Methods and Systems for Characterizing Cardiac Signal Morphology Using K-Fit Analysis

• US 20110077541A1
• Filed: 09/10/2010
• Published: 03/31/2011
• Est. Priority Date: 09/25/2009
• Status: Active Grant

• Current Assignee: Cardiac Pacemakers Incorporated (Boston Scientific Corporation)
• Original Assignee: Cardiac Pacemakers Incorporated (Boston Scientific Corporation)
• Inventors: Zhang, Shijie, Lin, Yayun, Li, Dan, Dong, Yanting
• Granted Patent: US 8,483,808 B2
• Time in Patent Office: Days
• US Class Current: 600/515
• CPC Class Codes: A61B 5/35 by template matching
• CPC Class Codes: G06F 2218/12 Classification; Matching

Abstract

A system and method for automatically analyzing a cardiac signal, including the step of providing an episode database on a computer storage medium including a plurality of episode data records of one or more patients. Each episode data record includes a cardiac signal from at least one data-generating device. The method also includes the step of selecting one or more of the N beats to be one or more beat templates, for at least a first cardiac signal having N beats. Another step is determining a value K for the cardiac signal using a computer system where K beat templates can represent all the N beats in the cardiac signal.

Claims

1. A method for automatically analyzing a cardiac signal, comprising the steps of:

providing an episode database on a computer storage medium including a plurality of episode data records of one or more patients, wherein each episode data record includes a cardiac signal from at least one data-generating device;

for at least a first cardiac signal having N beats, selecting one or more of the N beats to be one or more beat templates; and

determining a value K for the cardiac signal using a computer system where K beat templates can represent all the N beats in the cardiac signal.

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

assigning a episode characterization for each episode data record based at least partially on the value K and how well the episode is represented by K beat templates.

3. The method of claim 2 wherein the episode characterization is an arrhythmia episode characterization.

4. The method of claim 3 wherein the arrhythmia characterization is selected from a group comprising at least polymorphic ventricular tachycardia (PVT/VF), monomorphic ventricular tachycardia (MVT), atrial fibrillation (AF), and other supraventricular tachycardia (other SVT).

5. The method of claim 1 further comprising the step of:

calculating a morphology distance between each beat and a corresponding beat template candidate.

6. The method of claim 5 wherein the morphology distance is selected from a group comprising at least Euclidean distance, Mahaladobis distance, and Manhattan distance.

7. The method of claim 1 wherein for a particular cardiac signal, K beat templates are presented to a user on a display device to represent the signal morphology of the particular cardiac signal.

8. The method of claim 1 wherein for a particular cardiac signal, M beat templates are presented to a user on a display device to represent the signal morphology of the particular EGM, where M is less than K.

9. The method of claim 1 wherein the step of determining a value K for the cardiac signal comprises:

for each beat template, determining a morphology distance between each beat in the beat template and each beat N in the cardiac signal;

for each beat template, calculating a morphology distance compilation value which represents how well the beat template represents the cardiac signal, wherein the morphology distances are used in calculating the morphology distance compilation value;

outputting the beat template that results in the best representation of the cardiac signal based on the morphology distance compilation value for the cardiac signal;

outputting the minimum morphology compilation value and the beat templates for each value of K; and

selecting a K value for the cardiac signal that is a smallest K value that gives good performance.

10. The method of claim 9 wherein:

calculating the morphology distance compilation value for each beat template comprises summing the morphology distances between each beat in the beat template and each beat N in the cardiac signal to calculate a morphology distance sum, and

outputting the beat template comprises outputting the beat template with a lowest morphology distance sum.

11. The method of claim 1 wherein each beat comprises multiple channels of the cardiac signal and wherein the method further comprising the step of:

calculating a morphology distance between each beat and a corresponding complex beat template candidate, wherein the morphology distance is based on a channel morphology distance for each channel and a beat length difference.

12. A method for automatically analyzing a cardiac signal comprising the steps of:

i) providing an episode database on a computer system including at least one episode data record, wherein each episode data record includes a cardiac signal from at least one data-generating device, wherein the cardiac signal has N beats;

ii) selecting a range of possible values for K;

iii) for each value of K, performing the following steps using a computer system;

(a) for each set S of K beats, perform the following steps;

1. for each beat b not in Set S, find which member s(b) of the set S is closest to b and calculate a morphology distance between s(b) and b;

2. for each beat b, calculate a morphology distance compilation value for each set S, wherein the morphology distances are used in calculating the morphology distance compilation value;

(b) compare the morphology distance compilation values to identify which set S of K beats results in the best representation of the cardiac signal based on the morphology distance compilation value; and

(c) record the morphology distance compilation value and record set S of K beats; and

iv) based on the morphology distance compilation values for each value of K, determine a best value of K for the EGM where K beat templates can represent all the beats in the EGM signal.

13. The method of claim 12 wherein:

calculating the morphology distance compilation value for each beat template comprises summing the morphology distances between each beat in the beat template and each beat N in the cardiac signal to calculate a morphology distance sum, and

outputting the beat template comprises outputting the beat template with a lowest morphology distance sum.

14. A system for automatically analyzing a cardiac signal, comprising:

an episode database including a plurality of episode data records of one or more patients, wherein each episode data record includes a cardiac signal from at least one data-generating device;

a adjudication processor configured to;

for at least a first cardiac signal having N beats, selecting one or more of the N beats to be one or more beat templates; and

determining a value K for the cardiac signal where K beat templates can represent all the N beats in the cardiac signal.

15. The system of claim 14 further comprising an output database comprising an episode characterization for each episode data record that is determined by the adjudication processor based at least partially on the value K.

16. The system of claim 15 wherein the episode characterization is an arrhythmia episode characterization.

17. The system of claim 16 wherein the arrhythmia characterization is selected from a group comprising at least polymorphic ventricular tachycardia (PVT/VF), monomorphic ventricular tachycardia (MVT), atrial fibrillation (AF), and other supraventricular tachycardia (other SVT).

18. The system of claim 14 wherein the adjudication processor is further configured to calculate a morphology distance between each beat and a corresponding beat template candidate.

19. The system of claim 18 wherein the morphology distance is selected from a group comprising at least Euclidean distance, Mahaladobis distance, and Manhattan distance.

20. The system of claim 14 further comprising a display device, wherein for a particular cardiac signal, K beat templates are presented to a user on the display device to represent the signal morphology of the particular cardiac signal.