Selection of cardiac signal features detected in multiple classification intervals for cardiac pacing response classification

• US 7,711,424 B2
• Filed: 02/14/2008
• Issued: 05/04/2010
• Est. Priority Date: 04/28/2005
• Status: Active Grant

• Current Assignee: Cardiac Pacemakers Incorporated (Boston Scientific Corporation)
• Original Assignee: Cardiac Pacemakers Incorporated (Boston Scientific Corporation)
• Inventors: Meyer, Scott A., Dong, Yanting, Stalsberg, Kevin John
• Primary Examiner(s): Layno; Carl H.
• Assistant Examiner(s): Heller; Tammie K
• Application Number:
  US12/070,004
• Publication Number:
  US 20080140145A1
• Time in Patent Office: 810 Days
• Field of Search: 607/27, 607/28, 600/510
• US Class Current: 607/27
• CPC Class Codes: A61B 5/349 Detecting specific parameters of the electrocardiograph cycle
• CPC Class Codes: A61B 5/7264 Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00
• CPC Class Codes: A61N 1/371 Capture, i.e. successful stimulation
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

Cardiac devices and methods involve the detection of cardiac signals features in adjacent classification intervals. Portions of the cardiac signal features detected in adjacent classification intervals are associated and are used to classify the cardiac response to a pacing pulse. Associating the portions of the cardiac signal features may be based on expected signal morphology.

Claims

1. A method of operating a cardiac rhythm management device for classifying a cardiac response to a pacing pulse, comprising:

timing a plurality of cardiac pacing response classification intervals following delivery of the pacing pulse;

sensing a cardiac signal in at least two classification intervals of the plurality of classification intervals;

detecting at least one cardiac signal feature that occurs within a boundary region, the boundary region at least partially overlapping the at least two classification intervals;

classifying the cardiac response to the pacing pulse based on the detected cardiac signal feature; and

delivering pacing therapy based at least in part on classification of the cardiac response to the pacing pulse.

2. The method of claim 1, wherein the at least two classification intervals have a common boundary.

3. The method of claim 1, wherein the at least two classification intervals are separated by a time interval.

4. The method of claim 1, wherein classifying the cardiac response to the pacing pulse comprises:

associating portions of the at least one cardiac signal feature that occur within the boundary region; and

classifying the cardiac response to the pacing pulse based on the associated portions.

5. The method of claim 4, wherein associating the portions of the cardiac signal feature comprises associating the portions based on expected signal morphology.

6. The method of claim 5, wherein the expected signal morphology includes an expectation that a negative peak occurs before a positive peak.

7. The method of claim 4, wherein associating the portions of the cardiac signal feature comprises associating based on an expected time of the cardiac signal feature.

8. The method of claim 1, wherein the cardiac signal feature comprises a signal peak.

9. The method of claim 1, wherein the at least one cardiac signal feature comprises one or both of a positive cardiac signal peak and a negative cardiac signal peak.

10. The method of claim 1, wherein:

detecting the at least one cardiac signal feature in the boundary region comprises;

detecting a first portion of a cardiac signal peak in a first classification interval;

detecting a second portion of the cardiac signal peak in a second classification interval; and

associating the first and the second portions of the cardiac signal peak; and

classifying the cardiac response to the pacing pulse comprises classifying the cardiac response to the pacing pulse based on the associated portions of the cardiac signal peak.

11. A system for characterizing a cardiac response to pacing, comprising:

a sensing system configured to sense cardiac signals following pacing pulses delivered to a heart;

a processor coupled to the sensing system, the processor configured to time a plurality of classification intervals following delivery of a pacing pulse, detect a cardiac signal feature that occurs within a boundary region, the boundary region at least partially overlapping at least two classification intervals of the plurality of classification intervals, and classify the cardiac response to the pacing pulse based on the cardiac signal feature; and

a pacing circuit configured to deliver pacing therapy based at least in part on classification of the cardiac pacing response.

12. The system of claim 11, wherein the processor is configured to associate portions of the cardiac signal feature detected within the boundary region.

13. The system of claim 12, wherein the portions of the cardiac signal feature comprise portions of a negative signal peak or portions of a positive signal peak.

14. The system of claim 11, wherein the processor is configured to evaluate the cardiac signal feature across the plurality of classification intervals to determine feature information associated with the cardiac signal feature.

15. The system of claim 11, wherein the processor is configured to associate portions of the cardiac signal feature that occur within the two classification intervals if timing of the portions satisfies an expected signal morphology.

16. The system of claim 11, wherein the at least two classification intervals have a common temporal border.

17. The system of claim 11, wherein the at least two classification intervals are temporally non-continuous.

18. A system for characterizing a cardiac response to pacing, comprising:

means for sensing a cardiac signal in at least two classification intervals;

means for classifying a cardiac response to a pacing pulse, including means for detecting at least one cardiac signal feature that occurs within a boundary region that at least partially overlaps the at least two classification intervals and means for classifying the cardiac response to the pacing pulse based on the detected cardiac signal feature; and

circuitry configured to deliver pacing therapy based at least in part on classification of the cardiac response to the pacing pulse.

19. The system of claim 18, wherein the means for classifying associates portions of one or more cardiac signal peaks that fall within the boundary region based on expected signal morphology.

20. The system of claim 18, wherein the at least one cardiac signal feature comprises one or more cardiac signal peaks.