Energy-frequency-time heart sound analysis

• US 4,378,022 A
• Filed: 01/15/1981
• Issued: 03/29/1983
• Est. Priority Date: 01/15/1981
• Status: Expired due to Fees

• Current Assignee: California Institute of Technology
• Original Assignee: California Institute of Technology
• Inventors: Corcoran, William H., Harrison, Earl C., Suobank, David W.
• Primary Examiner(s): Howell, Kyle L.
• Assistant Examiner(s): Jaworski, Francis J.
• Application Number:
  US06/225,251
• Time in Patent Office: 803 Days
• Field of Search: 128/696, 128/701, 128/715, 128/773
• US Class Current: 600/528
• CPC Class Codes: A61B 10/00 Other methods or instruments for diagnosis, e.g. instruments for taking a cell sample, for biopsy, for vaccination diagnosis vaccination prophylaxis, vaccination therapy A61B17/20; Sex determination; Ovulation-period determination menstruation tables G06C3/00; Throat striking implements
• CPC Class Codes: A61B 7/005 Detecting noise caused by implants, e.g. cardiac valves
• CPC Class Codes: A61B 7/04 Electric stethoscopes microphones, acoustic transducers therefor H04R
• CPC Class Codes: A61F 2/2472 Devices for testing
• CPC Class Codes: G16H 15/00 ICT specially adapted for medical reports, e.g. generation or transmission thereof
• CPC Class Codes: G16H 40/63 for local operation

Abstract

A method for determining the operability status of heart valves IN VIVO. Accoustical data emanating from a heart valve to be analyzed is gathered. The data are divided into time slices, and the power, energy, and frequency relations for each time slice are calculated. The energy ratio of known peak frequencies is compared between the analysis data as gathered and known energy levels for a properly operating valve. The energy ratio of known frequency bands is also compared between the analysis data as gathered and known energy levels for a properly operating valve. Valve malfunctioning is indicated by the presence of a high change in the power and/or energy ratio at the selected key frequencies and bands.

Claims

1. The method of determining the operability status of a heart valve in vivo comprising the steps of:

(a) disposing a phonocardiographic transducer in a positon to gather time-amplitude-acoustical information emanating from the heart valve and adjacent areas;

(b) gathering the output signal from the transducer from a cardiac cycle;

(c) determining the location of a pre-established point in the cardiac cycle within the data;

(d) dividing the data of the cardiac cycle into time slices beginning with the pre-established point;

(e) calculating the power v. frequency curve for each time slice;

(f) comparing the power level associated with the valve being tested at pre-established frequencies within each time slice to corresponding power levels for the same frequencies at the same time slice for a known good valve; and

,

(g) indicating valve malfunction if the power ratio of measured data to known good data is greater than a pre-established amount over a pre-established threshold number of compared values.

2. The method of claim 1 wherein:

the power ratio is a Normalized Power Distribution (NPD) defined by the relationship;

##EQU3## for n=0, 1, 2, . . . , N and where, Δ

f is the frequency resolution (9.8 Hz) and a(ω

) and b(ω

) are the real and imaginary part of the complex Fourier coefficient for frequency ω

.

3. The method of claim 1 wherein:

the known good data are obtained from the same valve from steps (a) through (e) of claim 1 performed at a time when the valve is known to be properly operating.

4. The method of claim 1 or claim 3 wherein:

step (b) of gathering the output signal includes filtering the signal to remove data associated with frequencies below 100 Hz and above 1000 Hz.

5. The method of claim 1 or claim 4 wherein said step (d) of dividing the data into time slices comprises:

dividing the data into overlapping time windows having a length of duration longer than the increment between windows.

6. The method of claim 5 wherein:

each window is about 100 ms in duration, and the windows are taken at 20 ms increments.

7. The method of claim 1 or claim 3 wherein said step (c) of determining the location of a pre-established point in the cardiac cycle within the data includes the steps of:

(c1) gathering simultaneous blood-pressure data with said acoustical data;

(c2) storing said pressure data in association with said acoustical data as an identifier thereof;

(c3) searching said pressure data for a pre-established point in the cardiac cycle identifiable by its pressure characteristics; and

(c4) using the acoustical data associated with said pressure identifiable point in the cardiac cycle as a starting point for step (d).

8. The method of determining the operability status of a heart valve in vivo comprising the steps of:

(a) gathering simultaneously and retaining time-amplitude acoustical information emanating from the heart valve and adjacent area as well as the carotid or Korotkoff pressure within the cardiovascular system as produced by the heart for a cardiac cycle;

(b) searching the retained pressure data for a pre-established point within the cardiac cycle;

(c) using the pre-established point within the data found in step (b) as a starting point and dividing the acoustical data into a series of overlapped data windows for one cardial cycle wherein the duration of each window is longer than the increment between windows;

(d) calculating the power v. frequency curve for each window;

(e) comparing the power level associated with the valve being tested at pre-established known resonant frequencies to the power level at the same frequencies for the same windows in the cardiac cycle for a known good valve; and

,

(f) indicating valve malfunction if the ratio of the compared power levels is greater than a pre-established amount for greater than a pre-established number of values.

9. The method of claim 8 wherein:

the power ratio is a Normalized Power Distribution (NPD) defined by the relationship;

##EQU4## for n=0, 1, 2, . . . , N and where, Δ

f is the frequency resolution (9.8 Hz) and a(ω

) and b(ω

) are the real and imaginary part, respectively, of the complex Fourier coefficient for frequency ω

.

10. The method of claim 8 wherein;

the known good data are obtained from the same valve by steps (a) through (e) of claim 8 performed at a time when the valve is known to be properly operating.

11. The method of claim 8 or claim 10 wherein:

step (b) of gathering the output signal includes filtering the signal to remove data associated with frequencies below 100 Hz. and above 1000 Hz.

12. The method of claim 8 wherein:

each window is about 100 ms in duration and the windows are taken at 20 ms increments.

13. The method of claim 8 wherein:

only data associated with windows wherein the valve is in a non-transition state between open and closed is employed for step (e) of comparing.

14. The method of determining the operability status of a heart valve in vivo comprising the steps of:

(a) gathering simultaneously and retaining time-amplitude acoustical information emanating from the heart valve and adjacent area as well as the carotid or Korotkoff pressure within the cardiovascular system as produced by the heart for one cardiac cycle;

(b) searching the retained pressure data for a pre-established point within the cardiac cycle;

(c) using the pre-established point within the data found in step (b) as a starting point and dividing the acoustical data into a series of overlapped data windows for one cardiac cycle wherein the duration of each window is longer than the increment between windows;

(d) calculating the power v. frequency curve for each window;

(e) finding the maximum power level associated with the valve being tested at pre-established frequencies during the cardiac cycle;

(f) comparing the maximum power level for the valve being tested to the maximum power for a known good valve; and

,

(g) indicating valve malfunction if the ratio of the compared maximum power levels is greater than a pre-established amount.

15. The method of claim 14 wherein:

the power ratio is a Normalized Power Distribution (NPD) defined by the relationship;

##EQU5## for n=0, 1, 2, . . . , N and where, Δ

f is the frequency resolution (9.8 Hz) and a(ω

) and b(ω

) are the real and imaginary part of the complex Fourier coefficient for frequency ω

.

16. The method of claim 14 wherein:

the known good data are obtained from the same valve from steps (a) through (e) of claim 14 performed at a time when the valve is known to be operating properly.

17. The method of claim 14 or claim 16 wherein:

step (b) of gathering the output signal includes filtering the signal to remove data associated with frequencies below 100 Hz and above 1000 Hz.

18. The method of claim 14 wherein:

each window is about 100 ms in duration and the windows are taken at 20 ms increments.

19. The method of determining the operability status of a heart valve IN VIVO comprising the steps of:

(a) disposing a phonocardiographic transducer in a positon to gather time-amplitude-acoustical information emanating from the heart valve and adjacent areas;

(b) gathering the output signal from the transducer from a cardiac cycle;

(c) determining the location of a pre-established point in the cardiac cycle within the data;

(d) dividing the data of the cardiac cycle into time slices beginning with the pre-established point;

(e) calculating the energy v. frequency curve for each time slice;

(f) comparing the energy level associated with the valve being tested at pre-established frequencies within each time slice to corresponding energy levels for the same frequencies at the same time slice for a known good valve; and

,

(g) indicating valve malfunction if the energy ratio of measured data to known good data is greater than a pre-established amount over a pre-established threshold number of compared values.

20. The method of claim 19 wherein:

the energy ratio is a Normalized Energy Distribution (NED) defined by the relationship;

##EQU6## for n=0, 1, 2, . . . , N and where, Δ

f is the frequency resolution (9.8 Hz) and a(ω

) and b(ω

) are the real and imaginary part of the complex Fourier coefficient for frequency ω

.

21. The method of claim 19 wherein:

the known good data are obtained from the same valve from steps (a) through (e) of claim 19 performed at a time when the valve is known to be properly operating.

22. The method of claim 19 or claim 21 wherein:

step (b) of gathering the output signal includes filtering the signal to remove data associated with frequencies below 100 Hz and above 1000 Hz.

23. The method of claim 19 or claim 22 wherein said step (d) of dividing the data into time slices comprises:

dividing the data into overlapping time windows having a length of duration longer than the increment between windows.

24. The method of claim 23 wherein:

each window is about 100 ms in duration, and the windows are taken at 20 ms increments.

25. The method of claim 19 or claim 21 wherein said step (c) of determining the location of a pre-established point in the cardiac cycle within the data includes the steps of:

(c1) gathering simultaneous blood-pressure data with said acoustical data;

(c2) storing said pressure data in accordance with said acoustical data as an identifier thereof;

(c3) searching said pressure data for a pre-established point in the cardiac cycle identifiable by its pressure characteristics; and

(c4) using the acoustical data associated with said pressure identifiable point in the cardiac cycle as a starting point for step (d).

26. The method of determining the operability status of a heart valve in vivo comprising the steps of:

(a) gathering simultaneously and retaining time-amplitude acoustical information emanating from the heart valve and adjacent area as well as the carotid or Korotkoff pressure within the cardiovascular system as produced by the heart for a cardiac cycle;

(b) searching the retained pressure data for a pre-established point within the cardiac cycle;

(c) using the pre-established point within the data found in step (b) as a starting point and dividing the acoustical data into a series of overlapped data windows for one cardial cycle wherein the duration of each window is longer than the increment between windows;

(d) calculating the energy v. frequency curve for each window;

(e) comparing the energy level associated with the valve being tested at pre-established known resonant frequencies to the energy level at the same frequencies for the same windows in the cardiac cycle for a known good valve; and

,

(f) indicating valve malfunction if the ratio of the compared energy levels is greater than a pre-established amount for greater than a pre-established number of values.

27. The method of claim 26 wherein:

the energy ratio is a Normalized Energy Distribution (NED) defined by the relationship;

##EQU7## for n=0, 1, 2, . . . , N and where, Δ

f is the frequency resolution (9.8 Hz) and a(ω

) and b(ω

) are the real and imaginary part of the complex Fourier coefficient for frequency ω

.

28. The method of claim 26 wherein:

the known good data are obtained from the same valve by steps (a) through (e) of claim 26 performed at a time when the valve is known to be properly operating.

29. The method of claim 26 or claim 28 wherein:

step (b) of gathering the output signal includes filtering the signal to remove data associated with frequencies below 100 Hz. and above 1000 Hz.

30. The method of claim 26 wherein:

each window is about 100 ms in duration and the windows are taken at 20 ms increments.

31. The method of claim 26 wherein:

only data associated with windows wherein the valve is in a non-transition state between open and closed is employed for step (e) of comparing.

32. The method of determining the operability status of a heart valve IN VIVO comprising the steps of:

(a) gathering simultaneously and retaining time-amplitude acoustical information emanating from the heart valve and adjacent area as well as the carotid or Korotkoff pressure within the cardiovascular system as produced by the heart for one cardiac cycle;

(b) searching the retained pressure data for a pre-established point within the cardiac cycle;

(c) using the pre-established point within the data found in step (b) as a starting point and dividing the acoustical data into a series of overlapped data windows for one cardiac cycle wherein the duration of each window is longer than the increment between windows;

(d) calculating the energy v. frequency curve for each window;

(e) finding the maximum energy level associated with the valve being tested at pre-established frequencies during the cardiac cycle;

(f) comparing the maximum energy level for the valve being tested to the maximum energy for a known good valve; and

,

(g) indicating valve malfunction if the ratio of the compared maximum power levels is greater than a pre-established amount.

33. The method of claim 32 wherein:

the energy ratio is a Normalized Energy Distribution (NED) defined by the relationship;

##EQU8## for n=0, 1, 2, . . . , N and where, Δ

f is the frequency resolution (9.8 Hz) and a(ω

) and b(ω

) are the real and imaginary part of the complex Fourier coefficient for frequency ω

.

34. The method of claim 32 wherein:

the known good data are obtained from the same valve from steps (a) through (e) of claim 32 performed at a time when the valve is known to be operating properly.

35. The method of claim 32 or claim 34 wherein:

step (b) of gathering the output signal includes filtering the signal to remove data associated with frequencies below 100 Hz and above 1000 Hz.

36. The method of claim 32 wherein:

each window is about 100 ms in duration and the windows are taken at 20 ms increments.