Apparatus and method for detecting heartbeat using PPG

• US 20030212336A1
• Filed: 04/15/2003
• Published: 11/13/2003
• Est. Priority Date: 04/15/2002
• Status: Active Grant

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: Kim, Kyung Hwan, Lee, Mi Hee, Bang, Seok Won
• Granted Patent: US 6,905,470 B2
• Time in Patent Office: Days
• US Class Current: 600/504
• CPC Class Codes: A61B 5/02416 using photoplethysmograph signals, e.g. generated by infrared radiation A61B5/14552 takes precedence
• CPC Class Codes: A61B 5/7207 of noise induced by motion artifacts
• CPC Class Codes: A61B 5/7221 Determining signal validity, reliability or quality A61B5/7207 and A61B5/7217 take precedence
• CPC Class Codes: A61B 5/726 using Wavelet transforms

Abstract

The present invention relates to an apparatus and method capable of detecting a heartbeat using a photoplethysmography (PPG) even under a motion of a user. A primary object of the present invention is to allow the heartbeat of the user to be effectively detected from a PPG signal even under the motion. The object of the present invention is accomplished by the following process of acquiring the PPG signal from the user through a PPG sensor unit, detecting candidate heartbeat sequences from the PPG signal after calculating an expected mean interbeat interval through a signal processing technique based on the wavelet transformation and an autocorrelation function, and extracting an optimal heartbeat sequence or mean heart rate from the candidate heartbeat sequences.

Claims

1. An apparatus for detecting a heartbeat using a photoplethysmography (PPG), comprising:

a sensor unit for detecting a PPG signal from a body of a user;

a mean interbeat interval estimating unit for calculating an expected mean interbeat interval based on the detected PPG signal; and

a heartbeat sequence/mean heart rate extracting unit for detecting candidate heartbeat sequences from the PPG signal using the expected mean interbeat interval and extracting an optimal heartbeat sequence or a mean heart rate from the candidate heartbeat sequences.

2. The apparatus as claimed in claim 1, wherein the mean interbeat interval estimating unit includes a hard thresholding part for hard thresholding the PPG signal with a predetermined level, an autocorrelation part for calculating an autocorrelation value of the hard thresholded PPG signal, and a peak detecting part for detecting a time when a first peak occurs from the autocorrelation value.

3. The apparatus as claimed in claim 2, wherein the predetermined hard thresholding level of the hard thresholding part is expressed as 0.5×

max(input).

4. The apparatus as claimed in claim 2, wherein the mean interbeat interval estimating unit further includes a clipping part for clipping the PPG signal based on a predetermined level.

5. The apparatus as claimed in claim 1, wherein the heartbeat sequence/mean heart rate extracting unit includes a peak detecting part for detecting peaks from the PPG signal, a candidate heartbeat sequence forming part for forming a plurality of the candidate heartbeat sequences based on the detected peaks, a heartbeat sequence selecting part for selecting the optimal heartbeat sequence among the plurality of the candidate heartbeat sequences, and a mean heat rate calculating part for calculating the mean heart rate using the optimal heartbeat sequences selected by the heartbeat sequence selecting part.

6. The apparatus as claimed in claim 1, further comprising a motion artifact level determining unit for determining a level of a motion artifact in the PPG signal based on the PPG signal detected from the PPG sensor unit so that a warning message is produced when the level of the motion artifact is greater than a predetermined threshold value.

7. The apparatus as claimed in claim 6, further comprising a display unit for causing both the optimal heartbeat sequence or mean heart rate extracted from the heartbeat sequence/mean heart rate extracting unit and the warning message produced from the motion artifact level determining unit to be displayed thereon.

8. The apparatus as claimed in claim 6, wherein the motion artifact level determining unit includes a TEO operating part for calculating an output value related to a rate of change in an amplitude or frequency of the PPG signal, a convolution part for convoluting a signal outputted from the TEO operating part with a predetermined window function, and a motion artifact threshold measuring part for performing a hard thresholding process of a signal inputted from the convolution part based on the predetermined level of the motion artifact so that the warning message is produced when the hard thresholded signal is greater than the threshold value.

9. The apparatus as claimed in claim 1, further comprising a filtering unit for selectively passing only a signal within a specific frequency band among the PPG signal detected by the PPG sensor unit.

10. The apparatus as claimed in claim 1, further comprising a PPG signal approximating unit for decomposing the PPG signal detected from the PPG sensor unit into an approximation signal and detail signals at different scales using wavelet transformation and combining the approximation signal and the detail signals greater than a predetermined threshold value so that a resultant approximated signal can be generated.

11. The apparatus as claimed in claim 10, wherein the PPG signal approximating unit includes a wavelet transforming part for decomposing the PPG signal into the detail signals and the approximation signal through wavelet transformation, a second hard thresholding part for extracting only the signal greater than the predetermined threshold value among the detail signals generated through the wavelet transformation, and a signal combining part for combining the signal greater than the predetermined threshold value with the approximation signal generated through the wavelet transformation.

12. A method of detecting a heartbeat using a photoplethysmography (PPG), comprising the steps of:

(a) detecting a PPG signal received from a user through a PPG sensor unit;

(b) calculating an expected mean interbeat interval (mean IBI) from the PPG signal; and

(c) detecting determining an optimal heartbeat sequence or mean heart rate from candidate heartbeat sequences which have been detected from the PPG signal using the expected mean interbeat interval.

13. The method as claimed in claim 12, wherein the step (b) comprises the steps of:

(b1) performing a hard thresholding process of the PPG signal on the basis of a predetermined level;

(b2) calculating an autocorrelation function value of the hard thresholded PPG signal; and

(b3) detecting a time when a first peak of the autocorrelation function value occurs.

14. The method as claimed in claim 13, wherein the predetermined hard thresholding level is expressed as 0.5×

max(input).

15. The method as claimed in claim 13, wherein the step (b) further comprises the step of clipping the PPG signal with a predetermined level.

16. The method as claimed in claim 12, wherein the step (c) comprises the steps of:

(c1) detecting peaks from the PPG signal;

(c2) forming a plurality of candidate heartbeat sequences based on the detected peaks;

(c3) selecting the optimal heartbeat sequence among the plurality of candidate heartbeat sequences; and

(c4) calculating the mean heart rate based on the heartbeat sequences selected by the heartbeat sequence selecting part.

17. The method as claimed in claim 16, wherein the step (c2) comprises the steps of:

(c2-1) selecting respective peaks within a predetermined period of time as starting points; and

(c2-2) selecting peaks within intervals sequentially spaced apart from the starting points by a value of mean IBI±

Δ

T and then setting the selected peaks as the plurality of candidate heartbeat sequences.

18. The method as claimed in claim 17, wherein the step (c2-4) comprises the steps of:

selecting a peak close to the expected mean interbeat interval if there are a plurality of peaks within the intervals sequentially spaced apart from the starting point by the value of mean IBI±

Δ

T; and

expanding an interval range of Δ

T by a predetermined increment if there are no peaks within the intervals spaced apart from the starting points by the value of mean IBI±

Δ

T or the signal is lower than a predetermined value.

19. The method as claimed in claim 17, wherein the step (c3) comprises the steps of:

selecting window signals sequentially spaced apart from the selected starting points by a value of mean IBI; and

selecting the optimal heartbeat sequence by combining the window signals with the candidate heartbeat sequences having the same starting points as the window signals and then comparing combined results thereof with one another.

20. The method as claimed in claim 12, further comprising the step of:

(d) determining a level of a motion artifact based on the PPG signal detected from the PPG sensor unit and then generating a warning message when the level of the motion artifact is greater than a predetermined threshold value.

21. The method as claimed in claim 20, further comprising the step of:

(e) displaying the detected optimal heartbeat sequence or mean heart rate and the warning message on a screen.

22. The method as claimed in claim 20, wherein the step (d) comprises:

(d1) a TEO operating step of calculating a sudden rate of change in an amplitude and frequency of the PPG signal;

(d2) a convolution step of convoluting the signal outputted from the step (d1) with a predetermined window function; and

(d3) a warning message generating step of performing the hard thresholding process of the convoluted signal on the basis of a predetermined level and then generating a warning message when the hard thresholded signal is greater than the predetermined threshold value.

23. The method as claimed in claim 12, further comprising:

(f) a filtering step of selectively passing only a signal within a specific frequency band among the PPG signal detected by the PPG sensor unit.

24. The method as claimed in claim 12, further comprising the steps of:

(g) decomposing the PPG signal detected from the PPG sensor unit into an approximation signal and detail signals through wavelet transformation; and

(h) combining the approximation signal and the detail signals greater than a predetermined threshold value so that a resultant approximated signal can be generated.

25. The method as claimed in claim 24, wherein the step (h) comprises:

(h1) a hard thresholding step of extracting only the signal greater than the predetermined threshold value among the detail signals generated through the wavelet transformation; and

(h2) a signal combining step of combining the approximation signal with the detail signal extracted from the step (h1) and greater than the predetermined threshold value.