Wavelet transform of a plethysmographic signal

• US 20070004977A1
• Filed: 06/29/2005
• Published: 01/04/2007
• Est. Priority Date: 06/29/2005
• Status: Active Grant

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: Norris, Mark
• Granted Patent: US 7,515,949 B2
• Time in Patent Office: Days
• US Class Current: 600/336
• CPC Class Codes: A61B 5/14551 for measuring blood gases
• CPC Class Codes: A61B 5/7203 for noise prevention, reduction or removal
• CPC Class Codes: A61B 5/7207 of noise induced by motion artifacts
• CPC Class Codes: A61B 5/726 using Wavelet transforms

Abstract

A photoplethysmographic system and method is provided for filtering a photoplethysmographic (pleth) signal to reduce the effects of noise in the signal. The system and method utilize a combination of frequency, time and/or magnitude information, to identify and separate transient signal components within a pleth signal from repeating signal components within the pleth signal. Typically, signal components of interest repeat over a period that corresponds with a patient'"'"'s heartbeat. Such periodically repeating signals may be identified as stationary signals/objects within a frequency and time-based analysis. In contrast, motion artifacts or other sources of noise are often isolated (i.e., non-repeating) transient events and may be identified as non-stationary objects in a frequency and time-based analysis. Data associated with identified transient events may be filtered from or otherwise removed from a given signal. In this regard, a pleth signal may be cleansed prior to its use for, e.g., blood analyte determinations.

Claims

1. A method for use in pulse oximetry, comprising the steps of:

receiving a time-based signal reflective of one or more optical signals incident on a detector of a pulse oximeter, said time-based signal being attenuated by patient tissue;

transforming a series of time periods of said time-based signal to generate a series of transformed signals each having at least a frequency component;

monitoring said series of transformed signals to identify one or more frequency bands of one or more of said transformed signals that is potentially corrupted by artifact;

upon identifying one or more of said frequency bands that is potentially corrupted by artifact for at least one of said series of transformed signals, removing said one or more frequency bands from said at least one of said series of transformed signals to generate at least one modified transformed signal; and

processing said at least one modified transformed signal to obtain physiological information regarding said patient tissue.

2. The method of claim 1, wherein processing comprises performing an inverse transform on said at least one modified transformed signal to generate a modified time-based signal.

3. The method of claim 1, wherein obtaining comprises identifying a blood analyte value.

4. The method of claim 1, wherein said transforming step comprises generating transformed signals having at least a frequency component and a time component.

5. The method of claim 4, wherein said transforming step comprises performing a wavelet transform.

6. The method of claim 1, wherein frequency bands potentially corrupted by artifact for said series of transformed signals are substantially removed from each said modified transformed signal.

7. The method of claim 1, wherein a first frequency band is removed from a first transformed signal of said series of transformed signals and a second frequency band is removed from a second transformed signal of said series of transformed signals, wherein said first and second frequency bands are different.

8. The method of claim 1, wherein said receiving step comprises:

receiving first and second optical signals having different wavelengths; and

separating said first and second optical signals into first and second time-based signals, wherein said transforming, monitoring, removing and processing steps are individually performed on said first and second time-based signals.

9. The method of claim 8, wherein said receiving step comprises receiving red and infrared optical signals.

10. The method of claim 1, further comprising:

performing a second transform on said time-based signal for a time window including said series of time periods to generate a transformed base signal, wherein said transformed base signal includes at least a frequency component and a time component.

11. The method of claim 10, wherein said monitoring step further comprises:

comparing each of said series of transformed signals to a corresponding time and frequency component of said transformed base signal.

12. The method of claim 11, wherein said comparing step identifies objects within each of said series of transformed signals as at least one of:

stationary objects in relation to said transformed base signal; and

transient objects in relation to said transformed base signal.

13. The method of claim 12, wherein said transient objects define said frequency band of each said transformed signal that is potentially corrupted by artifact.

14. The method of claim 1, wherein said series of time periods are successive.

15. The method of claim 1, wherein said series of time periods at least partially overlap.

16. A method for use in pulse oximetry, comprising the steps of:

receiving a time-based signal reflective of one or more optical signals incident on a detector of a pulse oximeter;

first generating an energy map for a time window of said time-based signal, wherein said energy map includes time, duration and magnitude information for said time-based signal within said time window;

identifying at least one transient object within said energy map;

removing said at least one transient object from said energy map to generate a modified energy map; and

using said modified energy map to generate physiological patient data.

17. The method of claim 16, further comprising:

second generating a modified time-based signal using said modified energy map.

18. The method of claim 17, further comprising:

using said modified time-based signal to identify a blood analyte value.

19. The method of claim 17, wherein energy associated with said at least one transient object is substantially removed from said modified time-based signal.

20. The method of claim 16, wherein said first generating step comprises performing a first transform on at least a portion of said time window.

21. The method of claim 16, wherein said first generating step comprises performing a first wavelet transform on at least a portion of said time window.

22. The method of claim 21, further comprising:

performing a second wavelet transform on at least a portion of said time window.

23. The method of claim 22, wherein said first wavelet transform comprises a wavelet having a narrow time component relative to frequency and said second wavelet transform comprises a wavelet having a wide time component relative to frequency.

24. A method for use in pulse oximetry, comprising the steps of:

receiving a time-based signal reflective of one or more optical signals incident on a detector of a pulse oximeter;

performing a first transform on said time-based signal over a first time window to generate a first transformed signal having at least a frequency component;

performing a second transform on said time-based signal over a second time window to generate a second transformed signal having at least a frequency component, wherein said second time window is disposed within said first time window;

comparing said first and second transformed signals to identify noise components in said first time window;

subtracting said noise components from said first transformed signal to generate a modified transformed signal; and

using said modified transformed signal to generate physiological patient data.

25. The method of claim 24, further comprising:

inverse transforming said modified transformed signal to generate a modified time-based signal.

26. The method of claim 25, further comprising:

using said modified time-based signal to identify a blood analyte value.

27. A method for use in pulse oximetry, comprising the steps of:

receiving a time-based signal reflective of one or more optical signals incident on a detector of a pulse oximeter;

performing a wavelet transform on said time-based signal to generate a transformed signal having time, frequency and magnitude components;

removing at least a first object from said transformed signal based on at least a first criteria to generate a modified transformed signal; and

using said modified transformed signal to generate physiological patient data.

28. The method of claim 27, further comprising:

inverse transforming said modified transformed signal to generate a modified time-based signal that is substantially free of information associated with said first object.

29. The method of claim 28, further comprising:

using said modified time-based signal to identify a blood analyte value.

30. The method of claim 27, wherein said removing step comprises masking at least one duration of said transformed signal.

31. The method of claim 30, wherein masking comprises at least one of:

graphically removing a portion of an energy map associated with said transformed signal; and

numerically removing a duration of said transformed signal.

32. The method of claim 27, wherein said removing step comprises:

identifying at least one non-repeating peak in said transformed signal; and

subtracting information associated with said non-repeating peak form said transformed signal.

33. A method for use in pulse oximetry, comprising the steps of:

receiving a time-based signal reflective of one or more optical signals incident on a detector of a pulse oximeter;

performing a first wavelet transform on a first portion of said time-based signal, wherein said first wavelet transform utilizes a narrow time window to fit peaks within said time-based-signal;

performing a second wavelet transform on a second portion of said time-based signal, wherein said second wavelet utilizes a wide time window, wherein said wide time window spans at least first and second peaks in said time-based signal;

comparing results of said first and second wavelet transforms to identify transient events in said time-based signal;

generating an output substantially free of said transient events.

34. An apparatus for use in pulse oximetry, comprising:

a port for receiving an electronic signal reflective of one or more optical signals incident on a detector of a pulse oximeter;

a transformer for first transforming said electric signal for a first time window to generate a first transformed signal and for second transforming said electric signal for a second time window to generate a second transformed signal;

a processor operative to;

compare said first and second transformed signals to identify one or more noise components in at least said first time window;

subtract said one or more noise components from said first transformed signal to generate a modified transformed signal; and

use said modified transformed signal to generate physiological patient data.

35. An apparatus as set forth in claim 34, wherein said port comprises input structure associated with a digital signal processing unit.

36. An apparatus as set forth in claim 34, wherein said transformer comprises a wavelet transformer for applying wavelet transforms to said first and second time windows.

37. An apparatus as set forth in claim 36, wherein said wavelet transformer is operative to apply first and second different wavelet transforms to said first and second time windows, respectively.