Accuracy of heart rate estimation from photoplethysmographic (PPG) signals

• US 10,123,746 B2
• Filed: 05/05/2016
• Issued: 11/13/2018
• Est. Priority Date: 05/08/2015
• Status: Active Grant

• Current Assignee: Texas Instruments, Inc.
• Original Assignee: Texas Instruments, Inc.
• Inventors: Gunturi, Sarma Sundareswara, Chomal, Sunil
• Primary Examiner(s): Flory, Christopher A
• Application Number:
  US15/147,720
• Publication Number:
  US 20160324477A1
• Time in Patent Office: 922 Days
• Field of Search: 600479
• CPC Class Codes: A61B 2562/0219 Inertial sensors, e.g. accelerometers, gyroscopes, tilt switches
• CPC Class Codes: A61B 5/02416 using photoplethysmograph signals, e.g. generated by infrared radiation A61B5/14552 takes precedence
• 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
• CPC Class Codes: A61B 5/725 using specific filters therefor, e.g. Kalman or adaptive filters A61B5/316 take precedence
• CPC Class Codes: A61B 5/7257 using Fourier transforms
• CPC Class Codes: A61B 5/7278 Artificial waveform generation or derivation, e.g. synthesising signals from measured signals A61B5/327 takes precedence

Abstract

The disclosure provides a heart rate monitor (HRM) circuit. The HRM circuit includes an analog front end (AFE). The AFE receives a Photoplethysmographic (PPG) signal, and generates a fine PPG signal. A motion cancellation circuit is coupled to the AFE, and receives the fine PPG signal and an accelerometer signal. The motion cancellation circuit subtracts the accelerometer signal from the fine PPG signal to generate a coarse heart rate. A peak detector is coupled to the motion cancellation circuit and the AFE, and generates an instantaneous heart rate. A filter is coupled to the peak detector, and generates an estimated heart rate from the instantaneous heart rate, the estimated heart rate is provided as a feedback to the peak detector.

Claims

1. A heart rate monitoring (HRM) circuit comprising:

an analog front end (AFE) configured to receive a Photoplethysmographic (PPG) signal and configured to generate a fine PPG signal;

a motion cancellation circuit coupled to the AFE and configured to receive the fine PPG signal and an accelerometer signal, the motion cancellation circuit configured to subtract the accelerometer signal from the fine PPG signal to generate a coarse heart rate;

a peak detector coupled to the motion cancellation circuit and the AFE, and configured to generate an instantaneous heart rate; and

a filter coupled to the peak detector, and configured to generate an estimated heart rate from the instantaneous heart rate, wherein the estimated heart rate is provided as a feedback to the peak detector, and the peak detector generates the instantaneous heart rate from the coarse heart rate, the fine PPG signal and the estimated heart rate.

2. The HRM circuit of claim 1, wherein the peak detector comprises:

a first periodicity detector coupled to the motion cancellation circuit, and configured to receive the coarse heart rate and the estimated heart rate;

a second periodicity detector configured to receive the fine PPG signal and the estimated heart rate; and

an arbitration block coupled to the first periodicity detector and the second periodicity detector and configured to generate the instantaneous heart rate.

3. The HRM circuit of claim 2, wherein the estimated heart rate is provided as feedback to the first periodicity detector, the second periodicity detector and the arbitration block.

4. The HRM circuit of claim 2, wherein the first periodicity detector is configured to estimate peak in the coarse heart rate, and the second periodicity detector is configured to estimate peak in the fine PPG signal.

5. The HRM circuit of claim 2, wherein:

the first periodicity detector is configured to estimate a first quality metric from the estimated peak in the coarse heart rate and from the estimated heart rate; and

the second periodicity detector is configured to estimate a second quality metric from the estimated peak in the fine PPG signal and from the estimated heart rate.

6. The HRM circuit of claim 2, wherein the arbitration block is configured to generate the instantaneous heart rate from the estimated peak in the coarse heart rate, the estimated peak in the fine PPG signal, the first quality metric, the second quality metric, the accelerometer signal and the estimated heart rate.

7. The HRM circuit of claim 1, wherein the AFE comprises:

an analog to digital converter (ADC) configured to receive the PPG signal, the ADC configured to generate a digital PPG signal from the PPG signal;

a DC offset cancellation block configured to subtract DC offset from the digital PPG signal to generate an offset PPG signal; and

a decimator configured to decimate the offset PPG signal to generate the fine PPG signal.

8. The HRM circuit of claim 1, wherein the filter is a kalman filter.

9. A method for monitoring heart rate of a person comprising:

receiving a Photoplethysmographic (PPG) signal;

generating a fine PPG signal from the PPG signal;

subtracting an accelerometer signal from the fine PPG signal to generate a coarse heart rate;

generating an instantaneous heart rate in a peak detector; and

filtering the instantaneous heart rate to generate an estimated heart rate, wherein the estimated heart rate is provided as a feedback to the peak detector, and the instantaneous heart rate is generated from the coarse heart rate, the fine PPG signal and the estimated heart rate.

10. The method of claim 9, wherein generating the fine PPG signal from the PPG signal comprises:

generating a digital PPG signal from the PPG signal;

subtracting DC offset from the digital PPG signal to generate an offset PPG signal; and

decimating the offset PPG signal to generate the fine PPG signal.

11. The method of claim 10 further comprising decimating a coarse accelerometer signal in proportion to the decimation of the offset PPG signal to generate the accelerometer signal.

12. The method of claim 9, wherein generating the instantaneous heart rate further comprises operating the peak detector in an acquisition phase and a tracking phase.

13. The method of claim 12, wherein in the acquisition phase:

performing FFT on the coarse heart rate to estimate a range of peaks in the coarse heart rate;

discarding one or more peaks whose value is less than a threshold peak value;

selecting an estimated peak which is of a highest value over a defined range of fundamental frequencies and harmonic frequencies, the estimated peak is the instantaneous heart rate generated by the peak detector; and

estimating a first quality metric from the estimated peak.

14. The method of claim 13, wherein the instantaneous heart rate is filtered to generate the estimated heart rate.

15. The method of claim 12, wherein in the tracking phase:

performing FFT on the coarse heart rate to generate a first power spectral density;

performing FFT on the fine PPG signal to generate a second power spectral density; and

searching peaks which are within a defined threshold range of the estimated heart rate and based on a state of the person, the state of the person includes a person at rest and a person in motion.

16. The method of claim 15, wherein searching peaks when the person is at rest further comprises:

selecting a peak in a harmonic range when the estimated heart rate is less than a threshold and no peaks occur in a fundamental range; and

selecting a fundamental peak closest to the estimated heart rate when multiple harmonic pairs are present.

17. The method of claim 15, wherein searching peaks when the person is in motion further comprises:

selecting a peak on a positive side of the estimated heart rate in the first power spectral density when a magnitude of the accelerometer signal is above a first threshold, and the estimated heart rate is above a second threshold;

selecting a peak closest to the estimated heart rate when multiple peaks are present on the positive side of the estimated heart rate; and

selecting a peak closest to the estimated heart rate in a set of peaks when the magnitude of the accelerometer signal is below the first threshold, the set of peaks are peaks in the first power spectral density and the second power spectral density which are within the defined threshold range of the estimated heart rate.

18. A device comprising:

a processing unit; and

a heart rate monitoring (HRM) circuit coupled to the processing unit, the HRM circuit comprising;

an analog front end (AFE) configured to receive a Photoplethysmographic (PPG) signal and configured to generate a fine PPG signal;

a motion cancellation circuit coupled to the AFE and configured to receive the fine PPG signal and an accelerometer signal, the motion cancellation circuit configured to subtract the accelerometer signal from the fine PPG signal to generate a coarse heart rate;

a peak detector coupled to the motion cancellation circuit, and configured to generate an instantaneous heart rate; and

a filter coupled to the peak detector, and configured to generate an estimated heart rate from the instantaneous heart rate, wherein the estimated heart rate is provided as a feedback to the peak detector, and the peak detector generates the instantaneous heart rate from the coarse heart rate, the fine PPG signal and the estimated heart rate.

19. The device of claim 18, wherein the peak detector comprises:

a first periodicity detector coupled to the motion cancellation circuit, and configured to receive the coarse heart rate and the estimated heart rate;

a second periodicity detector configured to receive the fine PPG signal and the estimated heart rate; and

an arbitration block coupled to the first periodicity detector and the second periodicity detector and configured to generate the instantaneous heart rate.

20. The device of claim 19, wherein the estimated heart rate is provided as feedback to the first periodicity detector, the second periodicity detector and the arbitration block.