Method and apparatus for determining a respiration signal

• US 9,675,282 B2
• Filed: 02/07/2011
• Issued: 06/13/2017
• Est. Priority Date: 02/11/2010
• Status: Active Grant

• Current Assignee: Koninklijke Philips Electronics N.V. (Koninklijke Philips N.V.)
• Original Assignee: Koninklijke Philips N.V.
• Inventors: Morren, Geert Guy Georges
• Primary Examiner(s): Cheng, Jacqueline
• Assistant Examiner(s): PORTILLO, JAIRO H
• Application Number:
  US13/575,969
• Publication Number:
  US 20120296221A1
• Time in Patent Office: 2,318 Days
• Field of Search: None
• CPC Class Codes: A61B 2562/0219 Inertial sensors, e.g. accelerometers, gyroscopes, tilt switches
• CPC Class Codes: A61B 5/113 occurring during breathing
• CPC Class Codes: A61B 5/721 using a separate sensor to detect motion or using motion information derived from signals other than the physiological signal to be measured

Abstract

The invention relates to a method and apparatus for determining a respiration of a subject (305) in which, with a single multi-axial accelerometer (310) positioned on a body of the subject (305), accelerometer signals are generated (101) indicative of the acceleration of the subject (305) along different spatial axes, a vector magnitude signal of the acceleration of the subject (305) along the different spatial axes is calculated (102) from the accelerometer signals, a non-respiratory motion contribution to the acceleration along the different spatial axes is identified (103, 203) from the vector magnitude signal, which non-respiratory motion contribution is not caused by the respiration, and a respiration signal indicative of the respiration of the subject is determined (104, 204) by filtering the non-respiratory motion contribution from at least one of the accelerometer signals. In this way a method is provided which determines the respiration of a subject (305) with a single accelerometer (310) in an efficient and, for a patient, comfortable way.

Claims

1. A method for monitoring a respiration of a subject comprising the steps of:

generating, with a single multi-axial accelerometer positioned on a body of the subject, accelerometer signals indicative of the acceleration of the subject along different spatial axes,

calculating, with at least one processor, a vector magnitude signal of the acceleration of the subject as a magnitude of the vector sum of the accelerometer signals along the different spatial axes,

identifying, with the at least one processor, from the vector magnitude signal of the acceleration a non-respiratory motion contribution to the acceleration along the different spatial axes which non-respiratory motion contribution is not caused by the respiration, the non-respiratory motion being indicative of at least a motion of a thorax due to cardiac activity of the subject,

filtering, with the at least one processor, the non-respiratory motion contribution from at least one of the accelerometer signals;

determining, with the at least one processor, a respiration signal indicative of the respiration of the subject with the at least one filtered accelerometer signal; and

displaying the determined respiration signal on a display.

2. The method as defined in claim 1, wherein the step of determining the respiration signal includes the steps of:

filtering the non-respiratory motion contribution from each of the accelerometer signals separately,

determining the respiration signal from a combination of the filtered accelerometer signals.

3. The method as defined in claim 1, wherein the step of identifying the non-respiratory motion contribution comprises the steps of:

calculating a power spectrum of the vector magnitude signal,

extracting a characteristic frequency of the non-respiratory motion from the power spectrum,

and wherein the step of determining the respiration signal comprises a step of filtering the characteristic frequency from the accelerometer signals with an adaptive notch filter.

4. The method as defined in claim 1, wherein the step of identifying the non-respiratory motion contribution comprises the steps of:

calculating a coherence spectrum of the vector magnitude signal and one of the accelerometer signals,

extracting a characteristic frequency of the non-respiratory motion contribution from the coherence spectrum,

and wherein the step of determining the respiration signal comprises a step of filtering the characteristic frequency from the accelerometer signals with an adaptive notch filter.

5. The method as defined in claim 1, further comprising a step of extracting a respiration rate of the subject from the respiration signal.

6. The method as defined in claim 1, further comprising a step of filtering a frequency range from the vector magnitude signal which filtered vector magnitude signal is used in the step of identifying the non-respiratory motion contribution.

7. The method as defined in claim 1, wherein the step of identifying the non-respiratory motion contribution comprises a step of extracting a characteristic frequency of the non-respiratory motion contribution from the vector magnitude signal.

8. The method as defined in claim 7, wherein the characteristic frequency of the non-respiratory motion contribution comprises a heart beat frequency of the subject.

9. The method as defined in claim 7, wherein the characteristic frequency of the non-respiratory motion contribution comprises a step frequency of a moving subject.

10. The method as defined in claim 1, wherein the step of identifying the non-respiratory motion contribution comprises a step of extracting a noise reference signal representative for the unwanted noise contribution from the vector magnitude signal.

11. The method as defined in claim 10, wherein the noise reference signal is extracted from the vector magnitude signal with a digital filtering technique.

12. The method as defined in claim 10, wherein the noise reference signal comprises a cardiac interference signal.

13. The method as defined in claim 10, wherein the step of determining the respiration signal comprises a step of filtering the accelerometer signals with an adaptive noise filter with the noise reference signal.

14. The method as defined in claim 10, wherein the step of identifying the non-respiratory motion contribution further comprises a step of extracting a characteristic frequency of the non-respiratory motion contribution from the noise reference signal and wherein the step of determining the respiration signal comprises a step of filtering the characteristic frequency from the accelerometer signals with an adaptive notch filter.

15. A respiration monitoring apparatus for monitoring a respiration of a subject, wherein the respiration monitoring apparatus comprises:

a single multi-axial accelerometer configured to be positioned on a body of the subject, wherein the multi-axial accelerometer is adapted to generate accelerometer signals x(t), y(t), and z(t) indicative of the acceleration of the subject along orthogonal x, y, and z spatial axes respectively,

a signal processing unit programmed to calculate a vector magnitude signal m(t) of the acceleration of the subject as m(t)=√

{square root over (x(t)²+y(t)²+z(t)²)} and for identifying a non-respiratory motion contribution to the acceleration along different spatial axes from the vector magnitude signal m(t), the non-respiratory motion being indicative of at least a motion of a thorax due to a cardiac activity of the subject,

a respiration signal determination unit programmed to;

filter the non-respiratory motion contribution from at least one of the accelerometer signals; and

determine a respiration signal indicative of the respiration of the subject with the at least one filtered accelerometer signal; and

a display on which the determined respiration signal is displayed.

16. The respiration monitoring apparatus of claim 15, wherein the respiration signal determination unit is further programmed to:

filter the non-respiratory motion contribution from each of the accelerometer signals separately,

determine the respiration signal from a combination of the filtered accelerometer signals.

17. The respiration monitoring apparatus of claim 15, wherein the signal processing unit is further programmed to

extract a characteristic frequency of the non-respiratory motion contribution from the vector magnitude signal;

wherein the characteristic frequency of the non-respiratory motion contribution comprises a heart beat frequency of the subject.

18. The respiration monitoring apparatus of claim 15, wherein the signal processing unit is further programmed to:

extract a noise reference signal representative for the unwanted noise contribution from the vector magnitude signal

wherein the noise reference signal comprises a cardiac interference signal.

19. A respiration monitoring apparatus for determining a respiration of a subject, wherein the respiration monitoring apparatus comprises:

a single multi-axial accelerometer configured to be positioned on a body of the subject, wherein the multi-axial accelerometer is adapted to generate accelerometer signals indicative of the acceleration of the subject along different spatial axes,

a signal processor programmed to calculate a vector magnitude signal of the acceleration of the subject as a magnitude of the vector sum of the accelerometer signals along the different spatial axes and for identifying a non-respiratory motion contribution to the acceleration along different spatial axes from the vector magnitude signal,

a respiration signal determination processor programmed to filter the non-respiratory motion contribution from at least one of the accelerometer signals, and determine a respiration signal indicative of the respiration of the subject with the at least one filtered accelerometer signal; and

a display on which the determined respiration signal is displayed;

wherein the non-respiratory motion contribution comprises a cardiac interference signal.

20. The respiration monitoring apparatus of claim 19, wherein the respiration signal determination processor is further programmed to:

filter the non-respiratory motion contribution from each of the accelerometer signals separately,

determine the respiration signal from a combination of the filtered accelerometer signals.