signal until a predetermined number of beats have been evaluated or a predetermined time period has expired. On the other hand, if a beat is not identified as a noise beat, but is identified as a PNS beat, the beat is classified as a PNS beat. In either case, i.e., the current beat is a PNS beat or a non-PNS beat, the process continues for the next beat until a predetermined number of beats have been evaluated, a predetermined time period has expired, or a PNS episode is detected based on the predetermined sequence of beats being identified as a PNS beat.

To avoid PNS when setting pacing parameters, the IMD may set the pacing pulse amplitude and/or width from the IMD's minimum pacing capture threshold to maximum output. In some examples, the IMD may stop the incrementing process when PNS is detected. In some examples in which PNS detection is implemented after pacing parameters have been set, the IMD decrements the amplitude of the pacing pulse after the initial determination of PNS until PNS is no longer detected so long as cardiac capture is maintained.

In some examples, it may be desirable to determine whether a preferred or selected pacing vector or modality will result in PNS in a particular patient. This can be done by first applying pacing stimulation at the maximum output of the stimulation generator and looking to see if PNS is present. If PNS is present, the IMD may gradually decrement the amplitude of the pacing pulse until a minimum pacing amplitude (PNS threshold) is determined that still results in PNS. If the PNS threshold is above the threshold for capturing the ventricle to provide adequate pacing, the selected pacing vector may still be used. Otherwise, another vector or electrode configuration may be tested until one is found, in which configuration a pacing pulse may be delivered to provide pacing capture without also stimulating the phrenic nerve.

In some examples, once PNS is detected, the IMD or another device in communication with the IMD may modify pacing parameters to provide pacing that does not compromise the hemodynamics of the patient while avoiding PNS. In some examples, the heart sound signal is used to evaluate pacing parameters not only for PNS but also for overall cardiac function.

In some examples, the methods of detecting PNS described below may be performed by another device in communication with the IMD, and automatic PNS detection is achieved using only heart sound signals. The PNS detection feature may thus reduce the time required to assess PNS at implantation, and potentially reduce post-symptomatic implantation PNS by detecting indications that are not detected using manual PNS assessment (i.e., visual assessment, palpation, or under fluoroscopy).

In some examples, phrenic nerve stimulation may be desirable. For example, it may be desirable to provide PNS as a replacement for mechanical ventilation in patients with neurological disorders such as central sleep apnea. In such examples, the amount of stimulation applied may become different every few pulses in order to simulate a normal breathing pattern. PNS detection using heart sounds can be used to confirm the effectiveness of attempted phrenic nerve stimulation.

Fig. 1 is a conceptual diagram illustrating an example system 10 that may detect phrenic nerve stimulation. In some examples, system 10 monitors both cardiac electrical activity and heart sound-based signals. In some examples, system 10 provides stimulation to cardiac tissue based on a set of parameters, and monitors signals representative of cardiac electrical activity (e.g., Electrograms (EGMs)) and heart sound signals. The system 10 determines whether the cardiac stimulation at the current stimulation parameters resulted in undesired phrenic nerve stimulation based at least on the heart sound signals.

System 10 includes an Implantable Medical Device (IMD)16, which IMD16 is connected to leads 18, 20, and 22, and optionally communicatively coupled to a programmer 24. IMD16 senses electrical signals attendant to the depolarization and repolarization of heart 12, e.g., a cardiac EGM, via one or more leads 18, 20, and 22 or electrodes on the housing of IMD 16. In some examples, IMD16 also delivers cardiac therapy to heart 12 in the form of electrical signals via electrodes located on one or more leads 18, 20, and 22 or a housing of IMD 16. The cardiac therapy may be pacing, cardioversion, and/or defibrillation pulses. IMD16 may also provide respiration induction therapy. Respiration-inducing therapy includes electrical stimulation to one or more phrenic nerves 36 and 38 via electrodes located on leads 18, 20, and 22, other electrodes not shown in fig. 1, or on one or more of the housings of IMD 16. In some examples, the electrodes used to stimulate phrenic nerves 36 and 38 may be used for both cardiac and phrenic nerve stimulation. IMD16 also includes one or more heart sound sensors (not shown in fig. 1) for detecting the occurrence of phrenic nerve stimulation in patient 14. IMD16 may similarly include or be coupled to other sensors, such as one or more accelerometers, for detecting other physiological parameters of patient 14, such as activity or posture.

In some examples, programmer 24 takes the form of a handheld computing device, computing workstation, or networked computing device that includes a user interface for presenting information to a user and receiving input from the user. A user (such as a physician, technician, surgeon, electrophysiologist, or other clinician) may interact with programmer 24 to retrieve physiological or diagnostic information from IMD 16. The user may also interact with programmer 24 to program IMD16, e.g., select values for operating parameters of the IMD or initiate a phrenic nerve stimulation detection sequence.

IMD16 and programmer 24 may communicate via wireless communication using any technique known in the art. Examples of communication techniques may include, for example, low frequency or Radio Frequency (RF) telemetry. Other techniques are also contemplated. In some examples, programmer 24 may include a programming head that may be placed proximate to the body of the patient near the IMD16 implantation site in order to improve the quality or safety of communications between IMD16 and programmer 24. In other examples, programmer 24 may be located remotely from IMD16 and communicate with IMD16 via a network.

Leads 18, 20, 22 extend into heart 12 of patient 14 to sense electrical activity of heart 12 and/or deliver electrical stimulation to heart 12. The lead may also deliver electrical stimulation to the phrenic nerve 38. In the example shown in fig. 1, Right Ventricular (RV) lead 18 extends through one or more veins (not shown), the superior vena cava (not shown), and right atrium 26, and into right ventricle 28. Left Ventricular (LV) coronary sinus lead 20 extends through one or more veins, the vena cava, right atrium 26 and into coronary sinus 30 to a region adjacent to the free wall of left ventricle 32 (free wall) of heart 12. Right Atrial (RA) lead 22 extends through one or more veins and the vena cava, and into right atrium 26 of heart 12. In some examples, RA lead 22 may be used to stimulate right phrenic nerve 36. In some examples, LV coronary sinus lead 20 may be used to stimulate left phrenic nerve 38.

Techniques for detecting stimulation of one or more of phrenic nerves 36 and 38 are primarily described herein as being performed by IMD16, e.g., by processing circuitry of a processor of IMD 16. In other examples, some or all of the functionality attributed to IMD16 or its processor may be performed by one or more devices, such as programmer 24 or its processor. For example, IMD16 may process cardiac and/or heart sound signals to determine whether therapy should continue to be delivered based on current parameters, or whether parameters should be adjusted and control parameters used by IMD16 for delivering therapy. Alternatively, programmer 24 may process cardiac and/or heart sound signals received from IMD16 to determine whether therapy should continue to be delivered based on current parameters or whether parameters should be adjusted, and control therapy according to what IMD16 delivers. Further, although described herein with respect to IMDs, in other examples, the techniques described herein may be performed or implemented in an external medical device that may be coupled to a patient via a percutaneous or percutaneous (percutaneous) lead. In some examples, various functions of IMD16 may be performed by multiple IMDs in communication with each other.

Fig. 2 is a conceptual diagram illustrating IMD16 and leads 18, 20, and 22 of system 10 in more detail. In the illustrated example, bipolar electrodes 40 and 42 are positioned adjacent the distal end of lead 18. Further, bipolar electrodes 44 and 46 are positioned adjacent the distal end of lead 20, and bipolar electrodes 48 and 50 are positioned adjacent the distal end of lead 22. In an alternative example, not shown in fig. 2, one or more of leads 12, 20, and 22 (such as left ventricular lead 20) may include a quadrupolar electrode positioned adjacent a distal end of the lead.

In the illustrated example, electrodes 40, 44 and 48 take the form of ring electrodes, and electrodes 42, 46 and 50 may take the form of extendable helix tip electrodes mounted retractably within insulative electrode heads 52, 54 and 56, respectively. The leads 18, 20, 22 also include elongated electrodes 62, 64, 66, respectively, which may take the form of coils. In some examples, each of electrodes 40, 42, 44, 46, 48, 50, 62, 64, and 66 is electrically coupled within a respective conductor of the lead body of its associated lead 18, 20, 22, and thereby to circuitry within IMD 16.

In some examples, IMD16 includes one or more housing electrodes, such as housing electrode 4 shown in fig. 2, which may be integrally formed with an outer surface of hermetically sealed housing 8 of IMD16 or otherwise coupled to housing 8. In some examples, housing electrode 4 is defined by an uninsulated portion of an outward facing portion of housing 8 of IMD 16. Other separators between the insulated and non-insulated portions of the housing 8 may be employed to define two or more housing electrodes. In some examples, the housing electrode includes substantially all of the housing 8.

As described in further detail with reference to fig. 3, housing 8 encloses a signal generator that generates therapeutic stimulation (such as cardiac pacing, cardioversion, and defibrillation pulses) and a sensing module that senses electrical signals that accompany the depolarization and repolarization of heart 12. IMD16 may also include a heart sound sensor that monitors acoustic noise including, for example, heart sounds and sounds produced by phrenic nerve stimulation. The heart sound sensor may be, for example, an accelerometer or a microphone. The heart sound sensor may be enclosed within a housing 8. Alternatively, the heart sound sensor may be integrally formed with or carried on an outer surface of housing 8, carried on a lead coupled to IMD16, or within a lead coupled to IMD16 (such as one or more leads 18, 20, and 22), or may be a separate remote sensor that wirelessly communicates with IMD16, programmer 24, or any other device described herein.

IMD16 senses electrical signals attendant to the depolarization and repolarization of heart 12 via electrodes 4, 40, 42, 44, 46, 48, 50, 62, 64, and 66. IMD16 may sense such electrical signals via any bipolar combination of electrodes 40, 42, 44, 46, 48, 50, 62, 64, and 66. Further, any of the electrodes 40, 42, 44, 46, 48, 50, 62, 64, and 66 may be used for unipolar sensing in conjunction with the housing electrode 4.

In some examples, IMD16 delivers stimulation pulses via a bipolar combination of electrodes selected based on the EGM signal and/or the heart sound signal (analyzed by a signal analyzer within the IMD). For example, a bipolar combination of electrodes 40, 42, 44, 46, 48 and 50 is used to produce depolarization of cardiac tissue of heart 12. Further, phrenic nerve stimulation pulses may be delivered by various electrodes used to provide cardiac stimulation, and the electrodes that may be selected for delivering phrenic nerve stimulation are based on the location of the electrodes. In some examples, IMD16 delivers stimulation to cardiac tissue or the phrenic nerve in a monopolar configuration via any of electrodes 40, 42, 44, 46, 48, and 50 in combination with housing electrode 4. In some examples, the electrodes delivering cardiac and phrenic nerve electrical stimulation may be selected based on default settings-furthermore, the IMD may deliver cardioversion or defibrillation pulses to heart 12 or pulses to phrenic nerves 36 and 38 via any combination of elongated electrodes 62, 64, 66 and housing electrode 4.

The number and configuration of leads 18, 20, and 22 and electrodes shown are merely examples. Other configurations, i.e., the number and location of leads and electrodes, are possible. In some examples, system 10 may include additional leads or lead segments with one or more electrodes located at different locations in the cardiovascular system for sensing and/or delivering therapy to patient 14. For example, instead of, or in addition to, intracardiac leads 18, 20, and 22, system 10 may include one or more epicardial or subcutaneous leads that are not positioned within the heart. For example, a lead may be positioned for providing one or more electrodes proximate to or in contact with the phrenic nerve 36 or 38. As another example, the system 10 may include additional leads carrying a heart sound sensor positioned such that a signal generated by the heart sound sensor includes information about the respiratory activity of the patient (including, for example, inhalation and exhalation).

Fig. 3 is a block diagram illustrating an example configuration of IMD 16. In the example shown, IMD16 includes a processor 70, a memory 72, a signal generator 74, a sensing module 76, a telemetry module 78, a signal analyzer 80, a heart sound sensor 82, and an activity/posture sensor 84. Memory 72 includes computer readable instructions that, when executed by processing circuitry of processor 70, cause IMD16 and processor 70 to perform various functions attributed herein to IMD16 and processor 70. The memory 72 may include any volatile, non-volatile, magnetic, optical, or electrical media, such as Random Access Memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), Electrically Erasable Programmable ROM (EEPROM), flash memory, or any other digital or analog media.

The processor 70 may include any one or more of the following: a microprocessor, a controller, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA), or equivalent discrete or analog logic circuitry. In some examples, processor 70 may include multiple components, such as any combination of: one or more microprocessors, one or more controllers, one or more DSPs, one or more ASICs, or one or more FPGAs, as well as other discrete or integrated logic circuitry. The functions attributed to processor 70 herein may be implemented as software, firmware, hardware, or any combination thereof. In general, the processor 70 controls the signal generator 74 to deliver stimulation therapy to the heart 12 of the patient 14 according to a selected one or more of the therapy programs or parameters (which may be stored in the memory 72). As an example, processor 70 may control signal generator 74 to deliver electrical pulses having an amplitude, pulse width, frequency, or electrode polarity specified by the selected one or more therapy programs. The treatment program may be selected by the processor 70 based on information from the signal analyzer 80.

Signal generator 74 is configured to generate and deliver electrical stimulation therapy to patient 12. As shown in fig. 3, the signal generator 74 is coupled to the electrodes 4, 40, 42, 44, 46, 48, 50, 62, 64, and 66, e.g., via conductors of the respective leads 18, 20, and 22, and in the case where the electrode 4 is a housing electrode 4, the signal generator 74 is electrically coupled to the housing electrode 4 within the housing 8. For example, the signal generator 74 may deliver stimulation pulses to the phrenic nerves 36 and 38 via at least two of the electrodes 4, 40, 42, 44, 46, 48, 50, 62, 64, and 66. Further, in some examples, signal generator 74 may deliver pacing pulses, defibrillation shocks, or cardioversion shocks to heart 12 via at least two of electrodes 4, 40, 42, 44, 46, 48, 50, 62, 64, and 66. In some examples, signal generator 74 delivers stimulation in the form of a signal other than pulses, such as a sine wave, square wave, or other substantially continuous time signal.

The signal generator 74 may include a switching module (not shown), and the processor 70 may use the switching module to select which of the available electrodes are used to deliver electrical stimulation, e.g., via a data/address bus. The switch module may include a switch array, a switch matrix, a multiplexer, or any other type of switching device suitable for selectively coupling stimulation energy to selected electrodes. Electrical sensing module 76 monitors electrical cardiac signals from any combination of electrodes 4, 40, 42, 44, 46, 48, 50, 62, 64, and 66. Sensing module 76 may also include a switching module that processor 70 controls to select which of the available electrodes are used to sense cardiac activity, depending on which electrode combinations are used in the current sensing configuration.

Sensing module 76 may include one or more detection channels, each of which may include an amplifier. The detection channel may be used to sense cardiac signals. Some detection channels may detect events (such as R-waves or P-waves) and provide an indication of the occurrence of such events to processor 70 and/or signal analyzer 80. One or more other detection channels may provide signals to an analog-to-digital converter for conversion to digital signals for processing or analysis by processor 70 or signal analyzer 80.

For example, sensing module 76 may include one or more narrow band channels, each of which may include a narrow band filtered sense amplifier that compares a detected signal to a threshold. If the filtered and amplified signal is greater than the threshold, the narrowband channel indicates that a certain electrical signal event (e.g., depolarization) has occurred. The processor 70 then uses this detection in measuring the frequency of the sensed events. The signal analyzer 80 may use the detection in conjunction with the sensed heart sounds to determine one or more cardiac metrics.

In one example, the at least one narrowband channel may comprise an R-wave or P-wave amplifier. In some examples, the R-wave and P-wave amplifiers may take the form of automatic gain controlled amplifiers that provide adjustable sensing thresholds as a function of measured R-wave or P-wave amplitude. Examples of R-wave and P-wave amplifiers are described in the following U.S. patents: U.S. Pat. No. 5,117,824 to Keimel et al, entitled "APPARATUS FOR MONITORING physiological electrical SIGNALS", published on 2.6.1992, and incorporated herein by reference in its entirety.

In some examples, sensing module 76 includes a wideband channel, which may include an amplifier having a wider passband relative to a narrowband channel. The signals from the electrodes selected for coupling to the broadband amplifier may be converted to a multi-bit digital signal by an analog-to-digital converter (ADC) provided by, for example, sensing module 76, processor 70, or signal analyzer 80. Processor 70 may analyze the digitized version of the signal from the wideband channel. Processor 70 may employ digital signal analysis techniques to characterize the digitized signals from the wideband channel for use in, for example, detecting and classifying a patient's heart rhythm. In some examples, signal analyzer 80 employs digital signal analysis techniques to characterize the digitized signal from the wideband channel. The digitized signal may be used in conjunction with the heart sound signal to determine whether phrenic nerve stimulation has occurred.

Processor 70 may initiate a phrenic nerve stimulation detection sequence in response to detecting an activation event. In some examples, prior to initiating phrenic nerve stimulation detection, processor 70 may receive activation signals from programmer 24 via telemetry module 78, which may be activation events. In some examples, activation events may be one or more of the activities/gestures detected via activity gesture sensor 84, signal analyzer 80, memory 72, and sensing module 76. In some examples, processor 70 may initiate phrenic nerve stimulation detection at a given time. For example, the memory 72 may provide the processor 70 with a program in which phrenic nerve stimulation detection occurs at a predetermined time each day. In such cases, the activation event is a time of day. In other examples, processor 70 initiates phrenic nerve stimulation detection during a predetermined time range when predefined parameters are met. For example, when an activation event occurs, such as activity/posture sensor 84 indicating that patient 12 is lying down, processor 70 may initiate phrenic nerve stimulation detection between 10 pm and 5 am. In some particular examples, processor 70 may initiate phrenic nerve stimulation detection in response to an activation event, such as receiving an indication from activity/posture sensor 84 that the patient is lying on his or her left side. In some examples, processor 70 may initiate a phrenic nerve stimulation detection sequence based on an activation event (such as one or more pacing parameter changes). In some examples, processor 70 may initiate a phrenic nerve stimulation detection sequence in conjunction with a pacing parameter optimization process.

In the example of fig. 3 (e.g., for detecting the presence of phrenic nerve stimulation), IMD16 also includes a heart sound sensor 82 and a signal analyzer 80. The heart sound sensor 82 generates electrical signals based on sensed acoustic noise or vibration originating from, for example, heart movement and diaphragm movement. In some examples, heart sound sensor 82 may include more than one sensor. For example, the heart sound sensor may comprise a plurality of individual sensors. In some examples, the heart sound sensor 82 is an acoustic sensor, such as an accelerometer, microphone, or piezoelectric device. In addition to the heart sounds S1-S4, the acoustic sensor also receives sounds produced by diaphragmatic activation.

In the example shown in fig. 3, heart sound sensor 82 is enclosed within housing 8 of IMD 16. In some examples, the heart sound sensor 82 may be integrally formed with the outer surface of the housing 8, or formed on the outer surface of the housing 8. In some examples, heart sound sensor 82 is located on one or more leads coupled to IMD16, or may be implemented in a remote sensor that wirelessly communicates with IMD 16. In such cases, heart sound sensor 82 may be electrically or wirelessly coupled to circuitry contained within housing 8 of IMD 16. In some examples, remote heart sound sensor 82 may be wirelessly connected to programmer 24.

The signal analyzer 80 receives the electrical signal generated by the heart sound sensor 82. In one example, the signal analyzer 80 may process the sensor signal generated by the heart sound sensor 82 to detect the occurrence of phrenic nerve stimulation. In some examples, signal analyzer 80 processes the heart sound sensor signals to generate envelope signals, detect the occurrence of phrenic nerve stimulation, detect other heart sounds, extract heart sound features from the detected heart sound signals, and evaluate various cardiac metrics. Cardiac metrics may provide a method for assessing the electromechanical operation of heart 12. In some examples, the detected heart sound features (both those associated with phrenic nerve stimulation and those associated with other cardiac activity) may be compared to the values of each feature stored in memory 72. The heart sound features may then be classified based on the deviation from the stored values. The heart sound features and/or their classification may be used to determine whether phrenic nerve stimulation has occurred and to assess the function of heart 12.

An indication based on heart sound characteristics may be shown from the signal analyzer 80 to the processor 70. In some examples, the heart sound characteristics are output to the processor 70. The processor 70 may determine whether phrenic nerve stimulation occurred based on information received from the signal analyzer 80. In some examples, the processor may adjust the stimulation provided by the signal generator 74 based on the received heart sound-based information.

In various examples, one or more of the functions attributed to signal analyzer 80 may be performed by processor 70. In some examples, the signal analyzer 80 may be implemented as hardware, software, or some combination thereof. For example, the functions of the signal analyzer 80 described herein may be implemented in a software process executed by the processor 70.

Fig. 4 is a conceptual diagram illustrating an example system 100 for detecting phrenic nerve stimulation using heart sounds. The system 100 includes an IMD16, the IMD16 monitoring a heart sound based signal and determining whether a phrenic nerve is being stimulated based on the heart sound based signal. In some examples, IMD16 may also monitor cardiac electrical activity signals (e.g., EGM signals) and may provide cardiac tissue stimulation. In some examples, detection of phrenic nerve stimulation may trigger an optimization scheme for pacing parameters used to provide cardiac tissue stimulation. In some examples, system 100 detects the occurrence of phrenic nerve stimulation and provides an indication of the phrenic nerve stimulation to a remote device (e.g., programmer 24).

System 100 includes IMD16, IMD16 is connected to leads 104, 106, 112, and 114, and optionally communicatively coupled to a programmer (not shown in fig. 4). In response to electrical stimulation of phrenic nerves 36 and 38, IMD16 senses various signals accompanying activation of diaphragm 102. In some examples, leads 104 and 106 are positioned near the phrenic nerve. Stimulation may be provided to phrenic nerves 36 and 38 via electrodes 108 and 110. Leads 104 and 106 may be intracardiac leads that include additional electrodes (not shown) that provide cardiac stimulation, and leads 112 and 114 may be, for example, intracardiac leads for providing cardiac stimulation. In some examples, electrodes 108 and 110 may be snap ring electrodes (cuffelectors) that at least partially surround phrenic nerves 36 and 38, respectively.

In some examples, IMD16 senses electrical signals, e.g., cardiac EGMs, that accompany the depolarization and repolarization of heart 12 via electrodes on one or more of leads 104, 106, 112, and 114 or on the housing of IMD 16. In some examples, IMD16 delivers cardiac therapy to heart 12 in the form of electrical signals via electrodes located on one or more of leads 104, 106, 112, and 114. The IMD may also include or be coupled to other sensors (such as one or more accelerometers) for detecting other physiological parameters of the patient (such as activity or posture).

Techniques for monitoring stimulation of one or more of phrenic nerves 36 and 38 are primarily described herein as being performed by IMD16, e.g., by a processor of IMD 16. For example, IMD16 may process the heart sound signals to determine whether IMD16 should continue to deliver based on current parameters or whether the parameters should be adjusted. A processor in IMD16 may also control parameters used by IMD16 for delivering therapy. It can be appreciated that in another example, the techniques of the present disclosure may also be performed by another device in communication with IMD16, such as a programming and/or monitoring device at a clinic.

Fig. 5 is a block diagram illustrating an example system including one or more computing devices 212A-210N and an external device (such as server 206) coupled to IMD16 and programmer 24 shown in fig. 1 via network 204. Network 204 may generally be used to communicate diagnostic information (e.g., the occurrence of phrenic nerve stimulation) from IMD16 to a remote external computing device. In some examples, the heart sounds and/or ECG signals may be transmitted to an external device for processing.

In some examples, IMD16 transmits information during a predetermined time window. In some examples, the transmission window is aligned with a window during which an activation event may result in initiation of a phrenic nerve detection sequence. In some examples, network 204 may also transmit information from IMD16 regarding activation events that triggered phrenic nerve stimulation to a remote external computing device.

In some examples, the information transmitted by IMD16 may allow a clinician or healthcare professional to remotely monitor patient 14. In some examples, IMD16 may use its telemetry module 78 to communicate with programmer 24 via a first wireless connection and, for example, communicate with access point 202 at a different time via a second wireless connection. In the example of fig. 5, access point 202, programmer 24, server 206, and computing devices 212A-212N are interconnected and capable of communicating with each other over network 204. In some cases, one or more of access point 202, programmer 24, server 206, and computing devices 212A-212N may be coupled to network 204 via one or more wireless connections. IMD16, programmer 24, server 206, and computing devices 212A-212N may each include one or more processors, such as one or more microprocessors, DSPs, ASICs, FPGAs, programmable logic circuitry, and the like, that may perform various functions and operations, such as those described herein.

Access point 202 may include devices that connect to network 204 via any of a variety of connections, such as telephone dial-up, Digital Subscriber Line (DSL), or cable modem connections. In other examples, access point 202 may be coupled to network 204 through different forms of connections (including wired or wireless connections). In some examples, access point 202 may be co-located with patient 14 and may include one or more programming units and/or computing devices (e.g., one or more monitoring units) that may perform the various functions and operations described herein. For example, access point 202 may include a home monitoring unit that is co-located with patient 14 and may monitor activity of IMD 16. In some examples, the server 206 or computing device 212 may control or perform any of the various functions or operations described herein, e.g., determining whether the phrenic nerve is being stimulated based on the heart sounds.

In some cases, server 206 may be configured to provide a secure storage site for archiving diagnostic information (e.g., the occurrence of phrenic nerve stimulation and accompanying circumstances, such as pacing parameters) that has been collected and generated from IMD16 and/or programmer 24. The network 204 may include a local area network, a wide area network, or a global network, such as the internet. In some cases, programmer 24 or server 206 may assemble (assign) PNS information in a web page or other document for review by a trained professional, such as a clinician, via viewing a terminal associated with computing device 212. The system of fig. 5 may be implemented in some aspects using common networking technologies and functionality similar to that provided by the Medtronic carelink.rtm. network developed by Medtronic corporation of minneapolis, minnesota.

In the example of fig. 5, external server 206 may receive heart sound information from IMD16 via network 204. Based on the received heart sound information, the one or several processors 210 may perform one or more of the functions described herein with respect to the signal analyzer 80 and the processor 70. In some examples, an external device, such as server 206 or computing device 212, may provide activation signals to IMD16 via network 204. In response to the activation signal, IMD16 may initiate a phrenic nerve detection sequence consistent with, for example, one or more of the methods described below. In some examples, the heart sound signal is transmitted to an external device that transmits the activation signal. An external device, such as server 206, processes the signals using the phrenic nerve detection features described below to determine whether phrenic nerve stimulation has occurred.

Figure 6 is a flow chart of a method for determining the presence of phrenic nerve stimulation in a medical device according to one example of the present disclosure. As shown in fig. 6, to determine the presence of phrenic nerve stimulation, a device (such as, for example, IMD16 or a device or monitoring system external to IMD 16) may perform this determination on a beat-by-beat basis by sensing PNS test signals sensitive to the patient's diaphragm contractions (such as, for example, acoustic signals or accelerometer signals) (block 300). The device determines a signal artifact of the sensed PNS test signal, the signal artifact occurring during a post-cardiac signal window following the predetermined cardiac signal and during a pre-cardiac signal window preceding the predetermined cardiac signal (block 302). Based on the confirmed signal artifact of the PNS test signal, the device determines whether the beat has satisfied the PNS detection criteria (block 304), and if the beat has satisfied the PNS beat criteria (i.e., the "yes" branch of block 304), the beat is identified as a PNS beat (block 306). On the other hand, if the beat does not satisfy the PNS beat criteria (i.e., the "no" branch of block 304), then the beat is identified as not being a PNS beat (block 308) and the process repeats for the next beat (block 310).

When the current beat is identified as being a PNS beat (block 306), the device determines whether PNS episode criteria have been met (block 312). If the PNS episode criteria have not been met (i.e., the "No" branch of block 312), the process repeats for the next beat (block 310). If the PNS episode criteria have been met (i.e., the "YES" branch of block 312), the apparatus determines that a PNS episode is occurring (block 314). In one example, once a particular number or sequence of PNS beats has been detected, the PNS episode detection criteria is determined to have been satisfied (i.e., the "yes" branch of block 312). For example, the PNS detection episode criteria may be satisfied if a predetermined number of consecutive beats (such as, for example, three beats) are identified as PNS beats on a beat-by-beat basis. According to another example, the PNS episode detection criteria may be met if PNS is detected up to a total number of beats for each predetermined order of beats, such as up to a total of three beats for each third beat, i.e., for example, beat (i), beat (i-3), beat (i-6). Once the PNS episode detection criteria are met and thus the PNS episode has been detected (block 314), the device may generate an alert, adjust a pacing vector for delivering pacing therapy, or suspend delivery of pacing therapy. In one example, the device may suspend determining the presence of PNS in response to a determination as to whether PNS episode detection criteria have been made for 11 consecutive beats without detecting one PNS episode.

Thus, in the example in fig. 6, the PNS test signal characteristics within the PNS window of the beat are evaluated for detecting or determining whether the beat is associated with the presence of PNS. If the beat is not determined to be a PNS beat, the beat is classified as a non-PNS beat and the process continues for the next beat identified by the PNS test signal until a predetermined number of beats have been evaluated or a predetermined time period has expired. On the other hand, if the beat is determined to be a PNS beat, the beat is classified as a PNS beat. In either case, i.e., whether the current beat is a PNS beat or not, the process continues for the next beat until a predetermined number of beats have been evaluated, a predetermined time period has expired, or a PNS episode is detected based on a predetermined number or sequence of beats being identified as PNS beats on a beat-by-beat basis.

Figure 7 is a flowchart of a method for determining the presence of phrenic nerve stimulation in a medical device according to one example of the present disclosure. In some cases, it may be desirable to ensure that: the beat is not caused by noise occurring in the PNS test signal as a result of the PNS beat determination. For example, as shown in fig. 7, to determine the presence of phrenic nerve stimulation, the device may sense a PNS test signal, such as an acoustic signal or an accelerometer signal, that is sensitive to contraction of the patient's diaphragm (block 300). The device determines signal artifacts for the test signal for each beat on a beat-by-beat basis (block 302) that occur during a noise window that extends a time period after the sensed predetermined cardiac signal, in addition to a cardiac signal pre-window that extends a time period before the sensed predetermined cardiac signal and a cardiac signal post-window that extends a time period after the sensed predetermined cardiac signal. Based on the determined signal artifacts of the determined noise window, the device determines whether the PNS noise criteria have been met (block 303). If the PNS noise criteria has been met (i.e., the "yes" branch of block 303), the beat is determined to be a noise beat (block 305), and the process repeats for the next beat (block 310). If the PNS noise criteria have not been met (i.e., the "no" branch of block 303), the device determines whether the beat has met the PNS beat criteria (block 304). If the beat has satisfied the PNS beat criteria (i.e., the "YES" branch of block 304), the beat is identified as a PNS beat (block 306). On the other hand, if the beat does not satisfy the PNS beat criteria (i.e., the "no" branch of block 304), then the beat is identified as not being a PNS beat (block 308) and the process repeats for the next beat (block 310).

Whenever the current beat is not identified as a noise beat and is identified as a PNS beat (block 306), the device determines whether PNS episode criteria have been met (block 312). If the PNS episode criteria have not been met (i.e., the "No" branch of block 312), the process repeats for the next beat (block 310). If the PNS episode criteria have been met (i.e., the "YES" branch of block 312), the apparatus determines that a PNS episode is occurring (block 314).

In this way, using only the PNS test signal, each beat is first evaluated to determine whether the beat is a noise beat, and if the beat is not a noise beat, the presence of the PNS of the beat is evaluated based on the PNS test signal to determine whether the beat is a PNS beat. If the beat is determined not to be associated with noise, but is determined to be a PNS beat, the device determines whether PNS episode detection criteria have been met. The determination of whether PNS episode detection criteria have been met may be based on a particular number or sequence of PNS beats being detected. For example, the PNS episode detection criteria may be satisfied if a predetermined number of consecutive beats (such as, for example, three beats) are identified as PNS beats on a beat-by-beat basis. According to another example, the PNS episode detection criteria may be met if PNS is detected up to a total number of beats for each predetermined order of beats, such as up to a total of three beats for each third beat, i.e., for example, beat (i), beat (i-3), beat (i-6). Once the PNS episode detection criteria are met and thus the PNS episode has been detected (block 312), the device may generate an alert, adjust a pacing vector for delivering pacing therapy, or suspend delivery of pacing therapy. In one example, the device may pause determining the PNS episode in response to a determination as to whether the PNS noise criteria and PNS beat criteria have been performed for 11 consecutive beats.

Thus, in the example in fig. 7, the PNS test signal characteristics within the noise window of a beat are first evaluated for detecting or determining whether the beat is associated with noise. If no noise is detected for the beat, the PNS test signal characteristics within the PNS window of the beat are evaluated for detecting or determining whether the beat is associated with the presence of PNS. If the beat is neither a noise beat nor a PNS beat, then the beat is classified as a non-PNS beat, and the process continues for the next beat identified by the PNS test signal until a predetermined number of beats have been evaluated, or a predetermined time period has expired. On the other hand, if the beat is not determined to be a noise beat and is determined to be a PNS beat, the beat is classified as a PNS beat. In either case, i.e., whether the current beat is a PNS beat or not, the process continues for the next beat until a predetermined number of beats have been evaluated, a predetermined time period has expired, or a PNS episode is detected based on the predetermined number or sequence of beats being identified as PNS beats on a beat-by-beat basis.

Fig. 8 is a graphical representation of determining acoustic artifacts of a heart sound signal for determining the presence of phrenic nerve stimulation in a medical device according to an example of the present disclosure. According to one example, the PNS test signal may be a heart sound signal sensed by the heart sound sensor 82 and the predetermined heart signal may be Vp beats sensed by the sensing module 76. In other embodiments, the predetermined cardiac signal utilized may be, for example, a right atrial pacing (Ap) beat or a right atrial sensing (As) beat. In examples where the PNS test signal is a heart sound signal and the predetermined heart signal utilized is a Vp beat, a device (such as, for example, IMD16 or a device or monitoring system external to IMD 16) determines the presence of a PNS episode on a beat-by-beat basis using acoustic artifacts of the heart sound signal that occur during the Vp beat. The sensed heart sound signal data may be digitized by a 16-bit ADC sampling 64mV at 256Hz and bandpass filtered. The absolute value of the filtered heart sound signal (| FHS |) is determined and a predetermined number of beats is evaluated for PNS at a voltage and/or polarity that can be selected by the user.

In particular, as shown in fig. 8, the device may determine acoustic artifacts from the sensed heart sound signal 318 associated with VP beat 320 and identify PNS windows for VP beat 320, including a pre-ventricular pacing (PreVp) window 322 and a post-ventricular pacing (PostVp) window 326. According to another example, which includes detecting whether a current beat is a noise beat, to determine acoustic artifacts from the heart sound signal 318 associated with VP beats, the device identifies a noise window 328 for the beat 320 in addition to a PNS window including a pre-ventricular pacing (PreVp) window 322 and a post-ventricular pacing (PostVp) window 326.

According to one example, a prev window 322 may extend along the heart sound signal 318 from a point 324 to the beat 320, the prev window 322 being a predetermined time period prior to the beat 320, such as an atrial pacing (Ap) beat. Both the PostVp window 326 and the noise window 328 extend during a predetermined period of time of the heart sound signal 318 after the beat 320, and the noise window 328 extends beyond the PostVp window. In one example, the PreVp window 322 may extend for 25 samples (98ms) before the beat 320 and including 320, the PostVp window 326 may extend for a period extending between 27ms and 83ms after the beat 320, i.e., for a sampling period between the 7 th and 21 st samples after the beat 320, and the noise signal 328 may extend for a period extending between 27ms and 313ms after the beat 320, i.e., for a sampling period between the 7 th and 80 th samples after the beat 320. In another example, if the paced AV delay (PAV) or the perceived AV delay (SAV) is less than 25 samples, the PreVp window 322 may precede the beat 320 and include the beat 320 by less than 25 samples. In another example, the PreVp window 322 may be 150 ms.

Fig. 9 is a flow chart of a method for determining the presence of phrenic nerve stimulation in a medical device according to one example of the present disclosure. As shown in fig. 8 and 9, during phrenic nerve stimulation detection in the example where the PNS test signal is a heart sound signal and the predetermined heart signal is a Vp beat, the device senses the heart sound signal 318 and determines a heart sound processing window for the currently beating heart sound signal 318 (block 340). For example, the device determines the heart sound signal based PNS windows, i.e., the prev window 322 and the PostVp window 326, of the currently perceived Vp beat. Signal artifacts (e.g., one or more of maximum, minimum, range, average, sum, and absolute difference) of the heart sound signal are calculated within the PNS window based on the heart sound signal, i.e., the prevvp window 322 and the PostVp window 326. The absolute difference is similar to the Standard Deviation (SD) and is calculated by subtracting the mean from the signal, adding the absolute value of the resulting time series and dividing by the length of the signal. Using the determined heart sound signal artifact, the device determines whether the beat meets PNS beat criteria. If the PNS beat criteria are met, then the beat is determined to be a PNS beat. On the other hand, if the PNS beat criterion is not met, the beat is not determined to be a PNS beat.

To determine whether the PNS beat criterion is met, the device determines whether a maximum criterion of the absolute value of the filtered heart sound signal | FHS | within the postVp window 326 is met and a sum criterion of the absolute value of the filtered heart sound signal | FHS | within the postVp window 326 is met. For example, the device may determine whether the maximum of the absolute value of the filtered heart sound signal | FHS | within postVp window 326 is greater than the PNS maximum threshold (block 342).

In one example, the PNS maximum threshold in block 342 of fig. 9 may be set to the sum of three times the average of the absolute values | FHS | of the filtered heart sound signals within the PreVp window 322 and two times the standard deviation (i.e., absolute difference) of the filtered heart sound signals | FHS | within the PreVp window 322.

In another example, the PNS max threshold in block 342 may be set to a variable PNS max threshold a. For example, as shown in table 1 below, the variable PNS maximum threshold α may be a function of the range of the filtered heart sound signal FHS in the noise window 328. For example, as shown in table 1, if the range of the filtered heart sound signal FHS is less than 1070 ADC units, the variable PNS maximum threshold α may be set to 80 ADC units; if the filtered heart sound signal FHS ranges between 1070 ADC units and 7000 ADC units, the variable PNS maximum threshold α may be set to 500 ADC units; and if the range of the filtered heart sound signal FHS is greater than or equal to 7000 ADC units, the variable PNS maximum threshold a may be set to 1800 ADC units.

Range of FHS	α	β
			<1070	80	250
1070 to<7000	350	1000
			≥7000	1800	400

TABLE 1

In another example, if the absolute value of the filtered heart sound signal | FHS | within PostVp window 326 is determined to be greater than both three times the average of the absolute value of the filtered heart sound signal | FHS | within prev window 322 plus two times the standard deviation of the filtered heart sound signal | FHS | within prev window 322 and greater than the variable PNS maximum threshold α, the apparatus may determine that the maximum value of the absolute value of the filtered heart sound signal | FHS | within PostVp window 326 is greater than the PNS maximum threshold (i.e., "yes" branch of block 342).

If the maximum value of the absolute value of the filtered heart sound signal | FHS | within postVp window 326 is greater than the PNS maximum sum threshold (i.e., "yes" branch of block 342), the apparatus determines whether the sum of the absolute value of the filtered heart sound signal | FHS | within postVp window 326 is greater than the PNS sum threshold (block 344). If the sum of the absolute values of the filtered heart sound signals | FHS | within postVp window 326 is greater than the variable PNS sum threshold (i.e., "yes" branch of block 344), the current beat is determined to be a PNS beat (block 346). On the other hand, if the maximum value of the absolute value of the filtered heart signal | FHS | within postVp window 326 is not greater than the PNS maximum sum threshold (i.e., "no" branch of block 342), or the sum of the absolute value of the filtered heart sound signal | FHS | within postVp window 326 is not greater than the PNS sum threshold (i.e., "no" branch of block 344), then the current Vp beat is not determined to be a PNS beat (block 348).

In one example, the PNS sum threshold of block 344 may be the sum of the absolute value of the filtered heart sound signal | FHS | within the PreVp window 322 and the PNS beat sum variable threshold β. For example, as shown in table 1, the variable PNS sum threshold β may be a function of the range of the filtered heart sound signal FHS in the noise window 328. For example, as shown in table 1, if the range of the filtered heart sound signal FHS is less than 1070 ADC units, the variable PNS sum value threshold β may be set to 250 ADC units; if the filtered heart sound signal FHS ranges between 1070 ADC units and 7000 ADC units, the variable PNS sum value threshold β may be set to 1000 ADC units; and if the range of the filtered heart sound signal FHS is greater than or equal to 7000 ADC units, the variable PNS sum value threshold β may be set to 400 ADC units.

In another example, the PNS sum threshold of block 344 may be a multiple of the sum of the absolute values | FHS | of the filtered heart sound signals within the PreVp window 322, such as, for example, 1.25 times the sum of the absolute values | FHS | of the filtered heart sound signals within the PreVp window 322. In one example, if the absolute value of the filtered heart sound signal | FHS | within the PostVp window 326 is determined to be both greater than the sum of the absolute values of the filtered heart sound signals | FHS | within the prev window 322 and greater than a multiple of the sum of the absolute values of the filtered heart sound signals | FHS | within the prev window 322 (i.e., e.g., 1.25 times the sum of the absolute values of the filtered heart sound signals | FHS | within the prev window 322), the apparatus may determine that the sum of the absolute values of the filtered heart sound signals | FHS | within the PostVp window 326 is greater than the PNS sum threshold (i.e., "yes" branch in block 344).

Whenever the current beat is identified as being a PNS beat (block 346), the device determines whether PNS episode criteria have been met (block 352). If the PNS episode criteria has not been met (i.e., the "no" branch in block 352) or the current Vp beat is determined not to be a PNS beat (block 348), the process continues with the next beat (block 350), such that once the confirmation as to whether the current beat 330 is a PNS beat has been completed, the process repeats for the next beat 330 until a predetermined number of beats have been evaluated for the presence of PNS. In one example, to ensure that PNS assessment is completed in less than 3 minutes (176 s cycle length 11x 16x 1000 ms) for 16 vectors at an estimated heart rate of 60bpm, up to 11 beats may be assessed for PNS at a voltage output selectable by the user. Thus, the device may pause determining a PNS episode in response to the PNS criteria having been determined for 11 consecutive beats and no PNS episode being detected.

If the PNS episode criteria have been met (i.e., "Yes" branch of block 352), a PNS episode is detected (block 354). In one example, if a predetermined number of consecutive beats (such as, for example, three consecutive beats) are identified as PNS beats on a beat-by-beat basis, it is determined that the PNS episode criteria are met (i.e., the "yes" branch of block 352), and thus a PNS episode is detected (block 354). In another example, if PNS is detected up to a total of one beat for each predetermined order of beats, such as up to a total of three beats for each third beat, i.e., for example, beat (i), beat (i-3), beat (i-6) are determined to be PNS beats, then it is determined that the PNS episode criteria are met (i.e., the "yes" branch of block 352), and thus a PNS episode is detected (block 354). Once a PNS episode is detected (block 354), the device may store a determination that a PNS episode was detected, generate an alert, adjust a pacing vector for delivering pacing therapy, or suspend delivery of pacing therapy.

Figure 10 is a flowchart of a method for determining the presence of phrenic nerve stimulation in a medical device according to one example of the present disclosure. In some cases, it may be desirable to ensure that the determination that the beat is a PNS beat is not due to noise occurring in the test signal. For example, as shown in fig. 8 and 10, during detection of phrenic nerve stimulation, the device senses the heart sound signal 318 and, in determining the heart sound processing window for the currently beating heart sound signal 318 (block 340), determines a heart sound based noise window in addition to the heart sound based PNS windows described above (i.e., the prev window 322 and the PostVp window 326). Thus, signal characteristics of the heart sound signal (such as one or more of a maximum, a minimum, a range, an average, a sum, and an absolute difference) are calculated within each of the heart sound based noise window 328 and the heart sound based PNS window (i.e., the prev window 322 and the PostVp window 326). The absolute difference is similar to the Standard Deviation (SD) and is calculated by subtracting the mean from the signal, adding the absolute value of the resulting time series and dividing by the length of the signal. In the example of fig. 10, the heart sound signal characteristics within the noise window 328 of a beat are first evaluated to detect or determine whether the beat is associated with noise. If no noise is detected for the beat, the heart sound signal characteristics within the PreVp window 322 and PostVp window 326 of the beat are then evaluated to detect or determine whether the beat is associated with the presence of PNS.

As shown in the example of fig. 10, during evaluation of the noise window 328, the device determines whether the sum of the absolute values | FHS | of the filtered heart sound signals within the noise window 328 is greater than a noise sum threshold (block 356). If the sum of the absolute values of the filtered heart sound signals, FHS, within the noise window 328 is greater than the noise sum threshold (i.e., the "yes" branch of block 356), the apparatus determines whether the range of the filtered heart sound signals, FHS, in the noise window 328 (i.e., the difference between the maximum and minimum values) is less than the noise range threshold (block 358). If the range of the filtered heart sound signal FHS in the noise window 328 is less than the noise threshold range (i.e., the "YES" branch of block 358), the current beat is identified as a noise beat (block 360), and the process continues with the next beat (block 350).

If the sum of the absolute values of the filtered heart sound signals FHS within the noise window 328 is not greater than the noise sum threshold (i.e., the "no" branch of block 356), or the range of the filtered heart sound signals FHS in the noise window 328 is not less than the noise range threshold (i.e., the "no" branch of block 358), the apparatus determines whether the combination of the sum of the absolute values of the filtered heart sound signals FHS and the range of the filtered heart sound signals FHS is greater than the combination threshold (block 362). In one example, to determine whether the combination is greater than the combination threshold in block 362, the device determines whether the product of the sum of the absolute values of the filtered heart sound signals | FHS | in the noise window 328 and the range of the filtered heart sound signals FHS is greater than the combination threshold.

If the sum of the absolute value of the filtered heart sound signal | FHS | in the noise window 328 and the range of the filtered heart sound signal FHS are greater than the combination threshold (i.e., the "YES" branch of block 362), the current beat is identified as a noise beat (block 360) and the process continues with the next beat (block 350). On the other hand, if the sum of the absolute value of the filtered heart sound signal | FHS | in the noise window 328 and the range of the filtered heart sound signal FHS are not greater than the combination threshold (i.e., the "no" branch of block 362), the current beat is identified as not being a noise beat (block 364). According to one example, if the combination is a product of two thresholds, the noise sum threshold may be set to 22900 ADC units, the noise range threshold may be set to 1000 ADC units, and the combination threshold may be set to 170000000 ADC units.

In this way, if the sum of the absolute values of the filtered heart sound signals FHS within the noise window 328 is greater than the noise sum threshold (i.e., the "yes" branch in block 356) and the range of the filtered heart sound signals FHS in the noise window 328 is less than the noise range threshold (i.e., the "no" branch in block 358), or if the product of the sum of the absolute values of the filtered heart sound signals FHS in the noise window 328 and the range of the filtered heart sound signals FHS is greater than the combination threshold (i.e., the "yes" branch in block 362), the noise beat criterion for the current beat is satisfied and the beat is determined to be a noise beat (block 360). On the other hand, if the sum of the absolute values of the filtered heart sound signals FHS within the noise window 328 is not greater than the noise sum threshold (i.e., the "no" branch in block 356), or the range of the filtered heart sound signals FHS in the noise window 328 is not less than the noise range threshold (i.e., the "no" branch in block 358), and the product of the sum of the absolute values of the filtered heart sound signals FHS in the noise window 328 and the range of the filtered heart sound signals FHS is not greater than the combination threshold (i.e., the "no" branch in block 362), then the current beat does not satisfy the noise beat criterion and the beat is not determined to be a noise beat (block 364).

When the noise beat criteria for the current beat are not met (blocks 356, 358, and 362) and thus the current beat is not determined to be a noise beat (block 364), the device determines whether the beat meets the PNS beat criteria using the method for such determination as described above in fig. 9, and this determination is not repeated here for simplicity. Whenever the current beat is identified as being a PNS beat (block 346), the device determines whether PNS episode criteria have been met (block 352). If the PNS episode criteria has not been met (i.e., the "no" branch in block 352) or the current Vp beat is determined not to be a PNS beat (block 348), the process continues with the next beat (block 350), such that once the confirmation as to whether the current beat 330 is a PNS beat has been completed, the process repeats for the next beat 330 until a predetermined number of beats have been evaluated for the presence of PNS. In one example, to ensure that PNS assessment is completed in less than 3 minutes (176 s cycle length 11x 16x 1000 ms) for 16 vectors at an estimated heart rate of 60bpm, an assessment of PNS may be made for up to 11 beats at a voltage output selectable by the user. Thus, the device may pause determining a PNS episode in response to the PNS criteria having been determined for 11 consecutive beats and no PNS episode being detected.

If the PNS episode criteria have been met (i.e., "Yes" branch of block 352), a PNS episode is detected (block 354). In one example, if a predetermined number of consecutive beats (such as, for example, three consecutive beats) are identified as PNS beats on a beat-by-beat basis, it is determined that the PNS episode criteria are met (i.e., the "yes" branch of block 352), and thus a PNS episode is detected (block 354). In another example, if PNS is detected up to a total of one beat for each predetermined order of beats, such as up to a total of three beats for each third beat, i.e., for example, beat (i), beat (i-3), beat (i-6) are determined to be PNS beats, then it is determined that the PNS episode criteria are met (i.e., the "yes" branch of block 352), and thus a PNS episode is detected (block 354). Once a PNS episode is detected (block 354), the device may store a determination that a PNS episode was detected, generate an alert, adjust a pacing vector for delivering pacing therapy, or suspend delivery of pacing therapy.

The PNS beat criteria and the noise beat criteria for determining phrenic nerve stimulation in a medical device according to one example of the present disclosure may be summarized as follows:

noise beat criterion ═ a and B or C

A: summation of | FHS | in noise window >22900

B: range of FHS in noise window <1000

C：A x B>170000000

PNS beat criteria D and E and F and G

D: maximum value of | FHS | in PostVp >3x PreVp average value of | FHS | + SD of | FHS | in 2x PreVp

E: maximum value of | FHS | in PostVp > α

F: the sum of | FHS | in PostVp > the sum of | FHS | in PreVp + β

G: the sum of | FHS | in PostVp >1.25x PreVp

In the example shown in fig. 9, the device determines acoustic artifacts of the sensed heart sound signal within a pre-ventricular pacing (PreVP) window 322 and a post-ventricular pacing (PostVP) window 326 of the sensed heart sound signal during ventricular pacing beats 320 delivered by the device. In response to acoustic artifacts of the sensed heart sound signal within a pre-ventricular pacing (PreVP) window and a post-ventricular pacing (PostVP) window, the device determines whether PNS criteria are met. In response to the PNS criteria being met, the device determines whether the PNS episode criteria have been met, and detects a PNS episode in response to the PNS episode criteria being met. It should be understood that while any combination of PNS beat criteria D-G may be utilized for determining whether the PNS beat criteria have been met, in one example, if the maximum value of | FHS | within the PostVp window is determined to be greater than the sum of three times the average value of | FHS | within the PreVp window and two times the standard deviation of | FHS | within the PreVp window (PNS beat criteria D); the maximum value of FHS within the PostVp window is determined to be greater than a variable PNS maximum threshold value alpha (PNS beat criteria E), where the variable PNS maximum threshold value is a function of the range of the filtered heart sound signal (FHS) in the noise window; the sum of FHS within the PostVp window is determined to be greater than the sum of FHS within the PreVp window and a variable PNS sum threshold β (PNS beat criteria F), where the variable PNS beat sum threshold is a function of the range of the filtered heart sound signal FHS in the noise window; and the sum of FHS within the PostVp window is determined to be greater than a multiple of the sum of FHS within the PreVp window (PNS beat criteria G), the device may identify the current beat as a PNS beat.

In the example shown in fig. 10, the device determines acoustic artifacts using each of a noise window 328 of the heart sound signal sensed during a ventricular pacing beat 320 delivered by the device, a pre-ventricular pacing (PreVP) window 322, and a post-ventricular pacing (PostVP) window 326. The device determines whether a noise criterion has been met in response to acoustic artifacts of the sensed heart sound signal within a noise window, and determines whether a PNS criterion has been met in response to acoustic artifacts of the sensed heart sound signal within a pre-ventricular pacing (PreVP) window and a post-ventricular pacing (PostVP) window. In response to the PNS criteria being met, the device determines whether the PNS episode criteria have been met, and detects a PNS episode in response to the PNS episode criteria being met.

It should be understood that while any combination of PNS noise beat criteria a-C may be utilized for determining whether the PNS noise criteria have been met, in one example, the device may identify the current beat as a noise beat in response to both the sum of | FHS | within the noise window being greater than the noise sum threshold (PNS noise beat criteria a) and the range of FHS within the noise window being less than the noise range threshold (PNS noise beat criteria B), or in response to the product of the sum of | FHS | in the noise window and the range of FHS in the noise window being greater than the combination threshold (PNS noise beat criteria C). Further, the device identifies the current beat as not being a noise beat in response to at least one of the sum of FHS within the noise window not being greater than a noise sum threshold and the range of FHS within the noise window not being less than a noise range threshold and the product of the sum of FHS in the noise window and the range of FHS in the noise window not being greater than a combination threshold.

Techniques for detecting stimulation of one or more of phrenic nerves 36 and 38 are primarily described herein as being performed by IMD16, e.g., by a processor of IMD 16. In other examples, some or all of the functionality attributed to IMD16 or its processor may be performed by one or more devices, such as programmer 24 or its processor. For example, IMD16 may process cardiac and/or heart sound signals to determine whether therapy should continue to be delivered based on current parameters, or whether parameters should be adjusted and control parameters used by IMD16 for delivering therapy. Alternatively, programmer 24 may process cardiac and/or heart sound signals received from IMD16 to determine whether therapy should continue to be delivered based on current parameters, or whether parameters should be adjusted and controlled according to parameters for which IMD16 delivers therapy. Further, although described herein with respect to IMDs, in other examples, the techniques described herein may be performed or implemented in an external medical device that may be coupled to a patient via a percutaneous or percutaneous lead. In some examples, various functions of IMD16 may be performed by multiple IMDs in communication with each other.

Illustrative embodiments