Method and apparatus for third heart sound detection

• US 8,317,717 B2
• Filed: 09/16/2008
• Issued: 11/27/2012
• Est. Priority Date: 12/24/2003
• Status: Active Grant

• Current Assignee: Cardiac Pacemakers Incorporated (Boston Scientific Corporation)
• Original Assignee: Cardiac Pacemakers Incorporated (Boston Scientific Corporation)
• Inventors: Siejko, Krzysztof Z., Carlson, Gerrard M.
• Primary Examiner(s): Schaetzle, Kennedy
• Application Number:
  US12/283,760
• Publication Number:
  US 20090018461A1
• Time in Patent Office: 1,533 Days
• Field of Search: 600/528
• US Class Current: 600/528
• CPC Class Codes: A61B 7/04 Electric stethoscopes microphones, acoustic transducers therefor H04R

Abstract

A cardiac rhythm management system includes a heart sound detector providing for detection of the third heart sounds (S3). An implantable sensor such as an accelerometer or a microphone senses an acoustic signal indicative heart sounds including the second heart sounds (S2) and S3. The heart sound detector detects occurrences of S2 and starts S3 detection windows each after a predetermined delay after a detected occurrence of S2. The occurrences of S3 are then detected from the acoustic signal within the S3 detection windows.

Claims

1. A heart sound detection system, comprising:

an acoustic signal input adapted to receive an acoustic signal indicative of heart sounds including second heart sounds (S2) and third heart sounds (S3); and

a heart sound detector coupled to the acoustic signal input, the heart sound detector including;

an S2 detector adapted to detect S2 using the acoustic signal;

an S3 window generator adapted to generate an S3 window in response to each detection of S2;

an S3 detector adapted to detect S3 using the acoustic signal and an S3 threshold during the S3 windows;

a measurement module adapted to detect an acoustic energy using the acoustic signal; and

an S3 threshold generator adapted to determine the S3 threshold using the acoustic energy, wherein the S3 threshold generator is adapted to normalize the S3 threshold using the acoustic energy.

2. The system of claim 1, wherein the S3 window generator is adapted to generate the S3 window after a programmed delay starting with each of the detected S2.

3. The system of claim 2, wherein the measurement module is adapted to detect a total acoustic energy during a cardiac cycle, and the S3 threshold generator is adapted to dynamically adjust the S3 threshold using the total acoustic energy detected during the cardiac cycle.

4. The system of claim 3, wherein the S3 threshold generator is adapted to dynamically adjust the S3 threshold using a mean and a standard deviation of the total acoustic energy during the cardiac cycle.

5. The system of claim 2, wherein the measurement module is adapted to detect a total acoustic energy during systole of a cardiac cycle, and the S3 threshold generator is adapted to dynamically adjust the S3 threshold using the total acoustic energy during systole of the cardiac cycle.

6. The system of claim 5, wherein the S3 threshold generator is adapted to dynamically adjust the S3 threshold using a mean and a standard deviation of the total acoustic energy during the systole of the cardiac cycle.

7. A heart sound detection method, comprising:

receiving an acoustic signal indicative of heart sounds including second heart sounds (S2) and third heard sounds (S3);

detecting S2 using the acoustic signal;

generating an S3 window following each of the detected S2;

detecting S3 during the S3 windows by comparing the acoustic signal to an S3 threshold;

detecting an acoustic energy using the acoustic signal; and

dynamically adjusting the S3 threshold using the acoustic energy, including normalizing the S3 threshold using the acoustic energy.

8. The method of claim 7, wherein generating the S3 window comprises generating the S3 window after a programmed delay starting with each of the detected S2.

9. The method of claim 8, comprising:

receiving a cardiac signal indicative of ventricular events;

detecting ventricular events from the cardiac signal; and

terminating the S3 window if one of the ventricular events occurs during the S3 window.

10. The method of claim 8, comprising determining the S3 threshold using at least a mean and a standard deviation of the acoustic energy.

11. The method of claim 8, wherein detecting the acoustic energy comprises detecting a total acoustic energy during a cardiac cycle as the acoustic energy.

12. The method of claim 8, wherein detecting the acoustic energy comprises detecting a total acoustic energy during systole of a cardiac cycle as the acoustic energy.

13. A heart sound detection system, comprising:

an acoustic signal input to receive an acoustic signal indicative of heart sounds including first heart sounds (S1), second heart sounds (S2), and third heart sounds (S3); and

a heart sound detector coupled to the acoustic signal input, the heart sound detector including;

an S2 detector adapted to detect S2 using the acoustic signal;

an S3 detection preparation module adapted to align segments of the acoustic signal using the detected S2 and ensemble average the acoustic signal using the aligned segments;

an S3 window generator adapted to generate an S3 window in response to each detection of S2;

an S3 detector adapted to detect S3 during the S3 windows using the averaged acoustic signal and an S3 threshold; and

an S3 threshold generator adapted to determine the S3 threshold using an estimated background sound level (μ

_B) measured during a period between an occurrence of S1 and an adjacent occurrence of S2.

14. The system of claim 13, wherein the S3 threshold generator is adapted to dynamically adjust the S3 threshold using μ

_B.

15. The system of claim 14, wherein the S3 threshold generator is adapted to calculate the S3 threshold as μ

_B*K, where K is an adjustable multiplier.

16. The system of claim 15, wherein the S3 threshold generator is adapted to scale the period between the occurrence of S1 and the adjacent occurrence of S2 for a heart rate.

17. The system of claim 16, comprising:

a cardiac signal input adapted to receive a cardiac signal indicative of ventricular events; and

an S2 window generator adapted to generate an S2 window in response to the detection of each of the ventricular events, and wherein the S2 detector is adapted to detect S2 during the S2 windows.

18. The system of claim 17, wherein the S3 window generator is adapted to generate the S3 window after a second programmed delay starting with each of the detected S2.

19. The system of claim 13, wherein the S3 threshold generator is adapted to determine the S3 threshold using the μ

_B measured during a period set to be substantially equal to a time interval between a latest point in time where S1 energy is expected and a point in time where the S2 window begins and to be subjected to a maximum duration and a minimum duration.

20. The system of claim 13, wherein the S3 threshold generator is adapted to dynamically adjusting the S3 threshold using the μ

_B measured during a period centered between a latest point in time where S1 energy is expected and a point in time where the S2 window begins, wherein the latest point in time where S1 energy is expected is empirically estimated based on a patient population.

21. A heart sound detection method, comprising:

receiving an acoustic signal indicative of heart sounds including first heart sounds (S1), second heart sounds (S2), and third heart sounds (S3);

detecting S2 using the acoustic signal;

aligning segments of the acoustic signal using the detected S2;

ensemble averaging the acoustic signal using the aligned segments;

generating an S3 window following each of the detected S2;

detecting S3 during the S3 windows by comparing the ensemble averaged acoustic signal to a dynamically adjustable S3 threshold; and

dynamically adjusting the S3 threshold using an estimated background sound level (μ

_B) measured during a period between an occurrence of S1 and an adjacent occurrence of S2.

22. The method of claim 21, wherein dynamically adjusting the S3 threshold comprises calculating the S3 threshold as μ

_B*K, where K is an adjustable.

23. The method of claim 22, comprising:

receiving a cardiac signal indicative of ventricular events;

detecting a heart rate using the cardiac signal; and

scaling the period between the occurrence of S1 and the adjacent occurrence of S2 for the heart rate.

24. The method of claim 23, comprising:

detecting ventricular events from the cardiac signal; and

terminating the S3 window if one of the ventricular events occurs during the S3 window.

25. The method of claim 21, comprising:

receiving a cardiac signal indicative of ventricular events; and

generating an S2 window in response to the detection of each of the ventricular events,

and wherein detecting S2 comprises detecting S2 during the S2 windows.

26. The method of claim 25, wherein generating the S3 window comprises generating the S3 window after a programmed delay starting with each of the detected S2.

27. The method of claim 21, wherein dynamically adjusting the S3 threshold comprises dynamically adjusting the S3 threshold using the μ

_B measured during a period set to be substantially equal to a time interval between a latest point in time where S1 energy is expected and a point in time where the S2 window begins and to be subjected to a maximum duration and a minimum duration.

28. The system of claim 21, wherein dynamically adjusting the S3 threshold comprises:

dynamically adjusting the S3 threshold using the measured during a period centered between a latest point in time where S1 energy is expected and a point in time where the S2 window begins; and

estimating the latest point in time where S1 energy is expected empirically based on a patient population.