T-wave oversensing

• US 8,886,296 B2
• Filed: 10/14/2011
• Issued: 11/11/2014
• Est. Priority Date: 10/14/2011
• Status: Active Grant

• Current Assignee: Medtronic Incorporated (Medtronic Plc)
• Original Assignee: Medtronic Incorporated (Medtronic Plc)
• Inventors: Patel, Amisha S.
• Primary Examiner(s): Kahelin, Michael
• Assistant Examiner(s): FAIRCHILD, MALLIKA DIPAYAN
• Application Number:
  US13/274,010
• Publication Number:
  US 20130096445A1
• Time in Patent Office: 1,124 Days
• Field of Search: 607/9, 607/14, 607/30
• US Class Current: 600/509
• CPC Class Codes: A61B 5/35 by template matching
• CPC Class Codes: A61B 5/363 Detecting tachycardia or bradycardia
• 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 1/3702 Physiological parameters A61N1/365 takes precedence; evoked response A61N1/371
• CPC Class Codes: G16H 40/67 for remote operation
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16H 50/70 for mining of medical data, e.g. analysing previous cases of other patients

Abstract

The present disclosure is directed to the classification of cardiac episodes using an algorithm. In various examples, an episode classification algorithm evaluates electrogram signal data to determine whether T-wave oversensing has occurred. The T-wave oversensing analysis may include, for example, identifying beat runs within the cardiac episode whether the beats within the run have at least one characteristic that alternates beat to be or clustering beats within the cardiac episode based on beat to beat interval length. The T-wave oversensing determination may be based on probabilistic analysis in some examples.

Claims

1. A method for determining whether T-wave over-sensing (TWOS) occurred during a cardiac episode comprising a plurality of sensed beats, the method comprising:

identifying at least one beat run of at least a predetermined number of consecutive ones of the beats during the cardiac episode, wherein each of the beats within the run have at least one characteristic that alternates from beat to beat;

clustering the beats into two or more clusters based on beat-to-beat interval length; and

determining, based on at least one of the runs and the clusters, whether TWOS occurred during the cardiac episode.

2. The method of claim 1, further comprising determining whether the identified beat run comprises alternating interval lengths for at least a portion of the beat run.

3. The method of claim 1, further comprising determining whether the identified beat run comprises alternating widths for at least a portion of the beat run.

4. The method of claim 1, further comprising determining whether the identified beat run comprises alternating morphologies for at least a portion of the beat run.

5. The method of claim 1, further comprising determining whether the identified beat run comprises alternating amplitude intervals for at least a portion of the beat run.

6. The method of claim 1, wherein the determination of TWOS is a probabilistic determination.

7. The method of claim 6, wherein evidence for TWOS comprises

interval lengths clustered into two distinct bands;

at least one beat run with alternating intervals for the length of the beat run;

a run of alternating intervals greater than or equal to five intervals and a total count for alternating interval lengths is greater than or equal to three; and

a run of alternating widths is greater than four intervals;

wherein at least interval lengths clustered into two distinct bands and a run of alternating widths greater than four are given different weights.

8. The method of claim 1, further comprising determining whether there is a single transition between two distinct clusters, and, in response to a single transition between two distinct clusters adding weight to evidence of no TWOS.

9. The method of claim 1, wherein clustering the beats into two or more clusters comprises clustering the beats into a first cluster, a second cluster, and a third cluster based on beat-to-beat interval length, the method further comprising:

determining a representative interval length for each of the three clusters, wherein the first cluster interval length is longer than the second cluster interval length and the first cluster interval length is longer than the third cluster interval length;

summing the second cluster interval length and the third cluster interval length;

comparing the sum of the second cluster interval length and the third cluster interval length to the first cluster interval length; and

adding weight to evidence of TWOS in response to the sum of the second cluster interval length and the third cluster interval length being approximately equal to the first cluster interval length.

10. The method of claim 9, wherein the first cluster interval length is a mean interval length for each of the beats in the first cluster.

11. The method of claim 1, wherein each of the clusters is separated by a zone of interval lengths in which there is no more than one interval.

12. The method of claim 11, wherein the dead zone is between approximately 5 ms and 25 ms.

13. The method of claim 1, further comprising determining a count of a total number of alternating beat runs within the cardiac episode, and in response to the count being above a predetermined threshold, adding weight to evidence of TWOS.

14. The method of claim 1 further including:

determining mean values of each of the clusters;

determining a difference between the mean values; and

determining whether the at least two clusters are distinct clusters based at least in part on the difference between the mean values of each of the two clusters.

15. The method of claim 1, further including determining a number of alternating interval lengths for the at least one beat run.

16. A system for determining whether T-wave over-sensing (TWOS) occurred during a cardiac episode comprising a plurality of sensed beats, the system comprising:

a processor configured to;

identify at least one beat run of at least a predetermined number of consecutive ones of the beats during the cardiac episode, wherein each of the beats within the run have at least one characteristic that alternates from beat to beat;

cluster the beats into two or more clusters based on beat-to-beat interval length; and

determine, based on at least one of the runs and the clusters, whether TWOS occurred during the cardiac episode.

17. The system of claim 16, wherein the processor is further configured to determine whether the identified beat run comprises alternating interval lengths for at least a portion of the beat run.

18. The system of claim 16, wherein the processor is further configured to determine whether identified beat run comprises alternating widths for at least a portion of the beat run.

19. The system of claim 16, wherein the processor is further configured to determine whether the identified beat run comprises alternating morphologies for at least a portion of the beat run.

20. The system of claim 16, wherein the processor is further configured to determine whether the identified beat run comprises alternating amplitude intervals for at least a portion of the beat run.

21. The system of claim 16, wherein the processor is further configured to perform a probabilistic determination of whether TWOS occurred.

22. The system of claim 21, wherein evidence of TWOS comprises

interval lengths clustered into two distinct bands;

at least one beat run with alternating intervals for the length of the beat run;

a run of alternating intervals greater than or equal to five intervals and a total count for alternating interval lengths is greater than or equal to three; and

a run of alternating widths is greater than four intervals; and

wherein the processor is further configured to give different weights to at least interval lengths clustered into two distinct bands and a run of alternating widths greater than four.

23. The system of claim 16, wherein the processor is further configured to determine whether there is a single transition between two distinct clusters, and in response to a determination of a single transition, the processor is configured add weight to evidence of no TWOS.

24. The system of claim 16, wherein the processor is further configured to:

cluster the beats into a first cluster, a second cluster, and a third cluster based on beat-to-beat interval length;

determine an interval length for each of the three clusters, wherein the first cluster e interval length is longer than the second cluster interval length and the first cluster interval length is longer than the third cluster average interval length;

sum the second cluster interval length and the third cluster interval length;

compare the sum of the second cluster interval length and the third cluster interval length to the first cluster interval length; and

add weight to evidence of TWOS in response to the sum of the second cluster interval length and the third cluster interval length being approximately equal to the first cluster interval length.

25. The system of claim 24, wherein the first cluster interval length is a mean interval length for each of the beats in the first cluster.

26. The system of claim 16, wherein the processor is further configured to separate each of the clusters by a dead zone.

27. The system of claim 16, wherein the processor is further configured to determine a count of a total number of alternating beat runs within the cardiac episode, and in response to the count being above a predetermined threshold, add weight to evidence of TWOS.

28. The system of claim 16, wherein the processor is further configured to:

determine the mean values for each of the clusters;

determine a difference between the mean values of each of the two clusters; and

determine whether the at least two clusters are distinct clusters based at least in part on the difference between the mean values of each of the two clusters.

29. The system of claim 16, wherein the processor is further configured to determine a number of alternating intervals for the at least one beat run.

30. A non-transitory computer-readable medium comprising instructions for causing a programmable processor to determine whether T-wave over-sensing (TWOS) occurred during a cardiac episode comprising a plurality of sensed beats, the instructions causing the programmable processor to:

identify at least one beat run of at least a predetermined number of consecutive ones of the beats during the cardiac episode, wherein each of the beats within the run have at least one characteristic that alternates from beat to beat;

cluster the beats into two or more clusters based on beat-to-beat interval length; and

determine, based on at least one of the runs and the clusters, whether TWOS occurred during the cardiac episode.

31. A system for determining whether T-wave over-sensing (TWOS) occurred during a cardiac episode comprising a plurality of sensed beats, the system comprising:

means for identifying at least one beat run of at least a predetermined number of consecutive ones of the beats during the cardiac episode, wherein each of the beats within the run have at least one characteristic that alternates from beat to beat;

means clustering the beats into two or more clusters based on beat-to-beat interval length; and

means determining, based on at least one of the runs and the clusters, whether TWOS occurred during the cardiac episode.

32. A system for determining whether T-wave over-sensing (TWOS) occurred during a cardiac episode comprising a plurality of sensed beats, the system comprising:

a processor configured to;

cluster the beats into two or more clusters based on beat-to-beat interval length; and

determine, based on the clusters, whether TWOS occurred during the cardiac episode.