Nonlinear method and apparatus for electrocardiogram pacemaker signal filtering

• US 7,383,079 B2
• Filed: 04/06/2005
• Issued: 06/03/2008
• Est. Priority Date: 04/08/2004
• Status: Expired due to Fees

• Current Assignee: Welch Allyn Incorporated
• Original Assignee: Welch Allyn Incorporated
• Inventors: Holland, Alexander
• Primary Examiner(s): Layno; Carl H.
• Assistant Examiner(s): Bertram; Eric D
• Application Number:
  US11/099,849
• Publication Number:
  US 20050234361A1
• Time in Patent Office: 1,154 Days
• Field of Search: 600/508
• US Class Current: 600/509
• CPC Class Codes: A61B 5/349 Detecting specific parameters of the electrocardiograph cycle
• CPC Class Codes: A61N 1/3704 Circuits specially adapted therefor, e.g. for sensitivity control

Abstract

Signal “peak” and “valley” removal properties of mathematical morphology operators are exploited in a method and apparatus for detecting, removing, or improving fidelity of pacemaker signal components of the Electrocardiogram (ECG) at sampling rates well below the pacemaker signal Nyquist rate. The method works for the wide variability in pacemaker signal amplitude, width, firing frequency, and other characteristics encountered in practice, and it works for malfunctioning pacemakers that may fire at any point relative to the QRS complex of the ECG signal. Filtering operations require minimal digital storage and are computationally inexpensive (no multiplications), mainly involving “maximum” and “minimum” type signal detections over a finite time history of the input signal. This implies that the method can be inexpensively implemented on small instruments in either hardware or software. The method may also be used to estimate the height and polarity of the pacemaker voltage spike, even though the base sampling period may be longer than the width of the spike.

Claims

1. A method to at least one of detecting, removing and estimating a pacemaker pacer signal from an electrocardiogram (“

ECG”

) signal of a human or animal to diagnose a physiological condition, said method comprising the steps of;

digitizing a plurality of ECG potentials (M channels of ECG lead data) with an analog to digital converter (“

ADC”

) to create a discrete signal representation of each potential in time;

filtering the plurality of ECG potentials using a digital filter to create a plurality of filtered ECG potentials;

operating on the plurality of filtered ECG potentials using a non-linear mathematical morphology function to detect and remove the pacer maker signal from the ECG signal to create a plurality of processed ECG potentials wherein the step of operating on the plurality of filtered ECG potentials using a non-linear mathematical morphology function further comprises the step of estimating the pacer signal height using sampling rates that may be below the Nyquist rate; and

diagnosing a physiological condition of the human or animal based on the processed ECG potentials.

2. The method of claim 1 wherein the step of operating on the plurality of filtered ECG potentials using a non-linear mathematical morphology function comprises the additional step of using a dilate mathematical morphology function.

3. The method of claim 1 wherein the step of operating on the plurality of filtered ECG potentials using a non-linear mathematical morphology function comprises the additional step of using an erode mathematical morphology function.

4. The method of claim 1 wherein the step of operating on the plurality of filtered ECG potentials using non-linear mathematical morphology functions comprises the additional step of using both a dilate and an erode mathematical morphology function.

5. The method of claim 4 wherein the step of operating on the plurality of filtered ECG potentials using the non-linear dilate and erode mathematical morphology functions removes ECG signal peaks and valleys caused by the pacer signal creating a clean QRS complex waveform free of significant distortion caused by the pacer signal.

6. The method of claim 5 wherein operating on the plurality of filtered ECG potentials using the non-linear dilate and erode mathematical morphology functions removes ECG signal peaks and valleys having widths smaller than a specified threshold while preserving the integrity of peaks and valleys wider than the specified threshold associated with a QRS complex in the ECG potential.

7. The method of claim 6 wherein operating on the plurality of filtered ECG potentials using a non-linear mathematical morphology function to detect and remove the pacer signal from the ECG signal to create a plurality of processed ECG potentials comprises detecting and removing a pacer signal from the ECG signal that has lost proper phasing or synchronization with the QRS complex.

8. The method of claim 1 wherein the step of diagnosing a physiological condition of the human or animal based on the processed ECG potentials further comprises the additional step of detecting and diagnosing a pacemaker failure condition based on the detected pacer signal.

9. The method of claim 8 wherein the step of detecting and diagnosing a pacemaker failure condition comprises the additional step of detecting and diagnosing loss of pacemaker phase reference to the QRS complex or loss of pacemaker synchronization with the QRS complex.

10. The method of claim 1 further comprising the step of comparing the plurality of filtered ECG potentials to the plurality of processed ECG potentials by delaying the filtered ECG potentials in order to match a signal delay caused by the mathematical morphology function.

11. The method of claim 1 wherein digitizing a plurality of ECG potentials (M channels of ECG lead data) with an analog to digital converter (“

ADC”

) to create a discrete signal representation of each potential in time further comprises the step of detecting a pacer signal using an analog pacer detector to associate a flag with nearest points in the discrete signal representation of the ECG potentials.

12. The method of claim 11 wherein detecting pacer signals using an analog pacer detector comprises detecting a pacer signal that coincidentally falls between two points in the discrete signal created by the ADC such that at least one instance of a sub-Nyquist pacer signal would otherwise go undetected by the ADC.

13. An electrocardiogram (“

ECG”

) apparatus to at least one of detecting, removing and estimating pacemaker pacer signals from an ECG waveform to diagnose a physiological condition of a human or animal, said apparatus comprising;

ECG electrodes to pickup a plurality of ECG potentials (M channels of ECG lead data) from the body of the human or animal being diagnosed, the electrodes electrically coupled to the body and connected to an ECG input;

an analog to digital converter (“

ADC”

) electrically connected to the ECG input, the ADC to digitize the ECG potentials and to create a discrete signal representation of each potential in time;

a filter block electrically coupled to the ADC to transfer the discrete signal representation of each potential in time to the filter block, the filter block to perform digital filtering on the discrete signal representation of each potential and to create a plurality of filtered ECG potentials;

a non-linear pacer filter electrically coupled to the filter block to receive the plurality of filtered ECG potentials, the non-linear pacer filter to perform at least one non-linear mathematical morphology function on the filtered ECG potentials to detect and remove the pacer signal from the ECG waveform and to create at least one processed ECG waveform; and

a display to allow viewing of the processed ECG waveform to diagnose a physiological condition of a human or animal based at least in part on the viewing of the processed ECG waveform wherein the non-linear pacer filter further comprises;

a delay operator to shift the at least one processed ECG waveform in time so that the signal can be compared to signals and events unmodified by the morphology operator;

a morphology operator to perform a series of concatenated dilate/erode mathematical morphology functions on the plurality of filtered ECG potentials; and

a pacer spike detector/estimator to calculate a mean estimate and a peak to peak variation estimate of a main biphasic pacer spike using an accumulated history of spike sample values.

14. The apparatus of claim 13 wherein the non-linear pacer filter performs at least one non-linear erode mathematical morphology function.

15. The apparatus of claim 13 wherein the non-linear pacer filter performs at least one non-linear dilate mathematical morphology function.

16. The apparatus of claim 13 wherein the non-linear pacer filter performs cascaded non-linear erode and dilate mathematical morphology functions.

17. The apparatus of claim 13 wherein the delay operator causes a linear delay suitable to the number and type of concatenated dilate/erode operators performed by the morphology operator.

18. The apparatus of claim 13 wherein the pacer spike detector/estimator comprises at least one FIR or at least one IIR digital filter to perform low pass filtering as part of the mean estimate and an average deviation calculation.

19. The apparatus of claim 13 wherein a deviation from the mean calculation performed in the pacer spike detector/estimator is a variance, the variance recursively estimated from a plurality of pacer signal samples.

20. The apparatus of claim 13 wherein differences between a time shifted signal from the delay operator and a processed signal from the morphology operator are used to indicate the presence of a pacer spike.

21. The apparatus of claim 20 further comprising an analog pacer detector to detect a sub Nyquist pacer signal wherein the pacer signal falls between two points in the discrete signal created by the ADC such that at least one instance of the pacer signal would otherwise go undetected by the ADC.

22. The apparatus of claim 21 wherein analog detection of a pacer signal causes a flag or marker that can be delayed by a delay operator to match in time a signal processed and delayed by a morphology operator.

23. The apparatus of claim 13 wherein a marker or a flag signal indicates the presence of a pacer spike.

24. The apparatus of claim 13 wherein the non-linear mathematical morphology function is carried out in software or in hardware.