Apparatus and method for processing signal

• US 9,314,176 B2
• Filed: 04/19/2012
• Issued: 04/19/2016
• Est. Priority Date: 05/31/2011
• Status: Active Grant

• Current Assignee: Industrial Technology Research Institute
• Original Assignee: Industrial Technology Research Institute
• Inventors: Pao, Sun-Hua, Hsieh, Tsung-Min, Shau, Yio-Wha, Young, Chieh-Neng
• Primary Examiner(s): BHAT, ADITYA S
• Application Number:
  US13/450,930
• Publication Number:
  US 20120310600A1
• Time in Patent Office: 1,461 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: A61B 2560/0475 Special features of memory means, e.g. removable memory cards
• CPC Class Codes: A61B 5/30 Input circuits therefor
• CPC Class Codes: A61B 5/316 Modalities, i.e. specific diagnostic methods
• CPC Class Codes: A61B 5/335 using integrated circuit memory devices
• CPC Class Codes: A61B 5/725 using specific filters therefor, e.g. Kalman or adaptive filters A61B5/316 take precedence
• CPC Class Codes: H03H 17/0261 Non linear filters

Abstract

An apparatus and a method for processing signal are provided. The signal processing apparatus comprises an input interface and a processing unit. The input interface receives smoothing parameters and a to-be-separated signal. The processing unit establishes an upper extreme envelope and a lower extreme envelope of the to-be-separated signal, and calculates a mean envelope between the upper extreme envelope and the lower extreme envelope. The processing unit performs smoothing according to the smoothing parameters and the mean envelope to generate a smoothed mean envelope, and determines a trend component or a non-trend component according to the smoothed mean envelope.

Claims

1. A signal processing method, comprising:

receiving smoothing parameters and a to-be-separated signal, wherein the to-be-separated signal is generated according to a physiological signal, and the smoothing parameters are adjusted according to the physiological signal;

establishing an upper extreme envelope and a lower extreme envelope of the to-be-separated signal, wherein the upper extreme envelope and the lower extreme envelope respectively represent every local maxima and every local minima of the to-be-separated signal;

calculating a mean envelope between the upper extreme envelope and the lower extreme envelope;

performing smoothing according to the smoothing parameters and the mean envelope to generate a smoothed mean envelope;

determining a trend component or a non-trend component of the to-be-separated signal according to the smoothed mean envelope, wherein the trend component represents tendency of the to-be-separated signal, and the non-trend component of the to-be-separated signal represents fluctuation on the trend component; and

using the trend component to represent a physiological signal of a patient representative of a health aspect of the patient.

2. The signal processing method according to claim 1, wherein the establishing step comprises:

searching for a plurality of upper extremes and lower extremes of the to-be-separated signal;

establishing the upper extreme envelope according to the upper extremes; and

establishing the lower extreme envelope according to the lower extremes.

3. The signal processing method according to claim 1, wherein the smoothing parameters comprise a smooth window width, a smoothing iteration and a boundary processing method.

4. The signal processing method according to claim 3, wherein the boundary processing method is reflection symmetry or 2-fold rotational symmetry.

5. The signal processing method according to claim 1, wherein the determination step comprises:

determining whether the smoothed mean envelope converges;

repeating the establishing step, the calculation step and the smoothing step after removing the smoothed mean envelope from the to-be-separated signal if the smoothed mean envelope does not converge; and

determining the trend component or the non-trend component of the to-be-separated signal according to the smoothed mean envelope if the smoothed mean envelope converges.

6. The signal processing method according to claim 5, wherein the determination step comprises:

determining whether the standard deviation of the smoothed mean envelope is smaller than a default threshold;

determining that the smoothed mean envelope converges if the standard deviation of the smoothed mean envelope is smaller than the default threshold; and

determining that the smoothed mean envelope does not converge if the standard deviation of the smoothed mean envelope is not smaller than the default threshold.

7. The signal processing method according to claim 5, wherein in the determination step, whether the smoothed mean envelope converges is determined according to Cauchy convergence test.

8. The signal processing method according to claim 1, wherein the trend component accumulates smoothed mean envelopes.

9. The signal processing method according to claim 1, wherein the trend component is obtained from a weight computation according to the difference between smoothed mean envelopes and the to-be-separated signal.

10. A signal processing apparatus, comprising:

an input interface used for receiving smoothing parameters and a to-be-separated signal, wherein the to-be-separated signal is generated according to a physiological signal, and the smoothing parameters are adjusted according to the physiological signal; and

a processing unit used for

establishing an upper extreme envelope and a lower extreme envelope of the to-be-separated signal, wherein the upper extreme envelope and the lower extreme envelope respectively represent every local maxima and every local minima of the to-be-separated signal,

calculating a mean envelope between the upper extreme envelope and the lower extreme envelope, wherein the processing unit performs smoothing according to the smoothing parameters and the mean envelope to generate a smoothed mean envelope, and determines a trend component or a non-trend component of the to-be-separated signal according to the smoothed mean envelope, wherein the trend component represents tendency of the to-be-separated signal, and the non-trend component of the to-be-separated signal represents fluctuation on the trend component; and

using the trend component to represent a physiological signal of a patient representative of a health aspect of the patient.

11. The signal processing apparatus according to claim 10, wherein the processing unit searches for a plurality of upper extremes and lower extremes of the to-be-separated signal, establishes the upper extreme envelope according to the upper extremes, and establishes the lower extreme envelope according to the lower extremes.

12. The signal processing apparatus according to claim 10, wherein the smoothing parameters comprise a smooth window width, a smoothing iteration and a boundary processing method.

13. The signal processing apparatus according to claim 12, wherein the boundary processing method is reflection symmetry or 2-fold rotational symmetry.

14. The signal processing apparatus according to claim 10, wherein the processing unit determines whether the smoothed mean envelope converges, if the smoothed mean envelope does not converge, then the processing unit re-establishes the upper extreme envelope and the lower extreme envelope and re-calculates the mean envelope after removing the smoothed mean envelope from the to-be-separated signal, and performs smoothing according to the smoothing parameters and the re-calculated mean envelope to generate a new smoothed mean envelope, and if the new smoothed mean envelope converges, then the processing unit determines the trend component or the non-trend component according to the smoothed mean envelope.

15. The signal processing apparatus according to claim 14, wherein the processing unit determines whether the standard deviation of the smoothed mean envelope is smaller than a default threshold, if the standard deviation of the smoothed mean envelope is smaller than the default threshold, then the processing unit determines that the mean envelope converges, and if the standard deviation of the smoothed mean envelope is not smaller than the default threshold, then the processing unit determines that the mean envelope does not converge.

16. The signal processing apparatus according to claim 14, wherein the processing unit determines whether the smoothed mean envelope converges according to Cauchy convergence test.

17. The signal processing apparatus according to claim 10, wherein the trend component accumulates smoothed mean envelopes.

18. The signal processing apparatus according to claim 10, wherein the trend component is obtained from a weight computation according to the difference between smoothed mean envelopes and the to-be-separated signal.

19. The signal processing apparatus according to claim 10, wherein the input interface and the processing unit are integrated as a functional chip.

20. The signal processing apparatus according to claim 19, further comprising a chip carrier used for carrying the functional chip.

21. The signal processing apparatus according to claim 20, wherein the chip carrier comprises:

a signal sensing unit used for generating the to-be-separated signal according to a physiological signal; and

an output interface used for outputting the to-be-separated signal to the input interface.

22. The signal processing apparatus according to claim 21, wherein the signal sensing unit comprises:

a signal sensing element used for measuring the physiological signal;

an analog front end (AFE) coupled to the signal sensing element for amplifying the physiological signal to generate an analog signal; and

an analog-to-digital converter coupled with the output interface and used for converting the analog signal into the to-be-separated signal.

23. The signal processing apparatus according to claim 22, wherein the signal sensing element is a passive sensing element.

24. The signal processing apparatus according to claim 22, wherein the signal sensing element is an active sensing element.

25. The signal processing apparatus according to claim 22, wherein the chip carrier further comprises:

a driving unit used for driving the signal sensing element; and

a digital-to-analog converter used for controlling the driving unit.

26. The signal processing apparatus according to claim 22, wherein the chip carrier further comprises:

a stretchable portion used for extending the signal sensing element.

27. The signal processing apparatus according to claim 22, wherein the chip carrier further comprises:

an adhesive positioning portion set around the signal sensing element.