Method and apparatus for classifying and localizing heart arrhythmias

• US 20040059237A1
• Filed: 12/18/2002
• Published: 03/25/2004
• Est. Priority Date: 09/19/2002
• Status: Active Grant

• Current Assignee: Regents of the University of California (University of California)
• Original Assignee: Regents of the University of California (University of California)
• Inventors: Narayan, Sanjiv Mathur, Bhargava, Valmik
• Granted Patent: US 7,123,954 B2
• Time in Patent Office: Days
• US Class Current: 600/509
• 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: A61N 1/3625 External stimulators

Abstract

Analyzes surface electrocardiographic and intracardiac signals to identify and separate electrical activity corresponding to distinct but superimposed events in the heart. Assesses the spatial phase, temporal phase, rate, spectrum and reproducibility of each event to determine uniformity of activation in all spatial dimensions. Uses numerical indices derived from these analyses to diagnose arrhythmias. Uses these indices to determine the location of an arrhythmia circuit, and to direct the movement of an electrode catheter to this location for ablation or permanent catheter positioning. Subsequently, uses these indices to determine whether ablation has successfully eliminated the circuit. Uses variability in these indices from the surface electrocardiogram to indicate subtle beat-to-beat fluctuations which reflect the tendency towards atrial and ventricular arrhythmias.

Claims

1. A method of analyzing a cardiac signal, comprising:

providing a digital template and a digitized cardiac signal, the digitized cardiac signal comprising a plurality of amplitudes corresponding to consecutive time samples, wherein the time samples span a plurality of cardiac cycles;

generating a plurality of correlation values correlating the digital template to successive time samples of the cardiac signal, wherein a plurality of correlation values are generated for each cardiac cycle; and

mapping the correlation values.

2. The method as set forth in claim 1, wherein the correlation values are mapped against time.

3. The method as set forth in claim 1, wherein the correlation values are plotted.

4. The method as set forth in claim 1, wherein successive time samples are consecutive time samples.

5. The method as set forth in claim 1, wherein the cardiac signal is an ECG).

6. The method as set forth in claim 1, wherein the cardiac signal is an intracardiac electrogram.

7. The method as set forth in claim 1, and further comprising computing at least one Eigen vector.

8. The method as set forth in claim 1, and further comprising:

comparing correlation values with a threshold value; and

identifying features as correlation values that exceed the threshold value.

9. The method as set forth in claim 1, and further comprising generating frequency representation of the digitized cardiac signal and of the correlation values.

10. The method as set forth in claim 9, wherein the frequency representation of the digitized cardiac signal comprises a plot of frequency versus power.

11. The method as set forth in claim 9, and further comprising identifying an non-harmonic frequencies having relatively low amplitudes.

12. The method as set forth in claim 9, wherein the frequency representation of the digitized cardiac signal comprises a plot of frequency versus power and wherein the identified non-harmonic frequencies have power amplitudes which less than about {fraction (1/10)} of a maximum power amplitude of the frequency representation.

13. The method as set forth in claim 12, and further comprising:

generating a time-domain representation of a selected band of the frequency representation;

identifying a feature, based upon the time-domain representation.

14. The method as set forth in claim 9, wherein the frequency representation of the correlation values comprises a plot of frequency versus power and wherein the identified non-harmonic frequencies have power amplitudes which less than about {fraction (1/10)} of a maximum power amplitude of the frequency representation.

15. The method as set forth in claim 14, and further comprising:

generating a time-domain representation of a selected band of the frequency representation;

identifying a feature, based upon the time-domain representation.

16. The method as set forth in claim 1, and wherein:

the cardiac signal is generated from a plurality of leads; and

the correlation values are plotted, each plot comprising a representation of correlation values of two leads from the plurality of leads, for each plot the correlation values of one of the two leads being plotted against the correlation values of the other of the two leads.

17. The method as set forth in claim 16, wherein the plurality of leads comprises at least three leads, and the correlation values are plotted to yield plots for a plurality of planes.

18. The method as set forth in claim 16, wherein the plurality of leads comprises at least three orthogonal leads, and the correlation values are plotted to yield three plots for three corresponding orthogonal planes.

19. The method as set forth in claim 17, and further comprising identifying a coherence of spatial phase for at least one of the three plots, when correlation values for that plot rise and fall simultaneously for both axes of that plot.

20. The method as set forth in claim 17, and further comprising identifying a coherence of spatial phase for the three plots, when correlation values for each of the three plots rise and fall simultaneously for both axes of each of the three plots.

21. The method as set forth in claim 20, and further comprising:

comparing correlation values for each plot with a threshold value; and

identifying features as correlation values that exceed the threshold value.

22. The method as set forth in claim 1, wherein the amplitudes are electrical potential.

23. The method as set forth in claim 1, and wherein the amplitudes are movements or hemodynamic tracings.

24. The method as set forth in claim 1, and wherein the digital template is a sequence of amplitudes spanning part of a cardiac cycle.

25. The method as set forth in claim 1, and wherein:

the cardiac signal is generated from a plurality of leads;

for each of at least three of the plurality of leads, a mapping of correlation values is generated so that at least three mappings of correlation values are generated;

the method further comprises identifying alternans of mappings.

26. The method as set forth in claim 25, wherein the alternans comprise cycles of correlation that alternate for successive heartbeats.

27. The method as set forth in claim 1, and wherein:

the cardiac signal is generated from a plurality of leads;

for each of at least three of the plurality of leads, a mapping of correlation values is generated so that at least three mappings of correlation values are generated;

the method further comprises identifying the mapping, which has a poorest correlation with the other mappings.

28. The method as set forth in claim 1, and wherein the providing is preceded by:

(a) placing a medical instrument into proximity of a heart; and

(b) stimulating the heart using the medical instrument to generate a paced cardiac signal.

(c) selecting at least three spatial axes and, for each of the spatial axes, determining a correlation of the cardiac signal to the paced cardiac signal;

(d) identifying a spatial axis of the at least three spatial axes that has a lowest determined correlation;

(e) moving the medical instrument in a direction of the identified axis.

29. The method as set forth in claim 28, wherein the moving comprise moving the medical instrument along the identified axis.

30. The method as set forth in claim 28, wherein the moving improves the lowest determined correlation.

31. The method as set forth in claim 30, and further comprising repeating (c), (d) and (e).

32. The method as set forth in claim 30, wherein the identified spatial axis for one occurrence of (c), (d) and (e) is different than an identified spatial axis for another occurrence of (c), (d) and (e).

33. The method as set forth in claim 1, and wherein the providing is preceded by:

placing a medical instrument into proximity of a heart;

stimulating the heart using the medical instrument to generate a paced cardiac signal;

the cardiac signal is generated from a plurality of leads;

for each of at least three of the plurality of leads, a mapping of correlation values is generated so that at least three cardiac mappings of correlation values are generated;

the paced cardiac signal is generated from a plurality of leads;

for each of at least three of the plurality of leads, a mapping of correlation values is generated for the paced cardiac signal so that at least three paced-cardiac mappings of correlation values are generated for the paced cardiac signal;

identifying a paced-cardiac mapping of the at least three paced-cardiac mappings that correlates poorest with a mapping of the at least three cardiac mappings.

34. The method as set forth in claim 33, wherein the identifying comprises identifying a paced-cardiac mapping of the at least three paced-cardiac mappings that correlates poorest with a corresponding mapping of the at least three cardiac mappings.

35. The method as set forth in claim 33, wherein:

each mapping of the at least three cardiac mappings corresponds to a spatial axis;

each mapping of the at least three paced-cardiac mappings corresponds to the spatial axis;

the identifying comprises identifying the spatial axis corresponding to the identified paced-cardiac mapping; and

the identifying is followed by moving the medical instrument in a direction of the identified spatial axis.

36. The method as set forth in claim 35, wherein the moving comprise moving the medical instrument along the identified axis.

37. The method as set forth in claim 34, wherein the correlation of the identified paced-cardiac mapping with the corresponding cardiac mapping is improved.