Arrhythmia classification and therapy selection

• US 7,894,893 B2
• Filed: 09/30/2004
• Issued: 02/22/2011
• Est. Priority Date: 09/30/2004
• Status: Expired due to Fees

• Current Assignee: Cardiac Pacemakers Incorporated (Boston Scientific Corporation)
• Original Assignee: Cardiac Pacemakers Incorporated (Boston Scientific Corporation)
• Inventors: Bocek, Joseph, Kim, Jaeho, Lovett, Eric G.
• Primary Examiner(s): Bertram; Eric D.
• Application Number:
  US10/955,831
• Publication Number:
  US 20060074331A1
• Time in Patent Office: 2,336 Days
• Field of Search: 600/515, 607/4
• US Class Current: 607/4
• CPC Class Codes: A61B 5/35 by template matching
• 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: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

Different types of cardiac arrhythmia are classified based on the morphology of the arrhythmic beats. Cardiac beats associated with an arrhythmic episode are compared to a plurality of representative beat morphologies, each representative beat morphology characterizing a type of arrhythmia of the heart. An arrhythmic episode may be classified as a particular type of arrhythmia if the morphology of the arrhythmic cardiac beats matches a representative beat morphology characterizing the particular type of arrhythmia. An appropriate therapy for the particular type of arrhythmia may be selected based on the arrhythmia classification. A particular type of arrhythmia may be associated with one or more therapies used to treat the arrhythmia. The therapy used to treat the arrhythmia may comprise a therapy identified as a previously successful therapy.

Claims

1. A method of operating an implantable cardiac system for discriminating between types of arrhythmia affecting one or more heart chambers, the method comprising:

detecting cardiac beats associated with an arrhythmic episode;

comparing a morphology of each cardiac beat to a plurality of representative beat morphologies stored in memory and respectively associated with a plurality of types of monomorphic arrhythmia;

classifying the arrhythmic episode as a particular type of the plurality of types of monomorphic arrhythmia if at least x out of y beat morphologies of the arrhythmic episode are consistent with a representative beat morphology of the particular type of monomorphic arrhythmia based on the comparison, wherein x is a number greater than 1 and y is a number greater than x;

determining that the arrhythmic episode is an unknown type of arrhythmia if the cardiac beats of the arrhythmic episode have been compared to each of the plurality of representative beat morphologies without being matched to any of the plurality of representative beat morphologies based on the comparison;

determining whether or not a representative beat morphology can be acquired for the arrhythmic episode if the arrhythmic episode is the unknown type of arrhythmia;

classifying the arrhythmic episode as a new type of arrhythmia if the arrhythmic episode is determined to be the unknown type of arrhythmia and if the representative beat morphology for the arrhythmic episode can be acquired for the arrhythmic episode; and

classifying the arrhythmic episode as an uncharacterizable polymorphic arrhythmia if the arrhythmic episode is determined to be the unknown type of arrhythmia and if the representative beat morphology for the arrhythmic episode cannot be acquired, wherein detecting, comparing, determining, and classifying is performed by the implantable cardiac system.

2. The method of claim 1, wherein comparing the morphology of each cardiac beat to the plurality of representative beat morphologies associated with the plurality of types of arrhythmia comprises comparing the morphology of each cardiac beat to the plurality of representative beat morphologies in an order, the order based on frequency of detection of the arrhythmia types that are respectively associated with the representative beat morphologies.

3. The method of claim 1, further comprising adjusting x as the number of beats of the arrhythmia episode used to classify the arrhythmic episode.

4. The method of claim 1, further comprising selecting between a plurality of therapies for delivery to treat the arrhythmic episode, wherein a first therapy is selected if the arrhythmic episode is classified as one of the plurality of monomorphic arrhythmia types and a different therapy is selected if the arrhythmic episode is classified as polymorphic.

5. The method of claim 1, wherein:

comparing the morphology of each cardiac beat to the plurality of representative beat morphologies comprises comparing the morphology of each cardiac beat to a plurality of representative beat morphology templates; and

classifying the arrhythmic episode as the particular type of arrhythmia comprises classifying the arrhythmic episode as the particular type of arrhythmia if the morphologies of the cardiac beats are consistent with a representative beat morphology template associated with the particular type of arrhythmia.

6. The method of claim 1, wherein:

comparing the morphology of each cardiac beat to the plurality of representative beat morphologies comprises calculating a feature correlation coefficient for each cardiac beat; and

classifying the arrhythmic episode as a particular type of the plurality of monomorphic arrhythmia types comprises comparing the feature correlation coefficient to a threshold to determine whether each cardiac beat is consistent.

7. The method of claim 6, further comprising modifying the threshold to change sensitivity of arrhythmia recognition.

8. The method of claim 1, wherein x is at least 3.

9. A cardiac arrhythmia classification system, comprising:

a sensor system comprising electrodes for electrically coupling to a heart, the sensor system configured to detect a plurality of cardiac beats associated with an arrhythmic episode of the heart;

memory; and

an arrhythmia classification processor coupled to the sensor system, the arrhythmia classification processor configured to compare a morphology of each cardiac beat of the arrhythmic episode to a plurality of representative beat morphologies stored in memory and respectively associated with a plurality of types of arrhythmia and classify the arrhythmic episode as a particular type of arrhythmia if at least x out of y beat morphologies of the arrhythmic episode are consistent with a representative beat morphology of a particular type of arrhythmia based on the comparison, the arrhythmia classification processor further configured to identify the arrhythmic episode as an unknown type of arrhythmia if the cardiac beats of the arrhythmic episode have been compared to each of the plurality of representative beat morphologies without being matched to any of the plurality of the representative beat morphologies based on the comparison and determine whether or not the beat-to-beat morphology of the cardiac beats of the arrhythmic episode are sufficiently consistent to allow a representative beat morphology for the arrhythmic episode to be acquired, the arrhythmia classification processor further configured to classify the arrhythmic episode as a new type of arrhythmia if the arrhythmic episode is identified as the unknown type of arrhythmia and if the representative beat morphology for the arrhythmic episode can be acquired and to classify the arrhythmic episode as an uncharacterizable polymorphic arrhythmia if the arrhythmic episode is the unknown type of arrhythmia and if the representative beat morphology cannot be acquired, wherein the arrhythmia classification system is implantable.

10. The system of claim 9, wherein the arrhythmia classification processor is configured to select between a plurality of therapies for delivery by the cardiac arrhythmia classification system to treat the arrhythmic episode, wherein a first therapy is selected if the arrhythmic episode is classified as one of the plurality of arrhythmia types and a different therapy is selected if the arrhythmic episode is classified as polymorphic.

11. The system of claim 9, wherein the arrhythmia classification processor is configured to compare the morphology of each cardiac beat to the plurality of representative beat morphology templates and classify the arrhythmic episode as the particular type of arrhythmia if the morphologies of the cardiac beats are consistent with a representative beat morphology template associated with the particular type of arrhythmia.

12. The system of claim 9, wherein the arrhythmia classification processor is configured to compare the morphology of each cardiac beat to the plurality of representative beat morphologies in a selected order, the order based on frequency of detection of the arrhythmia types that are respectively associated with the representative beat morphologies.

13. The system of claim 9, wherein the arrhythmia classification processor is configured to compare the morphology of each cardiac beat of the arrhythmic episode to the plurality of representative beat morphologies by calculating a correlation metric for each cardiac beat and classify the arrhythmic episode as a particular type of the plurality of arrhythmia types by comparing the correlation metric to a threshold to determine whether each cardiac beat is consistent.

14. The system of claim 9, wherein x is at least 3.

15. A system for classifying cardiac beats, comprising:

means for detecting cardiac beats associated with an arrhythmic episode;

means for comparing a morphology of each cardiac beat to plurality of representative beat morphologies respectively associated with plurality of types of arrhythmia;

means for classifying the arrhythmic episode as a particular type of the plurality of arrhythmia types if at least x out of y beat morphologies of the arrhythmic episode are consistent with a representative beat morphology of a particular type of arrhythmia based on the comparison, wherein x is a number greater than 1 and y is a number greater than x;

means for determining that the arrhythmic episode is an unknown type of arrhythmia if the cardiac beats of the arrhythmic episode have been compared to each of the plurality of representative beat morphologies without finding correspondence to any of the plurality of representative beat morphologies; and

means for classifying the arrhythmic episode as an uncharacterizable polymorphic arrhythmia if the arrhythmic episode is the unknown type of arrhythmia and if the representative beat morphology for the arrhythmic episode cannot be acquired.

16. The system of claim 15, further comprising:

means for comparing the morphology of each cardiac beat to the plurality of representative beat morphologies by calculating a feature correlation coefficient for each cardiac beat; and

means for classifying the arrhythmic episode as a particular type of the plurality of arrhythmia types by comparing the feature correlation coefficient to a threshold to determine whether each cardiac beat is consistent.

17. The system of claim 15, wherein x is at least 3.