Systems and methods for arranging and labeling cardiac episodes

• US 8,712,507 B2
• Filed: 12/21/2006
• Issued: 04/29/2014
• Est. Priority Date: 09/14/2006
• Status: Active Grant

• Current Assignee: Cardiac Pacemakers Incorporated (Boston Scientific Corporation)
• Original Assignee: Cardiac Pacemakers Incorporated (Boston Scientific Corporation)
• Inventors: Cazares, Shelley, McCabe, Aaron, Sathaye, Alok, Li, Dan
• Primary Examiner(s): Koharski, Christopher D
• Assistant Examiner(s): So, Elizabeth K
• Application Number:
  US11/643,220
• Publication Number:
  US 20080071182A1
• Time in Patent Office: 2,686 Days
• Field of Search: 600508-528, 128/897, 128/898, 128/899, 128/905, 128/920, 128/922, 128/923
• US Class Current: 600/509
• CPC Class Codes: A61B 5/363 Detecting tachycardia or bradycardia
• CPC Class Codes: A61B 5/726 using Wavelet transforms

Abstract

Methods and systems for arranging and labeling cardiac episodes based on acquired cardiac episode data are described. Cardiac episodes are algorithmically arranged based on one or more discriminating features of the episode data. A user is presented with at least one episode selected from the arrangement of cardiac episodes. The user inputs a label that characterizes the selected episode. The label is algorithmically appended to the data of the selected episode and to other episodes of the arrangement of cardiac episodes based on the discriminating features.

Claims

1. A method operable on data acquired by an implantable cardiac device and associated with a plurality of cardiac episodes, the method comprising:

algorithmically grouping, with at least one of a number of data processors, the cardiac episodes based on multiple discriminating features of the episodes, each cardiac episode including a plurality of consecutive cardiac beats, wherein the discriminating features include at least one feature from a cardiac signal and at least one feature from a non-cardiac signal;

algorithmically selecting, with at least one of the number of data processors, at least one episode from an arrangement of cardiac episodes;

graphically displaying an electrogram (EGM) signal of the at least one episode to a user through a patient-external user interface;

after displaying the EGM signal, receiving an input from a user through the patient-external user interface, the input from the user identifying a user-selected label that characterizes the EGM signal of the selected episode;

algorithmically appending, with at least one of the number of data processors, the user-selected label to data of the selected episode; and

algorithmically appending, with at least one of the number of data processors, the user-selected label to data of non-selected episodes of the arrangement of cardiac episodes.

2. The method of claim 1, wherein grouping the cardiac episodes comprises grouping the episodes based on similarity of the one or more discriminating features.

3. The method of claim 1, wherein a number of groups is determined by a user.

4. The method of claim 1, wherein a number of groups is determined algorithmically.

5. The method of claim 1, wherein grouping the cardiac episodes comprises ordering the episodes based on the one or more discriminating features and grouping the ordered episodes.

6. The method of claim 5, wherein grouping the ordered episodes comprises:

identifying, by the user, one or more ordered episodes defining boundary episodes between groups; and

algorithmically grouping, with at least one of the number of data processors, the ordered episodes based on the boundary episodes.

7. The method of claim 1, wherein receiving the input from the user comprises:

algorithmically selecting, with at least one of the number of data processors, a label for the selected episode; and

confirming, by the user, the algorithmically selected label or selecting, by the user, a label different from the algorithmically selected label.

8. The method of claim 1, further comprising receiving additional episode data acquired by the implantable cardiac device and repeating the algorithmically grouping and algorithmically appending processes using previously acquired episode data and the additional episode data.

9. The method of claim 8, wherein repeating the algorithmically grouping and algorithmically appending processes is initiated algorithmically or manually.

10. The method of claim 1, wherein the one or more discriminating features of the episodes characterize a conduction pattern of an arrhythmic episode or a type of arrhythmia.

11. The method of claim 1, wherein the label characterizes at least some of the cardiac episodes as tachyarrhythmia episodes that are ventricular in origin or supraventricular in origin.

12. The method of claim 1, wherein the label characterizes at least some of the cardiac episodes as tachyarrhythmia episodes that are hemodynamically stable or hemodynamically unstable.

13. The method of claim 1, wherein the label characterizes at least some of the arrhythmia episodes as tachyarrhythmia episodes that should be treated with anti-tachyarrhythmia therapy or those that should not be treated with anti-tachyarrhythmia therapy.

14. The method of claim 1, wherein at least one of algorithmically grouping, receiving the user-selected label, and algorithmically appending the label is implementable by the implantable cardiac device.

15. The method of claim 1, wherein at least one of algorithmically grouping, presenting the at least one episode to the user, receiving the user-selected label, and algorithmically appending is implemented by a processor external of the implantable cardiac device.

16. The method of claim 1, further comprising:

determining a morphological organization of the cardiac episodes; and

algorithmically grouping, with at least one of the number of data processors, the cardiac episodes based on the morphological organization of the cardiac episodes.

17. The method of claim 1, wherein algorithmically grouping, with at least one of the number of data processors, the cardiac episodes further comprises algorithmically grouping the cardiac episodes using a fuzzy clustering algorithm.

18. The method of claim 1, further comprising:

determining an angle of projection of the cardiac signal; and

algorithmically grouping, with at least one of the number of data processors, the cardiac episodes based on the angle of projection of the cardiac signal.

19. A system, comprising:

cardiac electrodes and sensing circuitry configured to acquire data associated with cardiac episodes, each cardiac episode including a plurality of consecutive cardiac beats;

a user interface configured to graphically display an electrogram (EGM) signal of a selected cardiac episode to a user and to subsequently receive an input from the user that identifies a user-selected label for the episode; and

a data processor configured to group the cardiac episode data based on multiple discriminating features of the episodes and to append the user-selected label to the data of the selected cardiac episode and to data of non-selected cardiac episodes having features that are similar to the discriminating feature, wherein the discriminating features include features from a cardiac signal and features from at least one non-cardiac signal.

20. The system of claim 19, further comprising at least one sensor other than the cardiac electrodes, the at least one sensor configured to acquire physiological or non-physiological data associated with the cardiac episodes.

21. The system of claim 19, wherein at least one of the cardiac electrodes and sensing circuitry and data processor are components of an implantable cardiac device.

22. The system of claim 19, wherein:

the cardiac electrodes and sensing circuitry are components of an implantable device; and

the user interface and data processor are components of a non-implantable device.

23. The system of claim 19, wherein the cardiac episodes are tachyarrhythmia episodes.

24. The system of claim 19, wherein the labeled episode data is used to form a morphology template for discriminating cardiac rhythms.

25. The system of claim 19, wherein grouping the cardiac episode data comprises ordering the cardiac episode data and grouping the ordered episodes.

26. A system comprising:

a number of data processors for;

algorithmically grouping cardiac episodes based on one or more discriminating features of the episodes, each cardiac episode including a plurality of consecutive cardiac beats, wherein the discriminating features include features from a cardiac signal and features from at least one non-cardiac signal;

algorithmically selecting at least one episode from an arrangement of cardiac episodes;

graphically displaying an electrogram (EGM) signal of at least one episode to a user through a patient-external user interface;

receiving, after the EGM signal of the at least one selected episode is displayed, an input from a user through the patient-external user interface, the input from the user identifying a user-selected label that characterizes the EGM signal of the at least one selected episode;

algorithmically appending the user-selected label to data of the at least one selected episode; and

algorithmically appending the user-selected label to data of non-selected episodes of the arrangement of cardiac episodes.

27. The system of claim 26, further comprising grouping the episodes based on similarity of the one or more discriminating features.

28. The system of claim 26, further comprising ordering the episodes based on the one or more discriminating features.