Model-Based Method for Assessing Acoustic Signal Quality in Heart Monitoring Device

• US 20140343446A1
• Filed: 05/15/2013
• Published: 11/20/2014
• Est. Priority Date: 05/15/2013
• Status: Active Grant

• Current Assignee: Sharp Kabushiki Kaisha (Hon Hai Precision Industry Co., Ltd.)
• Original Assignee: Sharp Laboratories of America Incorporated (Hon Hai Precision Industry Co., Ltd.)
• Inventors: Hallberg, Bryan Severt, Hill, Frederick Norman
• Granted Patent: US 9,370,336 B2
• Time in Patent Office: Days
• US Class Current: 600/528
• CPC Class Codes: A61B 7/04 Electric stethoscopes microphones, acoustic transducers therefor H04R

Abstract

A model-based method for assessing acoustic signal quality in a heart monitoring device. The personal heart sound interval distribution of a person being actively monitored is compared with a modeled global heart sound interval distribution shared by most human beings after which processing action is taken consistent with the quality assessment. The error in the best fit between the personal interval distribution and the global interval distribution is presumed to be caused predominantly by noise, allowing the quality of the fit to serve as a proxy for the level of noise in the acoustic signal and used in making processing decisions.

Claims

1. A heart monitoring device, comprising:

a sound capture element configured to capture heart sounds and generate an acoustic signal containing the heart sounds; and

an acoustic signal processing element operatively coupled with the capture element and configured to generate from the acoustic signal over a time window a personal heart sound interval distribution, determine a best fit between the personal interval distribution and a global heart sound interval distribution, determine a fit quality value indicative of quality of the best fit and select a processing action for the time window using the fit quality value.

2. The device of claim 1, wherein the processing element is configured to determine the best fit at least in part by generating a personal heart sound interval distribution curve and comparing fit errors between the personal interval distribution curve and a parameterized global heart sound interval distribution curve for different values of at least one parameter of the global interval distribution curve.

3. The device of claim 1, wherein the processing action comprises reconfiguring a signal processing element for application to the acoustic signal over a future time window.

4. The device of claim 1, wherein the processing action comprises generating a heart parameter estimate for the time window.

5. The device of claim 1, wherein the processing action comprises generating a confidence indicator for the time window.

6. The device of claim 1, wherein the processing action comprises inhibiting determination of a heart parameter estimate for the time window.

7. The device of claim 1, wherein the processing element is further configured to compare the fit quality value with a quality threshold and, based on the comparison, select the processing action.

8. The device of claim 1, wherein the processing action comprises determining a heart parameter estimate and confidence indicator for the time window, further comprising a heart data output element operatively coupled with the processing element and configured to display the heart parameter estimate and the confidence indicator.

9. The device of claim 1, wherein the processing element is configured to generate the personal interval distribution at least in part by detecting an envelope of the acoustic signal over the time window.

10. The device of claim 9, wherein the processing element is further configured to generate the personal interval distribution at least in part by generating a histogram including counts of intervals between peaks in the envelope for intervals of different durations.

11. The device of claim 10, wherein the processing element is further configured to generate the personal interval distribution at least in part by smoothing the histogram to generate a personal heart sound interval distribution curve.

12. The device of claim 1, wherein the device is a mobile device.

13. A model-based method for assessing acoustic signal quality in a heart monitoring device, comprising:

generating by the device an acoustic signal capturing heart sounds;

generating by the device from the acoustic signal over a time window a personal heart sound interval distribution;

determining by the device a best fit between the personal interval distribution and a global heart sound interval distribution;

determining by the device a fit quality value indicative of quality of the best fit; and

selecting by the device a processing action for the time window using the fit quality value.

14. The method of claim 13, wherein determining the best fit comprises generating a personal heart sound interval distribution curve and comparing fit errors between the personal interval distribution curve and a parameterized global heart sound interval distribution curve for different values of at least one parameter of the global interval distribution curve.

15. The method of claim 13, wherein the processing action comprises reconfiguring a signal processing element for application to the acoustic signal over a future time window.

16. The method of claim 13, wherein the processing action comprises generating a heart parameter estimate for the time window.

17. The method of claim 13, wherein the processing action comprises generating a confidence indicator for the time window.

18. The method of claim 13, wherein the processing action comprises inhibiting determination of a heart parameter estimate for the time window.

19. The method of claim 13, wherein generating the personal interval distribution comprises detecting an envelope of the acoustic signal over the time window and generating a histogram including counts of intervals between peaks in the envelope for intervals of different durations.

20. The method of claim 13, wherein the processing action comprises determining a heart parameter estimate and confidence indicator for the time window and wherein the method further comprises displaying the heart parameter estimate and confidence indicator.