Method of removing signal interference from sampled data and apparatus for effecting the same

• US 6,351,664 B1
• Filed: 11/12/1999
• Issued: 02/26/2002
• Est. Priority Date: 11/12/1999
• Status: Expired due to Fees

• Current Assignee: GE Marquette Medical Systems, Inc. (General Electric Company)
• Original Assignee: GE Medical Systems Information Technologies Incorporated (General Electric Company)
• Inventors: Brodnick, Don
• Primary Examiner(s): KAMM, WILLIAM E
• Application Number:
  US09/439,712
• Time in Patent Office: 837 Days
• Field of Search: 600/509
• US Class Current: 600/509
• CPC Class Codes: A61B 5/316 Modalities, i.e. specific diagnostic methods
• CPC Class Codes: A61B 5/318 Heart-related electrical modalities, e.g. electrocardiography [ECG]

Abstract

A method and apparatus that separates an interference signal from a sampled input signal is provided. The method and apparatus approximates the frequency, phase and amplitude of the interference signal through an optional reverse direction learning process and then separates that portion of the interference signal overlapped or combined with a sample of the input signal. The signals can be separated by generating a second signal of suitable approximation in frequency, amplitude and phase with the interference signal and then subtracting a sample of that signal from the sampled input signal. The second signal frequency and amplitude/phase may be derived through an iterative process that measures the amplitude of the interference signal and the phase angle between the second signal and an output signal and then adjusts the frequency and amplitude of the second signal based, in part, on these measurements until the second signal substantially matches the interference signal. If desired, the processing method and apparatus may be applied to a digitized sampled waveforms and can operate at any sample frequency.

Claims

1. A method for removing interference signals from a sampled signal comprising the steps of:

sampling an input signal comprising a selected signal combined with an interference signal having a frequency and an amplitude;

approximating the frequency, amplitude and phase of the interference signal; and

separating the interference signal from the sampled input signal.

2. The method of claim 1 wherein the act of approximating uses an iterative reverse direction learning process.

3. The method of claim 1 further comprising the act of:

generating a second signal of suitable approximation in frequency and amplitude to the interference signal.

4. The method of claim 3 wherein the separating act comprises subtracting a sample of the second signal from one of the samples of the input signal.

5. The method of claim 1 wherein the approximating act includes the act of estimating the frequency and amplitude of the interference signal using a bang—

bang control algorithm.

6. The method of claim 1 wherein the approximating act includes the act of estimating the frequency and amplitude of the interference signal using a piece-wise linear control function.

7. The method of claim 1 wherein the approximating act includes the act of estimating the frequency and amplitude of the interference signal using a sigmoid control function.

8. The method of claim 1 wherein the approximating act includes the act of estimating the phase of the interference signal through a cross-product function.

9. The method of claim 1 wherein the approximating act comprises the acts of:

generating a second signal;

measuring a phase angle between samples of the selected signal and samples of the second signal;

estimating the amplitude of a sample of the interference signal; and

adjusting the frequency and amplitude and phase of the second signal to substantially match the interference signal.

10. The method of claim 9 wherein the separating act comprises the act of subtracting at least one sample of the second signal from the sampled input signal after the adjusting act.

11. The method of claim 9 wherein the selected signal is an output signal.

12. The method of claim 11 wherein the act of measuring the phase angle comprises calculating the cross product of the samples of the output signal and the second signal.

13. The method of claim 1 wherein the approximating act is an iterative process of estimating an amplitude of the interference signal and a phase angle between the interference signal and an output signal and adjusting the frequency and amplitude of the second signal based in part on these measurements until a sample of the second signal substantially matches a sample of the interference signal.

14. The method of claim 1 wherein the separating act comprises filtering a sample of the interference signal from a sample of the input signal.

15. The method of claim 14 wherein the filtering act uses a notch filter.

16. The method of claim 14 wherein the separating act comprises positioning a filter null of a notch filter at the frequency of the sample of the interference signal.

17. The method of claim 1 wherein the interference signal is a sinusoid.

18. A method for removing interference signals from selected electrocardiogram signals comprising the acts of:

sampling an input signal comprising a selected electrocardiogram signal combined with a sinusoidal interference signal having a frequency and an amplitude;

comparing successive samples of an output signal backward in time using a reverse direction learning process to measure frequency, amplitude and phase of the interference signal;

generating a second signal of suitable approximation in frequency, amplitude and phase to the interference signal; and

subtracting a sample of the second signal from one sample of the input signal.

19. The method of claim 18 wherein the generating act comprises the acts of:

(a) adjusting the amplitude and phase of the second signal by a predetermined amount if the amplitude and phase of the second signal is not substantially equal to the amplitude and phase of the interference signal;

(b) adjusting the frequency of the second signal by a predetermined amount if the frequency of the second signal is not substantially equal to the frequency of the interference signal; and

then

(c) repeating acts b and c until the second signal is substantially equal in frequency, amplitude and phase to the interference signal.

20. The method of claim 18 wherein the subtracting act comprises shifting the sample of the second signal one hundred and eighty degrees out of phase with respect to the phase of the input signal and then adding the sample of the second signal to one sample of the input signal.

21. The method of claim 18 wherein the selected signal is an electrocardiogram signal.

22. The method of claim 18 wherein the comparing act includes the act of estimating the frequency and amplitude of the interference signal using a bang—

bang control algorithm.

23. The method of claim 18 wherein the comparing act includes the act of estimating the frequency and amplitude of the interference signal using a piece-wise linear control function.

24. The method of claim 18 wherein the comparing act includes the act of estimating the frequency and amplitude of the interference signal using a sigmoid control function.

25. The method of claim 18 wherein the comparing act includes the act of estimating the phase of the interference signal through a cross-product function.

26. An apparatus for removing power line interference from a sampled signal, the apparatus comprising:

an input circuit tuned to receive an input signal, said input signal comprising a selected signal and an interference signal;

a detection circuit connected to the input circuit to approximate the frequency, amplitude and phase of the interference signal using a reverse direction learning process; and

a separating circuit coupled to the input circuit and the detection circuit to remove the interference signal from the input signal.

27. The apparatus of claim 26, wherein the detection circuit comprises a noise measuring circuit and a phase measuring circuit.

28. The apparatus of claim 26, wherein the detection circuit comprises an oscillator circuit that generates a signal of suitable approximation in frequency, amplitude and phase to the interference signal.

29. The apparatus of claim 28, wherein the detection circuit comprises a noise measuring circuit and a phase measuring circuit having a common input coupled to the separating circuit and having an output coupled to the oscillator circuit.

30. The apparatus of claim 26, wherein the separating circuit comprises a summing circuit.

31. An electrocardiograph including a filter for removing power line interference from a sampled electrocardiograph signal, the electrocardiograph comprising:

an input circuit tuned to receive an input signal comprising an electrocardiograph signal and an interference signal;

a detection circuit connected to the input circuit to approximate the frequency, amplitude and phase of the interference signal using a reverse direction learning process; and

a separating circuit coupled to the input circuit and the detection circuit to remove the interference signal from the input signal.

32. The electrocardiograph of claim 31, wherein the detection circuit comprises a noise measuring circuit and a phase measuring circuit.

33. The electrocardiograph of claim 31, wherein the detection circuit comprises an oscillator circuit that generates a signal of suitable approximation in frequency, amplitude and phase to the interference signal.

34. The electrocardiograph of claim 33, wherein the detection circuit comprises a noise measuring circuit and a phase measuring circuit having a common input coupled to the separating circuit and having an output coupled to the oscillator circuit.

35. The electrocardiograph of claim 31, wherein the separating circuit comprises a summing circuit.

36. The electrocardiograph of claim 31 wherein the detection circuit and the separating circuit comprise the filter, and wherein the electrocardiograph includes a plurality of like filters cascaded to independently reject the fundamental interference frequency and harmonics of the fundamental frequency.