Method and apparatus for a morphology-preserving smoothing filter

• US 6,641,541 B1
• Filed: 07/27/2000
• Issued: 11/04/2003
• Est. Priority Date: 07/27/2000
• Status: Expired due to Term

• Current Assignee: Cardiac Pacemakers Incorporated (Boston Scientific Corporation)
• Original Assignee: Cardiac Pacemakers Incorporated (Boston Scientific Corporation)
• Inventors: Lovett, Eric G., Sweeney, Robert J.
• Primary Examiner(s): Evanisko, George R.
• Assistant Examiner(s): BRADFORD, RODERICK D
• Application Number:
  US09/626,361
• Time in Patent Office: 1,195 Days
• Field of Search: 600/508, 600/509, 600/520, 128/901, 607/9, 327/551, 327/552, 333/181
• US Class Current: 600/508
• CPC Class Codes: A61B 5/30 Input circuits therefor
• CPC Class Codes: A61B 5/308 for electrocardiography [ECG]
• CPC Class Codes: A61B 5/316 Modalities, i.e. specific diagnostic methods
• CPC Class Codes: A61B 5/346 Analysis of electrocardiograms
• CPC Class Codes: A61B 5/7203 for noise prevention, reduction or removal
• CPC Class Codes: A61N 1/3956 Implantable devices for applying electric shocks to the heart, e.g. for cardioversion
• CPC Class Codes: H03H 17/026 Averaging filters
• CPC Class Codes: H03H 21/0043 Adaptive algorithms

Abstract

A method and apparatus for implementing a signal morphology preservation smoothing scheme. The smoothing scheme includes generating an output signal representative of a filtered version of the input signal, wherein the output signal is generated by adaptively removing low amplitude, high frequency noise components while simultaneously preserving signal morphology of the input signal. The smoothing scheme includes comparison of a distance metric against a distance threshold to determine whether an initial smoothed version of the input signal would be oversmoothed or undersmoothed. Then the smoothing scheme appropriately increments, decrements, or maintains the initial level of smoothing to generate an optimal smoothed signal representative of the output signal.

Claims

1. A method for filtering an input signal, comprising:

generating an output signal representative of a filtered version of the input signal by adaptively removing noise components from the input signal, wherein each of the input signal and the output signal includes discrete samples; and

determining a desired filtering level from a number of filtering levels for each of the discrete samples of the input signal, wherein determining a desired filtering level includes determining an initial distance metric corresponding to an initial filtering level for each of the discrete samples of the input signal.

2. The method of claim 1, wherein the initial filtering level is a desired filtering level for an immediately preceding sample.

3. The method of claim 1, further comprising identifying a distance threshold representing a desired amount of signal smoothing, wherein determining a desired filtering level includes comparing the initial distance metric against the distance threshold.

4. The method of claim 3, further comprising selecting the distance threshold based on the input signal that balances removal of noise components with preserving a morphological representation of the input signal.

5. The method of claim 3, further comprising decreasing the initial filtering level by at least one filtering level to become the desired filtering level when the initial distance metric is at least equal to the distance threshold.

6. The method of claim 3, wherein determining a desired filtering level includes determining another distance metric corresponding to the initial filtering level incremented by one filtering level.

7. The method of claim 6, wherein at least one of determining an initial distance metric and determining another distance metric includes calculating a distance metric using D(i_n)=|y_in(n)−

x(n)|, where i_n represents a smoothing index, D(i_n) represents a distance metric, y_in(n) represents an nth sample smoothed signal, and x(n) represents an nth sample input signal.

8. The method of claim 6, wherein at least one of determining an initial distance metric and determining another distance metric includes calculating a distance metric using $D$

(in)=14

∑

m=n-2n+1

 





yin

(m)-x

(m)

,

where i_n represents a smoothing index, D(i_n) represents a distance metric, y_in represents a sample smoothed signal, and x represents a sample input signal.

9. The method of claim 6, further comprising increasing the initial filtering level by at least one filtering level to become the desired filtering level when the another distance metric is smaller than the distance threshold.

10. A system, comprising a processor to receive an input signal and generate an output signal representative of a filtered version of the input signal by adaptively removing noise components from the input signal, wherein:

the input signal includes a number of discrete samples;

the processor is configured to determine a desired filtering level for each of the discrete samples of the input signal; and

the processor is configured to determine an initial distance metric corresponding to an initial filtering level to determine the desired filtering level.

11. The system of claim 10, further comprising a distance threshold that represents a desired amount of signal smoothing, wherein the processor is configured to compare the initial distance metric against the distance threshold.

12. The system of claim 11, wherein the processor is configured to select the distance threshold based on the input signal that balances removal of noise components with preserving a morphological representation of the input signal.

13. The system of claim 11, wherein the processor is configured to decrease by at least one filtering level the initial filtering level to set the desired filtering level when the initial distance metric is at least equal to the distance threshold.

14. The system of claim 11, wherein the processor is configured to determine another distance metric corresponding to the initial filtering level incremented by one filtering level.

15. The system of claim 14, wherein at least one of the initial distance metric and the another distance metric is determined using D(i_n)=|y_in(n)−

x(n)|D(i_n)=|y_in(n)−

x(n)|, where i_n represents a smoothing index, D(i_n) represents a distance metric, y_in(n) represents an nth sample smoothed signal, and x(n) represents an nth sample input signal.

16. The system of claim 14, wherein at least one of the initial distance metric and the another distance metric is determined using $D$

(in)=14

∑

m=n-2n+1

 





yin

(m)-x

(m)

,

where i_n represents a smoothing index, D(i_n) represents a distance metric, y_in represents a sample smoothed signal, and x represents a sample input signal.

17. The system of claim 14, wherein the processor is configured to increase by at least one filtering level the initial filtering level to set the desired filtering level when the another distance metric is smaller than the distance threshold.

18. A system for filtering an input signal, comprising:

means for receiving the input signal;

means for generating an output signal representative of a filtered version of the input signal by adaptively removing noise components from the input signal; and

means for determining a desired filtering level from a number of filtering levels for each discrete sample of the input signal,

wherein the input signal includes a number of discrete samples, and the means for determining includes means for calculating an initial distance metric corresponding to an initial filtering level for each of the discrete samples of the input signal.

19. The system of claim 18, wherein the initial filtering level is a desired filtering level for an immediately preceding sample.

20. The system of claim 18, further comprising a distance threshold representing a desired amount of signal smoothing, wherein the means for determining includes means for comparing the initial distance metric against the distance threshold.

21. The system of claim 20, wherein the processor is adapted to select the distance threshold based on the input signal that balances removal of noise components with preserving a morphological representation of the input signal.

22. The system of claim 20, wherein the means for determining includes means for decrementing the initial filtering level by at least one filtering level to become the desired filtering level when the initial distance metric is at least equal to the distance threshold.

23. The system of claim 20, wherein the means for determining includes means for calculating another distance metric corresponding to the initial filtering level incremented by one filtering level.

24. The system of claim 23, wherein at least one of the initial distance metric and the another distance metric is determined using D(i_n)=|y_in(n)−

x(n)|D(i_n)=|y_in(n)−

x(n)|D (i_n)=|y_in(n)−

x(n)|, where i_n represents a smoothing index, D(i_n) represents a distance metric, y_in(n) represents an nth sample smoothed signal, and x(n) represents an nth sample input signal.

25. The system of claim 23, wherein at least one of the initial distance metric and the another distance metric is determined using $D$

(in)=14

∑

m=n-2n+1

 





yin

(m)-x

(m)

,

where i_n represents a smoothing index, D(i_n) represents a distance metric, y_in represents a sample smoothed signal, and x represents a sample input signal.

26. The system of claim 23, wherein the means for determining includes means for incrementing the initial filtering level by at least one filtering level to become the desired filtering level when the another distance metric is smaller than the distance threshold.