Method and apparatus for classifying heartbeats in an ECG waveform

• US 5,817,027 A
• Filed: 08/27/1997
• Issued: 10/06/1998
• Est. Priority Date: 11/29/1995
• Status: Expired due to Fees

• Current Assignee: Koninklijke Philips N.V.
• Original Assignee: Hewlett-Packard Company (HP Inc.)
• Inventors: Arand, Patricia A., Post, William L.
• Primary Examiner(s): Kamm, William E.
• Assistant Examiner(s): EVANISKO, GEORGE ROBERT
• Application Number:
  US08/919,876
• Time in Patent Office: 405 Days
• Field of Search: 600/509, 600/515, 600/517, 600/508
• US Class Current: 600/515
• CPC Class Codes: A61B 5/0245 by using sensing means generating electric signals, i.e. ECG signals
• CPC Class Codes: A61B 5/35 by template matching
• CPC Class Codes: A61B 5/352 Detecting R peaks, e.g. for synchronising diagnostic apparatus; Estimating R-R interval
• CPC Class Codes: A61B 5/7203 for noise prevention, reduction or removal
• 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

A method and apparatus for classifying heartbeats obtains ECG data from a plurality of ECG waveforms, which are in turn obtained from signals received from a plurality of ECG electrodes. QRS detection logic detects heartbeats in the ECG data. Classification logic classifies the detected heartbeats into categories based on shape and/or timing. This classification is done by comparing each heartbeat against a group of templates corresponding to one or more heartbeat classifications. The templates are updated to track changes in the morphology of the heartbeats. The heartbeat classification is displayed to the cardiologist or medical professional for diagnosis of the condition of the patient'"'"'s heart.

Claims

1. A method of classifying a plurality of heartbeats in an EOG waveform for a patient undergoing a stress test, said ECG waveform obtained from a plurality of ECG electrodes, said method comprising the steps of:

performing said stress test;

receiving heartbeat data from said stress test;

comparing a first heartbeat from said plurality of heartbeats with a first template stored in memory or storage, said first template corresponding to a first heartbeat classification;

determining if said first heartbeat matches said first template, responsive to said comparing step; and

if said first heartbeat matches said first template,

averaging said first heartbeat into said first template;

classifying said first heartbeat as said first heartbeat classification;

storing said first heartbeat classification for subsequent diagnosis;

repeating said comparing, determining, averaging, classifying, and storing steps for the remainder of the plurality of heartbeats in the EGG waveform determined to match the first template, wherein said repeating step continuously updates the first template to track changes in the morphology of the plurality of heartbeats while the patient is undergoing the stress test.

2. The method of claim 1, further comprising the step of:

displaying an indicia of said first heartbeat classification on a display.

3. The method of claim 1, wherein said averaging step is weighted in favor of said first template.

4. The method of claim 1, further comprising the steps of:

if said first heartbeat does not match said first template, comparing said first heartbeat with a second template stored in memory or storage corresponding to a second heartbeat classification;

determining if said first heartbeat matches said second template, responsive to said comparing step; and

if said first heartbeat matches said second template, classifying said first heartbeat as said second heartbeat classification.

5. The method of claim 4, further comprising the step of:

averaging said first heartbeat into said second template.

6. The method of claim 5, wherein said averaging step is weighted in favor of said second template.

7. The method of claim 4, further comprising the steps of:

if said first heartbeat does not match said second template, creating a third template for said first heartbeat; and

classifying said third template as a third heartbeat classification.

8. The method of claim 4, further comprising the steps of:

if said first heartbeat does not match said first template or said second template, determining whether said first template or said second template was last updated the longest time ago;

creating a third template for said first heartbeat;

overwriting said second template with said third template, responsive to said determining step determining that said second template was updated the longest time ago; and

classifying said third template as a third heartbeat classification.

9. The method of claim 1, wherein said first heartbeat matches said first classification and said first classification is a dominant/supraventricular ectopic classification, said method further comprising the steps of:

checking to see if said first heartbeat was early;

if said checking step determined that said first heartbeat was early, classifying said first heartbeat as a supraventricular ectopic heartbeat; and

if said checking step determined that said first heartbeat was not early, classifying said first heartbeat as a dominant heartbeat.

10. The method of claim 1 wherein said first heartbeat matches said first classification and said first classification is a ventricular ectopic classification.

11. The method of claim 1 wherein said first heartbeat matches said first classification and said first classification is a questionable classification.

12. A method of classifying a plurality of heartbeats in an ECG waveform for a patient undergoing a stress test, said EGG waveform obtained from a plurality of ECG electrodes, said method comprising the steps of:

performing said stress test;

receiving heartbeat data from said stress test;

comparing a first heartbeat from said plurality of heartbeats with a first template stored in memory or storage, said first template corresponding to a first heartbeat classification, said comparing step further comprising the steps of;

deriving an activity function from said first heartbeat;

sliding said activity function across said first template;

while performing said sliding step, computing a minimum area difference between said activity function and said first template;

determining if said first heartbeat matches said first template, responsive to said comparing step, said determining step further comprising the steps of;

analyzing said minimum area difference to see if it exceeds a predetermined threshold;

if said minimum area difference does not exceed a predetermined threshold, concluding that said first heartbeat matches said first template; and

if said minimum area difference does exceed a predetermined threshold, concluding that said first heartbeat does not match said first template;

if said first heartbeat matches said first template,

averaging said first heartbeat into said first template;

classifying said first heartbeat as said first heartbeat classification;

storing said first heartbeat classification for subsequent diagnosis; and

repeating said comparing, determining, averaging, classifying, and storing steps for the remainder of the plurality of heartbeats in the ECG waveform determined to match the first template, wherein said repeating step continuously updates the first template to track changes in the morphology of the plurality of heartbeats while the patient is undergoing the stress test.

13. A medical device for classifying a plurality of heartbeats in an ECG waveform for a patient undergoing a stress test, said ECG waveform obtained from a plurality of ECG electrodes, said medical device comprising:

a computing unit having a processor, memory and storage, said computing unit further comprising heartbeat classification logic, said heartbeat classification logic further comprising;

means for comparing a first heartbeat from said plurality of heartbeats with a first template stored in memory or storage, said first template corresponding to a first heartbeat classification;

means for determining if said first heartbeat matches said first template, responsive to said comparing means;

if said first heartbeat matches said first template,

means for averaging said first heartbeat into said first template;

means for classifying said first heartbeat as said first heartbeat classification;

means for storing said first heartbeat classification for subsequent diagnosis; and

means for repeating said comparing, determining, averaging, classifying, and storing means for the remainder of the plurality of heartbeats in the ECG waveform determined to match the first template, wherein said repeating means continuously updates the first template to track changes in the morphology of the plurality of heartbeats while the patient is undergoing the stress test.

14. The medical device of claim 13, further comprising:

a display for displaying an indicia of said first heartbeat classification.

15. The medical device of claim 14, wherein said first heartbeat matches said first classification and an indicia of a ventricular ectopic classification is displayed on said display.

16. The medical device of claim 14, wherein said first heartbeat matches said first classification and an indicia of a questionable classification is displayed on said display.

17. The medical device of claim 13, wherein said averaging means averages in a manner that is weighted in favor of said first template.

18. The medical device of claim 13, said heartbeat classification logic further comprising:

if said first heartbeat does not match said first template, means for comparing said first heartbeat with a second template stored in memory or storage corresponding to a second heartbeat classification;

means for determining if said first heartbeat matches said second template, responsive to said comparing means; and

if said first heartbeat matches said second template, means for classifying said first heartbeat as said second heartbeat classification.

19. The medical device of claim 18, said heartbeat classification logic further comprising:

means for averaging said first heartbeat into said second template.

20. The medical device of claim 19, wherein said averaging means averages in a manner that is weighted in favor of said second template.

21. The medical device of claim 18, said heartbeat classification logic further comprising:

if said first heartbeat does not match said second template, means for creating a third template for said first heartbeat; and

means for classifying said third template as a third heartbeat classification.

22. The medical device of claim 18, said heartbeat classification logic further comprising:

if said first heartbeat does not match said first template or said second template, means for determining whether said first template or said second template was last updated the longest time ago;

means for creating a third template for said first heartbeat;

means for overwriting said second template with said third template, responsive to said determining means determining that said second template was updated the longest time ago; and

means for classifying said third template as a third heartbeat classification.

23. The medical device of claim 13, wherein said first heartbeat matches said first classification and said first classification is a dominant supraventricular ectopic classification, said medical device further comprising:

means for checking to see if said first heartbeat was early;

if said checking means determined that said first heartbeat was early, means for classifying said first heartbeat as supraventricular ectopic; and

if said checking means determined that said first heartbeat was not early, means for classifying said first heartbeat as dominant.