“Blurred template” approach for arrhythmia detection

• US 8,428,697 B2
• Filed: 01/22/2009
• Issued: 04/23/2013
• Est. Priority Date: 01/22/2009
• Status: Active Grant

• Current Assignee: Medtronic Incorporated (Medtronic Plc)
• Original Assignee: Medtronic Incorporated (Medtronic Plc)
• Inventors: Zhang, Xin, Brown, Mark L., Zhang, Xusheng, Gillberg, Jeffrey M.
• Primary Examiner(s): Lavert, Nicole F
• Application Number:
  US12/357,868
• Publication Number:
  US 20100185109A1
• Time in Patent Office: 1,552 Days
• Field of Search: 600508-509, 607/5
• US Class Current: 600/509
• CPC Class Codes: A61B 5/35 by template matching
• CPC Class Codes: A61B 5/361 Detecting fibrillation
• CPC Class Codes: A61B 5/363 Detecting tachycardia or bradycardia
• 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 1/3956 Implantable devices for applying electric shocks to the heart, e.g. for cardioversion
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

An implantable medical device and associated method sense cardiac signals for deriving a template representing a known EGM waveform morphology and for classifying an unknown waveform morphology. A boundary of the template, offset from the template, is computed and compared to an unknown waveform morphology for classifying the unknown waveform morphology.

Claims

1. A method for use in an implantable medical device, comprising:

sensing a cardiac signal;

deriving a known template representing a known waveform morphology from the cardiac signal;

computing one or more adjusted templates, each of the one or more adjusted templates computed by shifting the known template in at least one of time and amplitude;

computing an upper boundary using the one or more adjusted templates;

computing a lower boundary using the one or more adjusted templates, wherein an area is defined between the upper and lower boundaries;

acquiring an unknown waveform morphology from the cardiac signal;

comparing the unknown waveform morphology to the upper and lower boundaries; and

classifying the unknown waveform morphology based on the comparison.

2. The method of claim 1 wherein the upper boundary and the lower boundary are each offset from the known template.

3. The method of claim 2 wherein computing the upper boundary comprises

selecting a highest amplitude of the one or more adjusted templates at each of a plurality of time points and

defining the upper boundary using the selected highest amplitudes, and wherein computing the lower boundary comprises selecting a lowest amplitude of the one or more adjusted templates at each of the plurality of time points and defining the lower boundary using the selected lowest amplitudes.

4. The method of claim 1 wherein comparing the unknown waveform to the upper and lower boundaries comprises

computing an estimated area of the unknown waveform falling outside the area defined by the upper boundary and the lower boundary.

5. The method of claim 4 wherein computing the estimated area comprises identifying a sample point of the unknown waveform falling outside the area defined by the upper and lower boundaries and computing a distance of the identified sample point from a nearest one of the upper boundary and the lower boundary.

6. The method of claim 5 wherein computing the distance comprises determining a smallest amplitude difference between the identified sample point and a plurality of sample points along the nearest one of the upper and lower boundaries.

7. The method of claim 5 wherein computing the estimated area comprises multiplying the distance by a weighting factor and summing the distance with other distances computed using other identified sample points of the unknown waveform falling outside the upper and lower boundaries.

8. The method of claim 7 further comprising:

computing an outer upper boundary that is offset from the upper boundary, wherein a first region is defined between the outer upper boundary and the upper boundary;

computing an outer lower boundary that is offset from the lower boundary, wherein a second region is defined between the outer lower boundary and the lower boundary; and

determining the weighting factor based on the location of the identified sample point relative to the outer upper boundary and the outer lower boundary.

9. The method of claim 8 wherein computing the outer upper boundary comprises shifting the upper boundary in one of time and amplitude, and wherein computing the outer lower boundary comprises shifting the lower boundary in one of time and amplitude.

10. The method of claim 5 further comprising:

determining an adjusted unknown waveform by shifting the unknown waveform in time relative to the known template;

computing an adjusted estimated area of the adjusted unknown waveform falling outside the area defined by the upper and lower boundaries; and

classifying the unknown waveform using the lowest one of the estimated area and the adjusted estimated area.

11. The method of claim 1 wherein at least one of the upper and lower boundaries includes a portion of the known template.

12. The method of claim 2 wherein at least a portion of the known template is included in the area between the upper and lower boundaries.

13. The method of claim 4 further comprising, prior to computing the estimated area, aligning in time a selected point of the unknown waveform with a corresponding selected point of the known template.

14. An implantable medical device, comprising;

electrodes for sensing a cardiac signal;

memory for storing a known template representing a known waveform morphology of the cardiac signal;

a processor coupled to the memory, the processor configured to;

receive the cardiac signal;

compute one or more adjusted templates, each of the one or more adjusted templates computed by shifting the known template in at least one of time and amplitude;

compute an upper boundary using the one or more adjusted templates;

compute a lower boundary using the one or more adjusted templates, wherein an area is defined between the upper and lower boundaries;

acquire an unknown waveform morphology from the cardiac signal;

compare the unknown waveform morphology to the upper and lower boundaries; and

classify the unknown waveform morphology based on the comparison.

15. The device of claim 14 wherein the upper boundary and the lower boundary are each offset from the known template.

16. The device of claim 15 wherein the processor is configured to compute the upper boundary

by selecting a highest amplitude of the one or more adjusted templates at each of a plurality of time points and

defining the upper boundary using the selected highest amplitudes, and wherein the processor is configured to compute the lower boundary by selecting a lowest amplitude of the one or more adjusted templates at each of the plurality of time points and defining the lower boundary using the selected lowest amplitudes.

17. The device of claim 14 wherein the processor is configured to compare the unknown waveform to the upper and lower boundaries

by computing an estimated area of the unknown waveform falling outside the area defined by the upper boundary and the lower boundary.

18. The device of claim 17 wherein the processor is configured to compute the estimated area by identifying a sample point of the unknown waveform falling outside the area defined by the upper and lower boundaries and computing a distance of the identified sample point from a nearest one of the upper boundary and the lower boundary.

19. The device of claim 18 wherein the processor is configured to compute the distance by determining a smallest amplitude difference between the identified sample point and a plurality of sample points along the nearest one of the upper and lower boundaries.

20. The device of claim 18 wherein the processor is configured to compute the estimated area by multiplying the distance by a weighting factor and summing the distance with other distances computed using other identified sample points of the unknown waveform falling outside the upper and lower boundaries.

21. The device of claim 20 wherein the processor is configured to compute the estimated area by:

computing an outer upper boundary that is offset from the upper boundary, wherein a first region is defined between the outer upper boundary and the upper boundary;

computing an outer lower boundary that is offset from the lower boundary, wherein a second region is defined between the outer lower boundary and the lower boundary; and

determining the weighting factor based on the location of the identified sample point relative to the outer upper boundary and the outer lower boundary.

22. The device of claim 21 wherein the processor is configured to compute the outer upper boundary by shifting the upper boundary in one of time and amplitude, and wherein the processor is configured to compute the outer lower boundary by shifting the lower boundary in one of time and amplitude.

23. The device of claim 17 wherein the processor is configured to compare the unknown waveform to the upper and lower boundaries by:

determining an adjusted unknown waveform by shifting the unknown waveform in time relative to the known template;

computing an adjusted estimated area of the adjusted unknown waveform falling outside the area defined by the upper and lower boundaries; and

classifying the unknown waveform using the lowest one of the estimated area and the adjusted estimated area.

24. The device of claim 14 wherein at least one of the upper and lower boundaries includes a portion of the known template.

25. The device of claim 15 wherein at least a portion of the known template is included in the area between the upper and lower boundaries.

26. The device of claim 17 wherein the processor is configured to align, in time, a selected point of the unknown waveform with a corresponding selected point of the known template prior to computing the estimated area.

27. A non-transitory computer readable storage medium storing a set of instructions which when implemented in a medical device system cause the system to:

sense a cardiac signal;

derive a known template representing a known waveform morphology from the cardiac signal;

compute one or more adjusted templates, each of the one or more adjusted templates computed by shifting the known template in at least one of time and amplitude;

compute an upper boundary using the one or more adjusted templates;

compute a lower boundary using the one or more adjusted templates, wherein an area is defined between the upper and lower boundaries;

acquire an unknown waveform morphology from the cardiac signal;

compare the unknown waveform morphology to the upper and lower boundaries boundary; and

classify the unknown waveform morphology based on the comparison.