CN117479980A

CN117479980A - Method and apparatus for establishing parameters for cardiac event detection

Info

Publication number: CN117479980A
Application number: CN202280042046.2A
Authority: CN
Inventors: T·J·谢尔顿; K·M·埃斯卡兰特; M·A·霍尔姆; P·R·索海姆
Original assignee: Medtronic Inc
Current assignee: Medtronic Inc
Priority date: 2021-06-22
Filing date: 2022-06-06
Publication date: 2024-01-30

Abstract

A medical device having a motion sensor is configured to set an atrial event sensing parameter for sensing an atrial contraction event from a motion signal generated by the motion sensor. The medical device sets atrial event sensing parameters by: applying a sensing window during each of a plurality of ventricular cycles; determining a characteristic of the motion signal during the sensing window for at least a portion of the ventricular cycle; and setting an atrial event sensing parameter based on the determined characteristic. The medical device may sense an atrial event from the motion signal according to an atrial event sensing parameter.

Description

Method and apparatus for establishing parameters for cardiac event detection

Technical Field

The present disclosure relates to a medical device and method for establishing parameters for detecting cardiac events from motion sensor signals.

Background

Implantable cardiac pacemakers are typically placed in a subcutaneous pocket and coupled to one or more transvenous medical electrical leads carrying pacing and sensing electrodes positioned in the heart. The subcutaneously implanted cardiac pacemaker may be a single-chamber pacemaker coupled to one transvenous medical lead to position the electrodes in one heart chamber (atrial or ventricular) or a dual-chamber pacemaker coupled to two intracardiac leads to position the electrodes in both the atrial and ventricular chambers. A multi-lumen pacemaker may also be used that may be coupled to three transvenous leads, for example, to position electrodes for pacing and sensing in one atrial chamber and both the left and right ventricles.

Recently, intracardiac pacemakers have been introduced that can be implanted within the ventricular chambers of a patient's heart to deliver ventricular pacing pulses. Such pacemakers may sense R-wave signals accompanying intrinsic ventricular depolarizations, and deliver ventricular pacing pulses in the absence of sensed R-waves. While single-chamber ventricular sensing and pacing by an intracardiac ventricular pacemaker may adequately address some patient conditions, some patients may benefit from atrial and ventricular (dual-chamber) sensing for providing atrial synchronized ventricular pacing in order to maintain a regular cardiac rhythm.

Disclosure of Invention

The technology of the present disclosure generally relates to a pacemaker having a motion sensor that generates motion signals including ventricular and atrial event signals. The pacemaker is configured to sense atrial events from the motion signal. In some examples, sensed atrial events may be used to control atrial synchronized ventricular pacing pulses delivered by a pacemaker. Pacemakers operating in accordance with the techniques disclosed herein determine one or more atrial event sensing parameters for sensing atrial event signals by determining characteristics of a motion signal and setting the atrial event sensing parameters based on the determined characteristics over a plurality of ventricular cycles. In some examples, the atrial event sensing parameters are set based on a distribution of the features (e.g., based on a percentile or median or other central metric of the distribution).

In one example, the present disclosure provides a medical device comprising: a motion sensor configured to sense a motion signal; and a control circuit coupled to the motion sensor for receiving the sensed motion signal. The control circuit is configured to: a sensing window applied during each of a plurality of ventricular cycles; determining that the sensed motion signal after the end time of the sensing window meets atrial event criteria for at least a portion of the ventricular cycle; determining a characteristic of the motion signal sensed during each of the sensing windows associated with the portion of the ventricular cycle for which the motion signal meets the atrial event criteria; and setting an atrial event sensing parameter based on the determined characteristic. The control circuit is configured to sense an atrial event from the motion signal according to an atrial event sensing control parameter, and may generate a sensed atrial event signal in response to sensing the atrial event.

In another example, the present disclosure provides a method performed by a medical device. The method comprises the following steps: sensing a motion signal; applying a sensing window during each of a plurality of ventricular cycles; determining that the sensed motion signal after the end time of the sensing window meets atrial event criteria for at least a portion of the ventricular cycle; determining a characteristic of a motion signal sensed during each of a plurality of sensing windows associated with portions of the ventricular cycles for which the motion signal meets atrial event criteria; and setting an atrial event sensing parameter based on the determined characteristic. The method may include: an atrial event signal is sensed from the motion signal according to an atrial event sensing control parameter, and a sensed atrial event signal is generated in response to the sensed atrial event signal.

In another example, the present disclosure provides a non-transitory computer-readable storage medium storing a set of instructions that, when executed by control circuitry of a medical device, cause the device to: sensing a motion signal; applying a sensing window during each of a plurality of ventricular cycles; determining that the sensed motion signal after the end time of the sensing window meets atrial event criteria for at least a portion of the ventricular cycle; determining a characteristic of a motion signal sensed during each of a plurality of sensing windows associated with portions of the ventricular cycles for which the motion signal meets atrial event criteria; and setting an atrial event sensing parameter based on the determined characteristic. The instructions may further cause the medical device to sense an atrial event from the motion signal according to an atrial event sensing control parameter, and generate a sensed atrial event signal in response to sensing the atrial event.

Further disclosed herein are the subject matter of the following clauses:

1. a medical device, the medical device comprising:

a motion sensor configured to sense a motion signal; and

control circuitry coupled to the motion sensor for receiving the motion signal and configured to:

Setting a sensing window during each of a plurality of ventricular cycles;

determining that the sensed motion signal after the end time of the sensing window meets atrial event criteria for at least a portion of a plurality of ventricular cycles;

determining a first characteristic of a motion signal sensed during each of a plurality of ventricular cycles for which a motion signal sensed after an end time of the sensing window meets atrial event criteria;

setting an atrial event sensing parameter based on the determined first characteristic;

sensing an atrial event signal from the motion signal according to the atrial event sensing control parameter; and

a sensed atrial event signal is generated in response to sensing the atrial contraction event.

2. The medical device of clause 1, wherein the control circuit is configured to:

determining a first characteristic of the motion signal during each of the sensing windows by determining a first maximum amplitude of the motion signal during each of the sensing windows associated with portions of the plurality of ventricular cycles;

setting an atrial event sensing parameter by setting an early atrial event sensing threshold amplitude based on the first maximum amplitude; and

Atrial contraction events are sensed from the motion signal in response to the motion signal crossing an early atrial event sensing threshold amplitude during a sensing window of the ventricular cycle.

3. The medical device of any one of clauses 1-2, wherein the control circuit is further configured to:

setting a first test threshold amplitude;

determining, by the motion signal, a crossing time of the first test threshold during each of the sensing windows associated with portions of the plurality of ventricular cycles;

setting an atrial event sensing parameter based on the determined first characteristic by adjusting an end time of the sensing window based on the crossing time; and

sensing an atrial event signal from the motion signal in response to the motion signal crossing one of:

a first sensing threshold amplitude prior to an end time of the sensing window; and a second sensing threshold amplitude after an end time of the sensing window.

4. The medical device of clause 3, wherein the control circuit is further configured to:

determining a second test threshold amplitude based on a motion signal sensed during each of the plurality of ventricular cycles after an end time of the sensing window;

Setting a sensing window according to the adjusted end time of the sensing window during each of a subsequent plurality of ventricular cycles;

determining a crossing time of the second test threshold amplitude during a sensing window of each of a subsequent plurality of ventricular cycles; and

a second adjusted end time of the sensing window is determined based on the determined crossing time of the second test threshold amplitude.

5. The medical device of clause 4, wherein the control circuit is further configured to:

setting a range of end times of the sensing window based on the second adjusted end time of the sensing window; and

the end time of the sensing window is adjusted within this range.

6. The medical device of clause 5, wherein the control circuit is further configured to:

setting a range according to the first offset when the second adjusted end time is less than the threshold end time; and

when the second adjusted end time is greater than or equal to the threshold end time, the range is set according to a second offset, the second offset being different from the first offset.

7. The medical device of any one of clauses 1-6, further comprising a pulse generator configured to generate pacing pulses in response to the sensed atrial event signal.

8. The medical device of any one of clauses 1-7, further comprising:

a sensing circuit including an R-wave detector for sensing R-waves from the cardiac electrical signal; and

a pulse generator configured to generate ventricular pacing pulses in a non-atrial tracking pacing mode;

wherein the control circuit is configured to: a sensing window is set during each of a plurality of ventricular cycles in response to one of a ventricular pacing pulse generated by a pulse generator and an R-wave sensed by a sensing circuit during a non-atrial tracking pacing mode.

9. The medical device of any one of clauses 1-8, wherein the control circuit is configured to determine that the sensed motion signal after the end time of the sensing window meets an atrial event criterion by:

determining a second maximum amplitude of the sensed motion signal after an end time of the sensing window; and

the second maximum amplitude is determined to be greater than a predetermined confidence atrial event threshold amplitude.

10. The medical device of any one of clauses 1-9, wherein the control circuit is configured to determine that the sensed motion signal after the end time of the sensing window meets an atrial event criterion by:

Determining a time interval from an end time of the sensing window to a maximum peak of the motion signal after the end time of the sensing window; and

the time interval is determined to be within the confidence atrial event time interval region.

11. The medical device of any one of clauses 1-10, wherein the control circuit is configured to determine that the sensed motion signal after the end time of the sensing window meets an atrial event criterion by: a motion signal received after an end time of the sensing window is determined to cross a confidence atrial event threshold amplitude.

12. The medical device of any one of clauses 1-11, wherein the control circuit is configured to determine that the sensed motion signal after the sensing window end time meets an atrial event criterion by: a motion signal received after the end time of the sensing window is determined to cross a confidence atrial event threshold amplitude during the atrial event time interval region.

13. The medical device of any one of clauses 1-12, wherein the control circuit is configured to determine that the sensed motion signal after the end time of the sensing window meets an atrial event criterion by:

Determining a morphological feature of the motion signal sensed after an end time of the sensing window; and

and determining that the morphology features match the morphology features of the atrial event.

14. The medical device of any one of clauses 1-13, wherein the control circuit is configured to determine that the sensed motion signal after the end time of the sensing window meets an atrial event criterion by:

determining a first maximum amplitude of the motion signal during the sensing window;

determining a second maximum amplitude of the motion signal after an end time of the sensing window;

determining a ratio of the first maximum amplitude to the second maximum amplitude;

determining that the ratio meets an amplitude ratio requirement; and

the sensed motion signal after the end time of the sensing window is determined to meet an atrial event criterion in response to the ratio meeting an amplitude ratio requirement.

15. The medical device of clause 14, wherein the control circuit is further configured to enable the amplitude ratio requirement by:

determining a first maximum amplitude of the motion signal sensed during the sensing window for each of the plurality of ventricular cycles;

determining a second maximum amplitude of the sensed motion signal for each of the plurality of ventricular cycles after an end time of the sensing window;

Determining that the determined first maximum amplitude and second maximum amplitude for each of the plurality of ventricular cycles meet an enablement criterion; and

the amplitude ratio requirement is enabled in response to the enablement criteria being met.

16. The medical device of any one of clauses 1-15, wherein the control circuit is configured to: the atrial event sensing parameter is set based on the first characteristic by setting the atrial event sensing parameter based on a predetermined percentile of the first characteristic.

17. The medical device of any one of clauses 1-16, wherein the control circuit is configured to:

determining a center metric of the first feature;

determining an expansion metric of the first feature;

determining an offset based on the expansion metric; and

the atrial event sensing parameter is set based on the first characteristic by setting the atrial event sensing parameter to a center metric plus an offset.

18. The medical device of any one of clauses 1-16, wherein the control circuit is further configured to:

determining a first characteristic of the motion signal during each sensing window from all of the plurality of ventricular cycles; and

the atrial event sensing parameters are set by:

Determining a first predetermined percentile of a first characteristic determined during each of a plurality of sensing windows associated with portions of ventricular cycles for which a sensed motion signal after an end time of the sensing window meets atrial event criteria;

identifying a first feature greater than a first predetermined percentile determined from all of the plurality of ventricular cycles;

determining a second predetermined percentile of the identified first features determined from all of the plurality of ventricular cycles, the second predetermined percentile being greater than the first predetermined percentile; and

an atrial event sensing parameter is set based on the second predetermined percentile.

19. The medical device of any one of clauses 1-18, wherein:

the motion sensor includes a plurality of axes, each axis of the plurality of axes configured to generate an axis signal; and

the control circuit is further configured to select the motion signal by selecting a first vector signal comprising at least one of the axis signals.

20. The medical device of clause 19, wherein the control circuit is further configured to:

determining a second characteristic of the motion signal sensed after an end time of the sensing window for each of the plurality of ventricular cycles;

Determining that the second characteristic meets a vector acceptance criterion; and

an atrial event sensing parameter is set based on the determined first characteristic in response to the second characteristic meeting the vector acceptance criterion.

21. The medical device of clause 20, wherein the control circuit is configured to:

determining a second characteristic by determining a maximum amplitude of the motion signal sensed after an end time of the sensing window; and

determining that the second characteristic meets the vector acceptance criterion by:

determining a count of maximum amplitudes greater than a noise threshold amplitude after an end time of the sensing window; and

the count of the maximum amplitude is determined to be greater than a threshold.

22. The medical device of any one of clauses 20-21, wherein the control circuit is configured to select a second vector signal as the motion signal in response to determining that the second characteristic does not meet the vector acceptance criterion, the second vector signal comprising at least one axis signal different from the first vector signal.

23. A method, the method comprising:

sensing a motion signal;

setting a sensing window during each of a plurality of ventricular cycles;

determining that the sensed motion signal after the end time of the sensing window meets atrial event criteria for at least a portion of the ventricular cycle;

a sensed atrial event signal is generated in response to sensing an atrial contraction event.

24. The method according to clause 23, comprising

atrial contraction events are sensed from the motion signal in response to the motion signal crossing an early atrial event sensing threshold amplitude during a sensing window of a ventricular cycle.

25. The method of any one of clauses 23 to 24, further comprising:

Setting a first test threshold amplitude;

a first sensing threshold amplitude prior to an adjusted end time of the sensing window; and

a second sensing threshold amplitude after the adjusted end time of the sensing window.

26. The method of clause 25, further comprising:

27. The method of clause 26, further comprising:

the end time of the sensing window is adjusted within this range.

28. The method of clause 27, further comprising:

29. The method of any of clauses 23-28, further comprising generating a pacing pulse in response to the sensed atrial event signal.

30. The method of any one of clauses 23 to 29, further comprising:

sensing R-waves from cardiac electrical signals;

generating ventricular pacing pulses in a non-atrial tracking pacing mode; and

a sensing window is set during each of a plurality of ventricular cycles in response to one of a ventricular pacing pulse generated during a non-atrial tracking pacing mode and an R-wave sensed by a sensing circuit.

31. The method of any of clauses 23-30, wherein determining that the sensed motion signal after the end time of the sensing window meets the atrial event criteria comprises:

determining a second maximum amplitude of the motion signal sensed after the end time of the sensing window; and

32. The medical device of any one of clauses 23-31, wherein determining that the sensed motion signal after the end time of the sensing window meets the atrial event criteria comprises:

33. The method of any of clauses 23-32, wherein determining that the sensed motion signal after the end time of the sensing window meets the atrial event criteria comprises: a motion signal received after an end time of the sensing window is determined to cross a confidence atrial event threshold amplitude.

34. The method of any of clauses 23-33, wherein determining that the sensed motion signal after the end time of the sensing window meets the atrial event criteria comprises: a motion signal received after the end time of the sensing window is determined to cross a confidence atrial event threshold amplitude during the atrial event time interval region.

35. The method of any of clauses 23-34, wherein determining that the sensed motion signal after the end time of the sensing window meets the atrial event criteria comprises:

determining a morphological feature of the motion signal after an end time of the sensing window; and

36. The method of any of clauses 23-35, wherein determining that the sensed motion signal after the end time of the sensing window meets the atrial event criteria comprises:

determining that the ratio meets an amplitude ratio requirement; and

37. The method of clause 36, further comprising enabling the amplitude ratio requirement by:

38. The medical device of any one of clauses 23-37, wherein setting the atrial event sensing parameter based on the first characteristic comprises: an atrial event sensing parameter is set based on a predetermined percentile of the first characteristic.

39. The method of any one of clauses 23 to 38, further comprising:

determining a center metric of the first feature;

determining an expansion metric of the first feature;

determining an offset based on the expansion metric; and

40. The method of any one of clauses 23 to 38, further comprising:

Wherein setting the atrial event sensing parameter comprises:

41. The method of any of clauses 23-40, further comprising selecting the motion signal by selecting a first vector signal comprising at least one of the plurality of motion sensor axis signals.

42. The method of clause 41, further comprising:

43. The method of clause 42, further comprising:

44. The method of any of clauses 42-43, further comprising selecting a second vector signal as the motion signal in response to determining that the second characteristic does not meet the vector acceptance criterion, the second vector signal comprising at least one motion sensor axis signal different from the first vector signal.

45. A non-transitory computer readable medium storing instructions that, when executed by a processor of a medical device, cause the device to:

sensing a motion signal;

setting a sensing window during each of a plurality of ventricular cycles;

determining a characteristic of a motion signal sensed during each of a plurality of ventricular cycles for which a motion signal sensed after an end time of the sensing window meets atrial event criteria;

setting an atrial event sensing parameter based on the determined characteristic;

46. The non-transitory computer-readable medium of clause 45, further comprising instructions that cause the device to generate pacing pulses in response to the sensed atrial event signal.

47. A medical device, the medical device comprising:

a motion sensor configured to sense a motion signal; and

identifying at least one ventricular cycle having an atrial event signal in the motion signal during diastole based on the analysis of the motion signal;

Determining at least one characteristic of the motion signal sensed during at least one ventricular cycle having an atrial event signal in the motion signal during diastole; and

a starting value of the atrial event sensing parameter is established based on the at least one characteristic.

The details of one or more aspects of the disclosure are set forth in the accompanying drawings and the description below. Other features, objects, and advantages of the technology described in this disclosure will be apparent from the description and drawings, and from the claims.

Drawings

Fig. 1 is a conceptual diagram illustrating a medical device system that may be used to sense cardiac electrical and motion signals induced by cardiac motion and flowing blood and provide pacing therapy to a patient's heart.

Fig. 2 is a conceptual diagram of the intracardiac pacemaker shown in fig. 1.

Fig. 3 is a schematic diagram of an exemplary configuration of the pacemaker shown in fig. 1.

Fig. 4 is an example of a motion sensor signal that may be acquired over a cardiac cycle by a motion sensor included in the pacemaker of fig. 1.

Fig. 5 is an example of motion sensor signals acquired during two different cardiac cycles.

FIG. 6 is a flow chart of a method for establishing atrial event sensing parameters.

FIG. 7 is a flow chart of a method for selecting atrial event sensing vectors according to one example.

Fig. 8 depicts two exemplary histograms generated for two different motion sensor signal vectors.

Fig. 9 is a flow chart of a method for establishing an end time of a passive ventricular filling window (also referred to herein as an "A3 window").

FIG. 10 is an example of a histogram of recent threshold amplitudes across time during an extended A3 window.

FIG. 11 is a flow chart of a method for establishing early and late values of atrial event sensing threshold amplitudes applied during and after, respectively, a passive ventricular filling window.

FIG. 12 is one example of a histogram of the maximum amplitude of a motion signal used to establish an early atrial event sensing threshold amplitude.

FIG. 13 is one example of a histogram of the maximum amplitude of a motion signal used to establish a late atrial event sensing threshold amplitude.

Fig. 14 is a flow chart of a method for controlling atrial synchronized ventricular pacing according to one example.

Fig. 15 is a flow chart of a process performed by a pacemaker for setting atrial event sensing parameters according to another example.

FIG. 16 is a flow chart of a method for setting and adjusting atrial event sensing control parameters according to another example.

FIG. 17 is a flow chart of a method for setting atrial event sensing parameters according to another example.

FIG. 18 is a flowchart of a method for establishing an early atrial event sensing threshold amplitude value to be applied during an A3 window according to another example.

Fig. 19 is a graph of a motion sensor signal (shown as a rectified signal) during one ventricular cycle of a non-atrial tracked ventricular pacing mode according to one example.

Fig. 20 is a graph of motion signals during one ventricular cycle in another example.

FIG. 21 is a flowchart of a method for identifying a confidence atrial event period (also referred to herein as a "confidence A4 event period") according to some examples.

FIG. 22 is a graph of motion signals that do not meet the confidence A4 event criteria during one ventricular cycle according to one example.

FIG. 23 is a conceptual diagram of an A3 window histogram that may be accumulated in memory during a setup procedure for establishing a starting value of an early atrial event sensing threshold amplitude according to another example.

FIG. 24 is a flow chart of a method for establishing an initial atrial event sensing parameter according to another example.

FIG. 25 is a flow chart of a method for establishing atrial event sensing parameters according to another example.

Detailed Description

In general, this disclosure describes techniques for establishing cardiac event sensing parameters by an implantable medical device. As described below, atrial contraction events may be sensed from signals generated by motion sensors in response to heart motion such that the motion signals include atrial contraction event signals (also referred to herein as "atrial event signals") that correspond to atrial mechanical contractions and active phases of filling of the ventricles (sometimes referred to as "atrial kick"). The atrial event sensing parameters may include a selected directional quantity signal of the motion sensor, a sensing threshold amplitude, and a time window during which an atrial event may be sensed in response to the sensing threshold amplitude applied during the respective time window. The technology disclosed herein provides techniques for sensing atrial events from motion sensor signals according to one or more atrial event sensing parameters by a ventricular pacemaker that may be fully implantable within a ventricular heart chamber and has a motion sensor for generating an intra-ventricular motion signal. In this way, atrial events may be detected, for example, from within the ventricles for controlling atrial synchronized ventricular pacing. Atrial synchronized ventricular pacing pulses may be delivered by a pacemaker implanted in the ventricle without the need for a sensor located in or on the atrium of the patient's heart for detecting atrial events.

Fig. 1 is a conceptual diagram illustrating an Implantable Medical Device (IMD) system 10 that may be used to sense cardiac electrical signals and motion signals induced by cardiac motion and flowing blood and provide pacing therapy to a patient's heart 8. IMD system 10 includes a ventricular intracardiac pacemaker 14 in this example. Pacemaker 14 may be a leadless, transcatheter, intracardiac pacemaker adapted to be implanted entirely within a heart chamber (e.g., entirely within a Right Ventricle (RV) or entirely within a Left Ventricle (LV) of heart 8) for sensing cardiac signals and delivering ventricular pacing pulses. Pacemaker 14 may be reduced in size as compared to a subcutaneously implanted pacemaker and may be generally cylindrical in shape to enable transvenous implantation through a delivery catheter.

Pacemaker 14 is shown positioned in the RV along the endocardial wall, e.g., near the apex of the RV, although other locations are possible. The techniques disclosed herein are not limited to the pacemaker locations shown in the example of fig. 1 and other locations within heart 8 are possible. For example, ventricular intracardiac pacemaker 14 may be positioned in the LV and configured to detect cardiac motion signals and deliver atrial-synchronized ventricular pacing to the LV using techniques disclosed herein. Pacemaker 14 may be positioned in or on the RV or LV to provide corresponding right or left ventricular pacing and for sensing cardiac motion signals by a motion sensor within the pacemaker (which may be in or on the ventricular chamber).

Pacemaker 14 is capable of generating electrical stimulation pulses (e.g., pacing pulses) delivered to heart 8 through one or more electrodes on the outer housing of the pacemaker. Pacemaker 14 is configured to deliver RV pacing pulses and sense RV cardiac electrical signals using housing-based electrodes to produce RV Electrogram (EGM) signals. Cardiac electrical signals can be sensed using a housing-based electrode that is also used to deliver pacing pulses to the RV.

Pacemaker 14 is configured to control delivery of ventricular pacing pulses to the ventricular pacing pulses in a manner that facilitates synchronization between atrial activation and ventricular activation (e.g., by maintaining a target Atrial and Ventricular (AV) pacing interval between atrial events and ventricular pacing pulses). That is, pacemaker 14 controls pacing pulse delivery to maintain a desired AV pacing interval between atrial contractions corresponding to atrial systoles and ventricular pacing pulses delivered to cause ventricular depolarizations and ventricular systoles.

In accordance with the techniques described herein, atrial contraction events that produce active ventricular filling phases are detected by pacemaker 14 from a motion sensor (such as an accelerometer) surrounded by the housing of pacemaker 14. The motion signals generated by the accelerometer implanted in the ventricular chamber (which may be referred to as "intra-ventricular motion signals") include motion signals caused by ventricular and atrial events. For example, the acceleration of blood flowing into the RV through tricuspid valve 16 between the RA and the RV, caused by atrial systole and referred to as "atrial kick", may be detected by pacemaker 14 from signals generated by an accelerometer included in pacemaker 14. Other motion signals that may be detected by pacemaker 14, such as motion caused by ventricular contractions and passive ventricular filling, are described below in connection with fig. 4.

Atrial P-waves accompanying atrial depolarization are relatively low amplitude signals in the near-field ventricular cardiac electrical signal received by pacemaker 14 (e.g., as compared to the near-field R-waves), and thus may be difficult to reliably detect from the cardiac electrical signal acquired by pacemaker 14 implanted in the ventricular chamber. When based solely on cardiac electrical signals received by pacemaker 14, atrial synchronized ventricular pacing by pacemaker 14 or other functions that rely on atrial sensing may be unreliable. In accordance with the techniques disclosed herein, pacemaker 14 includes a motion sensor (such as an accelerometer) and is configured to detect atrial events corresponding to atrial mechanical activation or atrial systole from signals generated by the motion sensor. The ventricular pacing pulses may be synchronized with atrial events detected from the motion sensor signal by setting a programmable AV pacing interval that controls the timing of the ventricular pacing pulses relative to the detected atrial contraction event. As described below, detection of an atrial contraction event for synchronizing a ventricular pacing pulse with an atrial contraction period may include detection of other cardiac event motion signals to positively identify the atrial contraction event.

The target AV pacing interval may be a default or programmed value selected by the clinician and is the time interval from detection of an atrial event until delivery of a ventricular pacing pulse. In some cases, the target AV pacing interval may begin at a time when an atrial contraction event is detected based on the motion sensor signal, or at an identified fiducial point of the atrial event signal. The target AV pacing interval may be identified as hemodynamic best for a given patient based on clinical testing or evaluation of the patient or based on clinical data from a population of patients. The target AV pacing interval may be determined to be optimal based on the relative timing of the electrical and mechanical events as identified from the cardiac electrical signals received by pacemaker 14 and the motion sensor signals received by pacemaker 14.

Pacemaker 14 may be capable of bi-directional wireless communication with external device 20 to program AV pacing intervals and other pacing control parameters and mechanical event sensing parameters that may be used to detect ventricular mechanical events and atrial contraction events from motion sensor signals. Aspects of the external device 20 may generally correspond to the external programming/monitoring unit disclosed in U.S. patent No. 5,507,782 (Kieval et al). External device 20 is commonly referred to as a "programmer" because it is typically used by a physician, technician, nurse, clinician, or other qualified user to program the operating parameters in pacemaker 14. The external device 20 may be located in a clinic, hospital, or other medical facility. The external device 20 may alternatively be embodied as a home monitor or handheld device that may be used in a medical facility, in a patient's home, or in another location. The operating parameters including sensing and therapy delivery control parameters may be programmed into pacemaker 14 using external device 20.

External device 20 may include a processor 52, a memory 53, a display 54, a user interface 56, and a telemetry unit 58. Processor 52 controls external device operation and processes data and signals received from pacemaker 14. Display unit 54 may generate a display (which may include a graphical user interface) of data and information related to pacemaker function to the user to review pacemaker operation and programmed parameters as well as cardiac electrical signals, cardiac motion signals, or other physiological data that may be acquired by pacemaker 14 and transmitted to external device 20 during an interrogation session.

User interface 56 may include a mouse, touch screen, keyboard, etc. to enable a user to interact with external device 20 to initiate a telemetry session with pacemaker 14 for retrieving data from pacemaker 14 and/or transmitting data to the pacemaker, including programmable parameters for controlling cardiac event sensing and therapy delivery. Telemetry unit 58 includes a transceiver and antenna configured for bi-directional communication with telemetry circuitry included in pacemaker 14 and is configured to operate in conjunction with processor 52 to transmit and receive data related to pacemaker function over communication link 24.

At the time of implantation, during patient follow-up, or at any time after pacemaker implantation, pacemaker 14 may perform a setup procedure to establish parameters for detecting atrial events from motion sensor signals. During the procedure, the patient may stand, sit, lie down or lie ambulatory. The setup procedure may include acquiring motion sensor signal data and generating a profile of motion sensor signal characteristics for use in establishing atrial event sensing parameters. In some examples, the motion sensor signal data may be transmitted to the external device 20 and displayed in the form of a histogram on the display unit 54 of the external device 20. Atrial event sensing parameters established based on the motion sensor signal data may be automatically set or may be transmitted to the external device 20 to generate a display on the display unit 54 as recommended parameters, allowing the clinician to review and accept or modify the recommended parameters, for example, using the user interface 56.

In some examples, external device processor 52 may perform the operations disclosed herein for establishing a starting value of an atrial event sensing parameter based on data retrieved from pacemaker 14. The processor 52 may cause the display unit 54 to generate a display of data related to the motion sensor signals, including a histogram distribution of metrics determined from the heart motion signals, for use in selecting a starting value of an atrial event sensing control parameter (also referred to herein as an "atrial event sensing control parameter"). The display unit 54 may be a graphical user interface that enables a user to interact with the display, for example, to select various displays or information for viewing. In some examples, a user may select one or more atrial event sensing control parameter settings to be automatically established by pacemaker 14 and/or may program starting sensing control parameters or other programmable parameters for controlling sensor operation and therapy delivery. Processing circuitry and/or processor 52 included in pacemaker 14 may determine a starting value for one or more atrial contraction event sensing control parameters based on data acquired from motion sensor signals generated by accelerometers included in pacemaker 14 and various thresholds and criteria (which may include user programmable thresholds or criteria used in setting starting parameter values).

External device telemetry unit 58 is configured for bi-directional communication with implantable telemetry circuitry included in pacemaker 14. Telemetry unit 58 establishes wireless communication link 24 with pacemaker 14. Communication link 24 may use, for exampleWi-Fi, medical Implant Communication Services (MICS), or other Radio Frequency (RF) links of communication bandwidth. In some examples, external device 20 may include a programming head placed in proximity to pacemaker 14 to establish and maintain communication link 24, and in other examples, external device 20 and pacemaker 14 may be configured to communicate using distance telemetry algorithms and circuitry that do not require the use of programming heads and do not require user intervention to maintain the communication link. An exemplary RF telemetry communication system that may be implemented in system 10 is generally disclosed in U.S. patent No. 5,683,432 (Goedeke et al).

It is contemplated that the external device 20 may be connected to the communication network, either wired or wireless, via telemetry circuitry including transceivers and antennas, or via hardwired communication lines for transmitting data to a centralized database or computer, to allow remote management of the patient. A remote patient management system including a centralized patient database may be configured to utilize the presently disclosed techniques to enable a clinician to review EGMs, motion sensor signals, and marker channel data and authorize programming of sensing and therapy control parameters in pacemaker 14, for example, after viewing visual representations of the EGMs, motion sensor signals, and marker channel data.

Fig. 2 is a conceptual diagram of the intracardiac pacemaker 14 shown in fig. 1. Pacemaker 14 includes electrodes 162 and 164 spaced along housing 150 of pacemaker 14 for sensing cardiac electrical signals and delivering pacing pulses. Electrode 164 is shown as a tip electrode extending from distal end 102 of pacemaker 14, and electrode 162 is shown as a ring electrode along a middle portion of housing 150 (e.g., adjacent proximal end 104). The distal end 102 is referred to as "distal" because it is intended to be the leading end when the pacemaker 14 is advanced through a delivery tool (such as a catheter) and placed against a target pacing site.

Electrodes 162 and 164 form anode and cathode pairs for bipolar cardiac pacing and sensing. In other examples, pacemaker 14 may include two or more ring electrodes, two tip electrodes, and/or other types of electrodes exposed along pacemaker housing 150 for delivering electrical stimulation to heart 8 and sensing cardiac electrical signals. Electrodes 162 and 164 may be, but are not limited to, titanium, platinum, iridium, or alloys thereof, and may include a low polarization coating such as titanium nitride, iridium oxide, ruthenium oxide, platinum black, and the like. Electrodes 162 and 164 may be positioned at locations other than those shown along pacemaker 14.

The housing 150 is formed of a biocompatible material, such as stainless steel or titanium alloy. In some examples, the housing 150 may include an insulating coating. Examples of insulating coatings include parylene, urethane, PEEK, polyimide, or the like. The entire housing 150 may be insulated, but only the electrodes 162 and 164 are uninsulated. Electrode 164 may act as a cathode electrode and be coupled to internal circuitry, such as a pacing pulse generator and cardiac electrical signal sensing circuitry, enclosed by housing 150 via an electrical feedthrough across housing 150. The electrode 162 may be formed as a conductive portion of the housing 150 defining a ring-shaped electrode that is electrically isolated from other portions of the housing 150 as generally shown in fig. 2. In other examples, instead of providing a partially annular electrode (such as anode electrode 162), the entire perimeter of housing 150 may act as an electrode that is electrically isolated from tip electrode 164. The electrode 162 formed along the conductive portion of the housing 150 acts as a return anode during pacing and sensing.

Housing 150 includes control electronics subassembly 152 containing electronics for sensing cardiac signals, generating pacing pulses, and controlling therapy delivery and other functions of pacemaker 14 as described below in connection with fig. 3. In some examples, the motion sensor may be implemented as an accelerometer enclosed within the housing 150. The accelerometer provides signals to a processor included in the control electronics subassembly 152 for signal processing and analysis to detect atrial contraction events, such as for controlling timed ventricular pacing pulses, as described below.

The accelerometer may be a three-dimensional accelerometer. In some examples, the accelerometer may have one "longitudinal" axis that is parallel or aligned with the longitudinal axis 108 of the pacemaker 14 and two orthogonal axes that extend in a radial direction relative to the longitudinal axis 108. However, practice of the techniques disclosed herein is not limited to a particular orientation of the accelerometer within or along the housing 150. In other examples, a one-dimensional accelerometer may be used to obtain an intra-cardiac motion signal from which atrial contraction events are detected. In yet other examples, a two-dimensional accelerometer or other multi-dimensional accelerometer may be used. Each axis of the Shan Weihuo multi-dimensional accelerometer may be defined by a piezoelectric element, microelectromechanical system (MEMS) device, or other sensor element capable of generating an electrical signal in response to a change in acceleration applied to the sensor element (e.g., by converting acceleration into a force or displacement that is converted into an electrical signal). In a multi-dimensional accelerometer, the sensor elements may be arranged orthogonally, with each sensor element axis being orthogonal relative to the other sensor element axes. However, orthogonal arrangements of the elements of the multi-axis accelerometer are not necessarily required.

Each sensor element may generate an acceleration signal corresponding to a vector aligned with an axis of the sensor element. As described below, the techniques disclosed herein include selecting vector signals for a multi-dimensional accelerometer (also referred to as a "multi-axis" accelerometer) that is used to sense atrial contraction events. In some cases, one, two, or all three axis signals generated by a three-dimensional accelerometer may be selected as vector signals for detecting atrial contraction events, e.g., for controlling atrial-synchronized ventricular pacing delivered by pacemaker 14.

The housing 150 further includes a battery subassembly 160 that provides power to the control electronics subassembly 152. The battery subassembly 160 may include the features of the batteries disclosed in commonly assigned U.S. patent No. 8,433,409 (Johnson et al) and U.S. patent No. 8,541,131 (Lund et al), both of which are hereby incorporated by reference in their entirety.

Pacemaker 14 may include a set of fixation tines 166 to secure pacemaker 14 to patient tissue, for example, by actively engaging the ventricular endocardium and/or interacting with the ventricular trabeculae. Fixation tines 166 are configured to anchor pacemaker 14 to operably position electrode 164 proximate to a target tissue to deliver therapeutic electrical stimulation pulses. Various types of active and/or passive fixation members may be employed to anchor or stabilize pacemaker 14 in an implanted position. Pacemaker 14 may include a set of stationary tines as disclosed in commonly assigned U.S. patent No. 9,775,872 (Grubac et al).

Pacemaker 14 may optionally include a delivery tool interface 158. The delivery tool interface 158 may be positioned at the proximal end 104 of the pacemaker 14 and configured to be connected to a delivery device (such as a catheter) for positioning the pacemaker 14 at an implantation location (e.g., within a heart chamber) during an implantation procedure.

Fig. 3 is a schematic diagram of an exemplary configuration of pacemaker 14 shown in fig. 1. Pacemaker 14 includes pulse generator 202, cardiac electrical signal sensing circuit 204, control circuit 206, memory 210, telemetry circuit 208, motion sensor 212, and power supply 214. The various circuits represented in fig. 3 may be combined on one or more integrated circuit boards comprising: an Application Specific Integrated Circuit (ASIC), an electronic circuit, a processor (shared, dedicated, or group) and memory that execute one or more software or firmware programs, a combinational logic circuit, a state machine, or other suitable components that provide the described functionality.

In the examples described herein, the motion sensor 212 may include an accelerometer. However, the motion sensor 212 is not limited to an accelerometer, and other motion sensors may be successfully used in the pacemaker 14 to detect cardiac motion signals according to the techniques described herein. Examples of motion sensors that may be implemented in motion sensor 212 include piezoelectric sensors and MEMS devices.

The motion sensor 212 may include a multi-axis sensor, such as a two-dimensional sensor or a three-dimensional sensor, wherein each axis provides an axis signal for detecting a cardiac mechanical event that may be analyzed independently or in combination. The motion sensor 212, for example, when subjected to flowing blood and cardiac motion, produces an electrical signal related to the motion or vibration of the sensor 212 (and pacemaker 14). The motion sensor 212 may include one or more filters, amplifiers, rectifiers, analog-to-digital converters (ADCs), and/or other components for generating motion signals that are communicated to the control circuit 206. For example, each vector signal generated by each individual axis of the multi-axis accelerometer may be filtered by a high pass filter (e.g., a 10Hz high pass filter). The filtered signal may be digitized by an ADC and rectified for use by the atrial event detector circuit 240 to detect atrial contraction events. The high pass filter may be lowered (e.g., to 5 Hz) if it is desired to detect atrial signals having lower frequency content. In some examples, the high pass filtering is performed without low pass filtering. In other examples, each accelerometer axis signal is filtered by a low pass filter (e.g., a 30Hz low pass filter) with or without high pass filtering.

One example of an accelerometer for use in an implantable medical device that may be implemented in conjunction with the techniques disclosed herein is generally disclosed in U.S. patent No. 5,885,471 (Ruben et al). Implantable medical device arrangements including piezoelectric accelerometers for detecting patient movement are disclosed, for example, in U.S. patent 4,485,813 (Anderson et al) and U.S. patent 5,052,388 (Sivula et al), both of which are hereby incorporated by reference in their entirety. Examples of three-dimensional accelerometers that may be implemented in pacemaker 14 using the presently disclosed techniques and used to detect cardiac mechanical events are generally disclosed in U.S. patent No. 5,593,431 (Sheldon) and U.S. patent No. 6,044,297 (Sheldon). Other accelerometer designs may be used to generate electrical signals related to motion imparted on pacemaker 14 due to ventricular and atrial events.

The sensing circuit 204 is configured to receive cardiac electrical signals through the electrodes 162 and 164 via the pre-filter and amplifier circuit 220. The pre-filter and amplifier circuit may include a high pass filter (e.g., a 2.5Hz to 5Hz high pass filter) for removing DC offset or a wideband filter having a passband of 2.5Hz to 100Hz to remove DC offset and high frequency noise. The pre-filter and amplifier circuit 220 may further include an amplifier to amplify the "raw" cardiac electrical signal that is passed to an analog-to-digital converter (ADC) 226. The ADC 226 may communicate multi-bit digital Electrogram (EGM) signals to the control circuit 206 for use by the atrial event detector circuit 240 to identify ventricular electrical events (e.g., R-waves or T-waves) and/or atrial electrical events (e.g., P-waves). The identification of cardiac electrical events may be used in algorithms for establishing atrial sense control parameters and for detecting atrial contraction events from motion sensor signals. The digital signal from ADC 226 may be passed to rectifier and amplifier circuit 222, which may include a rectifier, band pass filter, and amplifier for passing the cardiac signal to R-wave detector 224.

The R-wave detector 224 may include a sense amplifier or other detection circuit that compares the incoming rectified cardiac electrical signal to an R-wave sensing threshold (which may be an automatically adjusted threshold). When the incoming signal exceeds the R-wave sensing threshold, the R-wave detector 224 generates an R-wave sensing event signal (Rsense) that is passed to the control circuit 206. In other examples, the R-wave detector 224 may receive the digital output of the ADC 226 for detecting R-waves by a comparator, morphology signal analysis of the digital EGM signal, or other R-wave detection technique. The processor 244 may provide sense control signals (e.g., R-wave sense threshold, sensitivity, and various blanking and refractory periods applied to cardiac electrical signals to control R-wave sensing) to the sense circuitry 204. The R-wave sensed event signal communicated from R-wave detector 224 to control circuit 206 may be used to schedule ventricular pacing pulses by pacing timing circuit 242 and to identify the timing of ventricular electrical events in an algorithm executed by atrial event detector circuit 240 for detecting atrial contraction events from signals received from motion sensor 212.

The control circuit 206 includes an atrial event detector circuit 240, a pacing timing circuit 242, and a processor 244. Control circuitry 206 may receive R-wave sensed event signals and/or digital cardiac electrical signals from sensing circuitry 204 for detecting and confirming cardiac events and controlling ventricular pacing. For example, when pacemaker 14 operates in a non-atrial tracked ventricular pacing mode, R-wave sensed event signals may be passed to pacing timing circuit 242 for suppressing and/or scheduling ventricular pacing pulses. The R-wave sensed event signal may also be passed to the atrial event detector circuit 240 for setting a time window used by the control circuit 206 in detecting atrial contraction events from the motion sensor signal.

Atrial event detector circuit 240 is configured to detect atrial contraction events from signals received from motion sensor 212. Techniques for setting a time window for detecting an atrial contraction event are described below in connection with fig. 9-10. In some examples, one or more ventricular mechanical events may be detected from the motion sensor signal during a given cardiac cycle to facilitate positive detection of atrial contraction events from the motion sensor signal during the ventricular cycle.

Atrial event detector circuit 240 receives motion signals from motion sensor 212 and may begin an atrial "blanking" period in response to ventricular electrical events (e.g., R-wave sensed event signals from sensing circuit 204 or pacing pulses delivered by pulse generator 202). The blanking period may correspond to a period of time following a ventricular electrical event during which ventricular mechanical events, e.g., corresponding to ventricular systole and isovolumetric diastole, are expected to occur. When ventricular pacing is properly synchronized with an atrial event, an atrial event corresponding to a ventricular systole is not expected to occur during an atrial blanking period (also referred to herein as a "post-ventricular atrial blanking period"). Thus, motion signal peaks that occur during the atrial blanking period are not sensed as atrial events. The atrial "blanking" period may be used to define a period of time following a ventricular electrical event during which atrial event detector circuit 240 does not sense an atrial contraction event. However, in some examples, the motion sensor signal is not necessarily blanked during this period of time, as the control circuit 206 may still receive the motion sensor signal during the atrial blanking period and may process the motion signal for sensing ventricular events during the atrial blanking period.

The atrial event detector circuit 240 determines whether the motion sensor signal meets atrial contraction event detection criteria outside of the atrial blanking period. In some examples, the motion sensor signal during the blanking period may be monitored by the atrial event detector circuit 240 for the purpose of detecting ventricular mechanical events that may be used to confirm or verify the detection of atrial contraction events. Thus, the ventricular mechanical event detection window may be set during an atrial blanking period and may be set according to a predetermined time interval after identifying a ventricular electrical event. The atrial event detector circuit 240 may be configured to detect one or more ventricular mechanical events during a respective ventricular event detection window during an atrial blanking period. The timing and detection of ventricular mechanical events may be used to update the atrial blanking period and/or may be used to confirm detection of an atrial event that occurs after an expected ventricular mechanical event.

Atrial event detector circuit 240 may set a time window corresponding to the passive ventricular filling phase and the active ventricular filling phase based on the timing of the previous ventricular electrical event (R-wave sensed event signal or ventricular pacing pulse). The motion sensor signal crossing the atrial event sensing threshold during any of these windows may be detected as an atrial contraction event. As described below, two different atrial event sensing thresholds may be established for application during and after the passive filling phase window (during the active filling phase window (also referred to below as the "A4 window").

Atrial event detector circuit 240 communicates an atrial event detection signal to processor 244 and/or pacing timing circuit 242 in response to detecting an atrial event. Pacing timing circuit 242 (or processor 244) may additionally receive R-wave sense event signals from R-wave detector 224 for controlling the timing of pacing pulses delivered by pulse generator 202. Processor 244 may include one or more clocks for generating clock signals used by pacing timing circuit 242 to timeout an AV pacing interval that begins when an atrial event detection signal is received from atrial event detector circuit 240. Pacing timing circuit 242 may include one or more escape interval timers or counters for timing out AV pacing intervals (which may be programmable intervals stored in memory 210 and retrievable by processor 244) for setting AV pacing intervals used by pacing timing circuit 242. One application of the atrial sensed event signal generated by atrial event detector circuit 240 is for setting AV pacing intervals for controlling the timing of ventricular pacing pulses. However, the control circuitry 206 may use the atrial sensed event signal for other purposes.

Pacing timing circuit 242 may additionally include a lower pacing rate interval timer for controlling the minimum ventricular pacing rate. For example, if an atrial contraction event is not detected from a motion sensor signal that triggers a ventricular pacing pulse at a programmed AV pacing interval, the ventricular pacing pulse may be delivered by pulse generator 202 upon expiration of a lower pacing rate interval to prevent ventricular arrest and maintain a minimum ventricular rate. Sometimes, the control circuit 206 may control the pulse generator 202 in a non-atrial tracking ventricular pacing mode (also referred to as "asynchronous ventricular pacing") during a process for establishing a sensed parameter for detecting an atrial contraction event from a motion signal. The non-atrial tracking ventricular pacing mode may be denoted as a VDI pacing mode in which ventricular pacing pulses are delivered in the absence of sensed R-waves and suppressed in response to an R-wave sensed event signal from sensing circuit 204. Dual-chamber sensing may be performed during a non-atrial tracking ventricular pacing mode by sensing ventricular electrical events by sensing circuitry 204 and sensing atrial contraction events from motion signals received by atrial event detector circuitry 240 from motion sensor 212. As described below, for example, in connection with fig. 7-14, the atrial event sensing parameters established during the VDI pacing mode may include an atrial event sensing vector of a motion sensor generating a signal from which an atrial contraction event was detected, an end of a passive ventricular filling window, and an atrial event sensing threshold amplitude value applied during and after the passive ventricular filling window.

Pulse generator 202 generates electrical pacing pulses that are delivered to the RV of the patient's heart through cathode electrode 164 and return anode electrode 162. In addition to providing control signals to pacing timing circuit 242 and pulse generator 202 for controlling the timing of ventricular pacing pulses, processor 244 may retrieve programmable pacing control parameters, such as pacing pulse amplitude and pacing pulse width, from memory 210, which are passed to pulse generator 202 for controlling pacing pulse delivery. The pulse generator 202 may include a charging circuit 230, a switching circuit 232, and an output circuit 234.

Charging circuit 230 may include a hold capacitor that may be charged to the pacing pulse amplitude at a multiple of the battery voltage signal of power supply 214 under control of the voltage regulator. The pacing pulse amplitude may be set based on a control signal from the control circuit 206. Switching circuit 232 may control when the holding capacitor of charging circuit 230 is coupled to output circuit 234 for delivering pacing pulses. For example, switching circuit 232 may include a switch that is activated by a timing signal received from pacing timing circuit 242 upon expiration of an AV pacing interval or VV (ventricular) lower rate pacing interval and remains closed for a programmed pacing pulse width to enable the hold capacitor of charging circuit 230 to discharge. During the programmed pacing pulse duration, the hold capacitor, which was pre-charged to the pacing pulse voltage amplitude, is discharged across electrodes 162 and 164 through the output capacitor of output circuit 234. Examples of pacing circuitry generally disclosed in U.S. patent 5,507,782 (Kieval et al) and U.S. patent 8,532,785 (Crutchfield et al) may be implemented in pacemaker 14 for charging a pacing capacitor to a predetermined pacing pulse amplitude and delivering pacing pulses under control of control circuitry 206.

Memory 210 may include computer readable instructions that, when executed by control circuitry 206, cause control circuitry 206 to perform various functions attributed throughout this disclosure to pacemaker 14. Computer readable instructions may be encoded within memory 210. Memory 210 may include any non-transitory computer-readable storage medium, including any volatile, non-volatile, magnetic, optical, or dielectric medium, such as Random Access Memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), electrically Erasable Programmable ROM (EEPROM), flash memory, or other digital medium, with the sole exception of a transitory propagating signal. In accordance with the techniques disclosed herein, memory 210 may store: a fixed time period; and other data used by control circuitry 206 to control the delivery of pacing pulses by pulse generator 202 (e.g., by detecting atrial events from motion sensor signals by atrial event detector circuit 240 and setting escape interval timers contained in pacing timing circuit 242).

The power supply 214 may correspond to the battery subassembly 160 in fig. 2, providing power to each of the other circuits and components of the pacemaker 14 as needed. The power source 214 may include one or more energy storage devices, such as one or more rechargeable or non-rechargeable batteries. For clarity, the connections between the power supply 214 and other pacemaker circuits and components are not shown in fig. 3, but should be understood from the general block diagram of fig. 3. For example, the power supply 214 may supply power to the charging and switching circuitry included in the pulse generator 202, the amplifier, the ADC 226, and the sensing circuitry 204, the telemetry circuitry 208, the memory 210, and other components of the motion sensor 212, as desired.

Telemetry circuitry 208 includes transceiver 209 and antenna 211 for transmitting and receiving data via a Radio Frequency (RF) communication link. As described above, telemetry circuitry 208 may be capable of two-way communication with external device 20 (fig. 1). The motion sensor signals and cardiac electrical signals and/or data derived from these signals may be transmitted by telemetry circuitry 208 to external device 20. Programmable control parameters and algorithms for performing atrial event detection and ventricular pacing control may be received by telemetry circuitry 208 and stored in memory 210 for access by control circuitry 206.

The functionality attributed herein to pacemaker 14 may be embodied as one or more processors, controllers, hardware, firmware, software, or any combination thereof. Depiction of different features as specific circuits is intended to highlight different functional aspects and does not necessarily imply that such functions must be realized by separate hardware, firmware, or software components or by any particular circuit architecture. Rather, functionality associated with one or more circuits described herein may be performed by separate hardware, firmware, or software components or integrated within general purpose hardware, firmware, or software components. For example, atrial contraction event detection from motion sensor signals and ventricular pacing control operations performed by pacemaker 14 may be implemented in control circuit 206 executing instructions stored in memory 210 and dependent on inputs from sensing circuit 204 and motion sensor 212. It is within the ability of one of ordinary skill in the art, given the disclosure herein, to provide software, hardware, and/or firmware to accomplish such functionality in the context of any modern pacemaker.

Fig. 4 is an example of a motion sensor signal 250 that may be acquired by motion sensor 212 over a cardiac cycle. Vertical dashed lines 252 and 262 represent the timing of two consecutive ventricular events (intrinsic ventricular depolarizations or ventricular pacing pulses), marking the respective beginning and end of a cardiac cycle 251 (referred to herein as "ventricular cycle 251" because the beginning of the central cycle in this example is a ventricular electrical event). The cardiac cycle represented by ventricular cycle 251 includes one cycle of ventricular systole and ventricular diastole. The motion signals include an A1 event 254, an A2 event 256, an A3 event 258, and an A4 event 260. A1 event 254 is an acceleration signal that occurs during ventricular systole and marks an approximate onset of ventricular mechanical systole (in this example, when motion sensor 212 is implemented as an accelerometer). The A1 event is also referred to herein as a "ventricular contraction event". A2 event 256 is an acceleration signal that may occur as the aortic and pulmonary valves close, marking an approximate shift or end of ventricular mechanical systole. The A2 event may also mark the beginning of the isovolumetric diastole phase of the heart chamber that occurs with the aortic and pulmonary valves closed.

A3 event 258 is an acceleration signal that occurs during passive ventricular filling and marks ventricular mechanical diastole. The A3 event is also referred to herein as an "A3 signal" and is also referred to as a "ventricular passive filling event. A2 events occur at the end of ventricular systole, which is an indicator of ventricular diastolic onset. A3 events occur during ventricular diastole. Thus, the A2 event and the A3 event may be collectively referred to as ventricular mechanical diastolic events, as both are indicators of ventricular diastole.

A4 event 260 is an acceleration signal that occurs during atrial contraction and active ventricular filling and marks the mechanical systole of the atrium. The A4 event 260 is also referred to herein as an "A4 signal" and is an "atrial contraction event" or just an "atrial event" detected from the motion sensor signal 250. Atrial event detector circuit 240 detects an A4 event 260. Processor 244 may control pacing timing circuit 242 to trigger ventricular pacing pulses by starting an AV pacing interval in response to detecting A4 event 260. The control circuit 206 may be configured to detect one or more of an A1 event, an A2 event, and an A3 event from the motion sensor signal 250 for at least some ventricular cardiac cycles for positively detecting an A4 event 260 and setting atrial event detection control parameters. The A1 event, A2 event, and/or A3 event may be detected and characterized to avoid false detection of the A4 event and to facilitate reliable A4 event detection for proper timing of atrial synchronized ventricular pacing pulses.

The techniques described herein may be performed by pacemaker 14 for establishing parameters for detecting A4 events without necessarily needing to identify and distinguish A1 event from A4 event. In contrast, motion signals acquired during non-atrial tracking ventricular pacing modes may be characterized by determining characteristics of the motion signals during the sensing window and/or outside of the atrial blanking period. In some examples, the distribution of features is used to establish atrial event sensing parameters.

Fig. 5 is an example of motion sensor signals 400 and 410 acquired during two different cardiac cycles. Ventricular pacing pulses are delivered at a time of 0.0 seconds for both cardiac cycles. The top sensor signal 400 is received during one cardiac cycle and the bottom sensor signal 410 is received during a different cardiac cycle. The two signals 400 and 410 are aligned in time at 0.0 seconds (the time of ventricular pacing pulse delivery). While the motion signals 400 and 410 and the motion signal 250 of fig. 4 are shown as raw accelerometer signals, it should be appreciated that the control circuit 206 may receive digitized, filtered, amplified, and rectified signals from the motion sensor 212 for processing and analysis as described in connection with the flowcharts and histogram distributions presented in the figures.

The A1 events 402 and 412 of the respective motion sensor signals 400 and 410 that occur during ventricular systole are observed to be well aligned in time at time 0.0 seconds after the ventricular pacing pulse. Similarly, A2 events 404 and 414 (which may mark the end of ventricular systole and isoventricular diastole phases) and A3 events 406 and 416 (which occur during passive ventricular filling) are well aligned in time. Since the A1 event, the A2 event, and the A3 event are ventricular events that occur during ventricular systole, at the end of ventricular systole, and at the beginning of isovolumetric ventricular diastole, and during passive ventricular filling, respectively, these events are expected to occur at relatively uniform intervals following a ventricular electrical event (in this example, a ventricular pacing pulse) and relative to each other. The temporal relationship of the A1, A2, and A3 events may be different after a ventricular pacing pulse than after a sensed intrinsic R-wave, however, during a stable paced or intrinsic ventricular rhythm, the relative timing of the ventricular A1, A2, and A3 events to each other and the immediately preceding ventricular electrical event is expected to be consistent from beat to beat.

The A4 events 408 and 418 of the first motion sensor signal 400 and the second motion sensor signal 410 are not aligned in time. The A4 event occurs during atrial systole and, therefore, the time interval of the A4 event immediately following the previous ventricular electrical event (sensed R-wave or ventricular pacing pulse) and the previous A1 event to A3 event may vary between cardiac cycles. During the non-atrial tracking ventricular pacing mode, the A4 event timing during the cardiac cycle may vary from one cycle to the next.

The timing consistency of the A1 event to A3 events relative to each other and the immediately preceding ventricular electrical events may be used to determine the atrial blanking period 436 and increase the confidence of reliably detecting the A4 events 408 and 418. No atrial contraction event is detected during the atrial blanking period 436, which extends from the ventricular electrical event (at time 0.0) to the estimated onset of ventricular diastole, such that the atrial blanking period 436 includes both the A1 event and the A2 event. A3 window 424 may be set with a start time 420 and an end time 422 corresponding to the end of a post-ventricular atrial blanking period 436. The end time 422 may be established using techniques described below, for example, in connection with fig. 9 and 10. The end time 422 may also be considered the start time of the A4 sense window 450, although in some cases the A4 signal may be sensed during the A3 window. Thus, the A3 window may also be referred to as a "sensing window," however, a different A4 sensing threshold amplitude may be applied to the motion signal during the A3 window 424 instead of after the A3 window end time 422.

The A4 events 408 and 418 may be detected based on the multi-level A4 sensing threshold 444. As seen by the lower motion sensor signal 410, the A4 event 418 may occur earlier after the A3 window 424 due to a change in atrial rate. In some cases, as the atrial rate increases, an A4 event 418 may occur within the A3 window 424. When this occurs, the A3 event 416 and the A4 event 418 may merge when passive ventricular filling and active ventricular filling occur together. The fused A3/A4 event may have a high amplitude, even greater than the amplitude of the A3 event 416 or A4 event 418 when it occurs alone. Thus, in some examples, a first higher A4 sense threshold amplitude 446 may be established for detecting early A4 signals fused with the A3 signal during the A3 window 424. A second lower A4 sense threshold amplitude 448 may be established for detecting a relatively late A4 event after the end time 422 of the A3 window 424 during the A4 window 450. The A4 window 450 extends from the end time of the A3 window 424 to the next ventricular electrical event (sensed or paced). The earliest crossing of the A4 sensing threshold 444 by the motion sensor signal after the start time 420 of the A3 window (or after expiration of the atrial blanking period 436) may be detected as an atrial contraction event. Techniques for establishing an early A4 sense threshold amplitude 446 used during the A3 window 424 and a late A4 sense threshold amplitude 448 used after the end time 422 of the A3 window 424 during the A4 window 450 are described below.

FIG. 6 is a flowchart 300 of a method for establishing atrial event sensing parameters according to some examples. The control circuit 206 may perform the method of flowchart 300 to automatically select and set a start value of an atrial event sensing parameter for sensing an A4 event from the motion sensor signal during an atrial tracking ventricular pacing mode (sometimes referred to as an atrial synchronous ventricular pacing mode). The process of flowchart 300 may be performed by control circuit 206 when pacemaker 14 is implanted and may be performed at other post-implantation times to update or reset the A4 (atrial event) sensing parameters.

At block 302, control circuitry 206 sets the pacing mode to a non-atrial tracking ventricular pacing mode (e.g., VDI) such that ventricular pacing pulses are asynchronously delivered to atrial events. The pacing rate may be set to a nominal rate, for example 50 pulses per minute. In some examples, the ventricular pacing mode may be a rate response mode (e.g., VDIR), but the method for establishing atrial event sensing parameters may be performed when the pacing rate is at or near a programmed lower rate (e.g., 40 to 60 pulses per minute). In patients with AV block, atrial contraction events occur asynchronously to ventricular electrical events during non-atrial tracking pacing modes. In patients with AV block, the ventricular electrical event will typically be a delivered ventricular pacing pulse, but in some cases, and in patients with complete AV conduction, an intrinsic R-wave may be included. Thus, atrial events may experience ventricular cardiac cycles at different times during the VDI pacing mode.

Motion sensor signal data is acquired during a non-atrial tracked ventricular pacing mode for each available sensing vector of the multi-axis motion sensor or for one or more available sensing vectors of the multi-axis motion sensor. Aspects of the motion sensor signal outside of the post-ventricular atrial blanking period (or later than the minimum A3 window start time) may be determined to characterize the motion sensor signal for each available sensing vector within the passive and active filling phases of each ventricular cycle. For example, at block 304, the control circuit 206 may determine at least one maximum motion sensor signal amplitude in a plurality of ventricular cycles (outside of the post-ventricular atrial blanking period) and a time when the motion sensor signal most recently crossed a nominal threshold amplitude. The data acquired at block 304 may be acquired during each ventricular cycle or during a predetermined number of ventricular cycles within a predetermined time interval. For example, the control circuitry 206 may obtain data from the motion sensor signal over a few seconds, up to a minute, minutes, up to an hour, or up to 24 hours to characterize aspects of the motion sensor signal in each of one or more sensing vectors of the motion sensor. In other examples, at least N values of the motion signal characteristic during a respective number N of ventricular cycles are determined.

At block 306, the control circuit 206 generates one or more distributions of amplitude and/or timing data acquired at block 304. In some examples, the one or more distributions are generated as one or more histograms. A histogram of the maximum amplitude data may be generated for each available sensing vector, for example, for selecting a sensing vector or combination of vectors of a motion sensor from which atrial event sensing is performed during an atrial tracking ventricular pacing mode. Techniques for generating a distribution of motion sensor maximum amplitude data and selecting a sensing vector are described below in connection with fig. 7 and 8.

In another example, a histogram of the most recent crossing times of the amplitude threshold during each ventricular cycle may be generated at block 306 for establishing the end time of the A3 window (also referred to as a passive ventricular filling window). Exemplary techniques for generating the distribution and establishing the A3 window end time are described below, for example, in connection with fig. 9 and 10.

A histogram-form distribution of the maximum amplitude data may be generated at block 306 for establishing early A4 and late A4 sensing threshold amplitude values for multi-level A4 sensing thresholds used to sense A4 events during an atrial tracking ventricular pacing mode. Exemplary techniques for generating histograms for establishing atrial event sensing thresholds are described below, for example, in connection with fig. 11-13 and 17-21.

At block 308, the control circuitry 206 analyzes one or more distributions of motion sensor signal characteristic data for establishing one or more atrial event sensing control parameters. Based on the distribution analysis, the control circuit 206 may select atrial event sensing parameters that may include motion sensor vector signals or combinations of vector signals from which the A4 signal is sensed during the atrial tracking pacing mode. Other atrial event sensing parameters that may be derived from the generated one or more distributions of motion signal features include the end time of the A3 window, the early sensing threshold amplitude and the late sensing threshold amplitude of the multi-level A4 sensing threshold. Various examples of techniques for acquiring data from a motion signal, generating a distribution of motion signal data, and deriving one or more atrial event sensing parameters from the distribution are described below, for example, at least in connection with fig. 7-13 and 23.

FIG. 7 is a flow chart 500 of a method for selecting an A4 sense vector according to one example. Control circuitry 206 may control the process of flowchart 500 to set the initial sensing vector during an early post-operative period following pacemaker implantation, and may repeat the process to reset the A4 sensing vector selection after a specified time period or when an A4 event is being insufficiently sensed (e.g., when a threshold number of ventricular pacing pulses are delivered at a rate smoothing interval during an atrial tracking ventricular pacing mode). The rate smoothing interval is a ventricular pacing interval used by the control circuit 206 to control the timing of ventricular pacing pulses in the absence of sensed A4 events during the atrial tracking ventricular pacing mode. The rate smoothing interval may be set by control circuitry 206 based on the actual ventricular rate during atrial-synchronized ventricular pacing (e.g., based on the intermediate ventricular cycle length) to avoid abrupt changes in ventricular rate (e.g., to the LRI) when no A4 event is sensed during the ventricular cycle. The A4 sensing vector selection process of flowchart 500 is performed for determining which vector signals (from one axis or combination of axes) of the multi-axis motion sensor produce a motion signal from which atrial events can be most reliably sensed, e.g., based on atrial event signal strength.

At block 502, control circuitry 206 sets the pacing mode of pacemaker 14 to a non-atrial tracking ventricular pacing mode (e.g., VDI), also referred to herein as an "asynchronous pacing mode". Control circuitry 206 may set a ventricular pacing interval (VV interval) (also referred to herein as a "ventricular lower pacing rate interval" (LRI)) according to a nominal pacing rate (e.g., 50 pulses per minute) in order to maintain a minimum lower ventricular rate. During asynchronous pacing mode, the A4 event may occur at different times during the ventricular cycle. The control circuit 206 may set a nominal or default A4 window during which to detect the motion signal peak amplitude to characterize the motion signal at block 504. In one example, for a pacing rate of 50 pulses per minute, the A4 window may be set to: with 800 to 900 milliseconds after the ventricular pacing pulse (or sensed R-wave) delivered and extending until the start time of the next ventricular pacing pulse (or sensed R-wave).

In other examples, the end of the A3 window and the beginning of the A4 window may be set as a percentage of the LRI or actual ventricular rate interval. For example, the control circuitry 206 may determine an actual ventricular cycle length (which may be paced or sensed) for a specified number of recent ventricular cycles. The control circuit 206 may determine an average or median value of the determined ventricular cycle lengths and set the A4 window to begin at a percentage of the average or median value. In one example, the median ventricular cycle length of the eight most recent ventricular cycles is determined, and the A3 window is set to end and the A4 window is set to start at 80% of the median ventricular cycle length or the 4 th longest ventricular cycle length of the 8 ventricular cycles. The end of the A3 window and the beginning of the A4 window may be set to be between a specified minimum time interval and maximum time interval, for example, in some examples, no less than 650 milliseconds and no more than 900 milliseconds or no more than 1000 milliseconds. A minimum or maximum value may be used when a specified percentage of the median ventricular cycle length is outside of a limit.

In some examples, setting the A4 window at block 504 may be performed by setting a long post-ventricular atrial blanking period from the beginning of a ventricular pacing pulse. The A4 window extends from the end of the long post-ventricular atrial blanking period until the next ventricular pacing pulse. The post-ventricular atrial blanking period may be set to an extended time period that is expected to encompass each of the A1 ventricular event, the A2 ventricular event, and the A3 ventricular event that are expected to occur at relatively predictable intervals after a ventricular pacing pulse (as shown in fig. 5). In this way, any relatively large amplitude peaks of the motion signal that occur after an extended atrial blanking period are more likely to be an A4 signal and less likely to be a ventricular event (A1, A2, or A3).

At block 506, the control circuit 206 determines a maximum amplitude of the motion sensor signal during each A4 window for each motion sensor vector signal selected for analysis during the automatic sensing parameter selection process. The motion sensor vector signals selected for analysis may include one, two, or all three single axis vector signals; one, two, or all three combinations of the two axis vector signals, and/or all three accelerometer axes combinations in the three axis vector signals. The maximum amplitude of each of the analyzed vector signals may be determined for each ventricular cycle occurring during a predetermined time interval or within a predetermined number of ventricular cycles. For example, as an example, the maximum amplitude may be determined from each of one or more analyzed vector signals during each ventricular cycle, or more than 50 to 1000 ventricular cycles, for a few seconds, a minute, a few minutes, an hour, or more.

The control circuit 206 determines a distribution of maximum amplitude values at block 508, for example, by populating a histogram of the maximum amplitudes determined during the A4 window for each vector signal selected for analysis. For example, as described above, a histogram of maximum amplitude may be generated for one longitudinal vector and two radial vectors of a three-dimensional accelerometer with one axis aligned with the long axis of pacemaker 14. In other examples, a histogram of maximum amplitude may be generated for each single axis vector signal and/or each two axis vector signal and/or triaxial vector signal. When a combination of two or all three axes is used to generate the vector signal, the acceleration signal sampling points of the two or all three axis signals may be summed to generate a two or three axis vector signal. In other examples, vector math may be used to determine the generated vector signals. The maximum amplitude may be determined from the rectified vector signal or may be the maximum peak-to-peak amplitude of the unrectified vector signal.

Fig. 8 depicts two exemplary histograms 600 and 620 generated for two different motion sensor vector signals. The maximum amplitude of the vector signal detected during the A4 window is plotted along the horizontal axis 604 of each histogram 600 and 620. In the example shown, the horizontal axis 604 is shown in ADC units, where each ADC unit is 11.8milli-g (gravitational acceleration). A histogram of the amplitudes determined from the motion sensor vector signals received by the control circuit 206 from the ADCs included in the motion sensor 212 may be generated in ADC units, but may optionally be converted to units of acceleration (e.g., meters per second squared (meters per second) ² ) For display on an external device (e.g., external device 20). For example, the acceleration conversion ratio may be 1 m/s ² /100milli-g。

The maximum amplitude during each A4 window matches the histogram bin (bin) value or range. For each matching maximum amplitude value, the matching histogram bin count is incremented by one to track the frequency of occurrence of each maximum amplitude value or range. The frequency or count of the maximum amplitude values obtained over a predetermined time period of a given vector signal that pass through the bins is plotted along the y-axis 602 in each histogram. Since the ventricular cycles are asynchronous with the atrial rhythm, many ventricular cycles during which no true A4 event occurs during the A4 window may occur. Thus, relatively higher counts 612 and 622 of relatively lower maximum amplitude may occur when the atrial systole does not happen to coincide with the nominal A4 window. In this example, a low maximum amplitude may be defined as a maximum amplitude of less than 5 ADC units detected during the A4 window. The high frequency of the A4 window with a relatively low maximum amplitude of the vector signal reflects a relatively large number of ventricular cycles during which the atrial systole does not happen to coincide with the A4 window (during late ventricular diastole).

As shown in the exemplary histogram 600, the reliable sensing vector signal includes a relatively high count 614 at a relatively high maximum amplitude value (e.g., in the range of 7 to 15 ADC units). Moderately high frequencies (count 614) of higher maximum amplitudes (e.g., 8 ADC units or more) during the A4 window indicate the occurrence of an A4 signal that happens to coincide with the A4 window during the non-atrial tracking pacing mode. Since these A4 signals have relatively high amplitudes, the vector signals used to generate histogram 600 may be highly reliable vector signals for sensing atrial events.

In contrast, the exemplary histogram 620 shows a relatively low or zero count 624 of a relatively high maximum amplitude signal during the A4 window. The vector signal used to generate histogram 620 is considered unreliable for use in detecting atrial events because most or all of the maximum amplitudes during the A4 window are relatively low, e.g., within the baseline noise of the vector signal. This means that even over a period where the A4 signal coincides with the A4 window (during the ventricular period), the A4 amplitude is substantially within the range of the baseline noise of the vector signal, e.g., less than 5 ADC units. The orientation of one or more axes used to generate the vector signals associated with histogram 620 may result in a null signal during atrial systole.

Referring again to fig. 7 and with continued reference to fig. 8, at block 510, histogram bins corresponding to relatively low maximum amplitudes may be rejected due to low level noise in the vector signal. In the example of fig. 8, the histogram bins storing counts of maximum amplitude less than the noise threshold 606 (e.g., 5 ADC units) are discarded due to noise. In other examples, the noise threshold 606 may be set to a value greater than or less than 5 ADC units, and may be set according to the baseline noise of the vector signal. In some examples, the noise threshold 606 is optional, or the noise threshold may be set to zero.

At block 512, the control circuit 206 may determine whether the maximum amplitude meets a vector selection criterion for each vector signal tested. In one example, at block 512, a total number of maximum amplitudes greater than noise threshold 606 (e.g., all maximum amplitudes counted within ranges 616 and 626 in histograms 600 and 620, respectively) is determined by control circuit 206. At block 512, a number of high maximum amplitudes greater than the high amplitude threshold 608 is also determined by the control circuit 206. In the example shown, the high amplitude threshold 608 is set to 8 ADC units. The number of maximum amplitudes greater than 8 ADC units is determined as a high maximum amplitude count. At block 512, the ratio of the high maximum amplitude count to the count of all maximum amplitudes greater than the noise threshold 606 is determined as a high maximum amplitude ratio. The high maximum amplitude ratio is determined for each of the tested A4 sense vector signals and occurs during the A4 window and is likely to be the frequency of occurrence of the high amplitude signals of the actual A4 signal because they occur outside of the ventricular posterior atrial blanking period (as generally depicted by 614 in fig. 8).

The control circuit 206 may compare this high maximum amplitude ratio determined for each single axis vector signal analyzed to a single axis vector selection threshold at block 514 for a number of high maximum amplitudes (greater than threshold 608) to a number of all maximum amplitudes greater than noise threshold 606. In one example, if at least 50% of the maximum amplitude value is greater than the high amplitude threshold, then after rejecting the low maximum amplitude value that is less than the noise threshold, a single axis vector signal may be selected as the A4 sense vector and may be used as a reliable A4 sensed single axis vector signal. In the exemplary histogram 600 of fig. 8, more than 50% of the maximum amplitude acquired within the range 616 (greater than 4 ADC units) is greater than 8 ADC units (high amplitude threshold 608). Thus, when the vector signal is a single axis vector signal used to obtain the maximum amplitude data to generate the histogram 600, the single axis vector signal may be selected as the A4 event sensing vector signal at block 516. Using a single axis sensing vector signal for sensing an A4 event allows a single axis of the accelerometer to be powered to generate an A4 sensing vector signal, which may save power and may extend the functional life of pacemaker 14 as compared to using a combination of two or more axes for generating vector signals for sensing atrial events.

In the event that two or more single axis vector signals meet the criteria applied at block 514, the single axis vector signal having the highest maximum amplitude ratio may be selected as the single axis A4 sense vector signal at block 516. In general, the control circuit 206 may select a vector signal that produces a distribution of maximum amplitude during the A4 window that is skewed to the right (with its distribution tail on the right longer than its distribution tail on the left as the amplitude increases from left to right on the x-axis). At block 516, the vector signal with the largest median maximum amplitude, the highest rightward bias of the maximum amplitude distribution, or other metric indicative of the rightward bias of the maximum amplitude distribution may be selected as the sole vector signal for sensing the A4 event during the atrial tracking ventricular pacing mode.

In some examples, the maximum amplitude determined for each vector signal may be compared to a rejection criterion. An alert may be generated when the maximum amplitude of all vector signals meets the rejection criteria. For example, if the high maximum amplitude ratio (the ratio of the number of maximum amplitudes greater than the high amplitude threshold 608 to the number of maximum amplitudes greater than the noise threshold 606) is less than the rejection threshold (as determined at block 518), the control circuit 206 may select a combination of all three axis signals to produce an A4 sense vector signal or select an optimal combination of two axis signals from the three axis signals to generate a two axis vector signal for A4 sensing at block 522. Selecting the minimum number of accelerometer axes required to sense an A4 event may preserve the current demand from the power supply 214. Additionally or alternatively, an alarm may be generated to inform the clinician that no analyzed vector signals have met reliable A4 sensing criteria. At block 522, a notification may be transmitted by telemetry circuitry 208 to external device 20.

In some cases, all three single-axis vector signals may have a high maximum amplitude ratio that is less than the single-axis vector signal threshold ("no" branch of block 514), but at least one, two, or all three single-axis vector signals may have a high maximum amplitude ratio that is greater than the reject threshold ("no" branch of block 518). When at least one of the possible single axis vector signals has a high maximum amplitude ratio between the single axis vector threshold and the rejection threshold, the control circuit 206 may select two axis signals (corresponding to two different motion sensor axes) to be used in combination for A4 event sensing at block 520. The two uniaxial vector signals with the highest maximum amplitude ratios may be selected to be used in combination for sensing atrial events. The two uniaxial vector signals may be summed to produce a biaxial vector signal. In some examples, a single axis vector signal aligned with the longitudinal axis of the housing 150 of the pacemaker 14 is selected along with another vector (e.g., a radial vector). The atrial event sensing vector may be selected as a combination of the longitudinal uniaxial vector signal summed with one of the radial uniaxial vector signals having the highest maximum amplitude ratio determined at block 512 to produce a two-axis vector signal for A4 event sensing.

When two (or three) single axis vector signals are selected in combination as the A4 event sensing vector signals, the multi-axis vector signals may be determined by vector summation of two (or three) individual vector signals. The vector signals may be digitally summed using the digitized, rectified vector signals. The summation may be performed after rectification to avoid destructive summation of two or more vector signals. In other examples, two or three separate analog vector signals may be summed when the single axis vector signal does not produce a high maximum amplitude ratio that is greater than the single axis vector threshold.

In other examples, instead of determining the ratio as indicated at block 512, another measure of the distribution of maximum amplitudes may be determined and compared to vector selection and/or rejection criteria. For example, as described below in connection with fig. 15, a median, average, or specified percentile of all maximum amplitudes determined at block 506 may be determined (in some examples, after discarding maximum amplitudes less than a noise threshold). The median value (or other measure of distribution) of the maximum amplitudes determined for each of the vector signals being analyzed may be compared to one another, and the vector signal yielding the largest median maximum amplitude may be selected. In some examples, the uniaxial vector signal may be selected when it corresponds to a maximum median maximum amplitude. However, a vector signal that is a combination of at least two axis signals will typically have a median maximum amplitude that is greater than the uniaxial vector signal (due to the summation of the two uniaxial signals). Thus, in some examples, the two-axis vector signal with the largest median maximum amplitude may be selected as the atrial event sensing vector signal.

In the event that each of the two axis volume signals is rejected due to the maximum amplitude meeting the rejection criteria, a combination of all three motion sensor axis signals may be selected as the triaxial signals for atrial event sensing. As further described in connection with fig. 15, when a threshold count less than the maximum amplitude acquired by a given vector signal is greater than the minimum amplitude threshold, the two-axis vector signal (or any analyzed vector signal) may be rejected. In some examples, the minimum amplitude threshold may be set to the minimum available A4 sensing threshold amplitude.

Fig. 9 is a flowchart 700 of a method for establishing an end time of an A3 window (e.g., end time 422 of window 424 shown in fig. 5). At block 702, the control circuit 206 sets the pacing mode to a non-atrial tracking pacing mode. By setting the VV pacing interval (LRI), the pacing rate may be set to a nominal pacing rate, e.g., 50 pulses per minute. At block 704, an A4 sense vector (a single axis vector signal or a summation of two or more single axis vector signals) is selected. The control circuit 206 may select the vector signal established as the A4 sense vector in the process described above in connection with fig. 7 and 8. The process of flowchart 700 may be performed only for selected atrial event sensing vector signals (single axis vector or two or three axis vector signals from a summed combination of two or more single axis vector signals), and may be repeated if the A4 sensing vector signal selection changes. In other examples, at least a portion of the process of flowchart 700 may be performed for all available motion sensor vector signals or a plurality of selected vector signals to enable simultaneous motion sensor signal data acquisition for generating a distribution of motion sensor signal features.

In some examples, after the atrial event sensing vector is selected at block 520 of fig. 7, the non-atrial tracking pacing mode and rate remain active. The control circuit 206 may proceed directly to the process of flowchart 700 to establish an A3 window end time, for example, directly from block 520 of fig. 7 to block 706 of fig. 9. In examples that include one-dimensional single axis motion sensors or fixed A4 sense vector signal selection, the vector selection process of fig. 7 is unnecessary. At block 704, the process of establishing an A3 window end time according to flowchart 700 may be performed by selecting a manually programmed or default atrial event sensing vector signal without first performing the process of flowchart 500.

A nominal A3 window is set at block 706, for example, starting 600 milliseconds after a ventricular event (sensed or paced) and extending until the next ventricular event. The nominal A3 window may be set by setting a post-ventricular atrial blanking period extending from a ventricular electrical event (sensed or paced) for a predetermined time period (e.g., 600 milliseconds). The nominal A3 window may be set to begin after the expected time of the A1 signal and the A2 signal (e.g., as shown in fig. 4) and extend as late as possible to the next ventricular electrical event to increase the likelihood of capturing the A3 signal during the nominal A3 window. In some examples, at block 706, the A3 window is set to begin at a fixed interval (e.g., 600 milliseconds after a ventricular event (sensed R-wave or ventricular pacing pulse)) and extends to an A3 window end time that is set to a percentage (e.g., 80% of the LRI or 80% of the average or median ventricular cycle length determined from a specified number of most recent ventricular cycle lengths (e.g., 8 ventricular cycle lengths)).

At block 708, the control circuit 206 sets a nominal threshold amplitude. As one example, the nominal threshold amplitude may be 9 ADC units, which may correspond to 0.9 meters/second ² . At block 709, the timing of the most recent crossing of the nominal threshold amplitude by the motion signal in each A3 window is determined for a predetermined time interval or a predetermined number of ventricular cycles for each vector signal analyzed. When atrial systole (and A4 events) does not happen to occur during the A3 window, the latest threshold crossing may be a true A3 signal, and since the A3 signal is a ventricular event signal following a ventricular electrical event at a relatively consistent time interval, the latest threshold crossing is expected to occur relatively early in the nominal A3 window. When the atrial systole does happen to occur during the A3 window, the latest threshold crossing may be an A3 signal or an A4 signal, depending on when the atrial systole occurs during the A3 window. However, during the extended A3 window, the relatively late threshold crossing is more likely to be the A4 signal than the A3 signal, as the A3 signal is typically related to the timing of the ventricular electrical event. In some ventricular cycles, the motion sensor vector signal may cross the nominal threshold amplitude multiple times during the A3 window. For example, the true A3 signal and the true A4 signal may cross a nominal threshold amplitude. In this case, in some examples, only the timing of the most recent threshold crossing is stored.

When threshold amplitude spans accumulate over a large number of ventricular cycles (e.g., over at least 50 to 100 ventricular cycles or more), the relatively high frequency of the most recent threshold spans will represent an A3 signal, and when the atrial systole happens to occur during the A3 window, some frequencies will represent an A4 signal. In some cases, the fused A3/A4 event may occur as the most recent threshold amplitude crossing during the A3 window. In some cases, the relatively late threshold amplitude crossing may be the A1 signal, although the A1 and A2 signals may occur during the post-ventricular atrial blanking period. The distribution of the timing of the most recent threshold amplitude crossing may reveal the expected time of the A3 event and the expected time of the A4 event during the ventricular cycle.

At block 710, the histogram may be populated using the data accumulated at block 709. The latest nominal threshold amplitude crossing time accumulated beat by beat over a predetermined time interval or a predetermined number of ventricular cycles may be binned according to a histogram bin time interval range. A count of the number of most recent threshold crossings that occurred during each bin time interval range is determined.

FIG. 10 is an example of a histogram 750 of recent nominal threshold amplitude crossing times that may be generated at block 710 of FIG. 9 for one sensing vector signal. The most recent threshold crossing time is plotted in milliseconds (ms) along the x-axis 754 and represents the time after the ventricular electrical event at which the last nominal threshold amplitude crossing occurred during the ventricular cycle. The frequency or count of the most recent threshold crossing times is plotted along the y-axis 752. In the example shown, for an A3 window whose start time 756 is 600 milliseconds after the ventricular electrical event, the leftmost, lowest bin includes the most recent crossing time that occurs in the range of 601 milliseconds to 650 milliseconds after the ventricular electrical event. The rightmost bin includes the most recent crossing time that occurs in the range of 1001 milliseconds to 1050 milliseconds after a ventricular electrical event (pacing pulse or sensed R-wave). In this example, each bin includes a 50 millisecond time frame, and the A3 window extends from 600 milliseconds after the previous ventricular electrical event to the next ventricular electrical event. Other start times and/or end times of the A3 window may be selected (and may depend on the ventricular rate), and in other examples, other time ranges of the histogram bins may be used.

The distribution represented by histogram 750 exhibits a bimodal distribution with a left peak 772 corresponding to a possible A3 signal during the extended A3 window and a right peak 770 corresponding to a possible A4 signal during the extended A3 window. The A4 confidence time threshold 758 may be set to an expected nominal threshold amplitude after which to span a time that is A4 signal with high probability and is highly unlikely to be an A3 signal. In the example of fig. 10, the A4 confidence time threshold 758 is set to 900 milliseconds based on the ventricular pacing rate being set to 50 pulses per minute. The A4 confidence time threshold 758 may be set to a fixed value based on the ventricular rate during the accumulation of the most recent threshold crossing time. In other examples, the A4 confidence time threshold 758 may be set based on a predetermined percentage (e.g., a predetermined percentile) of all accumulated recent threshold crossing times.

After bins above the A4 confidence time threshold 758 are discarded, the leftmost peak of the bimodal distribution is represented by the highest bin count 772 within the remaining range 760 of the histogram bins. The leftmost peak 772 is likely to represent the occurrence of an A3 event during the extended A3 window and may include some fused A3/A4 signal. An appropriate end time for the A3 window may be any time from the median 762 of the most recent threshold crossing times within range 760 to the A4 confidence time threshold 758. The A3 window end time may be selected between the median 762 and the A4 confidence time threshold 758 such that the A3 window includes the A3 signal (and fused A3/A4 signal) with high probability. In some examples, the A3 window end time or A4 window start time for A4 sensing during the atrial tracking ventricular pacing mode is set to a time based on the left peak of the bimodal distribution of most recent nominal threshold crossing times. For example, the A3 window end time used during the atrial tracking ventricular pacing mode may be set to the left peak of the dual peak profile (as time increases from left to right) plus an offset, where the offset is a predetermined value ranging from zero to 200 milliseconds, as an example.

Returning to fig. 9, the counts in the filled histogram bins that are greater than the A4 confidence time threshold are discarded at block 712. At block 714, the control circuitry 206 selects an A3 window end time based on the remaining histogram bin data after discarding bins that are greater than the A4 confidence time threshold. The control circuitry 206 may determine the median time for the most recent nominal amplitude threshold crossing for the remaining bins. The A3 window end time may be established as the median of the distribution plus the offset. The offset may be a predetermined value ranging from 0 milliseconds to 200 milliseconds (e.g., 50 milliseconds to 100 milliseconds). In other examples, after discarding histogram bins that are greater than the A4 confidence time threshold, the A3 window end time may be set to the percentile of the remaining histogram bin counts (within range 760 in fig. 10). For example, after discarding bin counts greater than the A4 confidence time threshold 758, the A3 window end time may be set to a time at which at least 70%, 80%, 90%, or 95% of the most recent threshold crossing time is less than the A3 end time.

In the example of fig. 10, the A3 window begins at the expiration of the post-atrial ventricular blanking period (e.g., 600 milliseconds after a ventricular event) until the next ventricular event (pacing pulse or sensed R-wave). During this extended A3 window, both early and late threshold crossings are detected, resulting in the bimodal distribution shown in fig. 10, including a left peak 772 corresponding to a possible A3 event signal and a right peak 770 corresponding to a possible A4 event signal. Rather than setting an extended A3 window and then discarding bins greater than the A4 confidence time threshold, the A3 window may be set at block 706 to extend from 600 milliseconds (or the end of the post-ventricular atrial blanking interval) up to the A4 confidence time threshold 758 (which may be set as a percentage of the median ventricular cycle length as described above). For example, the A3 window may be set to extend from 600 milliseconds up to 80% of the median of the eight ventricular cycle lengths, and as an example, not less than 650 milliseconds or greater than 900 milliseconds when the pacing lower rate is set to 50 pulses per minute. In this way, the most recent threshold crossing during the A3 window is most likely an A3 event because they occur before the A4 confidence time threshold 758. After the A4 confidence time threshold 758, histogram bins corresponding to even later threshold crossings also do not need to be filled, as they are unlikely to contain true A3 event threshold crossing times. The A3 window end time may be established as the median of the distribution of the most recent threshold crossing times plus the offset.

Fig. 11 is a flowchart 800 of a method for establishing early and late values of an A4 sensing threshold applied to a motion signal during an A3 window and after an A3 window, respectively, for sensing an A4 signal during an atrial tracking ventricular pacing mode, for example. At block 802, control circuit 206 sets the pacing mode to a non-atrial tracking pacing mode and sets a VV pacing interval (LRI) according to the selected pacing rate (e.g., 50 pulses per minute). At block 804, the control circuit 206 selects the A4 sense vector (as a single axis vector signal or a combination of two or more acceleration axis signals), for example, based on the method of fig. 7.

In some examples, the process of flowchart 800 is performed using only the A4 sense vector signal selected by the method of fig. 7. In other examples, portions of the process of flowchart 800 may be performed for all available motion sensor vector signals (including single axis vector signals, two axis vector signals, and three axis vector signals, or any number of vector signals that are selected for analysis). For example, the data required to generate a histogram for setting the A4 sense threshold amplitude value may be obtained for a plurality of vector signals used in the process of fig. 7, but only data corresponding to the vector signal selected as the A4 sense vector signal at blocks 516, 520, or 522 of fig. 7 may be used to set the A4 sense threshold amplitude value.

An A3 window end time may be set at block 806. In one example, as described above, the A3 window end time is set to a percentage of the LRI or median ventricular cycle length. In other examples, the A3 window is set to the A3 window end time selected at block 714 of fig. 9 based on a histogram of the most recent threshold crossing times. The A3 window may extend from a fixed start time (which may be programmable or based on empirical data) to an A3 window end time that is automatically determined from the distribution of the most recent nominal threshold amplitude crossing times as described above in connection with fig. 9-10. In other examples, a manually programmed or default A4 sense vector signal and/or A3 window end time is set at blocks 804 and 806, respectively. In this case, the process of fig. 7 and 9 need not be performed before the process of fig. 11 in establishing the A4 sensing threshold amplitude value.

At block 808, the control circuit 206 determines a maximum amplitude of the motion signal during the A3 window of each ventricular cycle over a predetermined number of cycles or a predetermined time period. At block 810, for each ventricular cycle over a predetermined number of cycles or a predetermined time period, the control circuit 206 determines a maximum amplitude of the motion signal after the end time of the A3 window and before the next ventricular electrical event (e.g., during the A4 window).

At block 812, the maximum amplitude determined during the A3 window is used by the control circuit 206 to generate an A3 window maximum amplitude distribution, for example by populating the A3 window maximum amplitude histogram. The count of histogram bins corresponding to the maximum amplitude less than the noise threshold may be discarded at block 814. The very low maximum amplitude during the A3 window may not be indicative of a true A3 or A4 event and may be a baseline noise of the motion signal. At block 816, the control circuit 206 selects an early A4 sense threshold amplitude value to apply during the A3 sense window based on the remaining non-discarded histogram data. A method for selecting early A4 sensing threshold amplitude values based on A3 window histogram data is described below in connection with fig. 12.

The control circuit 206 may generate a distribution of A4 window maximum amplitude data, e.g., by populating the A4 window maximum amplitude histogram, at block 818 using the maximum amplitudes determined at block 810 during the A4 window (after the A3 window until the end of the ventricular cycle marked by the next ventricular electrical event). At block 820, the histogram bin storing the maximum amplitude less than the noise threshold may be discarded. At block 822, the control circuit 206 selects the late A4 sense threshold amplitude value based on the remaining (not discarded) distributions of the maximum amplitude values. A method for selecting late A4 sensing threshold amplitude values based on histogram data is described below in connection with fig. 13.

Fig. 12 is one example of a histogram 850 of the maximum amplitude of motion sensor vector signals during an A3 window that may be generated by the control circuit 206. The maximum motion signal amplitude of the selected motion sensor vector is plotted on the x-axis 856 (shown in ADC units in this example). The count for each histogram bin is plotted along the y-axis 854. Although other histogram bin resolutions may be used (and may be in g or meters/second ² Arranged in units), each histogram bin is shown in fig. 12 as including an amplitude range of one ADC unit. Each bin stores a count of the number of times the maximum amplitude of the vector signal during the A3 window falls within the corresponding bin range.

The noise threshold amplitude 858 may be a predetermined value or determined as a percentile of the histogram frequency distribution. Less than the noise thresholdThe maximum amplitude during the A3 window of amplitude 858 may be due to baseline noise (or very low amplitude A3 signal) during the A3 window. The histogram bins corresponding to the maximum amplitude that is less than the noise threshold amplitude 858 may be discarded for the purpose of selecting an early A4 sensing threshold amplitude (also referred to herein as an "early atrial event sensing threshold"). In the example of FIG. 12, the noise threshold may be 4 ADC units, or about 50milli-g or about 0.5 meters/second ² 。

After discarding histogram bins that are less than the noise threshold amplitude 858 (to the left thereof), the control circuit 206 may determine the early A4 sense threshold 860 as a percentile of the remaining distribution of maximum amplitudes determined during the A3 window. Since most of the maximum amplitude values determined during the A3 window and counted in the histogram 850 are expected to represent actual A3 events, most of the maximum amplitudes should be less than the early A4 sensing threshold 860 so that they are not falsely detected as A4 events.

A smaller percentage of the relatively high maximum amplitude that occurs during the A3 window (e.g., a maximum amplitude that is greater than the early A4 sense threshold 860) may represent fused A3/A4 events that should be sensed as A4 events. According to one example, the early A4 sensing threshold 860 may be set to a relatively high percentile, for example, eighty percent, eighty-five percent, or ninety percent of the maximum amplitude of the rest of the histogram 850 (which is not discarded). In one example, the higher 15% of the maximum amplitude signal during the A3 window (after discarding bins below the noise threshold) would meet the atrial event sensing criteria and would be sensed as an A4 signal, for example, during the atrial tracking pacing mode. The lower 85% of the maximum amplitude signal during the A3 window is not sensed as the A4 signal.

Fig. 13 is one example of an A4 window maximum amplitude histogram 870 that may be generated by the control circuit 206 for one motion sensor vector signal being analyzed at block 818 of fig. 11. The maximum amplitude during the A4 window of the vector signal being analyzed is plotted on the x-axis 876. The count for each histogram bin is plotted along the y-axis 874. Each histogram bin is shown in fig. 13 as including a range of one ADC unit, although other histogram bin resolutions may be used. Each bin stores a count of the number of times the maximum amplitude of the vector signal after the A3 window (during the A4 window) falls within the corresponding bin range.

The noise threshold amplitude 878 may be a predetermined value or determined as a percentile of the histogram frequency distribution. Histogram bin counts less than noise threshold amplitude 878 may be discarded for the purpose of selecting late A4 sensing threshold amplitude values. These relatively low maximum amplitude signals during the A4 window may be baseline noise and are not true A4 signals. In the example of fig. 13, the noise threshold is 5 ADC units (or about 0.6 meters/second ² ). Counts in bins of less than 5 ADC units are discarded for the purpose of selecting late A4 sensing threshold amplitude values (also referred to herein as "late atrial event sensing thresholds").

After discarding the histogram bins corresponding to maximum amplitudes less than noise threshold amplitude 878, control circuit 206 may determine late A4 sense threshold amplitude value 880 as a percentile of the remaining maximum amplitude distribution. Since most of the maximum amplitude values above the noise threshold 878, determined during the A4 window and counted in the histogram 870, are expected to represent the actual A4 signal, most of the maximum amplitudes should be above the late A4 sensing threshold 880 to avoid insufficient sensing of the A4 event. A smaller percentage of the maximum amplitude that occurs during the A4 window may be noise or even late A3 signals. The late A4 sensing threshold 880 may be set to a relatively low percentile (e.g., fifty percent) of the maximum amplitude of the rest of the histogram 870 (not discarded). In this way, after discarding the maximum amplitude less than noise threshold 878, a lower 5% of the remaining maximum amplitude signal during the A4 window will not be sensed, but 95% of the maximum amplitude signal will be greater than late A4 sensing threshold 880 and meet atrial event sensing criteria, for example, during an atrial tracking ventricular pacing mode. The exemplary percentiles and noise thresholds presented herein are illustrative in nature, and it should be understood that other percentiles and noise thresholds may also be used to establish early and late A4 sensing thresholds.

A histogram or other type of graphical representation of a distribution of motion sensor signal characteristics that determines a motion sensor for setting atrial event sensing parameters, generally depicted in the figures presented herein, may be generated for display on the display unit 54 of the external device 20. The generated display may include a value or graphical depiction (e.g., a line, bar, or icon overlaid on a histogram) to indicate that atrial event sensing parameter values or settings are determined by the control circuit 206 (or external processor 52) based on the determined distribution of features.

Fig. 14 is a flow chart 900 of a method for controlling atrial synchronized ventricular pacing according to one example. At block 902, the control circuit 206 selects an A4 sense vector. A4 sensing vector (uniaxial, biaxial, or triaxial) may be selected using the methods described above in connection with fig. 7 and 8. At block 904, the A3 window is set according to the start time and the end time. The end time may be set to a time after a previous ventricular electrical event, which is determined using the methods described above in connection with fig. 9 and 10. At block 906, the multi-level A4 sensing threshold sets the multi-level A4 sensing threshold by establishing an early A4 sensing threshold amplitude value and a late A4 sensing threshold amplitude value using the methods described in connection with fig. 11-13.

In this example, at least one atrial sense parameter of the A4 sense vector, the A3 window end time, the early A4 sense threshold amplitude value, and/or the late A4 sense threshold amplitude value is established by: determining a characteristic of the motion sensor signal, generating at least one histogram or one or more other representations of a distribution of one or more determined characteristics of the motion sensor signal, and selecting a corresponding A4 sensing parameter based on an analysis of the distribution. In some examples, one or more of the atrial event sensing parameters, including the A4 sensing vector, the A3 window end time, and the early and late A4 sensing thresholds, are set to default or user programmable values without generating a histogram or other representative distribution of motion signal characteristics.

In some examples, as shown in block 902, an A4 sense vector is first determined. After the A4 sensing vector is selected, an A3 window end time may be established (block 904), and then early and late A4 sensing threshold amplitude values are established (block 906). However, in other examples, the A4 sensing parameters may be determined in a different order than that shown in fig. 14 or partially or fully simultaneously. For example, when pacing in a non-atrial tracking pacing mode, a nominal A3 window end time of 800 milliseconds to 900 milliseconds may be set, enabling simultaneous data acquisition during multiple ventricular pacing periods to generate a distribution of motion sensor signal data for simultaneously establishing two or more atrial event sensing parameters. The maximum amplitude during the A3 window, the maximum amplitude after the A3 window, and the latest nominal threshold crossing that occurs after the A3 window begins (e.g., after 600 milliseconds) may be determined from a plurality of ventricular cycles for use in generating a distribution of histograms as shown in the figures. The early and late values of the A4 sense vector, the A3 window end time, and the atrial event sensing threshold may be established based on the simultaneously acquired motion sensor signal characteristics and the corresponding distributions generated therefrom.

At block 908, the control circuitry 206 may set the pacing mode to an atrial tracking ventricular pacing mode, such as a VDD pacing mode. At block 910, the atrial event detector circuit 240 senses an A4 event during the ventricular cycle using the A4 sense parameters established at blocks 902-906. The control circuit 206 may generate an atrial sense event signal in response to the atrial event detector circuit 240 sensing an A4 event. The atrial sensed event signal may be used to control the timing of ventricular pacing pulses during the VDD pacing mode. The atrial tracking ventricular pacing mode set at block 908 may be referred to as an atrial synchronized ventricular pacing mode because the AV pacing interval may be started (block 912) in response to sensing an A4 signal for controlling the timing of ventricular pacing pulses (block 910). If an R-wave is sensed before the AV pacing interval expires ("Yes" branch of block 918), then at block 910, the control circuitry 206 senses the next A4 signal. The A4 signal is sensed in response to an earliest crossing of the multi-level atrial event sensing threshold during the A3 window or the A4 window.

Upon expiration of the AV pacing interval ("yes" branch of block 914), a ventricular pacing pulse is generated by pulse generator 202 and delivered (block 920). In this way, ventricular pacing pulses are synchronized with atrial contraction events to provide a more normal cardiac rhythm in patients experiencing AV block.

Fig. 15 is a flow chart 1000 of a process performed by pacemaker 14 for setting atrial event sensing parameters according to another example. In some examples, after a time delay following a telemetry session with external device 20, an automatic selection of a starting value of an atrial event sensing parameter may be initiated by control circuit 206 at block 1001. For example, in a clinic or hospital or remote, the feature of automatic selection of atrial event sensing parameters during an implantation procedure or any patient follow-up procedure may be programmed to be "on" or enabled by a user interacting with the external device 20. After pacemaker telemetry circuitry 208 no longer receives telemetry communications, control circuitry 206 may detect inactivity of or termination of a telemetry session with external device 20 and initiate an automatic selection process at block 1001 by waiting for a time delay (e.g., after one to ten minutes or three minutes in one example). This time delay after reception of the telemetry communication signal has ceased may allow other programming or procedures to be completed before the automatic atrial event sensing parameter selection process begins, such as during an implantation procedure or programming and interrogation session.

The control circuit 206 may apply other criteria at block 1001 before starting the selection process. For example, the control circuitry 206 may verify that the patient activity level is less than a threshold level, verify that a target heart rate for rate responsive pacing is less than a threshold rate based on the patient activity level, and/or verify that the actual ventricular rate is not greater than the threshold rate. In some examples, the control circuitry 206 may be configured to determine the patient activity metric from motion sensor signals related to the patient's physical activity level. In some examples, the patient activity metric may be used by the control circuitry to control rate responsive ventricular pacing to provide ventricular rate support during times when patient activity increases or rises.

Upon initiating the atrial event sensing parameter selection process, the control circuitry 206 may set test values for a plurality of control parameters to enable motion sensor signal analysis and data collection to generate distribution data for setting the atrial event sensing control parameters. At block 1002, the control circuitry 206 may switch from a programmed atrial tracking ventricular pacing mode (e.g., VDD pacing mode) to a temporary non-atrial tracking ventricular pacing mode (e.g., VDI pacing mode) that includes generating a motion sensor signal for analysis. Control circuitry 206 may set a temporary lower pacing rate during motion sensor signal analysis to control the ventricular pacing rate. In one example, the lower pacing rate is set to 50 pulses per minute, but in various examples may be set to 40 pulses per minute or higher.

At block 1002, the control circuit 206 may set a test value for detecting a test A3 threshold amplitude spanned by a most recent motion signal threshold during the A3 window for establishing an end time of the A3 window. Additionally or alternatively, the control circuit 206 may set one or both of the early and late A4 sensing threshold amplitude values to the maximum possible value to avoid actual A4 sensing during motion sensor signal analysis. In an illustrative example, the early (during the A3 window) and late (during the A4 window) A4 sense threshold amplitude values may be set to the maximum limit of the ADC or to a value that exceeds the maximum of the generated distribution range or the maximum bin range of the generated histogram of the maximum amplitude of the vector signal being analyzed. In this way, A4 event detection is avoided, which might otherwise terminate the A3 or A4 window, thereby eliminating additional analysis of the motion sensor signal during a given ventricular cycle. The control circuit 206 may further set the test settings for the post-ventricular atrial blanking period and the end time of the A3 window.

As described above, during motion sensor signal analysis, the end of the A3 window (and the beginning of the A4 window) may be set as a percentage of the LRI or ventricular rate interval. For example, at block 1002, upon switching to the VDI pacing mode, the control circuitry 206 may determine a ventricular cycle length (which may be paced or sensed) for a specified number of ventricular cycles. The control circuit 206 may determine an average or median of the determined ventricular cycle lengths and set the test end time of the A3 window to begin at a percentage of the average or median ventricular cycle length. In one example, after switching to the VDI pacing mode, the test A3 window end time is set to 80% of the fourth shortest ventricular cycle length of the first eight ventricular cycles. The end of the A3 window (and the beginning of the A4 window) may be set between a specified minimum time interval and maximum time interval, for example, in some examples, no less than 650 milliseconds and no more than 900 milliseconds from the most recent ventricular electrical event (sensed or paced). A minimum or maximum value may be used instead when a specified percentage of the median ventricular cycle length is outside of a limit.

The post-ventricular atrial blanking period may be set to a fixed value, a value based on a lower pacing rate set during the temporary VDI pacing mode, or a median ventricular cycle length. In one example, the post-ventricular atrial blanking period is set to 600 milliseconds when the lower pacing rate is set to 50 pulses per minute. The A3 window begins at the expiration of the post ventricular atrial blanking period and extends to the test A3 window end time.

The test A3 threshold amplitude used to detect the latest motion sensor signal threshold crossing during the test A3 window, for example, as described in connection with fig. 9 and 10, may be set to a relatively lower amplitude that is expected to be greater than the baseline motion sensor signal noise for use in setting the A3 window end time. For example, the test A3 threshold amplitude may be set to 9 ADC units, which may correspond to about 106milli-g or about 1 meter/second ² 。

At block 1002, the control circuitry 206 may establish vector signals from the motion sensor 212 that are to be analyzed to generate distribution data. In some examples, each available two-axis vector signal and three-axis vector signal is generated and analyzed to generate amplitude and timing distribution data for each of the vector signals. For example, each axis of the three-dimensional accelerometer is referred to as axis 1, axis 2, and axis 3, the combination of axis 1 and axis 2 may be referred to as a 1+2 vector signal, the combination of axis 1 and axis 3 may be referred to as a 1+3 vector signal, and the combination of axis 2 and axis 3 may be referred to as a 2+3 vector signal. The combination of all three axis signals from motion sensor 212 may be referred to as a 1+2+3 vector signal. In one example, each of the 1+2, 1+3, 2+3, and 1+2+3 vector signals may be processed and analyzed by the control circuit 206 to determine amplitude and timing data for the respective vector signals for selecting at least one atrial event sensing control parameter for use during the atrial tracking ventricular pacing mode.

After setting the test pacing mode, pacing rate, test vector signals, and other test control parameters, at block 1004, the control circuitry 206 may begin to acquire amplitude and timing data from the vector signals being analyzed. When the four vector signals listed above are selected for analysis during an automatic atrial sense parameter selection process, each vector signal may be analyzed on a rotational basis to remove or minimize the effect of confounding factors on a particular vector signal, such as posture-dependent changes that may occur in the vector signal due to changing patient positioning or activity. For example, one vector signal may be analyzed during a one minute time period to obtain amplitude and timing data, then the next vector signal may be analyzed during the next one minute time period, and so on. In this way, each of the four vector signals listed above may be analyzed in one minute every four minute time period. This process may be repeated a specified number of times (e.g., two, five, ten, etc.) until a desired number of minutes or data points are obtained for each of the vector signals being analyzed. In one example, the amplitude and timing data is obtained on a recurring basis from each of the four vector signals listed above for a total of five minutes or a total of at least 20 minutes per vector signal.

In other examples, one or more of the uni-axial vector signals (e.g., the axis 1 vector signal, the axis 2 vector signal, and/or the axis 3 vector signal) may be included as test vector signals in an analysis for obtaining vector signal data at block 1004. Any specified number of vector signals, where each vector signal may be obtained from a single accelerometer axis, a summation of two axis signals, or a summation of all three accelerometer axis signals, may be included in the analysis used to obtain the amplitude and timing data at block 1004. The number of vector signals analyzed and the accelerometer axes used to obtain each vector signal may be programmed by the user. When a combination of axes is used, the signals from each of the combination of axes may be sampled at specified time slots of the sampling rate such that the sampling points from two or all three axis signals may be summed to produce the desired vector signal that is a combination of two or all three accelerometer axis signals. For example, if the vector signal is sampled every 1 millisecond, each axis signal may be sampled within 333 millisecond time slots such that the sampling of each axis signal is available for summation with one or more other axis signals at each 1 millisecond sampling point time.

At block 1004, amplitude and/or timing data is accumulated as needed from the vector signals being analyzed for selecting one or more atrial event sensing parameters. For example, one or more of the A4 sense vector signal, the early A4 sense threshold amplitude value, the late A4 sense threshold amplitude value, and/or the A3 window end time, or any combination thereof, may be established by the control circuit 206 during the process of the flowchart 1000. When the process of flowchart 1000 is performed to select an A4 sense vector signal, the maximum amplitude during the A4 window is determined from each test vector signal for a desired number of ventricular cycles or time periods. By setting the late A4 sense threshold amplitude value to a relatively high value (e.g., the maximum available value), the A4 window does not terminate before the next ventricular event, allowing the control circuit 206 to determine the maximum vector signal amplitude until the next ventricular event (paced or sensed).

Other data determined from each of the vector signals may include the time that the most recent test A3 threshold was crossed during the test A3 window (e.g., as generally described in connection with fig. 9 and 10) and the maximum vector signal amplitude during the A3 window and/or the A4 window (e.g., as generally described in connection with fig. 11-13). The amplitude and/or timing data determined from each test vector signal depends on which atrial event sensing parameter value is being set by control circuitry 206 during the process of flowchart 1000.

At any time before, during, or after the completion of the determination of the amplitude and/or timing data from the motion sensor vector signal at block 1004, the control circuit 206 may detect a suspension condition as indicated at block 1006. One or more conditions may cause confounding effects on the amplitude and timing data, thereby ensuring a pause or delay in obtaining the data. For example, high or variable patient physical activity and/or high or variable heart rate may cause variations in the amplitude and timing data of a given vector signal, which confounds the distribution data generated for the vector signal. Accordingly, the amplitude and/or timing data acquired at block 1004 for selecting one or more atrial event sensing control parameters may be acquired over a time period associated with a relatively low stable heart rate (e.g., less than 80 heart beats per minute) and/or patient physical activity level (e.g., less than an activity threshold corresponding to daily life or rest).

The control circuit 206 may determine the patient physical activity metric from the motion sensor signal at regular time intervals. The target heart rate and the sensor-indicated pacing rate may be determined based on the physical activity metric to provide rate-responsive pacing to support the patient's physical activity level. The control circuit 206 may detect an abort condition based on at least one patient physical activity metric or level determined from the motion sensor signal being greater than a threshold activity level. In another example, the control circuit 206 may detect the abort condition based on variability of the patient's physical activity level, for example, by detecting a threshold change in the physical activity level over a predetermined time period. In other examples, control circuitry 206 may detect the abort condition in response to determining that the paced or sensed target heart rate, the sensor-indicated pacing rate, and/or the actual ventricular rate is greater than a threshold rate or is highly variable based on a threshold rate change over a predetermined time period.

In one example, a ventricular rate faster than 85 beats per minute may be the abort condition. The control circuit 206 may determine the ventricular cycle length during the amplitude and timing data acquisition at block 1004. After completing the data acquisition, at block 1006, the control circuit 206 may determine whether a percentage threshold (e.g., more than 20% to 30%) greater than the ventricular cycle length is shorter than a threshold interval. For example, if 20% or more of the ventricular cycle length is shorter than about 700 milliseconds (or faster than a rate of about 85 beats per minute) during data acquisition, an abort condition may be detected at block 1006 due to the high ventricular rate. The data may be discarded without generating a distribution or histogram of the data. In other examples, the data may be stored and compiled with data obtained during a next data acquisition time to generate and analyze a distribution of the data.

In other examples, the threshold ventricular cycle length interval for detecting a high heart rate may be 650 milliseconds to 750 milliseconds. In yet other examples, the threshold ventricular cycle length interval may be set based on an average or median ventricular cycle length determined from ventricular cycle lengths stored during data acquisition. For example, the threshold cycle length may be set to the median ventricular cycle length minus 100 milliseconds to 150 milliseconds, or set to a ventricular rate corresponding to 10 to 20 beats per minute faster than the median ventricular rate during data acquisition. In an illustrative example, when more than 20% (or other selected percentage) of the ventricular cycle length corresponds to a ventricular rate that is faster than the rate of the median cycle length during data acquisition plus 10 beats per minute, an abort condition may be detected at block 1006 due to a variable or high ventricular rate.

In some examples, an abort condition may be detected at block 1006 due to the variable heart rate, the abort condition including a threshold percentage of ventricular cycle length being longer than a long threshold cycle length (or slower than a corresponding ventricular rate). The control circuit 206 may set the long threshold period length based on a ventricular rate corresponding to a median ventricular period length determined from the ventricular period lengths stored during the data acquisition of block 1004. For example, the long threshold period length may be set to correspond to a ventricular rate interval of 10 beats per minute less a median ventricular rate determined from the median ventricular period length during data acquisition. When a ventricular cycle length exceeding a threshold percentage (e.g., 20%, 30%, or other percentage) is longer than a slow ventricular rate interval (e.g., corresponding to 10 beats per minute) that is less than the median rate, an abort condition may be detected at block 1006 due to the variable ventricular rate.

Additionally or alternatively, the high or variable patient activity level may be an abort condition detected at block 1006. The patient physical activity metric may be determined by integrating the absolute value of the selected accelerometer vector signal over a predetermined duration, such as 2 seconds. The metric may be referred to as an "activity count" and is related to acceleration exerted on the motion sensor due to patient body motion associated with physical activity during a predetermined time period. In some examples, the 2 second (or other time interval) activity count may be used directly to indicate patient physical activity level, or combined in further calculations to obtain other physical activity metrics. At block 1006, at least one activity count may be compared to a threshold count to determine whether an abort condition is met. In some examples, a threshold number of activity counts (e.g., 10 to 40 activity counts) greater than a threshold activity count may be detected as an abort condition. For example, when 30 activity counts (each activity count determined during a 2 second interval) are greater than a threshold activity count during data acquisition, the control circuit 206 may determine that the abort condition is met due to high patient activity at block 1006. The activity count threshold may correspond to a patient activity that exceeds daily life activity, corresponds to a fast walk or other activity level that may be associated with a change in body motion and/or posture of the motion sensor signal as compared to a relatively low body activity (e.g., rest or low level daily life activity).

In other examples, each activity count may be compared to a median activity count, and if the exceeding threshold percentage of each activity count determined during data acquisition differs from the median activity count by more than a threshold difference (positive or negative), an abort condition may be detected at block 1006 due to variable patient activity. In yet other examples, the target heart rate or sensor-indicated pacing rate may be determined by the control circuit 206 based on the activity count. In some examples, a target heart rate or sensor-indicated pacing rate determined based on the patient's physical activity metric may be compared to a threshold rate (e.g., 10 beats per minute) that is greater than the current ventricular rate and/or a rate variability criteria for detecting an abort condition.

In addition to or instead of a heart rate and/or patient physical activity based on the suspension condition, the control circuit 206 may detect the suspension condition in response to the telemetry circuit 208 receiving a communication signal, for example, from the external device 20. For example, when a telemetry session is initiated and telemetry circuitry 208 is receiving a programming command, control circuitry 206 may detect an abort condition. In some examples, telemetry circuitry 208 may be enabled to transmit motion sensor signals or related data during the atrial event sensing parameter setting procedure of flowchart 1000; however, other telemetry signals (such as programming commands) received from external device 20 may be detected as an abort condition. The programming commands include programmable control parameters for changing the use by the control circuitry 206 in controlling sensing, therapy delivery, or other pacemaker functions.

When an abort condition is detected at block 1006, the control circuit 206 may determine at block 1010 whether a maximum period of time has been reached for data acquisition and atrial event sensing parameter selection. If not, the control circuit 206 may suspend data acquisition at block 1012 and return to block 1004 to resume vector signal analysis to acquire amplitude and timing data. The data acquisition may be restarted on the next processor interrupt signal or after a predetermined time period (e.g., after one minute, five minutes, or other selected time period). In some examples, the control circuit 206 may monitor the ventricular cycle length and/or patient physical activity level until an abort condition is no longer detected, and resume data acquisition at block 1004 when a relatively stable ventricular rate and/or relatively low stable patient physical activity is detected. In still other examples, when an abort condition is detected, the control circuit 206 may wait a predetermined time period (e.g., one minute) and then resume motion sensor signal data collection. In some cases, data acquired before the abort condition is detected will be discarded. In other examples, data acquired before the abort condition is detected is saved and combined with data obtained after the data collection is resumed.

The data acquisition process may be restarted multiple times to a maximum number of attempts or within a maximum period of time, such as up to one hour, four hours, 24 hours, or other selected maximum period of attempts. If the maximum number of attempts or maximum time period expires at block 1010 for successfully acquiring the amplitude and timing data of all vector signals being analyzed, the process may terminate at block 1020. At block 1002, the control circuitry 206 may terminate any temporary control parameters previously set to test values. For example, at block 1024, the control circuit 206 may switch back to a programmed pacing mode (e.g., VDD pacing mode) using any default or programmed atrial sense control parameters. The control circuitry 206 may generate a notification at block 1022 indicating the termination of the automatic sensing parameter selection process. The notification generated at block 1022 may include the number of data acquisition attempts or restarts and/or one or more associated abort conditions detected. Telemetry circuitry 208 may transmit any generated notifications to external device 20 for display on display unit 54. The user may program the atrial sense control parameters for use during the VDD pacing mode.

When data acquisition is completed within a maximum period of time (e.g., five minutes per test vector signal), control circuit 206 proceeds to block 1014 to generate a data distribution, for example, in the format of histograms as generally shown in fig. 8, 10, 12, and 13. Based on the histogram distribution, one or more atrial event sensing parameter values may be selected and set at block 1016. Although an abort condition is indicated at block 1106, after the amplitude and/or timing data is acquired from the motion sensor vector signal and before the histogram distribution is generated, it should be appreciated that the abort condition may be detected by the control circuit 206 before, during, or after the amplitude and timing data is acquired (block 1004), the histogram distribution is generated (block 1014), or the A4 sense control parameter is selected (block 1016). Monitoring one or more abort conditions (such as any of the examples described above) may occur during the process of blocks 1001-1016 and is not limited to a particular point in time during the process of setting test parameters, acquiring data, generating a distribution of data, and selecting A4 sense control parameters.

In some examples, an A4 sense vector signal is selected based on a histogram of the maximum vector signal amplitude during an A4 window generated for each vector signal at block 1016. For each vector signal being analyzed, the control circuit 206 may determine a valid maximum amplitude count. For example, a maximum amplitude that is greater than a minimum threshold amplitude (e.g., a minimum programmable value of late atrial event sensing threshold amplitude) may be counted. Any vector signal having a rejection threshold number of valid maximum amplitude samples counts less than the valid maximum amplitude may be rejected due to a possible A4 sensed vector signal. For example, when it is determined that less than 20 of the maximum amplitude acquired for a given vector signal is greater than the minimum programmable late atrial event sensing threshold, the vector signal may be rejected from the selection process because the maximum amplitude acquired for that vector signal meets the rejection criteria.

The control circuitry 206 may determine a median maximum amplitude of all maximum amplitudes acquired that is greater than a minimum programmable late atrial event sensing threshold (or other minimum threshold). For each non-rejected vector signal, a median maximum amplitude may be determined after discarding the maximum amplitudes that are less than the noise threshold. For example, for each vector signal that does not meet the rejection criteria, a histogram bin storing a number of maximum amplitudes less than or equal to the minimum programmable A4 sense threshold amplitude value may be discarded. The median value of the maximum amplitude may be determined from the distribution of the maximum amplitude in the histogram bins that are greater than the minimum A4 sensing threshold amplitude value. At block 1016, the vector signal corresponding to the highest median maximum amplitude determined from the A4 window may be selected as the A4 sense vector signal.

In the illustrative example given above where each of the four vector signals specified by the accelerometer axis combinations of 1+2, 1+3, 2+3, and 1+2+3 are analyzed, if the effective maximum amplitude sampling points of all three of the two axis signals are less than the threshold number, then at block 1016, the three axis vector signal may be selected as the A4 sense vector signal. When at least one of the three two-axis magnitude signals has the requisite number of effective maximum amplitude sampling points, the two-axis magnitude signal having the highest median maximum amplitude may be selected as the A4 sense vector signal (e.g., to save current consumption required to power the third accelerometer axis). When two or more of the two-axis volume signals have the same median maximum amplitude during the A4 window, the two-axis volume signals sharing a single axis for determining patient physical activity may be selected. In other examples, the uniaxial vector with the highest median maximum amplitude may be identified, and any two-axis vector signal including the uniaxial vector signal with the highest median maximum amplitude may be selected. In one example, when axis 2 is generally aligned with longitudinal axis 108 of pacemaker 14 (see fig. 2), priority is given to the 1+2 vector signal, then the 2+3 vector signal, then the 1+3 vector signal when the median maximum amplitude of the A4 window matches between two or all three of the vector signals. The uniaxial vector signal generally aligned with the longitudinal axis of pacemaker 14 may correspond to the highest A4 signal amplitude, although this may vary with implant location and orientation. The uniaxial vector signal with the highest A4 signal amplitude may be identified from empirical data and any biaxial vector comprising the identified uniaxial vector may be given priority when two or more biaxial vector signals have equal median maximum amplitudes.

At block 1016, the control circuit 206 may set an advanced A4 sensing threshold amplitude value (applied during the A4 window during atrial-tracked ventricular pacing) based on the median of the maximum amplitudes determined for the selected A4 sensing vector signals during the A4 window. In one example, the late A4 sense threshold amplitude value is set to the median of the maximum amplitudes determined for the selected A4 sense vector signal during the A4 window. In other examples, the late A4 sense threshold amplitude value is set to a percentage of the median maximum amplitude value(e.g., 60% to 80%). In some cases, the method of setting the late A4 sense threshold amplitude value depends on the median maximum amplitude during the A4 window. For example, if the median maximum amplitude during the A4 window is 1.2 meters/second ² (or other threshold acceleration), the late A4 sense threshold amplitude may be set to the median maximum amplitude. If the median maximum amplitude is greater than 1.2 m/s ² (or other threshold acceleration), the late A4 sense threshold amplitude may be set to 70% of the median maximum amplitude of the A4 window, but not less than 1.2 meters/second ² (or other minimum limit).

The method for setting the late A4 sense threshold amplitude value may vary depending on the selected A4 sense vector signal. For example, when two axis vector signals are selected as the A4 sense vector signals, the late A4 sense threshold amplitude may be set according to a percentage of the median value or at least a minimum limit. The late A4 sense threshold amplitude may be set according to different percentages or limits when the triaxial signal is selected as the A4 sense vector signal. The late A4 sense threshold amplitude may be set within a lower limit and an upper limit (e.g., between 0.5 meters/second ² And 5.0 m/s ² Between). The minimum A4 sensing threshold amplitude may vary depending on the number of accelerometer axis signals combined in the selected vector signal. For example, a relatively low minimum threshold amplitude (e.g., 0.6 meters/second ² ) May be implemented for single axis vector signals; an intermediate minimum threshold amplitude (e.g., 0.7 m/s) ² ) May be implemented for two-axis magnitude signals, and a highest minimum threshold amplitude (e.g., 0.8 meters/second ² ) May be implemented for triaxial signals. A relatively higher minimum threshold amplitude setting may be allowed for vector signals that are a combination of two or all three accelerometer axis signals as compared to a single axis vector signal, as additional noise is included in the summed axis signals with each axis signal added.

At block 1016, the control circuit 206 may set an early A4 sensing threshold amplitude (applied during the A3 window during atrial tracking ventricular pacing). Early A4 sensing threshold amplitude may be based onThe maximum amplitude determined for the selected A4 sense vector signal is set during the A3 window. In one example, the early A4 sense threshold amplitude value may be set by: the median maximum amplitude during the A3 window (for the selected A4 sensing vector signal) is determined, multiplied by a multiplication factor (e.g., 1.5) and the product of the median maximum amplitude and the multiplication factor is added to the late A4 sensing threshold amplitude. In some examples, setting the early A4 sense threshold amplitude value may include adding an offset, such as by adding 0.3 meters/second ² . The early A4 sensing threshold amplitude may be set based on a median maximum amplitude during the A4 window, a median maximum amplitude during the A3 window, or a combination of both (which may be a weighted combination). In one example, the early A4 sense threshold amplitude may be set within a lower limit and an upper limit (e.g., between 0.8 meters/second ² And 18.8 m/s ² Between).

The control circuit 206 may set the A3 window end time at block 1016 based on a distribution of timing of the most recent test threshold crossing during the A3 window determined for the selected A4 sense vector signal. In one example, the A3 window end time is set based on the median time spanned by the most recent test threshold during the A3 window plus an offset (e.g., plus 50 to 150 milliseconds). The A3 window end time may be set within a minimum limit and a maximum limit (e.g., not less than 650 milliseconds and not more than 1000 milliseconds).

Upon completion of atrial event sensing control parameter selection at block 1016, the control circuitry 206 may generate a notification of the completion of the parameter selection at block 1022, the notification including the selected parameter. The selected parameters may be transmitted by telemetry circuitry 208 to external device 20 to generate a display of the results of the automatic setup procedure. In the event that the selected A4 sense control parameter is valid, the control circuitry 206 may switch to an atrial tracking ventricular pacing mode (e.g., VDD pacing mode) at block 1024.

In some cases, the maximum amplitude during the A4 window may be too low to reliably select the A4 sense control parameters. For example, when less than the threshold number of sampling points exceeds the minimum programmable late A4 sense threshold amplitude value for all vector signals, the control circuit 206 may set the A4 sense control parameter to a default or previous setting and generate a notification indicating a low A4 signal amplitude at block 1022. Notifications may be transmitted by pacemaker 14 and displayed by external device 20 allowing a user to select and program pacing modes and sensing control parameters.

FIG. 16 is a flowchart 1100 of a method for adjusting selected atrial event sensing control parameters, according to one example. The control circuit 206 may determine start values for one or more A4 sense control parameters, such as the A3 window end time and early and late A4 sense threshold amplitude values, at block 1101 using the techniques described above in connection with fig. 6-15. After the start value is selected, the start value may be adjusted according to the method of flowchart 1100 to an operational value for use when switching to the permanent atrial tracking ventricular pacing mode. At blocks 1102 and 1104, the control circuit 206 may continue to operate in a non-atrial tracking ventricular pacing mode (e.g., in a VDI pacing mode) to enable adjustment of the early A4 sensing threshold amplitude value from its selected starting value and adjustment of the A3 window end time from its selected starting value.

At block 1102, the control circuit 206 may continue to operate in the VDI pacing mode and update the early A4 threshold amplitude value based on the maximum amplitude of the selected vector signal during the A3 window determined from one or more ventricular cycles. In one example, the median maximum amplitude during the A3 window may be determined from a predetermined number of consecutive ventricular cycles (e.g., 3 to 12 ventricular cycles). For example, the early A4 threshold amplitude value may be adjusted after every eight ventricular cycles based on a median maximum amplitude of the selected vector signal determined during the eight ventricular cycles. In one example, the median maximum amplitude of the A3 window may be determined from the 8 consecutive ventricular cycles as the 4 th highest maximum amplitude. In some examples, a target value for the early A4 threshold amplitude value may be determined based on the determined median maximum amplitude. The initial value of the early A4 threshold amplitude determined during the setup process may be adjusted by a predetermined increment or decrement toward the target value. The predetermined increment or decrement may be 0.1 to 0.5 m/s ² And in one example, 0.3 meters/second ² . While operating in the VDI pacing mode, the process may be repeated every 8 ventricular cycles (or other predetermined number of cycles) for an adjustment time interval, such as for one minute, two minutes, five minutes, or other selected time interval. The adjusted initial early A4 threshold amplitude value is not validated until the control circuit 206 switches to the atrial tracking ventricular pacing mode, such that an A4 event is not detected until all atrial event sensing parameters are adjusted from a starting value determined based on the distribution of the data described above to an operating value.

At block 1104, the control circuit 206 may determine a most recent threshold crossing time for the nominal threshold during each A3 window by the selected A4 sense vector signal. The nominal threshold may be set to determine the most recent threshold crossing time during the A3 window for establishing the A3 window end time as described above in connection with fig. 9 and 10. The starting A3 window end time may be adjusted based on a most recent test threshold crossing time during the A3 window determined from one or more ventricular cycles. The test threshold amplitude may be set to a percentage (e.g., 75%) of the late A4 sense threshold amplitude value set during the setup procedure described above. The A3 window end time established during the setup procedure is based on a value that can be set to a predetermined fixed value (e.g., 0.9 meters/second ² ) Is the most recent crossing of the test threshold. However, the initial late A4 threshold amplitude determined for the selected vector during the setup procedure is tailored to the patient and the selected atrial event sensing vector and A4 signal amplitude. The test threshold set as a percentage of the starting late A4 threshold amplitude may be a more appropriate threshold for determining the most recent threshold crossing time and setting the A3 window end time. For example, if at block 1101, the start value of the late A4 threshold amplitude is set to 2.5 meters/second at the end of the setup process ² The test threshold set to 75% of the late A4 threshold amplitude is 1.9 meters/second ² . The test threshold may be used during the A3 window to detect the most recent test threshold crossing time for adjusting the A3 window end time to provide patient-tailored for the selected sensing vectorOptimization of the end time of the A3 window.

After every 3 to 12 ventricular cycles, the median of the latest test threshold crossing times during the A3 window can be updated. The median time for the most recent A3 threshold amplitude crossing may be determined as the 4 th shortest time in the 8 ventricular cycles. The median of the most recent test threshold crossing times may be used to update the A3 window end time established during the setup procedure described above. The target A3 window end time may be set based on the median time. The A3 window end time may be adjusted towards the target value by adding or subtracting an adjustment interval from the current value of the A3 window end time. The A3 window end time may be adjusted every 8 ventricular cycles or other selected number of ventricular cycles for 2 minutes (or other adjustment time interval) to reach an adjusted starting A3 window end time that is effective as an operating A3 window end time when switching to an atrial tracking ventricular pacing mode (e.g., VDD pacing mode).

After adjusting the A3 window end time and/or early A4 sensing threshold from their respective set start values to an operational value during the VDI pacing mode, the control circuitry 206 may switch to a temporary atrial tracking pacing mode (e.g., VDD pacing mode) at block 1106. The early A4 sense threshold amplitude value and the operational value of the A3 window end value may be validated when switching to the temporary VDD pacing mode.

At block 1108, the late A4 sensing threshold amplitude value may be adjusted from its starting value. During this temporary VDD pacing mode, the late A4 sensing threshold amplitude value may be adjusted from its starting value based on the maximum amplitude of the selected sensing vector signal during one or more A4 windows. In one example, the control circuit 206 determines the median maximum amplitude during the A4 window of the selected vector signal after every X ventricular cycles. An adjusted late A4 sensing threshold amplitude value may be determined based on the determined median value. In some examples, the target late A4 sensing threshold amplitude value may be determined based on a median value determined every eight ventricular cycles. The initial value of the late A4 sense threshold amplitude may be adjusted upward or downward toward the updated target late A4 sense threshold by a predetermined adjustment interval (e.g., + -0.3 meters/second ² ). During the VDD pacing mode, the process may be repeated every X ventricular cycles for a predetermined time interval (e.g., every eight ventricular cycles for two minutes) to reach the operational late A4 sensing threshold amplitude value.

At block 1110, control circuitry 206 may determine a rate smoothing interval based on one or more ventricular cycle lengths during the temporary VDD pacing mode. In some examples, the initial rate smoothing interval is set to a programmed LRI. The median ventricular cycle length over X ventricular cycles (e.g., eight ventricular cycles) may be determined. The adjusted rate smoothing interval may be set to a predetermined interval that is longer than the median ventricular cycle length (e.g., 100 milliseconds to 150 milliseconds longer than the median ventricular cycle length). The rate smoothing interval may be updated every X ventricular cycles over a predetermined time interval (e.g., two minutes).

After adjusting the starting value of the late A4 sense threshold amplitude and adjusting the rate smoothing interval during the temporary VDD pacing mode, at block 1112, the control circuit 206 may switch to the permanent atrial tracking pacing mode with the operating value of the A4 sense control parameter and the adjusted rate smoothing interval in effect. In this way, the starting value of the A4 sense control parameter and the rate smoothing interval determined during the automatic setup procedure described above may be adjusted to an operating value that is set and validated according to the simultaneous signal amplitude and timing characteristics of the selected vector signal and the current ventricular rate.

FIG. 17 is a flowchart 1050 of a method for setting atrial event sensing parameters according to one example. At block 1052, the control circuit 206 sets the pacing mode to a non-atrial tracking pacing mode (e.g., VDI pacing mode) and sets the ventricular pacing interval according to the selected pacing rate (e.g., 50 pulses per minute), although other pacing rates may be used. At block 1054, the control circuit 206 selects one or more A4 sense vector signals. In some examples, the process of flowchart 1050 is performed using only the A4 sense vector signal selected by the method of fig. 7 described above. In other examples, portions of the process of flowchart 1050 may be performed for some or all of the available motion sensor vector signals (including single axis vector signals, two axis vector signals, and/or three axis vector signals). In one example, the control circuit 206 selects each of the available two-axis and three-axis signals one at a time on a rotational basis for sensing a motion signal from which amplitude and/or timing data is acquired for selecting a starting value of at least one atrial event sensing parameter. The data needed to determine the starting value of the atrial event sensing parameter may be obtained for a plurality of vector signals used in the process of fig. 7 for selecting the A4 sensing vector signals. However, in some examples, only data corresponding to the vector signal selected as the A4 sense vector signal at block 516, block 520, or block 522 of fig. 7 may be used to determine a starting value of the atrial event sensing parameter.

At block 1056, the control circuit 206 sets an initial A3 window end time. In one example, the A3 window end time is set to a percentage of the median ventricular cycle length, as described above. In other examples, the A3 window end time is set to a default or programmed end time, which may be 800 milliseconds to 1000 milliseconds after the ventricular electrical event, which may be a ventricular pacing pulse or a sensed R-wave. The A3 window is set to begin at the expiration of the post-ventricular atrial blanking period as described above in connection with fig. 5, and may end 900 milliseconds after an ventricular electrical event, as an example.

At block 1058, the control circuit 206 determines at least one motion signal characteristic during the A3 window for each ventricular cycle over a predetermined time period or a predetermined number of ventricular cycles. In various examples, the motion signal characteristics determined during each A3 window are determined for at least one minute for each vector signal selected. As an example, the motion signal characteristics may include a most recent crossing time of the maximum amplitude of the motion signal and/or the test threshold amplitude during the A3 window.

At block 1060, the control circuit 206 may determine one or more motion signal characteristics after the end time of the A3 window (during the A4 window) for each of the ventricular cycles during a predetermined time period or a predetermined number of ventricular cycles. As an example, the control circuit 206 may determine a maximum peak amplitude during the A4 window, a time of the maximum peak amplitude, and/or a time of threshold crossing. Other examples of motion signal characteristics that may be determined by the control circuit 206 during the A4 window after the A3 window end time are described below in connection with fig. 19, 20, and 22.

At block 1062, the control circuit 206 identifies at least a portion of the ventricular cycles as confidence A4 event cycles. The control circuit 206 may identify the ventricular cycle as a confidence A4 event cycle by: the motion signal sensed after the end time of the A3 window is determined to meet the confidence A4 event criteria by comparing the motion signal characteristics determined during the A4 window at block 1060 to the confidence A4 event criteria at block 1062. Some exemplary techniques for determining that the A4 window motion signal characteristic meets the confidence A4 event criteria are described below in connection with fig. 19 and 20. For example, the maximum amplitude of the motion signal during the A4 window may need to be greater than the confidence A4 threshold amplitude and/or occur during the confidence A4 time interval. In some examples, the ratio of the maximum amplitude during the A3 window to the maximum amplitude during the A4 window may need to be less than a maximum ratio threshold, such as described below in connection with fig. 21 and 22.

At block 1064, the control circuit 206 determines an atrial event sensing parameter based on the motion signal characteristics determined during the A3 window associated with the ventricular cycle identified as the confidence A4 event cycle. The control circuit 206 may determine a characteristic of the motion signal during each of the A3 windows associated with the ventricular cycles identified as the confidence A4 event cycles. The control circuit 206 may set a starting value for the atrial event sensing parameter based on these determined A3 window features corresponding to the confidence A4 event period.

As described herein, after selecting a starting value of the atrial event sensing parameter, the control circuit 206 may switch to an atrial tracking ventricular pacing mode and sense an atrial event signal from the motion signals (the motion signals sensed using the selected A4 sensing vector) according to the atrial event sensing control parameter and generate a sensed atrial event signal in response to sensing the atrial event signal. The pulse generator 202 may generate pacing pulses in response to sensed atrial event signals, for example, at AV pacing intervals from the sensed atrial event signals. In some examples, the starting value of the atrial event sensing parameter set at block 1064 may be adjusted one or more times based on the motion signal during one or more subsequent ventricular cycles, for example as generally described above in connection with fig. 16, before sensing the A4 event based on the atrial event sensing parameter and triggering a ventricular pacing pulse in response to the A4 event sensed in the atrial tracking ventricular pacing mode. One or more atrial event sensing parameters may be updated based on motion signal characteristics during an atrial tracking ventricular pacing mode, for example, as generally disclosed in commonly assigned U.S. patent application No. 17/159,636 (Sheldon et al).

FIG. 18 is a flow chart 1150 of a method for establishing an early A4 sense threshold amplitude to be applied to sense atrial events during an A3 window, according to some examples. As generally described above, the process of flowchart 1150 may be performed during a setup procedure for acquiring motion signal data for establishing an A4 sensing parameter. The early A4 sensing threshold amplitude value may be applied to the motion signal during an A3 window for sensing the early A4 signal (e.g., during an atrial tracking ventricular pacing mode).

At block 1152, control circuit 206 sets the pacing mode to a non-atrial tracking pacing mode and sets the ventricular pacing interval according to the selected pacing rate (e.g., 50 pulses per minute), although other pacing rates may be used. At block 1154, the control circuit 206 selects one or more A4 sense vector signals. In some examples, the process of flowchart 1150 is performed using only the A4 sense vector signal selected by the method of fig. 7 described above. In other examples, portions of the process of flowchart 1150 may be performed for some or all of the available motion sensor vector signals (including single axis vector signals, two axis vector signals, and/or three axis vector signals). For example, the data required to generate a histogram for setting early (and late) A4 sensing threshold amplitude values may be obtained for a plurality of vector signals used in the process of fig. 7, but only the data corresponding to the vector signals selected as A4 sensing vector signals at block 516, block 520, or block 522 of fig. 7 may be used to populate the histogram and determine a starting value for the early A4 sensing threshold amplitude.

An A3 window end time may be set at block 1156. In one example, the A3 window end time is set to a percentage of the median ventricular cycle length. In other examples, the A3 window is set to the A3 window end time selected at block 714 of fig. 9 based on a histogram of the most recent threshold crossing time determined from the motion signal sensed using the selected A4 sensing vector signal. The A3 window may extend from a start time (e.g., the end of the post-ventricular atrial blanking interval), which may be programmable or based on empirical data, to an A3 window end time, which in some examples is automatically determined from a distribution of recent test threshold amplitude crossing times, e.g., as described above in connection with fig. 9-10. In other examples, a manually programmed or default A3 window end time is set at block 1156. In this case, establishing the early A4 sense threshold amplitude value does not have to perform the process of fig. 9 prior to the process of fig. 18. The A3 window end time may be set to a default value of 800 milliseconds to 1000 milliseconds after a ventricular electrical event (e.g., a ventricular pacing pulse or sensed R-wave) for determining a characteristic of the motion signal during the A3 window during the setup procedure.

At block 1158, the control circuit 206 may determine a maximum amplitude of the motion signal during an A3 window of each ventricular cycle over a predetermined number of cycles or a predetermined time period. At block 1160, the control circuit 206 determines a characteristic of the motion signal sensed after the end time of the A3 window. In one example, at block 1160, a maximum amplitude of the motion signal after the A3 window end time and before the next ventricular electrical event (e.g., during the A4 window) is determined for each ventricular cycle over a predetermined number of cycles or a predetermined time period.

At block 1164, the maximum amplitude determined during the A3 window is used by the control circuit 206 to generate an A3 window maximum amplitude distribution, for example by populating an A3 window maximum amplitude histogram allocated in the memory 210. However, in the method of fig. 18, the control circuit 206 may first filter the maximum amplitude determined during the A3 window by removing, discarding, or ignoring the A3 window maximum amplitude corresponding to the ventricular cycle during which the sensed motion signal after the A3 window does not meet the confidence A4 event criteria.

Thus, at block 1162, the control circuit 206 may determine whether the confidence A4 event criteria is met based on one or more motion signal characteristics (such as maximum amplitude) determined from motion signals sensed after the A3 window end time (and at the next sensed or paced ventricular electrical event). The term "confidence A4 event criteria" (also referred to herein as "atrial event criteria") refers to a criteria applied to a motion signal that, when satisfied, indicates a high likelihood of a true A4 signal during the A4 window. The confidence A4 event criteria may be applied to the motion signal sensed after the A3 window end time and, in some cases, during the A3 window and after the A3 window end time. The confidence A4 event criteria may include at least one of a confidence A4 threshold amplitude, a confidence A4 time interval region, and/or a maximum ratio of A3 window maximum amplitude to A4 window maximum amplitude. Confidence A4 event criteria is defined to identify ventricular cycles with A4 window peak amplitudes and/or timings with high probability as true A4 event signals, e.g., as shown in fig. 19.

Fig. 19 is a graph 1200 of a motion signal 1202 (shown as a rectified signal) during one ventricular cycle 1201 of a non-atrial tracked ventricular pacing mode according to one example. The motion signal 1202 is generated from the selected motion signal sensing vector. Ventricular cycle 1201 begins on ventricular pacing pulse 1204 and ends on second ventricular pacing pulse 1230, which is delivered at a programmed ventricular rate interval (LRI) during a non-atrial tracking pacing mode. The motion sensor signals 1202 include an A1 signal, an A2 signal, and an A3 signal corresponding to a ventricular event that occurs at a predictable time after a ventricular pacing pulse 1204, for example, as described above in connection with fig. 5. During the process of fig. 18, control circuitry 206 may set a post-ventricular atrial blanking period 1206 in response to pulse generator 202 delivering ventricular pacing pulses 1204. The control circuit 206 sets the A3 window 1210 to begin when the post-ventricular atrial blanking period 1206 expires and has an end time 1214, which may be set to a default value (e.g., 900 milliseconds) during the set-up procedure, which may depend on the pacing rate.

The control circuit 206 may determine a maximum amplitude 1218 of the motion signal 1202 during the A3 window 1210 for selecting an early A4 sense threshold amplitude. The maximum amplitude 1218 may be used to populate an A3 window histogram allocated in the memory 210 for determining a distribution of the maximum amplitudes and selecting a starting value for the early A4 sensing threshold amplitude based on at least a portion of the determined A3 window maximum amplitudes.

The control circuit 206 may determine the maximum amplitude 1220 during the A4 window 1216 (which ends on the next ventricular pacing pulse 1230) after the A3 window end time 1214. The control circuit 206 may populate the A4 window histogram with the maximum amplitude 1220 to determine a distribution of A4 window maximum amplitudes for selecting the A4 sensing vector signal and/or setting late A4 sensing threshold amplitudes. Additionally, the A4 window maximum amplitude 1220 may be analyzed by the control circuit 206 to determine whether the ventricular cycle 1201 includes a signal during the A4 window 1216, which may be a true A4 event signal with high confidence. In this way, the control circuitry 206 may identify the ventricular cycle 1201 as a confidence A4 event cycle.

Since ventricular pacing pulses 1204 and 1230 are not tracked to atrial events, a true A4 event may occur at any time during ventricular cycle 1201 and not necessarily during A4 window 1216. However, the A3 event signal is expected to follow the ventricular pacing pulse 1204 at a predictable interval as described above in connection with fig. 5. In some ventricular cycles, a true A4 event signal may happen to occur during the A4 window. However, in other cycles, the A4 event signal may occur randomly during the A3 window or earlier, or not at all during the ventricular cycle, depending on the relative atrial rate and paced ventricular rate, and the timing of the atrial cycle relative to the ventricular cycle.

Thus, in order to highly confidence that the maximum amplitude 1218 during the A3 window 1210 is associated with a true A3 signal instead of, for example, a fused A3/A4 signal, the control circuit 206 may apply a confidence A4 event criteria to the motion signal sensed during the A4 window after the A3 window end time 1214. In some examples, the control circuit 206 may determine an A4 window maximum amplitude 1220 and compare it to a confidence A4 threshold amplitude 1228. As an example, the confidence A4 threshold amplitude 1228 may be set to 0.8 meters/second ² 0.9 m/s ² 1.0 m/s ² Or 1.2 m/s ² . In some examples, the control circuit 206 may determine that the confidence A4 event criteria are met at block 1162 of fig. 18 in response to the maximum amplitude 1220 being greater than or equal to the confidence A4 threshold amplitude 1228. In other examples, the control circuit 206 may determine that the confidence A4 event criteria is met in response to detecting that the motion signal 1202 crosses the threshold 1228 after the A3 window end time 1214.

The control circuit 206 may additionally or alternatively compare the time of the A4 window maximum amplitude 1220 to the confidence A4 interval 1226. Confidence A4 interval 1226 may be defined by an early A4 time limit 1222 and/or a late A4 time limit 1224. As an example, the early A4 time limit 1222 may be defined as 0 ms, 20 ms, 30 ms, 40 ms, 50 ms, 60 ms, 75 ms, 85 ms, or 100 ms after the A3 window end time 1214. In other examples, early A4 time limit 1222 may be set to 950 milliseconds to 1000 milliseconds, as an example, after ventricular pacing pulse 1204. If the maximum peak amplitude 1220 of the motion sensor signal 1202 occurs earlier than the early A4 time limit 1222, the maximum peak may correspond to a late A3 event signal or a fused A3/A4 event signal, rather than a true A4 signal that is separated in time from the A3 event signal. As shown, when the maximum peak amplitude 1220 occurs after the early A4 time limit 1222, the maximum peak 1220 is more likely to be associated with a true A4 event.

As an example, the late A4 time limit 1224 may be set to 100 milliseconds to 400 milliseconds after the A3 window end time 1214. The late A4 time limit 1224 may be set to 100 milliseconds to 300 milliseconds after the early A4 time limit 1222. In various examples, the confidence A4 interval 1226 may extend from 900 milliseconds after the ventricular pacing pulse 1204 up to 1200 milliseconds after the ventricular pacing pulse 1204. In another example, when the A3 window end time 1214 is 900 milliseconds after the ventricular pacing pulse 1204, for example, the confidence A4 interval 1226 extends from 975 milliseconds to 1125 milliseconds after the ventricular pacing pulse 1204. In other examples, the confidence A4 interval 1226 may expire when the next ventricular pacing pulse 1230 is delivered, such that the confidence A4 interval 1226 extends from the early A4 time limit 1222 until the end of the ventricular cycle 1201. Thus, the confidence A4 interval 1226 may be part of the A4 window 1216, beginning later and/or ending earlier than the A4 window 1216.

When the maximum peak amplitude 1220 occurs during the confidence A4 interval 1226, the control circuit 206 may determine that the confidence A4 event criteria are met. When the maximum peak amplitude 1220 is greater than the threshold 1228 and/or occurs during the confidence A4 interval, the control circuit 206 may determine that the confidence A4 event criteria are met. In still other examples, the control circuit 206 may determine that the confidence A4 event criteria are met when the motion signal 1202 occurs during the confidence A4 interval crossing the threshold 1228. In the illustrative example, the maximum peak amplitude 1220 is required to be at least 1.2 meters/second ² And occurs between 975 milliseconds and 1125 milliseconds after ventricular pacing pulse 1204 in order to meet the confidence A4 event criteria and ventricular cycle 1201 is identified as a confidence A4 event cycle. If the maximum peak amplitude of the motion signal 1252 occurs earlier or later than the confidence A4 time interval 1226, in some examples, the control circuit 206 may determine that the ventricular cycle 1251 is not a confidence A4 event cycle, even if the maximum peak amplitude is greater than the threshold amplitude 1228.

In response to determining that the confidence A4 event criteria is met, the control circuit 206 includes a maximum amplitude 1218 of the motion signal 1202 during the A3 window 1210 in the A3 window histogram. When the confidence A4 event criteria are met, the maximum amplitude 1218 during the A3 window 1210 is likely to correspond to a true A3 event signal, and is unlikely to correspond to a fused A3/A4 event signal. Thus, the A3 window maximum amplitude 1218 associated with the ventricular cycle 1201 identified as the confidence A4 cycle is used to select an early A4 sensing threshold amplitude by the control circuit 206. However, when the confidence A4 event criteria is not met, the A3 window maximum amplitude may correspond to an early A4 event or fused A3/A4 event signal and may be ignored or discarded by the control circuit 206 when determining the distribution of A3 window maximum amplitudes and selecting an early A4 sensing threshold amplitude.

As described below in connection with fig. 24, the control circuit 206 may determine the latest crossing time 1240 of the test threshold 1212 by the motion signal 1202 during the A3 window 1210. The control circuit 206 may use the latest test threshold crossing time 1240 to adjust the initial, default A3 window end time 1214 to an A3 window end time tailored to the patient and the selected A4 sensing vector used to sense the motion signal 1202. In some examples, the most recent threshold crossing time determined for a ventricular cycle identified as a confidence A4 event cycle (such as ventricular cycle 1201) may be used to set an optimized A3 window end time tailored to the patient prior to switching to the atrial tracking ventricular pacing mode.

Fig. 20 is a graph 1250 of motion signal 1252 during one ventricular cycle 1251 starting with a ventricular pacing pulse 1254 and ending with a ventricular pacing pulse 1232. A3 window 1210 begins when post-ventricular atrial blanking period 1206 expires. In this example, the maximum peak amplitude 1270 of the motion signal 1252 sensed after the A3 window end time 1214 is during the confidence A4 interval 1226 but is less than the threshold 1228. The control circuit 206 may determine that the confidence A4 event criteria is not met in response to the maximum peak amplitude 1270 after the A3 window end time 1214 being less than the threshold 1228. In some examples, if the maximum peak amplitude 1270 is greater than the threshold 1228 but is earlier or later than the confidence A4 interval 1226, the control circuit 206 may determine that the confidence A4 event criteria is not met. In the example of fig. 20, the control circuit 206 does not use the maximum amplitude 1268 during the A3 window 1210 to populate the A3 window histogram or to select an early A4 sense threshold amplitude. In this example, the A4 event may occur during the A3 window 1210 such that the maximum amplitude 1268 does not correspond to a true A3 event signal that is separated in time from the A4 event signal. The maximum peak amplitude 1268 may correspond to the fused A3 and A4 (A3/A4) event signals.

The control circuit 206 may be configured to determine the crossing time 1242 of the test threshold 1212 as described above for adjusting the initial, default A3 window end time 1214 to an optimized A3 window end time for the patient. In some examples, when the sensed motion signal 1252 after the A3 window end time 1214 does not meet the confidence A4 event criteria, the most recent test threshold crossing time 1242 determined from the associated ventricular cycle 1251 is ignored or discarded by the control circuit 206 and is not used to determine an A3 window end time for A4 event sensing when switching to the atrial tracking ventricular pacing mode.

It should be appreciated that the one or more motion signal characteristics determined during the A4 window 1216 used to determine that the confidence A4 event criteria are met are not limited to the maximum amplitude, time of maximum amplitude, and/or threshold crossing time as described in connection with the examples of fig. 19 and 20. In other examples, the one or more features that may be determined by the control circuit 206 from the motion signal sensed after the A3 window end time 1214 to determine when the confidence A4 event criteria is met may include a maximum slope, timing of the maximum slope, number of peaks greater than a threshold amplitude, number of inflection points, signal width, area of signal spikes with maximum peak amplitude, integration of the motion signal within the A4 window 1216, integration of the motion signal within the confidence A4 interval 1226, overall motion signal waveform morphology during the confidence A4 interval (e.g., as compared to morphology templates of known A4 signals), or other morphological features. Various examples of features of the motion signal that may be determined for comparison to the confidence A4 event criteria may be determined by the control circuit 206 from the motion signal during the A4 window 1216 and/or during the confidence A4 interval 1226.

Referring again to fig. 18, the control circuit 206 determines at block 1162 whether the confidence A4 event criteria is met, for example, using any of the example techniques described in connection with fig. 19 and 20. When the confidence A4 event criteria is met for a ventricular cycle, the control circuit 206 may populate an A3 window histogram of the A4 sensing vector signal being evaluated with an A3 window maximum amplitude determined from the same ventricular cycle. If the confidence A4 event criteria is not met, the control circuitry 206 may not use the A3 window maximum amplitude (by skipping block 1164), but may use the maximum A4 window amplitude to populate an A4 window histogram of the corresponding A4 sensing vector signal being evaluated at block 1166. In other examples, the control circuit 206 may not populate the A4 window histogram with the maximum A4 window amplitude when the ventricular cycle is not identified as a confidence A4 event cycle based on the criteria at block 1162.

At block 1167, the control circuit 206 may determine whether a period of time for accumulating the motion signal data of the selected A4 sense vector signal has expired. For example, the control circuit 206 may accumulate data for populating the A3 window histogram and the A4 window histogram for 30 seconds, 1 minute, 2 minutes, 5 minutes, 10 minutes, 20 minutes, or another selected period of time, e.g., by collecting data for five minutes or less from each of the three available two-axis vector signals and the three-axis vector signals. In the illustrative example, five minutes of data for a given vector signal may be acquired in a 1 minute slot by cycling through four vector signals over a 20 minute period. In other examples, the data for a given vector signal may be acquired over one minute or another selected time period and evaluated for setting the starting value of the A4 sensing parameter without the need to collect and analyze data from multiple sensing vector signals.

If the time period for accumulating motion signal data has not expired, the control circuitry 206 may return to block 1158 to continue the process of accumulating data and evaluating ventricular cycles according to the confidence A4 event criteria. If the time interval has expired, the control circuit 206 may proceed to block 1168. In some examples, the specified time period for accumulating motion signal data is not required. The control circuitry 206 may analyze the motion signals sensed during one or more ventricular cycles until at least one or another specified number of ventricular cycles is identified as a confidence A4 event cycle based on the criteria applied at block 1162. When at least one trusted A4 event period is identified, the control circuitry 206 may proceed to block 1168.

In some examples, motion signal amplitudes less than a noise threshold may be discarded at block 1168, e.g., as generally described above in connection with fig. 11. Very low maximum amplitudes may not be indicative of a true A3 or A4 event and may be baseline noise of the motion signal. When performing the process of fig. 18 for multiple vector signals, the control circuit 206 may select an A4 sense vector signal based on the A4 window histogram at block 1170, e.g., according to the techniques described above in connection with fig. 7. In other examples, the control circuit 206 may select the A4 sense vector signal at block 1170 based on the maximum amplitude of the motion signal during the A4 window of one or more cardiac cycles identified as a confidence A4 event cycle according to the criteria applied at block 1162. For example, a vector signal having the highest A4 window maximum amplitude determined from one or more of the confidence A4 event periods in all vector signals being evaluated may be selected as the A4 sense vector signal at block 1170. In another example, a vector signal having a highest A4 window maximum amplitude and a highest ratio of A4 window maximum amplitude to A3 window maximum amplitude determined from at least one confidence A4 event period may be selected as the A4 sense vector signal. In other examples, the process of fig. 18 is performed for a single vector signal that has been selected as the A4 sense vector signal, and the selection from among the plurality of vector signals is not required at block 1170.

At block 1172, the control circuit 206 selects an early A4 sensing threshold amplitude based on a remaining, non-discarded A3 window maximum amplitude that has been filtered to include an A3 window maximum amplitude corresponding to a ventricular cycle during which the motion signal meets the confidence A4 event criteria. The early A4 sensing threshold amplitude may be set to a percentile of the A3 window maximum amplitude. For example, the early A4 sense threshold amplitude may be set to 85%, 90%, 95%, 100%, 110% of the maximum amplitude of the A3 window120%, 150%, 200%, or other selected percentile. The offset may optionally be added to (or subtracted from) a specified percentile value of the maximum amplitude of the A3 window. In one example, the early A4 sense threshold amplitude is set to 95 th percentile of the A3 window maximum amplitude plus 0 meters/second ² To 1 m/s ² Is set in the above-described range (a). In some examples, the early A4 sense threshold amplitude is set to be greater than all A3 window maximum amplitudes determined from the confidence A4 event period in order to minimize the likelihood of oversensing an A3 event as a false A4 event.

In other examples, the early A4 sense threshold amplitude may be set to the average A3 window maximum amplitude plus a predetermined number of standard deviations. The control circuit 206 may determine the average maximum amplitude and standard deviation and set the early A4 sense threshold amplitude to the average plus 2.8 standard deviation, plus 3 standard deviations, plus 3.2 standard deviations, or other specified number of standard deviations.

The control circuit 206 may generate a distribution of A4 window maximum amplitude data at block 1166 using the maximum amplitude determined at block 1160 during the A4 window, for example by populating the A4 window maximum amplitude histogram. The A4 window histogram bins storing maximum amplitudes less than the noise threshold may optionally be discarded at block 1168. The control circuit 206 may select an advanced A4 sense threshold amplitude value based on a distribution of the remaining (not discarded) of A4 window maximum amplitudes at block 1174. A method for selecting late A4 sensing threshold amplitude is described above in connection with fig. 13.

In some examples, all maximum amplitudes determined during the A4 window may be stored in memory 210, for example, for populating the A4 window histogram at block 1166 and for selecting late A4 sensing threshold amplitudes at block 1174. In other examples, as described below in connection with fig. 19 and 20 or fig. 21 and 22, an A4 window maximum amplitude determined from a ventricular cycle identified as a confidence A4 event cycle may be used to populate an A4 window histogram at block 1166. Late A4 sense threshold amplitude values may be selected at block 1174 based on these filtered A4 window maximum amplitudes such that a high percentage or all maximum amplitudes in the A4 window histogram corresponding to the confidence A4 event period are greater than the late A4 sense threshold amplitude to facilitate reliable sensing of all true A4 event signals.

Fig. 21 is a flow chart 1300 of a method for identifying a ventricular cycle as a confidence A4 event cycle according to some examples. The process of flowchart 1300 may be performed in conjunction with the setup procedure described above (e.g., in conjunction with fig. 18) for acquiring motion signal data for establishing A4 sensing parameters. As described above in connection with fig. 18, upon initiation of the setup procedure, control circuit 206 may set the pacing rate to a selected rate (e.g., 50 pulses per minute) and select one or more A4 sense vector signals for use in acquiring motion signal data. The A3 window end time is set to be used during a setup procedure using any of the techniques described above. During the non-atrial tracking pacing mode, control circuit 206 determines a maximum peak amplitude during each A3 window at block 1302. The control circuit 206 determines the maximum peak amplitude after each A3 window end time at block 1304. The maximum amplitude after the end time of the A3 window may be used to identify a confidence A4 event period.

In addition to or in lieu of the confidence A4 event criteria described above in connection with FIGS. 19 and 20, the control circuitry 206 may apply a maximum A3 to A4 amplitude ratio requirement for identifying a confidence A4 event period. At block 1306, the control circuit 206 may determine an a3:a4 amplitude ratio for a plurality of ventricular cycles during a predetermined time interval in which motion signal data is being acquired. The a3:a4 amplitude ratio may be determined as the ratio of the maximum peak amplitude of the motion signal during the A3 window to the maximum peak amplitude of the motion signal after the end time of the A3 window (during the A4 window). A3:a4 amplitude ratio may be determined for each ventricular cycle over a predetermined time period (or a predetermined number of ventricular cycles). In other examples, the a3:a4 amplitude ratio may be determined for only ventricular cycles identified as trusted A4 event cycles based on other A4 event criteria (such as the criteria described above in connection with fig. 19 and 20).

In some examples, the control circuit 206 may compare each determined a3:a4 amplitude ratio to a maximum ratio threshold for determining whether a confidence A4 event criteria is met for a given ventricular cycle. However, the control circuit 206 may first determine whether the A3:A4 amplitude ratio requirement is applied as part of the confidence A4 event criteria. To this end, the control circuit 206 may determine a representative measure of the a3:a4 amplitude ratio determined at block 1308. As examples, the representative metrics may be average, median, trimmed median, predetermined percentile. This representative measure of the a3:a4 amplitude ratio determined during the predetermined time interval may be compared to a maximum ratio threshold at block 1308.

When the representative measure of the a3:a4 amplitude ratio is less than the threshold ratio (referred to herein as the "enable threshold ratio") at block 1310, the control circuit 206 enables the maximum a3:a4 amplitude ratio requirement at block 1314. A3:a4 amplitude ratio requirements can be applied alone or in combination with any of the confidence A4 event criteria described above. In some examples, a percentage of the a3:a4 amplitude ratio that is greater than the maximum ratio threshold is determined and compared to the threshold percentage. In an illustrative example, the maximum a3:a4 amplitude ratio requirement may be enabled when a threshold percentage (e.g., at least 80%) of the a3:a4 amplitude ratio determined during the predetermined time period is less than two. The threshold percentage of the A3: A4 amplitude metric that needs to be less than the enable threshold ratio 2 (or other selected enable threshold ratio) may be 50%, 60%, 70%, 75%, 80%, 85%, 90% or other selected percentage.

In other examples, at block 1310, a median or average a3:a4 amplitude ratio may be determined and compared to an enable threshold ratio (e.g., ratio 1.75, 1.85, 2.0, 2.1, or other selected ratio). When the average or median a3:a4 amplitude ratio is less than the enable threshold ratio, the control circuit 206 enables the maximum a3:a4 amplitude ratio requirement at block 1314. When the maximum a3:a4 amplitude ratio requirement is enabled, the a3:a4 amplitude ratio for the given ventricular cycle may be less than a maximum ratio threshold, which may have the same value as the enablement threshold ratio applied at block 1310 or a different value, for ventricular cycles meeting the confidence A4 event criteria at block 1162 of fig. 18.

Referring again to fig. 18, when the maximum a3:a4 amplitude ratio requirement is enabled, when the control circuit 206 determines that the a3:a4 amplitude ratio of the ventricular cycles is greater than the maximum ratio threshold, the control circuit 206 does not populate the A3 window histogram with the A3 window maximum amplitude determined for the given ventricular cycle at block 1164. When the a3:a4 amplitude ratio for a given ventricular cycle is less than (or equal to) the maximum ratio threshold and meets any other applied criteria, such as the motion signal crossing the confidence A4 threshold amplitude during the confidence A4 interval as described above in connection with fig. 19, the control circuit 206 may determine that the confidence A4 event criteria are met at block 1162 of fig. 18.

Returning to FIG. 21, when the control circuit 206 determines at block 1308 that the measure of the A3:A4 amplitude ratio is not less than the enable threshold ratio, the control circuit 206 disables the maximum A3:A4 amplitude ratio requirement at block 1312. In some patients, the peak amplitude of the true A3 event signal may be regularly high and/or the peak amplitude of the true A4 event signal may be regularly low such that the a3:a4 amplitude ratio is continuously high, e.g., greater than the enable threshold ratio applied at block 1310. In these patients, the maximum A3:A4 amplitude ratio requirement for identifying the confidence A4 event period may be disabled. Other criteria, such as those described above in connection with fig. 19 and 20, may be applied at block 1162 of fig. 18 for identifying a confidence A4 event period for selecting an A3 window maximum amplitude for populating the A3 window histogram at block 1164.

It should be appreciated that instead of comparing the ratio of the A3 window maximum amplitude to the A4 window maximum amplitude to a maximum ratio threshold, the control circuit 206 may be configured to determine the ratio of the A4 window maximum amplitude to the A3 window maximum amplitude and compare the ratio to a minimum ratio threshold. When the determined a3:a4 amplitude ratio for the ventricular cycle is greater than the minimum ratio threshold, the control circuit 206 may determine that the confidence A4 event criteria is met and the ventricular cycle is a confidence A4 event cycle. Further, it should be appreciated that instead of determining the ratio, the difference between the A3 window maximum amplitude and the A4 window maximum amplitude may be determined by the control circuit 206 and compared to a difference threshold to identify ventricular periods that meet the confidence A4 event criteria.

FIG. 22 is a graph 1280 of a motion signal 1282 that does not meet a confidence A4 event criteria based on an A3 to A4 amplitude ratio during one ventricular cycle 1281, according to one example. Ventricular cycle 1281 begins at ventricular pace pulse 1284 and ends at ventricular pace pulse 1234. As described above in connection with fig. 19 and 20, the A3 window 1210 begins when the post-ventricular atrial blanking period 1206 expires and has an end time 1214. In this example, the maximum peak amplitude 1272 of the motion signal 1282 after the A3 window end time 1214 occurs during the confidence A4 interval 1226 and crosses the confidence A4 amplitude threshold 1228. When the maximum a3:a4 amplitude ratio requirement is not enabled (block 1310 of fig. 21), the control circuit 206 may determine that the confidence A4 event criteria are met (e.g., at block 1162 of fig. 18) in response to the maximum peak amplitude 1272 after the A3 window end time 1214 being greater than the amplitude threshold 1228 during the confidence A4 interval 1226.

However, in some cases, a true A4 event may occur in the A3 window, resulting in a large maximum peak amplitude 1286 during the A3 window 1210. Additionally, the motion signal peaks may occur during the A4 window and even within the confidence A4 interval region 1226, which may be greater than the confidence A4 amplitude threshold 1228. In this case, the confidence A4 event criteria based on the confidence A4 amplitude threshold 1228 and/or confidence A4 interval 1226 may be met, but the maximum peak amplitude 1286 may not represent the true A3 event amplitude. As shown in fig. 22, after the fused A3/A4 event signal during the A3 window 1210, relatively higher motion signal peaks may occur during the A4 window due to diastolic reflux or other hemodynamic conditions that may occur when the A4 event (and associated ventricular filling) occurs relatively early in the ventricular cycle (e.g., fused with the A3 event) and ventricular systole (evacuation) is delayed.

Because the relatively large maximum peak amplitude 1286 in the A3 window 1210 may correspond to the fused A3/A4 event signal, the maximum peak amplitude 1286 does not represent a true A3 event amplitude that is separated in time from a true A4 event signal, and thus may be discarded by the control circuit 206 when the early A4 sensing threshold is set. To detect this situation where the possible fused A3/A4 event signals subsequently meet the relatively high motion signal peaks of other confidence A4 event criteria during the A4 window, the control circuit 206 may apply the maximum a3:a4 ratio requirement as described in connection with fig. 21. Accordingly, the control circuit 206 may be configured to determine a ratio of the A3 window maximum peak amplitude 1286 to the A4 window maximum peak amplitude 1272. When the ratio is relatively high (indicating a very large A3 window signal), the confidence A4 event criteria may not be met.

Thus, when the maximum a3:a4 ratio requirement is enabled, the control circuit 206 may compare the ratio of the A3 window peak amplitude 1286 to the A4 window peak amplitude 1272 to a maximum ratio threshold. If the A3:A4 amplitude ratio is greater than the maximum ratio threshold, then the ventricular cycle 1281 is not identified as a confidence A4 event cycle. When populating the A3 window histogram, the A3 window maximum amplitude 1286 may be ignored or discarded by the control circuit 206, even though the maximum peak amplitude 1272 after the A3 window end time 1214 meets other confidence A4 event criteria. In some examples, the A4 window maximum amplitude 1272 may be discarded or ignored by the control circuit 206 when populating the A4 window histogram, and may not be used to select an A4 sensing vector signal and/or late A4 sensing threshold amplitude. As an example, the maximum ratio threshold may be 1.8 to 2.4, and may be set to 2 in one example.

In some patients, the ratio of true A3 event signal amplitude to true A4 event signal amplitude may be continually high, for example, due to regularly large A3 event signals and/or regularly small A4 event signals. Thus, the maximum a3:a4 amplitude ratio requirement may not apply to all patients or all times that a setup procedure is performed in a given patient. When the motion signal regularly exhibits a high ratio of the maximum amplitude of the A3 window to the maximum amplitude of the A4 window, the maximum a3:a4 amplitude ratio requirement may not be enabled for use in determining the confidence A4 event period. However, when the control circuit 206 determines that the a3:a4 amplitude ratio is regularly less than the specified ratio threshold, the control circuit 206 may enable the maximum a3:a4 amplitude ratio requirement for determining the confidence A4 event period.

For example, as described above in connection with FIG. 21, the maximum A3:A4 amplitude ratio requirement may be enabled only when the average, median, predetermined percentage, or other metric representing A3:A4 amplitude ratios determined during a predetermined time interval is less than an enabled threshold ratio. Once the maximum a3:a4 amplitude ratio requirement is enabled, the enabling threshold ratio for enabling the maximum ratio requirement may be the same as or different from the maximum ratio threshold applied to the respective a3:a4 amplitude ratio. For example, the average, median, or predetermined percentage of all a3:a4 amplitude ratios determined by the control circuit 206 during a predetermined time interval (or a predetermined number of ventricular cycles) may be required to be less than a first threshold (e.g., less than 2.0) in order to enable the maximum a3:a4 ratio requirement. Once enabled, each a3:a4 amplitude ratio may be required to be less than a second threshold, which may be equal to, greater than, or less than the first threshold. Further, in some examples, the second threshold, which serves as the maximum ratio threshold, may be determined by the control circuit 80 based on the a3:a4 amplitude ratio determined during a predetermined time interval (or a predetermined number of ventricular cycles). The maximum ratio threshold may be set as a percentage of the representative measure of the a3:a4 amplitude ratio, or may be set by adding an offset to the representative measure of the a3:a4 amplitude ratio.

The control circuit 80 may be configured to enable or disable a3:a4 amplitude ratio requirements applied to the respective a3:a4 amplitude ratios based on a maximum amplitude determined during the A3 window and a maximum amplitude determined after an A3 window end time during a predetermined time interval. The above-described method of determining a representative metric of the a3:a4 amplitude ratio determined over a predetermined time interval and comparing the representative metric to an enable threshold ratio is one example of how the control circuit 80 may be configured to enable or disable the a3:a4 amplitude ratio requirement of the confidence A4 event criteria. More generally, the control circuit 80 may determine the A3 window maximum amplitude and the A4 window maximum amplitude over a predetermined time period (or a predetermined number of ventricular cycles) to determine when the A3 window maximum amplitude and/or the A4 window maximum amplitude meets the enabling criteria. An enabling criterion may be defined for detecting when the A3 window maximum amplitude is continuously high compared to the A4 window maximum amplitude, e.g. based on a ratio or difference between the A3 window maximum amplitude and the A4 window maximum amplitude. The enabling criterion may be met when the ratio or difference between the A3 window maximum amplitude and the A4 window maximum amplitude indicates a continuously relatively low difference or a continuously low A3 window maximum amplitude and a continuously high A4 window maximum amplitude. When the ratio or difference between the A3 window maximum amplitude and the A4 window maximum amplitude indicates a continuously relatively high difference between the A3 window maximum amplitude and the A4 window maximum amplitude, the enablement criteria may not be met, where the A3 window maximum amplitude is relatively high compared to the A4 window maximum amplitude.

In various examples, the a3:a4 amplitude ratio (or difference) may be determined by the control circuit 206 for all ventricular cycles during a predetermined time interval for a given A4 sensing vector signal for determining when to enable the maximum a3:a4 amplitude ratio requirement. In other examples, the a3:a4 amplitude ratio may be determined by the control circuit 206 only for ventricular periods determined to meet all other confidence A4 event criteria. In these periods, a measure of the A3:A4 amplitude ratio may be determined and compared to a criterion for enabling or disabling the maximum A3:A4 amplitude ratio requirement. When the criteria for enabling the maximum a3:a4 amplitude ratio requirement are met, the control circuit 206 may then compare each of the individually determined a3:a4 amplitude ratios to a maximum ratio threshold. The control circuit 206 may compare the a3:a4 amplitude ratio to a maximum ratio threshold for all ventricular cycles during a predetermined time interval or only for ventricular cycles meeting its other confidence A4 event criteria in order to identify a confidence A4 event cycle and filter out ventricular cycles that are not confidence A4 event cycles for setting the early A4 sensing threshold and optionally other A4 sensing parameters.

Fig. 23 is a conceptual diagram 1350 of an A3 window histogram that may be accumulated in memory 210 during a setup procedure for establishing a starting value of an early A4 sensing threshold amplitude according to some examples. The lower histogram 1360 includes the A3 window maximum amplitude determined by the control circuit 206 from the motion signal during all ventricular cycles during the predetermined time interval for acquiring motion signal data. The control circuit 206 may populate an A3 window histogram for each A4 sensing vector signal being analyzed, the A3 window histogram including the A3 window maximum amplitudes for all ventricular cycles during a predetermined time interval. Upper histogram 1352 includes an A3 window maximum amplitude determined from the ventricular cycles that meet the confidence A4 event criteria. The control circuit 206 may determine whether the motion signal meets the confidence A4 event criteria according to any of the techniques described above in connection with fig. 18-22. The A3 window maximum amplitude corresponding to a ventricular cycle that does not meet the confidence A4 event criteria is filtered or removed from the A3 window maximum amplitude included in the (filled) top confidence A4 event cycle histogram 1352.

The control circuit 206 may set the A3 maximum amplitude limit 1354 based on the A3 window maximum amplitude stored in the confidence A4 event period histogram 1352. A3 maximum amplitude limit 1354 may be set to a percentile, for example, the 80 th, 85 th, 90 th, or 95 th percentile of the maximum amplitude stored in histogram 1352. In other examples, the A3 maximum amplitude limit 1354 may be set to an average plus a predetermined number of standard deviations, e.g., an average plus three standard deviations.

The control circuit 206 may discard or ignore the maximum amplitude less than the A3 maximum amplitude limit 1354 in the second histogram 1360 that includes the A3 window maximum amplitudes from all ventricular cycles during the predetermined time interval for data acquisition. The control circuit 206 may set the early A4 sense threshold amplitude 1364 based on a maximum amplitude within the range 1362 that is greater than the A3 maximum amplitude limit 1354. The maximum amplitude that occurs during ventricular cycles within range 1362 that are relatively high and do not meet the confidence A4 event criteria during the A4 window may correspond to the fused A3/A4 event signal during the A3 window. Most of these signals should be detected as an A4 event during the A3 window, while most of the signals corresponding to maximum amplitudes less than the A3 maximum amplitude limit 1354 may be true A3 event signals and should not be detected as A4 events during the A3 window. Thus, the early A4 sensing threshold amplitude 1364 may be set to a relatively low percentile of the maximum amplitude within the range 1362. For example, early A4 sensing threshold amplitude value 1364 may be set to the 10 th, 15 th, 20 th, or other percentile of the maximum amplitudes within range 1362 such that a majority of the maximum amplitudes in range 1362 will be sensed as A4 events during the A3 window during the atrial tracking ventricular pacing mode.

Thus, in some examples, the control circuit 206 may be configured to determine a first percentile of the motion signal characteristics determined during an A3 window associated with a ventricular cycle identified as a confidence A4 event cycle. The control circuit 206 may identify a motion signal feature value that is greater than a first percentile of the confidence A4 event cycle features, for example, from all ventricular cycles of a predetermined time interval. The control circuit 206 may set the A4 sensing parameter based on the first percentile, for example, set the A4 sensing parameter to a second percentile of motion signal characteristics from all ventricular cycles identified as being greater than the first percentile.

In other examples, the control circuit 206 may determine the A3 maximum amplitude limit 1354 as a predetermined percentile of the amplitudes stored in the histogram 1352 and set the early A4 sensing threshold amplitude 1364 to the A3 maximum amplitude limit 1354 plus the offset. As an example, the offset may be capable of being from 0.5 meters/second ² To 2.0 m/s ² Programming is performed. In one example, early A4 sense threshold amplitude 1364 may be set to an A3 maximum amplitude limit 1354 plus 1.0 meters/second ² Is set in the above-described range (a). A relatively low offset may be used when the A4 window maximum amplitude during the confidence A4 event period is relatively low, and a relatively high offset may be used when the A4 window maximum amplitude during the confidence A4 event period is relatively high. In some examples, the control circuit 206 may determine the offset based on a metric of A4 event amplitude determined from the A4 window histogram.

Fig. 24 is a flowchart 1400 of a method for establishing a start value of an A4 sensing vector parameter according to another example. As generally described above, for example, in connection with fig. 16, the control circuitry 206 may operate in a non-atrial tracking ventricular pacing mode (e.g., VDI pacing mode). The control circuitry 206 may operate in a non-atrial tracking ventricular pacing mode for a predetermined time interval while, for example, acquiring amplitude and timing information from motion signals for a plurality of sensing vectors. The sensing vector for sensing the A4 event signal during a subsequent atrial tracking ventricular pacing mode may be selected in accordance with the techniques described above in connection with fig. 7 and 8. When a sensing vector is selected, the amplitude data acquired for the selected sensing vector may be used to determine the starting values of the early and late A4 sensing threshold amplitudes. However, in the method of flowchart 1400, one vector signal is initially selected for sensing the motion signal and acquiring motion signal amplitude and timing data, and if the sensed vector meets vector acceptance criteria, the acquired data is used to establish starting values for other A4 sensed parameters without having to acquire motion signal data from multiple sensed vector signals, which may require a relatively long period of time.

At block 1402, the control circuit 206 sets the pacing mode to a non-atrial tracking pacing mode and sets an initial, nominal, or default A3 window end time (e.g., 800 milliseconds to 900 milliseconds) at block 1404. Other examples for setting the initial A3 window end time (corresponding to the start of the A4 window) during acquisition of the motion signal data are described above, e.g., in connection with fig. 7. At block 1406, the control circuit 206 may select a default A4 sense vector signal, which may be programmed by the clinician. The A4 sensing vector signal may be a combination of two axis signals (e.g., an axis signal aligned with the longitudinal axis of the pacemaker housing and one radial axis signal). In other examples, the A4 sense vector signal may be a single axis signal or a different two axis signal. In some cases, all three axis signals may be selected in combination at block 1406, however, typically either a single axis signal or a two axis signal may be selected first to save power. The triaxial signal may be selected only after other uniaxial and/or biaxial signals that are needed to power fewer axes of the multi-axis accelerometer do not meet vector acceptance criteria, as described below.

At block 1408, the control circuit 206 may start a timer set for a predetermined time period for determining motion signal amplitude and timing data for one selected A4 sensing vector signal. For example, the data may be determined from the motion signal for 1 minute, 2 minutes, 3 minutes, 4 minutes, or 5 minutes, or other selected time period. Typically, the time period for determining data from one motion signal is less than the time period required to determine data from a plurality of motion signals selected one at a time as described above, for example in connection with fig. 7.

The control circuit 206 determines amplitude and timing data from the motion signals at blocks 1410, 1412, and 1414 during each ventricular cycle or multiple ventricular cycles until the timer expires at block 1415. For example, the control circuit 206 may determine a maximum amplitude during each A3 window of each or more ventricular cycles at block 1410. As described above, the A3 window maximum amplitude stored in memory may be filtered to remove the A3 window maximum amplitude associated with a ventricular cycle during which the motion signal does not meet the confidence A4 event criteria.

The control circuit 206 may determine the most recent test threshold crossing time during the A3 window at block 1412. In some examples, control circuit 206 stores the most recent test threshold crossing times for all A3 windows (all ventricular cycles) during a predetermined time interval. In some examples, the most recent test threshold crossing time may be a negative-going threshold crossing. Memory 210 may be configured to store the most recent test threshold crossing time in a histogram, such as that described in connection with fig. 10. In other examples, the control circuit 206 may determine the most recent test threshold crossing time and store it in the memory 210 only for ventricular cycles determined to meet the confidence A4 event criteria, for example, as described in connection with any of fig. 18-22. For example, the control circuit 206 may determine that the maximum amplitude of the motion signal during the A4 window is greater than the confidence A4 threshold amplitude, that the maximum amplitude (or confidence A4 threshold crossing) occurs during the confidence A4 interval region, and/or that the A3 to A4 amplitude ratio is less than the maximum ratio threshold. When the confidence A4 event criteria is met for a given ventricular cycle, the most recent test threshold crossing time may be stored in memory 210 for use in selecting an A3 window end time. The most recent test threshold crossing time determined during the confidence A4 event ventricular cycle most likely corresponds to the most recent test threshold crossing time of the true A3 event signal, thereby enabling the control circuit 206 to set the A3 window end time that is most likely to occur after the A3 event signal to reliably avoid A3 event signal oversensing.

At block 1414, the control circuit 206 may determine and store a maximum amplitude of the motion signal during the A4 window after the end time of the A3 window for the plurality of ventricular periods. When the timer expires at block 1415, the control circuit 206 may determine whether the vector acceptance criteria are met at block 1416. The vector acceptance criteria may be a criteria applied to an A4 window maximum amplitude to facilitate selection of an A4 sensing vector signal with high A4 signal strength or signal-to-noise ratio for facilitating reliable sensing of an A4 event signal. For example, the vector acceptance criterion may require that at least a threshold number of A4 window maximum amplitudes be greater than a threshold amplitude. The control circuitry 206 may discard or ignore less than 0.8 meters/second at block 1416 ² Maximum amplitude of A4 window (or another selected threshold), and is determined to be greater than or equal to 0.8 m/s ² Whether the count of the remaining maximum amplitude of the threshold (or another selected threshold) is greater than the threshold number of samples. For example, an A4 sense vector signal may require at least 10 to 20 samples to meet the vector acceptance criteria. However, the number of samples required may depend on a total predetermined time period (or a total predetermined number of ventricular cycles) during which the A4 window maximum amplitude is determined for the selected A4 sensing vector.

Other criteria that may be applied at block 1416 to accept the currently selected A4 sensing vector signal are generally described above in connection with fig. 7 and 8. For example, the ratio of the count of all A4 window maximum amplitudes greater than the high amplitude threshold to the count of all A4 window maximum amplitudes greater than the noise threshold may be determined as a high maximum amplitude ratio that may be compared to a vector acceptance threshold ratio for accepting or rejecting the currently selected A4 sensing vector signal.

In response to determining that the vector acceptance criteria is not met, the control circuit 206 may return to block 1406 to select a different A4 sense vector signal and repeat the process of determining amplitude and threshold crossing time from the motion signals of the plurality of ventricular cycles, e.g., during a predetermined time period. When the vector acceptance criteria are met (which may be met after evaluating the first A4 sense vector signal or after evaluating two or more A4 sense vector signals), the control circuit 206 may proceed to block 1418, block 1420, and block 1422 to select an A4 sense control parameter.

In this way, when the first selected A4 sensing vector signal meets vector acceptance criteria, the time required to establish the starting A4 sensing parameter value may be significantly reduced by determining amplitude and timing data from a plurality of vector signals and then selecting one vector signal as the motion signal for sensing the A4 signal during the atrial tracking ventricular pacing mode. For example, when the predetermined period of time is one minute, the data accumulated over one minute may be used to establish a start A4 sensing threshold parameter value when the A4 window maximum amplitude meets the vector acceptance criteria at block 1416, thereby excluding the time and processing power spent for accumulating data from multiple A4 sensing vector signals.

At block 1418, the control circuit 206 may select an early A4 sense threshold amplitude value according to any of the techniques described, for example, in connection with fig. 18-23. The early A4 sensing threshold amplitude may be determined based on an A3 window maximum amplitude determined from the confidence A4 event period. At block 1420, the control circuit 206 may select the late A4 sense threshold amplitude value, for example, according to any of the techniques described above. In some examples, late A4 sensing threshold amplitude may be established in accordance with the techniques described above in connection with fig. 11 and 13. In other examples, the late A4 sense threshold amplitude may be determined based on an A4 window maximum amplitude determined from the confidence A4 event period. The control circuit 206 may select the A3 window end time at block 1422 based on the most recent test threshold crossing time during the A3 window, e.g., according to any of the examples described above in connection with fig. 9 and 10. In other examples, the A3 window end time may be set using only the most recent test threshold crossing time determined from the ventricular cycles identified as the confidence A4 event cycles according to the techniques described above in connection with fig. 19-22. In this case, the A3 window end time may be set to a percentile of the most recent test threshold crossing time determined from the ventricular cycle during which the motion signal meets the confidence A4 event criteria.

In other examples, the control circuit 206 may determine an average, median, or other center metric of the most recent test threshold crossing time during the A3 window associated with the confidence A4 event period. The control circuitry 206 may determine the A3 window end time as the center metric plus the offset. The offset may be based on an extended metric of the most recent test threshold crossing time after removing any test threshold crossing time associated with ventricular cycles that do not meet the confidence A4 event criteria. For example, the A3 window end time may be set to the center metric plus n standard deviations of the most recent threshold crossing time, where n is 2.5, 2.8, 3.0, or other selected number of standard deviations.

In still other examples, the control circuit 206 may determine a value of a first predetermined percentile of the latest test threshold crossing times associated with ventricular cycles meeting the confidence A4 event criteria. The control circuit 206 may determine the A3 window end time at block 1422 as a second predetermined percentile that is greater than all of the most recent test threshold crossing times (determined for all of the ventricular cycles during the predetermined time interval) of the first predetermined percentile. The first percentile may be set to a relatively high percentile (e.g., 90% to 100%) of the most recent test threshold crossing time of the confidence A4 event period. The second percentile may be set to a relatively low percentile (e.g., 5% to 20%) of the most recent test threshold crossing time determined from all ventricular cycles that are greater than the value of the first percentile. Alternatively, the A3 window end time may be set to the first percentile plus the offset (e.g., plus 20 to 60 milliseconds). In this way, the A3 window end time may be set longer than and before the expected time of the end of the true A3 event signal.

The control circuit 206 may set a minimum A3 window end time at block 1424. As an example, the minimum A3 window end time may be set to 600 milliseconds, 650 milliseconds, or 700 milliseconds. The A3 window end time set at block 1404 is set to a relatively long time interval to allow the control circuit 206 to determine a distribution of the maximum amplitude of the motion signal and the threshold crossing time during the A3 window that most likely includes the A3 signal. However, a minimum A3 window end time may be set at block 1424 to allow the control circuit 206 to adjust the A3 window end time from the end time selected at block 1422 to a shorter end time, as described further below, as appropriate.

The control circuit 206 may adjust the test threshold applied during the A3 window at block 1425 for accumulating the most recent threshold crossing time. The test threshold amplitude may be set to a percentage (e.g., 75%) of the late A4 sense threshold amplitude value selected at block 1420. As described above in connection with fig. 16, the A3 window end time selected at block 1422 during the setup procedure is based on the recent crossing of a nominal test threshold (shown in fig. 19 as threshold 1212), which may be set to a predetermined nominal value (e.g., 0.9 meters/second) ² ). However, the late A4 sensing threshold amplitude selected at block 1420 is tailored to the patient and the selected A4 sensing vector signal such that the second test threshold set to a percentage of the initial late A4 sensing threshold amplitude at block 1425 may be a more appropriate test threshold for determining the most recent test threshold crossing time for setting the A3 window end time for the selected A4 sensing vector signal.

For example, if the start value of the late A4 sense threshold amplitude is set to 2.5 meters/second at block 1420 ² Then the test threshold may be adjusted to 75% of the late A4 sensing threshold amplitude or 1.9 meters/second in this example at block 1425 ² Significantly higher than 0.9 m/s ² Is set to a nominal test threshold value. The adjusted second test threshold may be applied during one or more subsequent A3 windows to detect a recent test threshold crossing time in order to adjust a starting value of an A3 window end time (e.g., based on a recent test threshold crossing accumulated over a predetermined time period (or a predetermined number of ventricular cycles))Time) to further optimize the A3 window end time.

The A3 window end time may be adjusted at block 1426 over a predetermined adjustment time period (e.g., 10 seconds, 30 seconds, one to five minutes, or other time period, or other number of ventricular cycles) using the techniques described above in connection with fig. 16. For example, the A3 window end time may be adjusted every 8 th ventricular cycle during an adjustment time interval of X minutes (e.g., 1 minute to 2 minutes). The A3 window end time may be adjusted based on a median or other representative value of the most recent test threshold crossing times determined during the A3 windows of the first eight (or other selected number) ventricular cycles. The test threshold applied during the adjustment time interval during the A3 window is now the adjusted test threshold set at block 1425, which may be based on the late A4 sensing threshold.

In an example, the control circuit 206 may increase or decrease the A3 window end time every 8 ventricular cycles toward the target A3 window end time by a specified adjustment increment (e.g., 5 milliseconds to 50 milliseconds) based on the median most recent test threshold crossing time. The target A3 window end time may be determined by the control circuit 206 as the median most recent test threshold crossing time plus an offset or as a percentage of the median most recent test threshold crossing time. If the A3 window end time is equal to the target A3 window end time, no adjustment is made. If the A3 window end time is within an adjustment increment from the target A3 window end time, the control circuitry 206 may adjust the A3 window end time to the target A3 window end time. The A3 window end time is not adjusted to a value less than the minimum A3 window end time determined at block 1424. Each time the A3 window end time is adjusted, it may be stored or updated in memory 210, with or without the adjusted A3 window end time being validated after each adjustment is made during the adjustment time interval. At block 1428, when switching to the atrial tracking ventricular pacing mode (e.g., VDD pacing mode), the A3 window end time reached at the end of the X minute adjustment interval may function as the operational A3 window end time.

In some examples, in addition to the A3 window end time at block 1426, the early A4 sensing threshold amplitude may also be adjusted from its selected starting value during the X minute adjustment interval. For example, when the A3 window end time is adjusted every N (e.g., 8) ventricular cycles over an X (e.g., 1 to 2) minute time period, the early A4 sensing threshold amplitude may be adjusted based on the maximum amplitude of the motion signal sensed during the A3 window with the adjusted end time. As described above in connection with fig. 16, the early A4 sensing threshold amplitude may be adjusted toward a target value that is based on a median maximum amplitude determined during a predetermined number of A3 windows (e.g., eight A3 windows).

After adjusting the A3 window end time (and in some cases, the early A4 sensing threshold amplitude) for X minutes, the control circuitry 206 may switch to a temporary atrial tracking pacing mode (e.g., VDD pacing mode) at block 1428. The early A4 sense threshold amplitude and the operational value of the A3 window end time reached at block 1426 may be effective when switching to the temporary VDD pacing mode. The operational value of the late A4 sense threshold amplitude selected at block 1420 may be validated when switching to the temporary VDD pacing mode.

The late A4 sensing threshold amplitude may be adjusted from its starting value for the next predetermined time interval or a predetermined number of ventricular cycles (e.g., Y minutes) at block 1430. During the temporary VDD pacing mode, the control circuit 206 may adjust the late A4 sensing threshold amplitude value from its starting value at block 1430 based on the median value of the A4 window maximum amplitude determined after every X ventricular cycles, e.g., as described above in connection with fig. 16. In some examples, the target value of late A4 sensing threshold amplitude may be determined based on a median A4 window maximum amplitude determined after every eight ventricular cycles. The starting value of the late A4 sensing threshold amplitude may be adjusted toward the target value each time the target value is updated. The process may repeat for Y minutes (e.g., less than one minute or one to five minutes) every 8 ventricular cycles (or other selected number of cycles) during the temporary VDD pacing mode to reach the operational late A4 sensing threshold amplitude value at block 1430.

When adjusting the late A4 sensing threshold amplitude, a test threshold based on the late A4 sensing threshold and applied during the A3 window for determining the latest threshold crossing time may be adjusted. In this way, the A3 window end time may be adjusted one or more times during or upon expiration of Y minutes during which the late A4 sensing threshold amplitude is adjusted. In other examples, the A3 window end time is not adjusted during Y minutes and remains at a value reached before switching to the temporary VDD pacing mode.

The minimum and maximum A3 window end times may be set at block 1432 based on the adjusted A3 window end times. For example, the A3 window range may be defined by a minimum value set to the adjusted A3 window end time minus a first offset (e.g., 25 milliseconds or other selected value) and a maximum value set to the adjusted A3 window end time plus a second offset (e.g., 75 milliseconds (or other selected value)). In other examples, the A3 window end time range may be set to the adjusted A3 window end time plus or minus 50 milliseconds (or other selected value). In these examples, the A3 window end time may then be adjusted by the control circuit 206 within 100 milliseconds.

The offset used to set the minimum and maximum values at block 1432 may be scaled based on the A3 window end time reached at block 1430. For example, when the A3 window end time is relatively short (e.g., less than 800 milliseconds), an equal offset (e.g., plus or minus 50 milliseconds) may be used to set the A3 window end time range. When the A3 window end time is relatively long (e.g., greater than or equal to 800 milliseconds), unequal offsets (e.g., 75 milliseconds plus 25 milliseconds minus) may be used to set the A3 window end time range. The A3 window end time range limit may adjust the length of the A3 window end time based on the most recent test threshold crossing time determined by control circuit 206 after switching to VDD pacing mode at block 1434.

After switching to the atrial tracking pacing mode at block 1434, the control circuitry 206 senses an A4 event signal according to the established atrial event sensing parameters. Pulse generator 202 generates pacing pulses in response to sensed atrial event signals to provide atrial synchronized ventricular pacing, for example, by delivering ventricular pacing pulses upon expiration of an AV pacing interval that begins in response to the atrial event signals sensed by atrial event detector circuit 240.

Prior to switching to the atrial tracking pacing mode at block 1434, the process of flowchart 1400 is described as including adjusting the A3 window end time (and in some cases, the early A4 sensing threshold amplitude) at block 1426 and/or the late A4 sensing threshold amplitude at 1430. However, in some examples, the adjustment at block 1426 and/or block 1430 may be optional because the value of the A4 sensed parameter selected at block 1418, block 1420, and block 1422 is determined from the relatively recently sensed motion signal during a simplified setup procedure that may last as short as one to two minutes. These adjustments may be included when the time interval between determining the motion signal characteristics for the selected values at block 1418, block 1420, and block 1422 and switching to the permanent atrial tracking ventricular pacing mode at block 1434 is relatively long (e.g., 5 minutes, 10 minutes, 15 minutes, or longer). The motion signal may change over time such that a relatively long delay between acquisition of the motion signal for selection of the initial A4 sensing parameter and switching to the atrial tracking ventricular pacing mode may result in a suboptimal atrial event sensing parameter at the beginning of atrial synchronous pacing. However, when the time between acquiring the motion signal (at block 1410, block 1412, and block 1414) for setting the atrial event sensing parameter (at block 1418, block 1420, and block 1422) and switching to the atrial tracking ventricular pacing mode (at block 1434) is relatively short (e.g., less than five minutes), further adjustment of the sensing parameter may not be required.

In an illustrative example, the setup procedure may include a relatively short time period (e.g., one minute) for acquiring motion signal characteristics from a single A4 sensing vector signal for setting the A3 window end time and the start values of early and late A4 sensing threshold amplitudes (blocks 1402-1420). The setup procedure may further include a relatively short (e.g., one minute) time period for adjusting the start A3 window end time using an adjusted test threshold set based on the start value of the late A4 sensing threshold (blocks 1422-1426). After this relatively short setup procedure (e.g., two minutes total), the control circuit 206 may switch to the atrial tracking ventricular pacing mode (at block 1434) using the start values of the early and late A4 sensing thresholds and the adjusted start A3 window end time as the operational A4 sensing control parameters. The A3 window end time minimum to maximum range may be set at block 1432 based on the data determined at block 1426 for subsequent adjustment of the A3 window end time. However, in some examples, blocks 1428 and 1430 may be skipped because the time from setting the start value to completing the adjusted start A3 window end time is relatively short (only two minutes in this example) such that the start values of the early and late A4 sensing thresholds remain valid. When switching to the atrial tracking pacing mode at block 1434, these starting values of the operational A4 sensing control parameters may then be adjusted by the control circuit 206 during the atrial tracking ventricular pacing mode as needed based on the ongoing analysis of the motion signal.

Although the process of determining the motion signal characteristics at blocks 1410, 1412 and 1414 and the process of determining the A4 sense control parameters at blocks 1418, 1420 and 1422 are shown in a particular order in fig. 24, it should be understood that the control circuit 206 may determine the motion signal characteristics, verify that the vector acceptance criteria are met, and select the starting values of the A4 sense control parameters in a different order than shown or in a parallel process.

FIG. 25 is a flowchart 1500 of a method for establishing A4 event sensing control parameters according to another example. At block 1502, as an example, pacemaker 14 operates in a non-atrial tracked ventricular pacing mode by delivering asynchronous ventricular pacing pulses at an LRI corresponding to a pacing rate of 40, 50, or 60 pulses per minute, for example. At block 1504, the control circuit 206 may select an A4 sense vector signal, which A4 sense vector signal may be selected according to any of the techniques described above. The control circuitry 206 may set a nominal A3 window end time (e.g., 800 milliseconds to 1000 milliseconds), which may depend at least in part on the pacing rate.

At block 1506, the control circuit 206 identifies at least one confidence A4 event period by analyzing the motion signal received from the selected motion signal sensing vector using any of the techniques described above (e.g., using the examples described in connection with any of fig. 18-22). In some examples, when as few as one confidence A4 event period is identified at block 1506, the control circuit 206 may proceed to block 1507. In other examples, the control circuit 206 may analyze the motion signal at block 1506 until a specified number of confidence A4 event periods (e.g., 2, 3, 5, 8, or 12 confidence A4 event periods) are identified.

It should be appreciated that if at least one or another requisite number of confidence A4 event periods are not identified at block 1506, the control circuit 206 may return to block 1504 and select a different A4 sensing vector signal and adjust the ventricular pacing rate. The A4 event amplitude in the currently selected sensing vector signal may be too low to identify a confidence A4 event signal. In some cases, if the atrial rate and the ventricular rate are similar, and the atrial contraction happens to occur simultaneously with ventricular systole or early ventricular diastole, a confidence A4 event signal may not be identified because no true A4 event occurred during the A4 window. Adjustment of the ventricular pacing rate for one or more cardiac cycles may shift the A4 event into an A4 window of one or more ventricular cycles, thereby enabling the control circuit 206 to identify at least one confidence A4 event cycle.

Additionally or alternatively, the control circuit 206 may optionally verify the acceptability of the A4 sense vector signal at block 1507 based on the at least one trusted A4 event period identified at block 1506. For example, the control circuit 206 may verify that one or more A4 window maximum amplitudes (or a minimum, average, median, or other representative value of a plurality of A4 window maximum amplitudes determined from a plurality of confidence A4 event periods) are greater than a threshold amplitude. The control circuit 206 may additionally or alternatively verify that the ratio of the A4 window maximum amplitude to the A3 window maximum amplitude (determined from one or more confidence A4 event periods) is greater than an A4/A3 ratio threshold. After identifying at least one trusted A4 event period and optionally verifying that the A4 sensing vector signal meets criteria related to A4 signal strength, the control circuit 206 may proceed to block 1508 to establish a starting value of the A4 sensing parameter to be used during atrial-synchronized ventricular pacing.

At block 1508, the control circuit 206 may establish a late A4 sense threshold amplitude. When a single confidence A4 event period is identified at block 1506, the control circuitry 206 may determine a maximum amplitude of the motion signal during the A4 window of the confidence A4 event period. The control circuit 206 may set the late A4 sense threshold amplitude to be less than the A4 window maximum amplitude of the confidence A4 event period. The late A4 sensing threshold amplitude may be determined as a percentage (e.g., 50%, 60%, 70%, 75%, 80%, 85%, or other percentage) of the maximum amplitude of the A4 window. In other examples, the control circuit 206 may determine the starting A4 sense threshold amplitude established at block 1508 to be an offset that is less than the A4 window maximum amplitude. For example, the late A4 sense threshold amplitude may be set to 0.3 meters/second less than the A4 window maximum peak amplitude ² 0.4 m/s ² 0.5 m/s ² 0.75 m/s ² Or 1.0 m/s ² 。

When multiple trusted A4 event periods are identified at block 1506, the control circuitry 206 may determine a lowest value of A4 window maximum amplitude for the multiple trusted A4 event periods. The control circuit 206 may set the start value of the late A4 sense threshold amplitude to a minimum value that is less than the A4 window maximum peak amplitude.

At block 1510, the control circuit 206 may set a starting value of the early A4 sense threshold amplitude based on at least one confidence A4 event period. When a single confidence A4 event period is identified at block 1506, the control circuitry 206 may determine a maximum amplitude of the motion signal during the A3 window of the confidence A4 event period. The control circuit 206 may set the early A4 sense threshold amplitude to be greater than the A3 window maximum amplitude of the confidence A4 event period. The early A4 sensing threshold amplitude may be determined as a multiple (e.g., 1.25, 1.5, or 2.0, or other multiple) of the A3 window maximum amplitude. In other examples, the control circuit 206 may determine the starting A3 sense threshold amplitude established at block 1508 to be an offset greater than the A3 window maximum amplitude. For example, early stagesThe A4 sense threshold amplitude may be set to 0.5 meters/second greater than the A3 window maximum amplitude ² 0.75 m/s ² 1.0 m/s ² 1.5 m/s ² Or 2.0 m/s ² 。

When multiple trusted A4 event periods are identified at block 1506, the control circuitry 206 may determine a highest value for the A3 window maximum peak amplitude for the multiple trusted A4 event periods. The control circuit 206 may set the starting value of the early A4 sense threshold amplitude to a highest value that is greater than the A3 window maximum peak amplitude.

In some examples, control circuit 206 may set the early A4 sense threshold amplitude based on both the A3 window maximum amplitude and the A4 window maximum amplitude. For example, the multiple of the early A4 sense threshold amplitude applied to the A3 window maximum amplitude by the control circuit 206 may be scaled according to the A4 window maximum amplitude. When the A4 window maximum amplitude of the confidence A4 event period is relatively small (e.g., less than or equal to 3.0 meters/second) ² ) The multiple or offset applied to the A3 window maximum amplitude for determining the early A4 sensing threshold may be relatively small (e.g., a multiple of 1.25 to 1.75 or 0.5 meters/second) ² To 1.0 m/s ² Offset of (c). When the A4 window maximum amplitude of the confidence A4 event period is relatively large (e.g., greater than 3.0 meters/second) ² ) The multiple or offset applied to the A3 window maximum amplitude for determining the early A4 sensing threshold may be relatively large (e.g., a multiple of 2.0 or 1.5 meters/second) ² To 2.0 m/s ² Offset of (c). It should be appreciated that more than two ranges of A4 window maximum amplitudes (with corresponding exemplary multiples or offsets applied to the A3 window maximum amplitudes for setting the early A4 sensing threshold amplitudes) as given herein may be stored in the memory 210. The multiple or offset used to set the early A4 sense threshold may be scaled over the range of A4 window maximum amplitudes according to a linear, stepwise or other relationship.

In still other examples, the control circuit 206 may determine a sum of the A3 window maximum amplitude and the A4 window maximum amplitude at block 1510. The control circuit 206 may set the early A4 sense threshold amplitude to be less than the sum. The early A4 sense threshold amplitude may be set toA percentage of the sum (e.g., 75% or 80%) but not less than the A3 window maximum amplitude. The early A4 sense threshold amplitude may be set to an offset less than the sum but not less than the A3 window maximum amplitude (e.g., 0.5 meters/second ² To 1.0 m/s ² )。

At block 1512, the control circuitry 206 may establish an A3 window end time. The control circuitry 206 may set the test threshold to, for example, 0.8 meters/second ² To 1.0 m/s ² . In some examples, the test threshold is set to a percentage of the late A4 sense threshold amplitude established at block 1508 or a percentage of the early A4 sense threshold amplitude established at block 1510. For example, the test threshold may be set between 60% and 90% or at 75% of the late A4 sense threshold amplitude established at block 1510. The control circuit 206 may determine that the most recent negative test threshold crossing during the A3 window of at least one confidence A4 event period. When more than one confidence A4 event period is identified at block 1506, the control circuitry 206 may determine the largest one of the most recent test threshold crossing times.

At block 1512, the control circuitry 206 may set an A3 window end time based on the determined most recent test threshold crossing time. The A3 window end time may be set to the latest test threshold crossing time plus an offset. The offset may be between 25 milliseconds and 100 seconds, and in one example is 50 seconds. The control circuit 206 may establish an operating range of A3 window end times that defines minimum and maximum A3 window end times to which the A3 window end times may be adjusted. The operating range of the A3 window end time may be established as ±50 milliseconds or another range of values around the established A3 window end time. The operating range may or may not be centered on the established A3 window end time.

At block 1514, the control circuitry 206 may switch to an atrial tracking ventricular pacing mode in which the established A4 sensing parameter is valid. In other examples, control circuitry 206 may switch to a temporary pacing mode prior to switching to a permanent atrial tracking ventricular pacing mode during which one or more A4 sensing parameters may be adjusted from the established starting value to an operating value that is subsequently validated when switching to the permanent atrial tracking ventricular pacing mode. Exemplary techniques for adjusting the start value of the A4 sense parameter to the operational value are generally described above, for example, in connection with fig. 16 and 24.

Referring again to fig. 19, in an example, when the ventricular cycle 1201 is identified as a confidence A4 event cycle as described above, the control circuit 206 may set the late A4 sense threshold amplitude based on the A4 window maximum amplitude 1220. For example, the control circuit 206 may determine the starting value of the late A4 sense threshold amplitude to be 75% of the maximum amplitude 1220. The control circuit 206 may establish a starting value of the early A4 sense threshold amplitude as being greater than the A3 window maximum amplitude 1218, e.g., as the A3 window maximum amplitude plus 1.0 meters/second ² To 2.0 m/s ² . The control circuit 206 may determine the starting value of the A3 window end time to be longer than the most recent negative crossing time 1240 of the test threshold 1212. The test threshold 1212 may be a nominally set test threshold (e.g., 0.9 meters/second) ² ) Or set to a percentage (e.g., 75%) of the late A4 sense threshold amplitude established based on the A4 window maximum amplitude 1220.

Many examples of establishing a start value for an A4 sense control parameter are disclosed herein. It should be appreciated that the various examples given herein in connection with the figures may be performed by the medical device processing circuitry in any combination for establishing one or more of an A4 sensing vector signal, an early A4 sensing threshold amplitude, a late A4 sensing threshold amplitude, an A3 window end time, and a minimum and/or maximum A3 window end time, as examples.

It should be understood that, depending on the example, certain acts or events of any of the methods described herein can be performed in a different order, may be added, combined, or omitted entirely (e.g., not all such acts or events are necessary for the practice of the method). Further, in some examples, acts or events may be performed concurrently, e.g., through multi-line processing, interrupt processing, or multiple processors, rather than sequentially. In addition, for purposes of clarity, although certain aspects of the disclosure are described as being performed by a single circuit or unit, it should be understood that the techniques of the disclosure may be performed by a combination of units or circuits associated with, for example, a medical device.

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

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

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

Claim (modification according to treaty 19)

1. A medical device, the medical device comprising:

a motion sensor configured to sense a motion signal; and

control circuitry coupled to the motion sensor to receive the motion signal and configured to:

Setting a sensing window during each of a plurality of ventricular cycles;

determining a characteristic of the motion signal sensed during the sensing window and after an end time of the sensing window;

determining that the characteristic of the motion signal sensed during the sensing window and after the end time of the sensing window meets an atrial event criterion indicative of a real-house event signal after the end time of the sensing window for at least a portion of the plurality of ventricular cycles;

determining a first characteristic of the motion signal sensed during each of the sensing windows associated with the portion of the plurality of ventricular cycles for which the motion signal sensed during the sensing window and after the end time of the sensing window meets the atrial event criteria;

sensing an atrial event signal from the motion signal according to an atrial event sensing control parameter; and

a sensed atrial event signal is generated in response to sensing the atrial event signal.

2. The medical device of claim 1, wherein the control circuit is configured to:

determining the first characteristic of the motion signal during each of the sensing windows by determining a first maximum amplitude of the motion signal during each of the sensing windows associated with the portion of the plurality of ventricular cycles;

setting the atrial event sensing parameter by setting an early atrial event sensing threshold amplitude based on the first maximum amplitude; and

an atrial contraction event is sensed from the motion signal in response to the motion signal crossing the early atrial event sensing threshold amplitude during the sensing window of a ventricular cycle.

3. The medical device of claim 1, wherein the control circuit is further configured to:

setting a first test threshold amplitude;

determining a crossing time of the first test threshold by the motion signal during each of the sensing windows associated with the portion of the plurality of ventricular cycles;

setting the atrial event sensing parameter based on the determined first characteristic by adjusting the end time of the sensing window based on the crossing time; and

Sensing the atrial event signal from the motion signal in response to the motion signal crossing one of:

a first sensing threshold amplitude prior to the end time of the sensing window; and

a second sensing threshold amplitude after the end time of the sensing window.

4. The medical device of claim 3, wherein the control circuit is further configured to:

determining a second test threshold amplitude based on the motion signal sensed during each of the plurality of ventricular cycles after the end time of the sensing window;

setting the sensing window according to the adjusted end time of the sensing window during each of a subsequent plurality of ventricular cycles;

determining a crossing time of the second test threshold amplitude during the sensing window of each of the next plurality of ventricular cycles; and

a second adjusted end time of the sensing window is determined from the determined crossing time of the second test threshold amplitude.

5. The medical device of claim 4, wherein the control circuit is further configured to:

Setting a range of the end time of the sensing window based on the second adjusted end time of the sensing window; and

the end time of the sensing window is adjusted within the range.

6. The medical device of claim 5, wherein the control circuit is further configured to:

setting the range according to a first offset when the second adjusted end time is less than a threshold end time; and

7. The medical device of claim 1, further comprising:

a pulse generator configured to generate pacing pulses in response to sensed atrial event signals.

8. The medical device of claim 1, further comprising

A sensing circuit comprising an R-wave detector for sensing R-waves from cardiac electrical signals; and

Wherein the control circuit is configured to: the sensing window is set during each of the plurality of ventricular cycles in response to one of a ventricular pacing pulse generated by the pulse generator and an R-wave sensed by the sensing circuit during the non-atrial tracking pacing mode.

9. The medical device of claim 1, wherein the control circuit is configured to determine that the characteristic of the motion signal sensed during the sensing window and after the end time of the sensing window meets the atrial event criteria by:

determining a maximum amplitude of the motion signal sensed after the end time of the sensing window; and

determining that the maximum amplitude is greater than a predetermined confidence atrial event threshold amplitude.

10. The medical device of claim 1, wherein the control circuit is configured to determine that the characteristic of the motion signal sensed during the sensing window and after the end time of the sensing window meets the atrial event criteria by:

determining a time interval from the end time of the sensing window to a maximum peak of the motion signal after the end time of the sensing window; and

The time interval is determined to be within a confidence atrial event time interval region.

11. The medical device of claim 1, wherein the control circuit is configured to determine that the characteristic of the motion signal sensed during the sensing window and after the end time of the sensing window meets the atrial event criteria by:

determining that the motion signal received after the end time of the sensing window crosses a confidence atrial event threshold amplitude.

12. The medical device of claim 1, wherein the control circuit is configured to determine that the characteristic of the motion signal sensed during the sensing window and after the end time of the sensing window meets the atrial event criteria by:

determining that the motion signal received after the sensing window end time crosses a confidence atrial event threshold amplitude during an atrial event time interval region.

13. The medical device of claim 1, wherein the control circuit is configured to determine that the characteristic of the motion signal sensed during the sensing window and after the end time of the sensing window meets the atrial event criteria by:

Determining a morphological feature of the motion signal sensed after the end time of the sensing window; and

determining that the morphology feature matches an atrial event morphology feature.

14. The medical device of claim 1, wherein the control circuit is configured to determine that the characteristic of the motion signal sensed during the sensing window and after the end time of the sensing window meets the atrial event criteria by:

determining a second maximum amplitude of the motion signal after the end time of the sensing window;

determining that the ratio meets an amplitude ratio requirement; and

determining that the characteristic of the motion signal sensed during the sensing window and after the end time of the sensing window meets the atrial event criteria in response to the ratio meeting the amplitude ratio requirement.

15. The medical device of claim 1, wherein the control circuit is further configured to: the atrial event sensing parameter is set based on the first characteristic by setting the atrial event sensing parameter based on a percentile of the first characteristic.

Claims

1. A medical device, the medical device comprising:

a motion sensor configured to sense a motion signal; and

setting a sensing window during each of a plurality of ventricular cycles;

determining that the motion signal sensed after an end time of the sensing window meets atrial event criteria for at least a portion of the plurality of ventricular cycles;

determining a first characteristic of the motion signal sensed during each of the sensing windows associated with the portion of the plurality of ventricular cycles for which the motion signal sensed after the end time of the sensing window meets the atrial event criteria;

2. The medical device of claim 1, wherein the control circuit is configured to:

setting a first test threshold amplitude;

a second sensing threshold amplitude after the end time of the sensing window.

the end time of the sensing window is adjusted within the range.

7. The medical device of claim 1, further comprising:

8. The medical device of claim 1, further comprising

9. The medical device of claim 1, wherein the control circuit is configured to determine that the motion signal sensed after the end time of the sensing window meets the atrial event criteria by:

10. The medical device of claim 1, wherein the control circuit is configured to determine that the motion signal sensed after the end time of the sensing window meets the atrial event criteria by:

11. The medical device of claim 1, wherein the control circuit is configured to determine that the motion signal sensed after the end time of the sensing window meets the atrial event criteria by:

12. The medical device of claim 1, wherein the control circuit is configured to determine that the motion signal sensed after the end time of the sensing window meets the atrial event criteria by:

13. The medical device of claim 1, wherein the control circuit is configured to determine that the motion signal sensed after the end time of the sensing window meets the atrial event criteria by:

14. The medical device of claim 1, wherein the control circuit is configured to determine that the motion signal sensed after the end time of the sensing window meets the atrial event criteria by:

determining that the ratio meets an amplitude ratio requirement; and

determining that the motion signal sensed after the end time of the sensing window meets the atrial event criteria in response to the ratio meeting the amplitude ratio requirement.