MOTION BASED ESTIMATION OF BIOMETRIC SIGNALS

• US 20150342533A1
• Filed: 05/30/2014
• Published: 12/03/2015
• Est. Priority Date: 05/30/2014
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Kelner, Ilya
• Granted Patent: US 9,717,427 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: A61B 2560/0228 using calibration standards
• CPC Class Codes: A61B 5/02438 with portable devices, e.g. worn by the patient
• CPC Class Codes: A61B 5/0245 by using sensing means generating electric signals, i.e. ECG signals
• CPC Class Codes: A61B 5/1118 Determining activity level
• CPC Class Codes: A61B 5/4866 Evaluating metabolism A61B5/083 takes precedence
• CPC Class Codes: A61B 5/681 Wristwatch-type devices
• 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: G06F 1/32 Means for saving power
• CPC Class Codes: G06F 17/10 Complex mathematical operations function generation by table look-up G06F1/03; evaluation of elementary functions by calculation G06F7/544
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16Z 99/00 Subject matter not provided for in other main groups of this subclass

Abstract

A wearable electronic device, comprising a heart rate sensor to determine a heart rate of the user, a motion sensor to detect motion of a user, the motion sensor operated in an initial calibration phase during which the heart rate sensor is detecting the heart rate of the user, and the motion sensor is detecting the motion of the user, the calibration phase being used to compute a correspondence between motion of the user and the heart rate of the user, the motion sensor being operated in a low power phase during which the heart rate sensor is powered down, and a processor configured to run a heart rate estimation module configured to calculate an estimated heart rate of the user based upon the motion of the user and the correspondence between the heart rate of the user and the motion of the user determined in the calibration phase.

Claims

1. A wearable electronic device, comprising:

a heart rate sensor configured to determine a heart rate of the user;

a motion sensor configured to detect motion of a user, the motion sensor being operated in an initial calibration phase during which the heart rate sensor is powered and detecting the heart rate of the user, and the motion sensor is detecting the motion of the user, the calibration phase being used to compute a correspondence between motion of the user and the heart rate of the user, the motion sensor being operated in a low power phase during which the heart rate sensor is powered down and not detecting the heart rate of the user; and

a processor configured to run a heart rate estimation module being configured to calculate an estimated heart rate of the user based upon the motion of the user detected from the motion sensor during the low power phase and based upon the correspondence between the heart rate of the user and the motion of the user determined in the initial calibration phase.

2. The wearable electronic device of claim 1, wherein the processor includes a caloric burn estimation module configured to compute calories burned at least partially based on the estimate heart rate during the low power phase.

3. The wearable electronic device of claim 2, where the processor is further configured to display the calories burned on a display of the wearable electronic device.

4. The wearable electronic device of claim 1, wherein the wearable electronic device is a wearable computing device and the processor, heart rate sensor, and motion sensor are mounted thereto.

5. The wearable electronic device of claim 1, wherein the processor is further configured to operate in a remote mode in which the motion sensor is a remote sensor in a remote device, wherein the processor is configured to:

receive remote data from the remote sensor; and

compute the estimated heart rate based on the remote data.

6. The computing system of claim 1, wherein the heart rate estimation module is further configured to:

calculate a heart rate derivative and a representative motion value for each of a plurality of intervals during the initial calibration phase; and

apply a statistical filter to the heart rate derivatives and the representative motion values calculated in the initial calibration phase in order to compute a correspondence between the heart rate of the user and the motion of the user.

7. The computing system of claim 6, wherein the statistical filter is further configured to:

classify each heart rate derivative into one of a plurality of quantized derivative states;

sort the representative motion values into a plurality of lists based on the quantized derivative state of the heart rate derivative associated with each representative motion value;

compare the representative motion values in each list to the representative motion values in every other list in order to identify an overlapping segment in each list;

compute a characteristic value for each list by processing the representative motion values in the overlapping segment for each list.

8. The computing system of claim 7, wherein the heart rate estimation module is further configured to apply the statistical filter during the low power phase to determine the estimated heart rate of the user by calculating a representative motion value for each of a plurality of intervals during the low power phase, comparing the representative motion value to the characteristic value for each of the quantized derivative states, calculating an estimated heart rate derivative based on the characteristic values 106, and computing the estimated heart rate based on a last known heart rate 114 and the estimated heart rate derivative.

9. The computing system of claim 6, wherein the statistical filter is an Unscented Kalman Filter.

10. A method of estimating a heart rate of a user, comprising:

determining the heart rate of the user with a heart rate sensor;

operating a motion sensor in an initial calibration phase during which the heart rate sensor is powered and detecting heart rate, and the motion sensor is detecting motion of a user;

computing a correspondence between the motion of the user and the heart rate of the user during the initial calibration phase;

operating the motion sensor in a low power phase during which the heart rate sensor is powered down and not detecting heart rate;

calculating an estimated heart rate of the user with a heart rate estimation module configured to run on a processor based upon the motion of the user computed from the motion sensor during the low power phase and based upon the correspondence between the motion of the user and the heart rate of the user determined in the initial calibration phase.

11. The method of claim 10, further comprising:

computing calories burned with a caloric burn estimation module at least partially based on the estimated heart rate during the low power phase.

12. The method of claim 11, further comprising:

displaying the calories burned on a display device of a wearable electronic device.

13. The method of claim 11, further comprising:

logging at least one of the heart rate of the user and the calories burned to a memory of a wearable electronic device; and

uploading at least one of the heart rate of the user and the calories burned to a cloud based storage system.

14. The method of claim 10, further comprising:

calculating a heart rate derivative and a representative motion value for each of a plurality of intervals during the initial calibration phase.

15. The method of claim 14, further comprising:

applying a statistical filter to the heart rate derivatives and the representative motion values in the initial calibration phase to compute a correspondence between the heart rate of the user and the motion of the user.

16. The method of claim 15, further comprising:

classifying each heart rate derivative into one of a plurality of quantized derivative states;

dividing the representative motion values into a plurality of lists based on the quantized derivative state of the heart rate derivative associated with each representative motion value;

comparing the representative motion values in each list to the representative motion values in every other list in order to identify an overlapping segment in each list;

computing a characteristic value for each list by processing the representative motion values in the overlapping segment for each list.

17. The method of claim 16, further comprising:

applying the statistical filter during the low power phase to estimate the heart rate of the user by calculating a representative motion value for each of a plurality of intervals during the low power phase.

18. The method of claim 17, further comprising:

comparing the representative motion value to the characteristic value for each of the quantized derivative states, calculating an estimated heart rate derivative based on the characteristic values, and computing the estimated heart rate based on a last known heart rate and the estimated heart rate derivative.

19. The method of claim 15, wherein the statistical filter is an Unscented Kalman Filter.

20. An electronic device, comprising:

a first sensor configured to determine a first biometric factor of the user;

a second sensor configured to detect a second biometric factor of the user, the second sensor being operated in an initial calibration phase during which the first sensor is powered and detecting the first biometric factor, and the second sensor is detecting the second biometric factor, the calibration phase being used to compute a correspondence between the first biometric factor and the second biometric factor, the second sensor being operated in a low power phase during which the first sensor is powered down and not detecting the first biometric factor;

a processor configured to run an estimation module being configured to calculate an estimated first biometric factor of the user based upon the detected second biometric factor from the second sensor during the low power phase and based upon the correspondence between the first biometric factor of the user and the second biometric factor of the user determined in the initial calibration phase, and display the estimated first biometric factor on a display of the computing system.