Wavelet-based analysis of pulse oximetry signals

• US 7,035,679 B2
• Filed: 06/21/2002
• Issued: 04/25/2006
• Est. Priority Date: 06/22/2001
• Status: Expired due to Term

• Current Assignee: Nellcor Puritan Bennett Ireland
• Original Assignee: CardioDigital Ltd.
• Inventors: Addison, Paul Stanley, Watson, James Nicholas
• Primary Examiner(s): WINAKUR, ERIC FRANK
• Application Number:
  US10/480,983
• Publication Number:
  US 20050070774A1
• Time in Patent Office: 1,404 Days
• Field of Search: 600323- 24, 600/324, 600/330, 600/310, 600/322, 600/336, 600/529
• US Class Current: 600/323
• CPC Class Codes: A61B 5/0205 Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
• CPC Class Codes: A61B 5/14551 for measuring blood gases
• CPC Class Codes: A61B 5/6816 Ear lobe
• CPC Class Codes: A61B 5/6826 Finger
• CPC Class Codes: A61B 5/6838 Clamps or clips
• CPC Class Codes: A61B 5/7207 of noise induced by motion artifacts
• CPC Class Codes: A61B 5/726 using Wavelet transforms
• CPC Class Codes: A61B 5/7264 Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00
• CPC Class Codes: A61B 5/742 using visual displays A61B5/339 takes precedence
• CPC Class Codes: G06F 2218/00 Aspects of pattern recognition specially adapted for signal processing

Abstract

A pulse oximetry signal, suitably a photoplethysmogram (PPG), is decomposed by wavelet transform techniques, and the decomposed signal analysed to provide selected physiological data. The signal may be processed to remove noise, artefacts, or transient features. Information on respiration may also be recovered.

Claims

1. A method of measuring physiological parameters, comprising the steps of:

using a pulse oximeter to obtain a pulse oximetry signal;

decomposing the pulse oximetry signal by wavelet transform analysis;

deriving a continuous wavelet transform surface;

plotting the surface against a location parameter and a scale parameter; and

following a feature on the plotted surface to derive one or more physiological parameters.

2. A method according to claim 1, in which the pulse oximetry signal is a photoplethysmogram (PPG).

3. A method as claimed in claim 1, in which the step of deriving a continuous wavelet transform surface comprises deriving the wavelet energy surfaces of the pulse oximeter signal.

4. A method as claimed in claim 1, in which the step of deriving a continuous wavelet transform surface includes the step of deriving the wavelet transform modulus of the pulse oximeter signal.

5. A method as claimed in claim 1, in which the scale parameter is a characteristic frequency of the wavelet used in the decomposition.

6. A method as claimed in claim 1, in which the scale parameter is the wavelet dilation.

7. A method as claimed in claim 1, including visually displaying information derived from the pulse oximetry signal by the wavelet transform analysis.

8. A method as claimed in claim 7, in which said information is displayed in real time.

9. A method as claimed in claim 7, in which said information includes one or more of:

the distribution of energies within the pulse oximetry signal, coherent structures in the processed signal, a contour plot of the decomposed waveform, a surface plot of the decomposed waveform, and a 2D or 3D energy scalogram.

10. A method according to claim 7, in which the unprocessed pulse oximetry signal is also displayed.

11. A method according to claim 1, in which the step of following a feature on the plotted surface comprises identifying a ridge caused by respiration, and following said ridge to derive information relating to respiration.

12. A method according to claim 11, in which respiration information is derived from high amplitude ridges using ridge-following methods.

13. A method according to claim 11, in which the respiration information is derived by analysis of amplitude or frequency modulation.

14. A method according to claim 11, in which the information relating to respiration comprises a breathing rate.

15. A method according to claim 14, in which respiration information is derived from high amplitude ridges using ridge-following methods.

16. A method according to claim 14, in which the respiration information is derived by analysis of amplitude or frequency modulation.

17. A method according to claim 11, wherein said ridge following comprises the step of extracting phase information along the ridge to identify and/or analyze individual breaths.

18. A method according to claim 17, in which the respiration information is derived by cross scale correlation of phase values.

19. A method according to claim 1, in which decomposing the pulse oximetry signal is effective to remove, at least one of noise, artefact, and transient features.

20. A method according to claim 19, in which said removal employs inverse transformation of the cropped transform.

21. A method according to claim 19, in which said removal employs wavelet ridge methods.

22. A method according to claim 19, in which said removal employs modulus maxima methods.

23. A method according to claim 1, in which information from the transform is used to determine the present or predicted severity of illness of a subject.

24. A method according to claim 23, in which said information from the transform comprises features derived from the wavelet transform at a selected frequency level, and wherein said features are used as an input to a classification algorithm for the determination of the present or predicted severity of illness of a subject.

25. A method according to claim 24, wherein the features comprise at least one from the group including power, mean, skew, kurtosis and entropy.

26. A method according to claim 1, wherein the step of following a feature on the plotted surface comprises following the loci of maxima and/or minima of features on the surface.

27. A method according to claim 26, wherein the step of following the loci of maxima and/or minima of features on the surface comprises a ridge following technique.

28. A method according to claim 26, wherein the step of following the loci of maxima and/or minima of features on the surface comprises a modulus maxima technique.

29. A method according to claim 1, comprising the steps of identifying modulus maxima on the transform surface, using the modulus maxima following to determine the location and nature of pertinent points in the PPG, and determining a subject'"'"'s arterial compliance based on the pertinent points.

30. A physiological measurement system comprising:

a pulse oximeter which includes an optical probe and circuit means connected to the probe to derive a pulse oximetry signal from a subject when the probe is applied to the subject, and signal processing means arranged to;

receive the pulse oximetry signal;

decompose the pulse oximetry signal by wavelet transform analysis, said wavelet transform analysis employing a continuous wavelet function;

derive a wavelet time-scale energy surface;

plot the surface against a location parameter and a scale parameter; and

follow a feature on the plotted surface to derive one or more physiological parameters.

31. A system according to claim 30, further including a visual display unit operable to display the pulse oximetry signal and information derived therefrom in real time.

32. A method of measuring physiological parameters, comprising the steps of:

using a pulse oximeter to obtain a pulse oximetry signal;

decomposing the pulse oximetry signal by wavelet transform analysis;

deriving a continuous wavelet transform surface;

plotting the surface against a location parameter and a scale parameter;

identifying one or more ridges caused by respiration, then;

(a) identifying and following high amplitude ridges;

(b) extracting phase information from along the ridge;

(c) analyzing amplitude modulation;

(d) analyzing frequency modulation; and

(e) deriving information relating to respiration by polling between the results of methods (a) to(d).

33. The method of claim 32, in which the information relating to respiration comprises a breathing rate.

34. The method of claim 32, in which the information relating to respiration comprises the identification and/or analysis of individual breaths.

35. The method of claim 32, wherein the information relating to respiration is derived by cross scale correlation of phase values.