Device, method and computer-readable storage medium for SVT and VT classification

1. A method for classifying of supraventricular tachyarrhythmia or SVT from ventricular tachyarrhythmia or VT using signals provided by an electrogram comprising:

providing a template signal x(i) and a test signal y(i), wherein the template signal and the test signal comprise samples from an electrogram;

defining subspaces associated with said template signal wherein R denotes a signal space that spans from V min to V max , where V min is a minimum amplitude and V max is a maximum amplitude and R is divided into subspaces R P , R Z , and R N such that R=R P ∪R Z ∪R N and R P ∩R Z =R P ∩R N =R Z ∩R N =Ø, where ∪ is a union operator, ∩ is an intersection operator, and Ø represents null space;

transforming at least the test signal into a representation of the test signal ty(i) based on a trichotomization function:

ty

⁡

(

i

)

=

{

1

if

⁢

⁢

y

⁡

(

i

)

∈

R

P

0

if

⁢

⁢

y

⁡

(

i

)

∈

R

Z

-

1

if

⁢

⁢

y

⁡

(

i

)

∈

R

N

wherein ty(i) are integer values;

determining a correlation between the template signal and the test signal by summing the representation of the test signal ty(i):

∑

i

=

1

L

⁢

⁢

ty

⁡

(

i

)

wherein L represents a number of said samples; and,

classifying of SVT from VT based on the correlation, performed using an implantable cardiac device.

2. The method according to claim 1 , wherein the providing the template signal and the test signal comprises providing template and test signals from an intracardiac electrogram or IEGM, a surface electrocardiogram or ECG or a subcutaneous electrogram.

3. The method according to claim 1 , wherein providing the template signal comprises obtaining the template signal by averaging a plurality of cycles of signals.

4. The method according to claim 3 , wherein the averaging the plurality of cycles of signals comprises averaging a plurality of conducted ventricular IEGM signals.

5. The method according to claim 1 , further comprising updating the template signal periodically or continuously.

6. The method according to claim 1 , further comprising aligning the template and test signals based on at least one predefined fiducial point.

7. The method according to claim 1 , further comprising:

associating the template signal with at least two subspaces of template signal space; and,

transforming at least one of the template signal and the test signal with respect to the at least two subspaces.

8. The method according to claim 7 , further comprising defining the at least two subspaces by

defining a first subspace comprising values which differ from template signal values at most by a predefined first value; and,

defining a second subspace comprising values which differ from the template signal values at least by the predefined first value.

9. The method according to claim 7 , further comprising associating the template signal with three subspaces and defining the three subspaces by

defining a first subspace comprising values which differ from template signal values at most by a predefined first value;

defining a second subspace comprising values which differ from template signal values at least by the predefined first value and at most by a predefined second value; and,

defining a third subspace comprising values which differ from the template signal values at least by the predefined second value.

10. The method according to claim 7 , further comprising bounding the at least two subspaces by threshold vectors.

11. The method according to claim 10 , further comprising obtaining the threshold vectors by increasing or decreasing sample values of the template signal by a predefined value.

12. The method according to claim 1 , wherein the determining the correlation comprises using transformed test signals.

13. The method according to claim 1 , wherein the determining the correlation comprises using only a transformed test signal.

14. The method according to claim 1 , wherein the determining the correlation comprises determining an adapted signed correlation index by summing the sample values of a transformed test signal or by dividing a sum of the sample values of the transformed test signal by a number of samples.

15. The method according to claim 1 , wherein the classifying comprises classifying a ventricular IEGM as being of ventricular origin if the correlation is below a predefined threshold, or as being of supraventricular origin otherwise.

16. The method according to claim 1 , wherein the classifying further comprises performing an RR interval analysis.

17. The method according to claim 16 , further comprising incrementing a VT sample counter by 1 for a ventricular cycle that falls in a VT/VF zone if and only if the correlation between the test signal and the template signal is below a predefined threshold.

18. The method according to claim 1 , wherein the classifying further comprises performing an SVT-VT classification algorithm that involves both atrial and ventricular rate and rhythm analysis.

19. The method according to claim 1 , wherein the providing the template signal comprises providing an atrial IEGM waveform and wherein the determining the correlation further comprises distinguishing an intrinsic atrial event from a retrograde conducted atrial event when the correlation between an atrial test IEGM signal with the atrial IEGM waveform detects a 1:1 relationship between the SVT or the VT.