Cardiovascular parameter estimation in the presence of motion

Patent metadata

Field	Value
Publication number	US-10743777-B2
Application number	US-201715457702-A
Country	US
Kind code	B2
Filing date	Mar 13, 2017
Priority date	Dec 8, 2016
Publication date	Aug 18, 2020
Grant date	Aug 18, 2020

Abstract

Disclosed embodiments pertain to cardiovascular parameter (e.g. heart rate) measurements when motion is present. Biometric sensor signal measurements may be obtained based on cardiovascular parameters of a user; and motion sensor signal measurements may be obtained based on user motion. An activity type may be determined based on the motion sensor signals. For example, when non-motion related frequencies in a frequency domain representation of the biometric sensor signal are obscured by user motion, an activity type may be determined based on the motion sensor signals. Further, based on the activity type, for each cardiovascular parameter (e.g. heart rate), a corresponding likely cardiovascular parameter value (e.g. a likely heart rate) may be determined. A corresponding fundamental frequency associated with the biometric sensor signal may then be determined for each cardiovascular parameter based on the motion sensor signal measurements and the corresponding likely cardiovascular parameter value.

First claim

What is claimed is:

1. A processor-implemented method comprising:
obtaining, with a processor, a plurality of biometric sensor signal measurements of a biometric sensor signal output by a biometric sensor, the biometric sensor signal measurements based, in part, on one or more cardiovascular parameters of a user;
obtaining, with the processor, a plurality of motion sensor signal measurements of a motion sensor signal output by a motion sensor, the motion sensor signal measurements based, in part, on motion of the user;
determining, with the processor, an activity type based, in part, on a comparison of a spectral power distribution corresponding to the motion sensor signal measurements with at least one threshold, wherein the at least one threshold is based on cumulative probability distributions of activity related power spectral densities;
determining, with the processor, based on the activity type, for each cardiovascular parameter in a subset of the one or more cardiovascular parameters, a corresponding predicted cardiovascular parameter value; and
determining, with the processor, for each cardiovascular parameter in the subset, a corresponding fundamental frequency associated with the biometric sensor signal, based, in part, on the motion sensor signal measurements and the corresponding predicted cardiovascular parameter value.

2. The method of claim 1 , wherein the determination of the activity type is performed when non-motion related frequencies are not discernible in a frequency domain representation of the biometric sensor signal.

3. The method of claim 1 , further comprising:
determining, for each cardiovascular parameter in the subset, a corresponding estimated cardiovascular parameter value based, in part, on the corresponding fundamental frequency for the cardiovascular parameter.

4. The method of claim 1 , further comprising determining the spectral power distribution in the motion sensor signal comprising:
obtaining a spectral power value based, in part, on a first set of harmonics in a frequency domain representation of the motion sensor signal.

5. The method of claim 4 , wherein obtaining the spectral power value comprises:
determining a plurality of second sets of harmonics in the frequency domain representation of the motion sensor signal;
obtaining, corresponding to each second set of harmonics, a corresponding power measure;
comparing the corresponding power measures for the plurality of second sets of harmonics; and
selecting, based on the comparison, the first set of harmonics from the plurality of second sets of harmonics.

6. The method of claim 5 , wherein selecting the first set of harmonics from the plurality of second sets of harmonics comprises:
selecting, as the first set of harmonics, a second set of harmonics with a highest power measure relative to other second sets of harmonics in the plurality of second sets of harmonics.

7. The method of claim 5 , wherein obtaining the corresponding power measure for each second set of harmonics comprises at least one of:
obtaining the corresponding power measure based on a corresponding count of a number of spectral peaks in the second set of harmonics; or
obtaining the corresponding power measure based on a corresponding sum of amplitudes of spectral peaks in the second set of harmonics; or
obtaining the corresponding power measure based on a corresponding sum of integrated spectral peak signals in the second set of harmonics.

8. The method of claim 5 , wherein comparing the corresponding power measures for the plurality of second sets of harmonics comprises one of:
comparing the corresponding power measures for the plurality of second sets of harmonics to each other; or
comparing the corresponding power measures for the plurality of second sets of harmonics to the at least one threshold.

9. The method of claim 1 , wherein determining the activity type associated with the motion of the user comprises:
detecting a motion pattern based, in part, on the motion sensor signal measurements; and
determining the activity type based on the detected motion pattern.

10. The method of claim 1 , wherein the one or more cardiovascular parameters comprise a heart rate and the corresponding predicted cardiovascular parameter value comprises a predicted heart rate.

11. The method of claim 1 , wherein determining the corresponding predicted cardiovascular parameter value for each cardiovascular parameter in the subset comprises:
determining the corresponding predicted cardiovascular parameter value for each cardiovascular parameter in the subset using a physiological model, wherein, for each cardiovascular parameter in the subset, the physiological model correlates the activity type with the corresponding predicted cardiovascular parameter value.

12. The method of claim 11 , wherein the correlation of the activity type with the corresponding predicted cardiovascular parameter value is user-specific.

13. A device comprising:
a motion sensor, the motion sensor to output a motion sensor signal based, in part, on motion of a user,
a biometric sensor, the biometric sensor to output a biometric sensor signal based, in part, on one or more cardiovascular parameters of the user, and
a processor coupled to the motion sensor and the biometric sensor, wherein the processor is configured to:
obtain a plurality of biometric sensor signal measurements of the biometric sensor signal;
obtain a plurality of motion sensor signal measurements of the motion sensor signal;
determine an activity type based, in part, on a comparison of a spectral power distribution corresponding to the motion sensor signal measurements with at least one threshold, wherein the at least one threshold is based on cumulative probability distributions of activity related power spectral densities;
determine, based on the activity type, for each cardiovascular parameter in a subset of the one or more cardiovascular parameters, a corresponding predicted cardiovascular parameter value; and
determine, for each cardiovascular parameter in the subset, a corresponding fundamental frequency associated with the biometric sensor signal, based, in part, on the motion sensor signal measurements and the corresponding predicted cardiovascular parameter value.

14. The device of claim 13 , wherein the processor is configured to perform the determination of the activity type when non-motion related frequencies are not discernible in a frequency domain representation of the biometric sensor signal.

15. The device of claim 13 , wherein the processor is further configured to:
determine, for each cardiovascular parameter in the subset, a corresponding estimated cardiovascular parameter value based, in part, on the corresponding fundamental frequency for the cardiovascular parameter.

16. The device of claim 13 , wherein the processor is configured to determine the spectral power in the motion sensor signal by being configured to:
obtain a spectral power value based, in part, on a first set of harmonics in a frequency domain representation of the motion sensor signal; and
determine the activity type based on the spectral power value.

17. The device of claim 16 , wherein to obtain the spectral power value, the processor is configured to:
determine a plurality of second sets of harmonics in the frequency domain representation of the motion sensor signal;
obtain, corresponding to each second set of harmonics, a corresponding power measure;
compare the corresponding power measures for the plurality of second sets of harmonics; and
select, based

Assignees

• Qualcomm Inc

Inventors

• Martin Russel
• Narayanan Shashank
• Reddivari Hithesh
• Naware Vidyut
• Tchertkov Igor
• Czompo Joseph

Classifications

• A61B5/0075

  by spectroscopy, i.e. measuring spectra, e.g. Raman spectroscopy, infrared absorption spectroscopy (A61B5/0071 takes precedence) · CPC title

• A61B5/7221

  Determining signal validity, reliability or quality (preventing, reducing or removing noise induced by motion artefacts A61B5/7207; noise originating from a therapeutic or surgical apparatus A61B5/7217) · CPC title

• A61B5/024

  Measuring pulse rate or heart rate · CPC title

• A61B5/7257

  using Fourier transforms · CPC title

• A61B5/721

  using a separate sensor to detect motion or using motion information derived from signals other than the physiological signal to be measured · CPC title