Method for reducing baseline drift in a biological signal

• US 8,301,230 B2
• Filed: 09/25/2008
• Issued: 10/30/2012
• Est. Priority Date: 09/25/2008
• Status: Expired due to Fees

• Current Assignee: Air Products and Chemicals Incorporated (Linde Plc)
• Original Assignee: Air Products and Chemicals Incorporated (Linde Plc)
• Inventors: Toth, Michael S.
• Primary Examiner(s): Heller, Tammie K
• Application Number:
  US12/284,932
• Publication Number:
  US 20100076329A1
• Time in Patent Office: 1,496 Days
• Field of Search: 600/509, 607/18
• US Class Current: 600/509
• CPC Class Codes: A61B 5/0006 ECG or EEG signals
• CPC Class Codes: A61B 5/316 Modalities, i.e. specific diagnostic methods
• CPC Class Codes: A61B 5/352 Detecting R peaks, e.g. for synchronising diagnostic apparatus; Estimating R-R interval

Abstract

A computer-based method for reducing or eliminating baseline drift from a biological (bio) signal includes the steps of dividing the bio signal into a plurality of shorter signals having fixed time intervals, fitting a corresponding portion of a baseline function to a baseline of a respective one of each of the shorter signals, and subtracting the baseline function from the bio signal, resulting in a bio signal with a flat baseline.

Claims

1. A computer-based method for reducing baseline drift in a biological (bio) signal, the method comprising:

dividing the bio signal into a plurality of shorter signals having fixed time intervals;

fitting a corresponding portion of a baseline function to a baseline of a respective one of each of the shorter signals;

subtracting the baseline function from the bio signal, thereby flattening the baseline of the bio signal;

setting a total number of the time intervals to be equal to a total number of samples of the bio signal contained in a single cycle of a first frequency value associated with the baseline drift; and

repeating the steps of dividing, fitting, subtracting and setting, while substituting a second frequency value associated with the baseline drift for the first frequency value.

2. The method of claim 1, wherein the first frequency value lies in a range from 0.5 Hertz to 150 Hertz.

3. The method of claim 1, wherein the fitting is performed using cubic-spline interpolation.

4. The method of claim 1, wherein the fitting is performed using a least squares fit.

5. The method of claim 1, wherein the second frequency value is of higher value than the first frequency value.

6. The method of claim 1, wherein the steps are repeated based on an evaluation of a flatness of the bio signal after the subtracting is performed.

7. The method of claim 1, wherein the bio signal is an electrocardiogram signal.

8. The method of claim 1, in which baseline drift is eliminated resulting in a flat baseline of the bio signal.

9. The method of claim 1, wherein the first and the second frequency values are selected from a frequency range in which, independent of the bio signal and how the bio signal is sampled, the baseline drift is expected to occur.

10. A computer-based method for reducing baseline drift in a biological (bio) signal, the method comprising:

dividing the bio signal into a plurality of shorter signals having fixed time intervals;

fitting a baseline function to a baseline of each of the shorter signals;

subtracting the baseline function from the bio signal, thereby flattening the baseline of the bio signal;

setting a total number of the time intervals to be equal to a total number of samples of the bio signal contained in a single cycle of a first frequency value associated with the baseline drift;

repeating the steps of dividing, fitting, subtracting and setting, while substituting a second frequency value associated with the baseline drift for the first frequency value, wherein the steps are repeated based on an evaluation of a flatness of the bio signal after the subtracting is performed; and

evaluating the flatness by fitting an evaluation function to the bio signal and examining a magnitude of the evaluation function.

11. The method of claim 10, further comprising:

repeating the steps when the magnitude exceeds a predetermined threshold value.

12. A computer-based method for reducing baseline drift in a biological (bio) signal, the method comprising:

dividing the bio signal into a plurality of shorter signals having fixed time intervals;

fitting a baseline function to a baseline of each of the shorter signals;

subtracting the baseline function from the bio signal, thereby flattening the baseline of the bio signal;

setting a total number of the time intervals to be equal to a total number of samples of the bio signal contained in a single cycle of a first frequency value associated with the baseline drift; and

repeating the steps of dividing, fitting, subtracting and setting, while substituting a second frequency value associated with the baseline drift for the first frequency value, wherein the steps are repeated as a function of a comparison between a variation in a value of the baseline function to a resolution of an analog-to-digital converter used to generate the bio signal.

13. The method of claim 12, further comprising:

repeating the steps when the variation exceeds the resolution.

14. A device for reducing baseline drift in a biological (bio) signal, the device being constructed and arranged to:

divide the bio signal into a plurality of shorter signals having fixed time intervals;

fit a corresponding portion of a baseline function to a baseline of a respective one of each of the shorter signals;

subtract the baseline function from the bio signal, thereby flattening the baseline of the bio signal; and

set a total number of the time intervals to be equal to a total number of samples of the bio signal contained in a single cycle of a first frequency value associated with the baseline drift, the device comprising;

a communications arrangement configured to receive the bio signal; and

a processor configured to divide the bio signal into a plurality of shorter signals having fixed time intervals, fit the corresponding portion of the baseline function to the baseline of the respective one of each of the shorter signals, subtract the baseline function from the bio signal, thereby flattening the baseline of the bio signal, and set the total number of the time intervals to be equal to the total number of samples of the bio signal contained in the single cycle of the first frequency value associated with the baseline drift; and

a memory including instructions configuring the processor to divide the bio signal into the plurality of shorter signals having fixed time intervals, fit the corresponding portion of the baseline function to the baseline of the respective one of each of the shorter signals, subtract the baseline function from the bio signal, thereby flattening the baseline of the bio signal, and set the total number of the time intervals to be equal to the total number of samples of the bio signal contained in the single cycle of the first frequency value associated with the baseline drift,

wherein the instructions direct the processor to;

repeat the steps of dividing, fitting, subtracting and setting, while substituting a second frequency value associated with the baseline drift for the first frequency value.

15. The device of claim 14, wherein the first frequency value lies in a range from 0.5 Hertz to 150 Hertz.

16. The device of claim 14, wherein the processor is configured to perform the fitting using cubic-spline interpolation.

17. The device of claim 14, wherein the processor is configured to perform the fitting using a least squares fit.

18. The device of claim 14, wherein the second frequency value is of higher value than the first frequency value.

19. The device of claim 14, wherein the processor is configured to repeat the steps based on an evaluation of a flatness of the bio signal after performing the subtracting.

20. The device of claim 14, wherein the bio signal is an electrocardiogram signal.

21. The device of claim 14, wherein the first and the second frequency values are selected from a frequency range in which, independent of the bio signal and how the bio signal is sampled, the baseline drift is expected to occur.

22. A device for reducing baseline drift in a biological (bio) signal, the device being constructed and arranged to:

divide the bio signal into a plurality of shorter signals having fixed time intervals;

fit a baseline function to a baseline of each of the shorter signals;

subtract the baseline function from the bio signal, thereby flattening the baseline of the bio signal; and

set a total number of the time intervals to be equal to a total number of samples of the bio signal contained in a single cycle of a first frequency value associated with the baseline drift, the device comprising;

a communications arrangement configured to receive the bio signal;

a processor configured to divide the bio signal into the plurality of shorter signals having fixed time intervals, fit the baseline function to the baseline of each of the shorter signals, subtract the baseline function from the bio signal, thereby flattening the baseline of the bio signal, and set the total number of the time intervals to be equal to the total number of samples of the bio signal contained in the single cycle of the first frequency value associated with the baseline drift; and

a memory including instructions configuring the processor to divide the bio signal into the plurality of shorter signals having fixed time intervals, fit the baseline function to the baseline of each of the shorter signals, subtract the baseline function from the bio signal, thereby flattening the baseline of the bio signal, and set the total number of the time intervals to be equal to the total number of samples of the bio signal contained in the single cycle of the first frequency value associated with the baseline drift,

wherein the processor is configured to;

repeat the steps of dividing, fitting, subtracting and setting, while substituting a second frequency value associated with the baseline drift for the first frequency value, wherein the steps are repeated based on an evaluation of a flatness of the bio signal after performing the subtracting, and

evaluate the flatness by fitting an evaluation function to the bio signal and examining a magnitude of the evaluation function.

23. The device of claim 22, wherein the processor is configured to repeat the steps when the magnitude exceeds a predetermined threshold value.

24. A device for reducing baseline drift in a biological (bio) signal, the device being constructed and arranged to:

divide the bio signal into a plurality of shorter signals having fixed time intervals;

fit a baseline function to a baseline of each of the shorter signals;

subtract the baseline function from the bio signal, thereby flattening the baseline of the bio signal; and

set a total number of the time intervals to be equal to a total number of samples of the bio signal contained in a single cycle of a first frequency value associated with the baseline drift, the device comprising;

a communications arrangement configured to receive the bio signal;

a processor configured to divide the bio signal into the plurality of shorter signals having fixed time intervals, fit the baseline function to the baseline of each of the shorter signals, subtract the baseline function from the bio signal, thereby flattening the baseline of the bio signal, and set the total number of the time intervals to be equal to the total number of samples of the bio signal contained in the single cycle of the first frequency value associated with the baseline drift; and

a memory including instructions configuring the processor to divide the bio signal into the plurality of shorter signals having fixed time intervals, fit the baseline function to the baseline of each of the shorter signals, subtract the baseline function from the bio signal, thereby flattening the baseline of the bio signal, and set the total number of the time intervals to be equal to the total number of samples of the bio signal contained in the single cycle of the first frequency value associated with the baseline drift,

wherein the processor is configured to repeat the steps of dividing, fitting, subtracting and setting, while substituting a second frequency value associated with the baseline drift for the first frequency value, and wherein the steps are repeated as a function of a comparison between a variation in a value of the baseline function to a resolution of an analog-to-digital converter used to generate the bio signal.

25. The device of claim 24, wherein the processor is configured to repeat the steps when the variation exceeds the resolution.

26. A computer-readable storage medium having stored thereon a series of instructions executable by a processor for removing noise from a digital signal, the instructions configured to cause the processor to perform the steps of:

dividing the bio signal into a plurality of shorter signals having fixed time intervals;

fitting a corresponding portion of a baseline function to a baseline of a respective one of each of the shorter signals;

subtracting the baseline function from the bio signal, thereby flattening the baseline of the bio signal;

setting a total number of the time intervals to be equal to a total number of samples of the bio signal contained in a single cycle of a first frequency value associated with the baseline drift; and

repeating the steps of dividing, fitting, subtracting and setting, while substituting a second frequency value associated with the baseline drift for the first frequency value.

27. The computer-readable storage medium of claim 26, wherein the first and the second frequency values are selected from a frequency range in which, independent of the bio signal and how the bio signal is sampled, the baseline drift is expected to occur.