Method and system for generating a likelihood of cardiovascular disease, analyzing cardiovascular sound signals remotely from the location of cardiovascular sound signal acquisition, and determining time and phase information from cardiovascular sound signals

• US 8,600,488 B2
• Filed: 02/01/2007
• Issued: 12/03/2013
• Est. Priority Date: 03/18/2002
• Status: Active Grant

• Current Assignee: AusculSciences, Inc.
• Original Assignee: SonoMedica, Inc.
• Inventors: Mohler, Sailor Hampton, Mulholland, Dan, Figgatt, Alan, Holford, Warren
• Primary Examiner(s): GETZOW, SCOTT M
• Application Number:
  US11/700,827
• Publication Number:
  US 20080077029A1
• Time in Patent Office: 2,497 Days
• Field of Search: 600/493, 600/513, 600/514, 600/528, 600/586
• US Class Current: 600/514
• CPC Class Codes: A61B 5/02 Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
• CPC Class Codes: A61B 5/0205 Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
• CPC Class Codes: A61B 5/0245 by using sensing means generating electric signals, i.e. ECG signals
• CPC Class Codes: A61B 5/318 Heart-related electrical modalities, e.g. electrocardiography [ECG]
• CPC Class Codes: A61B 5/322 Physical templates or devices for measuring ECG waveforms, e.g. electrocardiograph rulers or calipers by template matching A61B5/35
• CPC Class Codes: A61B 5/7203 for noise prevention, reduction or removal
• CPC Class Codes: A61B 5/7246 using correlation, e.g. template matching or determination of similarity
• CPC Class Codes: A61B 5/725 using specific filters therefor, e.g. Kalman or adaptive filters A61B5/316 take precedence
• CPC Class Codes: A61B 5/7257 using Fourier transforms
• CPC Class Codes: A61B 5/7278 Artificial waveform generation or derivation, e.g. synthesising signals from measured signals A61B5/327 takes precedence
• CPC Class Codes: A61B 5/7282 Event detection, e.g. detecting unique waveforms indicative of a medical condition A61B5/4094, A61B5/4818, A61B5/349, A61B5/0823 take precedence
• CPC Class Codes: A61B 7/026 comprising more than one sound collector
• CPC Class Codes: A61B 7/04 Electric stethoscopes microphones, acoustic transducers therefor H04R
• CPC Class Codes: G16H 50/30 for calculating health indices; for individual health risk assessment

Abstract

A system, method and computer executable code for generating a likelihood of cardiovascular disease from acquired cardiovascular sound signals is disclosed, where the generated likelihood of cardiovascular disease is based at least on an overlapping in time of bruit candidates in one heart cycle or in different heart cycles. Also disclosed is a system, method, and computer executable code for collecting, forwarding, and analyzing cardiovascular sound signals, where the collecting and analyzing may occur at locations that are remote from each other. Further disclosed is a system, method, and computer executable code for determining the time and phase information contained in cardiovascular sound signals, for use in analyzing those cardiovascular sound signals.

Claims

1. A method comprising:

a. receiving cardiovascular sound signals of a patient that were previously captured at a point of care of the patient;

b. processing at a location remote from the point of care the received cardiovascular sound signals to generate a probability indicator indicative of the likelihood that the patient has cardiovascular disease, wherein the operation of processing at said location remote from the point of care the received cardiovascular sound signals to generate said probability indicator indicative of the likelihood that the patient has cardiovascular disease comprises;

i. parsing said received cardiovascular sound signals so as to generate a corresponding set of parsed heart cycle signals representative of said received cardiovascular sound signals relative to an associated heart cycle;

ii. generating high-pass filtered heart cycle data by high-pass filtering said parsed heart cycle signals so as to substantially retain frequency components above 300 Hz and so as to substantially attenuate frequency components below 200 Hz;

iii. segmenting said high-pass filtered heart cycle data so as to generate a plurality of associated time data segments;

iv. for each time data segment of said plurality of associated time data segments;

a) calculating a frequency spectrum of said time data segment, wherein said frequency spectrum comprises a plurality of spectrum values for each of a corresponding plurality of associated spectral components;

b) calculating a normalized frequency spectrum of said time data segment, wherein for each spectral component of said corresponding plurality of associated spectral components, each of said plurality of spectrum values is normalized with respect to a corresponding associated noise floor, so as to generate a corresponding plurality of normalized spectrum values;

c) calculating a skew of said normalized frequency spectrum, wherein said skew is responsive to a second moment of said normalized spectrum values of said normalized frequency spectrum;

d) comparing said skew with a skew threshold; and

e) for each said spectral component of said plurality of associated spectral components;

i) comparing a normalized spectrum value of said corresponding plurality of normalized spectrum values with a bruit power detection threshold; and

ii) if said normalized spectrum value exceeds said bruit power detection threshold AND said skew is less than said skew threshold, then a frequency associated with said spectral component, a time associated with said time data segment, said skew, and said normalized spectrum value are associated with a bruit candidate associated with said cardiovascular disease; and

v. identifying whether or not said bruit candidate is likely associated with said cardiovascular disease, wherein said probability indicator is responsive to whether or not said bruit candidate is identified as being a bruit likely associated with said cardiovascular disease, and if so, said probability indicator is responsive to at least one characteristic of said bruit candidate; and

c. forwarding the probability indicator to at least one of the point of care and another location.

2. The method of claim 1, the received cardiovascular sound signals further comprising digital signals that were previously converted from analog signals.

3. The method of claim 1, the received cardiovascular sound signals further comprising a plurality of heart waveform signals each corresponding to a different location on the patient.

4. The method of claim 3, each heart waveform signal of said plurality of heart waveform signals further comprising a plurality of heart cycle signals.

5. The method of claim 4, each heart cycle signal of said plurality of heart cycle signals corresponding to a heart cycle of the patient.

6. The method of claim 1, said receiving further comprising receiving a plurality of data files, each data file of said plurality of data files containing at least one heart waveform signal of a patient.

7. The method of claim 1, further comprising receiving an ECG signal of the patient that was previously captured at said point of care of the patient, and using said ECG signal at said location remote from said point of care to process said received cardiovascular sound signals to generate said probability indicator indicative of the likelihood that the patient has cardiovascular disease.

8. The method of claim 1, said forwarding further comprising forwarding the probability indicator to a location remote from said location remote from the point of care where the probability indicator was generated.

9. The method of claim 8, further comprising forwarding the probability indicator to the point of care over a data communications network.

10. The method of claim 9, wherein said data communications network further comprises the interne.

11. The method of claim 1, said receiving further comprising receiving the cardiovascular sound signals in an encrypted format.

12. The method of claim 1, wherein said probability indicator is responsive to a probability of at least one repetitive bruit.

13. The method of claim 1, further comprising receiving background noise signals that were previously captured simultaneously with said cardiovascular sound signals at said point of care of the patient, and processing at said location remote from said point of care the received background noise signals, wherein said probability indicator is generated responsive to both said received cardiovascular sound signals and said received background noise signals.

14. The method of claim 13, wherein the operation of identifying whether or not said bruit candidate is likely associated with said cardiovascular disease comprises:

a. determining whether or not said bruit candidate is also detected within said received background noise signals; and

b. if said bruit candidate is detected within said received background noise signals, then treating said bruit candidate as not being a bruit likely associated with said cardiovascular disease.

15. The method of claim 13, wherein the operation of identifying whether or not said bruit candidate is likely associated with said cardiovascular disease comprises:

a determining whether or not said bruit candidate is relatively close in time and frequency to a different bruit candidate that had been previously treated as not being a bruit likely associated with said cardiovascular disease; and

b. if said bruit candidate is relatively close in time and frequency to said different bruit candidate that had been previously treated as not being a bruit likely associated with said cardiovascular disease, then treating said bruit candidate as not being a bruit likely associated with said cardiovascular disease.

16. The method of claim 1, wherein the operation of calculating said frequency spectrum of said time data segment comprises transforming said time data segment with a Fast Fourier Transform.

17. The method of claim 13, further comprising:

a forming a sample set from said time data segment and portions of time data segments adjacent thereto, wherein said sample set overlaps said time data segment and associated time data segments adjacent thereto; and

b. suppressing the contribution of time samples at the beginning or end of said sample set with either a Blackman or Kaiser Window windowing function, wherein said frequency spectrum is calculated with said Fast Fourier Transform by transforming said sample set after operation of said Blackman or Kaiser Window windowing function thereupon.

18. The method of claim 1, further comprising for each time data segment of said plurality of associated time data segments:

a. calculating a mean time data power of said high-pass filtered heart cycle data of said time data segment; and

b. comparing said mean time data power with a mean time data power threshold for each time data segment of said plurality of associated time data segments, wherein for each said spectral component of said plurality of associated spectral components, if said normalized spectrum value exceeds said bruit power detection threshold AND if said mean time data power exceeds said mean time data power threshold AND said skew is less than said skew threshold, then said frequency associated with said spectral component, said time associated with said time data segment, said skew, and said normalized spectrum value are associated with said bruit candidate associated with said cardiovascular disease.

19. A system comprising:

a. means for receiving cardiovascular sound signals of a patient that were previously captured at a point of care of the patient;

b. means for processing at a location remote from the point of care the received cardiovascular sound signals to generate a probability indicator indicative of the likelihood that the patient has cardiovascular disease, wherein said means for processing at said location remote from the point of care the received cardiovascular sound signals to generate said probability indicator indicative of the likelihood that the patient has cardiovascular disease comprises;

i. means for parsing said received cardiovascular sound signals so as to generate a corresponding set of parsed heart cycle signals representative of said received cardiovascular sound signals relative to an associated heart cycle;

ii. means for generating high-pass filtered heart cycle data by high-pass filtering said parsed heart cycle signals so as to substantially retain frequency components above 300 Hz and so as to substantially attenuate frequency components below 200 Hz;

iii. means for segmenting said high-pass filtered heart cycle data so as to generate a plurality of associated time data segments;

iv. for each time data segment of said plurality of associated time data segments;

a) means for calculating a frequency spectrum of said time data segment, wherein said frequency spectrum comprises a plurality of spectrum values for each of a corresponding plurality of associated spectral components;

b) means for calculating a normalized frequency spectrum of said time data segment, wherein for each spectral component of said corresponding plurality of associated spectral components, each of said plurality of spectrum values is normalized with respect to a corresponding associated noise floor, so as to generate a corresponding plurality of normalized spectrum values;

c) means for calculating a skew of said normalized frequency spectrum, wherein said skew is responsive to a second moment of said normalized spectrum values of said normalized frequency spectrum;

d) means for comparing said skew with a skew threshold; and

e) for each said spectral component of said plurality of associated spectral components;

i) means for comparing a normalized spectrum value of said corresponding plurality of normalized spectrum values with a bruit power detection threshold;

wherein

ii) if said normalized spectrum value exceeds said bruit power detection threshold AND said skew is less than said skew threshold, then a frequency associated with said spectral component, a time associated with said time data segment, said skew, and said normalized spectrum value are associated with a bruit candidate associated with said cardiovascular disease; and

v. means for identifying whether or not said bruit candidate is likely associated with said cardiovascular disease, wherein said probability indicator is responsive to whether or not said bruit candidate is identified as being a bruit likely associated with said cardiovascular disease, and if so, said probability indicator is responsive to at least one characteristic of said bruit candidate; and

c. means for forwarding the probability indicator to at least one of the point of care and another location.

20. The system of claim 19, wherein said probability indicator is responsive to a probability of at least one repetitive bruit.

21. The system of claim 19, further comprising a means for receiving background noise signals that were previously captured simultaneously with said cardiovascular sound signals at said point of care of the patient, and a means for processing at said location remote from said point of care the received background noise signals, wherein said probability indicator is generated responsive to both said received cardiovascular sound signals and said received background noise signals.

22. The system of claim 21, wherein said means for identifying whether or not said bruit candidate is likely associated with said cardiovascular disease comprises:

a. means for determining whether or not said bruit candidate is also detected within said received background noise signals;

wherein

b. if said bruit candidate is detected within said received background noise signals, then said bruit candidate is treated as not being a bruit likely associated with said cardiovascular disease.

23. The system of claim 22, wherein said means for identifying whether or not said bruit candidate is likely associated with said cardiovascular disease comprises:

a. means for determining whether or not said bruit candidate is relatively close in time and frequency to a different bruit candidate that had been previously treated as not being a bruit likely associated with said cardiovascular disease;

wherein

b. if said bruit candidate is relatively close in time and frequency to said different bruit candidate that had been previously treated as not being a bruit likely associated with said cardiovascular disease, then said bruit candidate is treated as not being a bruit likely associated with said cardiovascular disease.

24. The system of claim 19, wherein said means for calculating said frequency spectrum of said time data segment comprises a means for transforming said time data segment with a Fast Fourier Transform.

25. The system of claim 24, further comprising:

a. means for forming a sample set from said time data segment and portions of time data segments adjacent thereto, wherein said sample set overlaps said time data segment and associated time data segments adjacent thereto; and

b. means for suppressing the contribution of time samples at the beginning or end of said sample set with either a Blackman or Kaiser Window windowing function, wherein said frequency spectrum is calculated with said Fast Fourier Transform by transforming said sample set after operation of said Blackman or Kaiser Window windowing function thereupon.

26. The system of claim 19, further comprising:

a. means for calculating a mean time data power of said high-pass filtered heart cycle data of said time data segment for each time data segment of said plurality of associated time data segments; and

b. means for comparing said mean time data power with a mean time data power threshold for each time data segment of said plurality of associated time data segments, wherein for each said spectral component of said plurality of associated spectral components, if said normalized spectrum value exceeds said bruit power detection threshold AND if said mean time data power exceeds said mean time data power threshold AND said skew is less than said skew threshold, then said frequency associated with said spectral component, said time associated with said time data segment, said skew, and said normalized spectrum value are associated with said bruit candidate associated with said cardiovascular disease.

27. A physical computer-readable medium containing computer program instructions for a method comprising:

a receiving cardiovascular sound signals of a patient that were previously captured at a point of care of the patient;

b. processing at a location remote from the point of care the received cardiovascular sound signals to generate a probability indicator indicative of the likelihood that the patient has cardiovascular disease, wherein the operation of processing at said location remote from the point of care the received cardiovascular sound signals to generate said probability indicator indicative of the likelihood that the patient has cardiovascular disease comprises;

i. parsing said received cardiovascular sound signals so as to generate a corresponding set of parsed heart cycle signals representative of said received cardiovascular sound signals relative to an associated heart cycle;

ii. generating high-pass filtered heart cycle data by high-pass filtering said parsed heart cycle signals so as to substantially retain frequency components above 300 Hz and so as to substantially attenuate frequency components below 200 Hz;

iii. segmenting said high-pass filtered heart cycle data so as to generate a plurality of associated time data segments;

iv. for each time data segment of said plurality of associated time data segments;

a) calculating a frequency spectrum of said time data segment, wherein said frequency spectrum comprises a plurality of spectrum values for each of a corresponding plurality of associated spectral components;

b) calculating a normalized frequency spectrum of said time data segment, wherein for each spectral component of said corresponding plurality of associated spectral components, each of said plurality of spectrum values is normalized with respect to a corresponding associated noise floor, so as to generate a corresponding plurality of normalized spectrum values;

c) calculating a skew of said normalized frequency spectrum, wherein said skew is responsive to a second moment of said normalized spectrum values of said normalized frequency spectrum;

d) comparing said skew with a skew threshold; and

e) for each said spectral component of said plurality of associated spectral components;

i) comparing a normalized spectrum value of said corresponding plurality of normalized spectrum values with a bruit power detection threshold; and

ii) if said normalized spectrum value exceeds said bruit power detection threshold AND said skew is less than said skew threshold, then a frequency associated with said spectral component, a time associated with said time data segment, said skew, and said normalized spectrum value are associated with a bruit candidate associated with said cardiovascular disease; and

v. identifying whether or not said bruit candidate is likely associated with said cardiovascular disease, wherein said probability indicator is responsive to whether or not said bruit candidate is identified as being a bruit likely associated with said cardiovascular disease, and if so, said probability indicator is responsive to at least one characteristic of said bruit candidate; and

c. forwarding the probability indicator to at least one of the point of care and another location.

28. The physical computer-readable medium of claim 27, wherein said probability indicator is responsive to a probability of at least one repetitive bruit.

29. The physical computer-readable medium of claim 27, said method further comprising receiving background noise signals that were previously captured simultaneously with said cardiovascular sound signals at said point of care of the patient, and processing at said location remote from said point of care the received background noise signals, wherein said probability indicator is generated responsive to both said received cardiovascular sound signals and said received background noise signals.

30. The physical computer-readable medium of claim 29, wherein the operation of identifying whether or not said bruit candidate is likely associated with said cardiovascular disease comprises:

a. determining whether or not said bruit candidate is also detected within said received background noise signals; and

b. if said bruit candidate is detected within said received background noise signals, then treating said bruit candidate as not being a bruit likely associated with said cardiovascular disease.

31. The physical computer-readable medium of claim 30, wherein the operation of identifying whether or not said bruit candidate is likely associated with said cardiovascular disease comprises:

a. determining whether or not said bruit candidate is relatively close in time and frequency to a different bruit candidate that had been previously treated as not being a bruit likely associated with said cardiovascular disease; and

b. if said bruit candidate is relatively close in time and frequency to said different bruit candidate that had been previously treated as not being a bruit likely associated with said cardiovascular disease, then treating said bruit candidate as not being a bruit likely associated with said cardiovascular disease.

32. The physical computer-readable medium of claim 27, wherein the operation of calculating said frequency spectrum of said time data segment comprises transforming said time data segment with a Fast Fourier Transform.

33. The physical computer-readable medium of claim 32, said method further comprising:

a. forming a sample set from said time data segment and portions of time data segments adjacent thereto, wherein said sample set overlaps said time data segment and associated time data segments adjacent thereto; and

b. suppressing the contribution of time samples at the beginning or end of said sample set with either a Blackman or Kaiser Window windowing function, wherein said frequency spectrum is calculated with said Fast Fourier Transform by transforming said sample set after operation of said Blackman or Kaiser Window windowing function thereupon.

34. The physical computer-readable medium of claim 27, said method further comprising for each time data segment of said plurality of associated time data segments:

a. calculating a mean time data power of said high-pass filtered heart cycle data of said time data segment; and

b. comparing said mean time data power with a mean time data power threshold for each time data segment of said plurality of associated time data segments, wherein for each said spectral component of said plurality of associated spectral components, if said normalized spectrum value exceeds said bruit power detection threshold AND if said mean time data power exceeds said mean time data power threshold AND said skew is less than said skew threshold, then said frequency associated with said spectral component, said time associated with said time data segment, said skew, and said normalized spectrum value are associated with said bruit candidate associated with said cardiovascular disease.

35. A method comprising:

a. capturing cardiovascular sound signals of a patient at a point of care of the patient;

b. forwarding to a location remote from the point of care the captured cardiovascular sound signals; and

c. receiving from the location remote from the point of care a probability indicator indicative of the likelihood that the patient has cardiovascular disease, the probability indicator being generated based on the captured cardiovascular sound signals that were forwarded to the location remote from the point of care, wherein the operation of generating said probability indicator indicative of the likelihood that the patient has cardiovascular disease comprises;

i. parsing said captured cardiovascular sound signals so as to generate a corresponding set of parsed heart cycle signals representative of said captured cardiovascular sound signals relative to an associated heart cycle;

ii. generating high-pass filtered heart cycle data by high-pass filtering said parsed heart cycle signals so as to substantially retain frequency components above 300 Hz and so as to substantially attenuate frequency components below 200 Hz;

iii. segmenting said high-pass filtered heart cycle data so as to generate a plurality of associated time data segments;

iv. for each time data segment of said plurality of associated time data segments;

a) calculating a frequency spectrum of said time data segment, wherein said frequency spectrum comprises a plurality of spectrum values for each of a corresponding plurality of associated spectral components;

b) calculating a normalized frequency spectrum of said time data segment, wherein for each spectral component of said corresponding plurality of associated spectral components, each of said plurality of spectrum values is normalized with respect to a corresponding associated noise floor, so as to generate a corresponding plurality of normalized spectrum values;

c) calculating a skew of said normalized frequency spectrum, wherein said skew is responsive to a second moment of said normalized spectrum values of said normalized frequency spectrum;

d) comparing said skew with a skew threshold; and

e) for each said spectral component of said plurality of associated spectral components;

i) comparing a normalized spectrum value of said corresponding plurality of normalized spectrum values with a bruit power detection threshold; and

ii) if said normalized spectrum value exceeds said bruit power detection threshold AND said skew is less than said skew threshold, then a frequency associated with said spectral component, a time associated with said time data segment, said skew, and said normalized spectrum value are associated with a bruit candidate associated with said cardiovascular disease; and

v. identifying whether or not said bruit candidate is likely associated with said cardiovascular disease, wherein said probability indicator is responsive to whether or not said bruit candidate is identified as being a bruit likely associated with said cardiovascular disease, and if so, said probability indicator is responsive to at least one characteristic of said bruit candidate.

36. The method of claim 35, the captured cardiovascular sound signals further comprising a plurality of heart waveform signals each corresponding to a different location on the patient.

37. The method of claim 36, each heart waveform signal of said plurality of heart waveform signals further comprising a plurality of heart cycle signals, each heart cycle signal of said plurality of heart cycle signals corresponding to a heart cycle of the patient.

38. The method of claim 35, further comprising capturing an ECG signal and forwarding said ECG signal to the location remote from the point of care.

39. The method of claim 35, further comprising encrypting the cardiovascular sound signals prior to said forwarding to the location remote from the point of care.

40. The method of claim 35, wherein said probability indicator is responsive to a probability of at least one repetitive bruit.

41. The method of claim 35, further comprising receiving from the location remote from the point of care background noise signals that were previously captured simultaneously with said captured cardiovascular sound signals at said point of care of the patient, and processing at said location remote from said point of care the captured background noise signals, wherein said probability indicator is generated responsive to both said captured cardiovascular sound signals and said captured background noise signals.

42. The method of claim 41, wherein the operation of identifying whether or not said bruit candidate is likely associated with said cardiovascular disease comprises:

a. determining whether or not said bruit candidate is also detected within said captured background noise signals; and

b. if said bruit candidate is detected within said captured background noise signals, then treating said bruit candidate as not being a bruit likely associated with said cardiovascular disease.

43. The method of claim 42, wherein the operation of identifying whether or not said bruit candidate is likely associated with said cardiovascular disease comprises:

a. determining whether or not said bruit candidate is relatively close in time and frequency to a different bruit candidate that had been previously treated as not being a bruit likely associated with said cardiovascular disease; and

b. if said bruit candidate is relatively close in time and frequency to said different bruit candidate that had been previously treated as not being a bruit likely associated with said cardiovascular disease, then treating said bruit candidate as not being a bruit likely associated with said cardiovascular disease.

44. The method of claim 35, wherein the operation of calculating said frequency spectrum of said time data segment comprises transforming said time data segment with a Fast Fourier Transform.

45. The method of claim 44, further comprising:

a. forming a sample set from said time data segment and portions of time data segments adjacent thereto, wherein said sample set overlaps said time data segment and associated time data segments adjacent thereto; and

b. suppressing the contribution of time samples at the beginning or end of said sample set with either a Blackman or Kaiser Window windowing function, wherein said frequency spectrum is calculated with said Fast Fourier Transform by transforming said sample set after operation of said Blackman or Kaiser Window windowing function thereupon.

46. The method of claim 35, further comprising for each time data segment of said plurality of associated time data segments:

a. calculating a mean time data power of said high-pass filtered heart cycle data of said time data segment; and

b. comparing said mean time data power with a mean time data power threshold for each time data segment of said plurality of associated time data segments, wherein for each said spectral component of said plurality of associated spectral components, if said normalized spectrum value exceeds said bruit power detection threshold AND if said mean time data power exceeds said mean time data power threshold AND said skew is less than said skew threshold, then said frequency associated with said spectral component, said time associated with said time data segment, said skew, and said normalized spectrum value are associated with said bruit candidate associated with said cardiovascular disease.

47. A system comprising:

a. means for capturing cardiovascular sound signals of a patient at a point of care of the patient;

b. means for forwarding to a location remote from the point of care the captured cardiovascular sound signals; and

c. means for receiving from the location remote from the point of care a probability indicator indicative of the likelihood that the patient has cardiovascular disease, the probability indicator being generated based on the captured cardiovascular sound signals that were forwarded to the location remote from the point of care, wherein the operation of generating said probability indicator indicative of the likelihood that the patient has cardiovascular disease comprises;

i. parsing said captured cardiovascular sound signals so as to generate a corresponding set of parsed heart cycle signals representative of said captured cardiovascular sound signals relative to an associated heart cycle;

ii. generating high-pass filtered heart cycle data by high-pass filtering said parsed heart cycle signals so as to substantially retain frequency components above 300 Hz and so as to substantially attenuate frequency components below 200 Hz;

iii. segmenting said high-pass filtered heart cycle data so as to generate a plurality of associated time data segments;

iv. for each time data segment of said plurality of associated time data segments;

a) calculating a frequency spectrum of said time data segment, wherein said frequency spectrum comprises a plurality of spectrum values for each of a corresponding plurality of associated spectral components;

b) calculating a normalized frequency spectrum of said time data segment, wherein for each spectral component of said corresponding plurality of associated spectral components, each of said plurality of spectrum values is normalized with respect to a corresponding associated noise floor, so as to generate a corresponding plurality of normalized spectrum values;

c) calculating a skew of said normalized frequency spectrum, wherein said skew is responsive to a second moment of said normalized spectrum values of said normalized frequency spectrum;

d) comparing said skew with a skew threshold; and

e) for each said spectral component of said plurality of associated spectral components;

i) comparing a normalized spectrum value of said corresponding plurality of normalized spectrum values with a bruit power detection threshold; and

ii) if said normalized spectrum value exceeds said bruit power detection threshold AND said skew is less than said skew threshold, then a frequency associated with said spectral component, a time associated with said time data segment, said skew, and said normalized spectrum value are associated with a bruit candidate associated with said cardiovascular disease; and

v. identifying whether or not said bruit candidate is likely associated with said cardiovascular disease, wherein said probability indicator is responsive to whether or not said bruit candidate is identified as being a bruit likely associated with said cardiovascular disease, and if so, said probability indicator is responsive to at least one characteristic of said bruit candidate.

48. The system of claim 47, wherein said probability indicator is responsive to a probability of at least one repetitive bruit.

49. The system of claim 47, wherein the operation generating said probability indicator indicative of the likelihood that the patient has cardiovascular disease further comprises receiving from the location remote from the point of care background noise signals that were previously captured simultaneously with said cardiovascular sound signals at said point of care of the patient, and processing at said location remote from said point of care the captured background noise signals, wherein said probability indicator is generated responsive to both said captured cardiovascular sound signals and said captured background noise signals.

50. The system of claim 49, wherein the operation of identifying whether or not said bruit candidate is likely associated with said cardiovascular disease comprises:

a. determining whether or not said bruit candidate is also detected within said captured background noise signals; and

b. if said bruit candidate is detected within said captured background noise signals, then treating said bruit candidate as not being a bruit likely associated with said cardiovascular disease.

51. The system of claim 50, wherein the operation of identifying whether or not said bruit candidate is likely associated with said cardiovascular disease comprises:

a. determining whether or not said bruit candidate is relatively close in time and frequency to a different bruit candidate that had been previously treated as not being a bruit likely associated with said cardiovascular disease; and

b. if said bruit candidate is relatively close in time and frequency to said different bruit candidate that had been previously treated as not being a bruit likely associated with said cardiovascular disease, then treating said bruit candidate as not being a bruit likely associated with said cardiovascular disease.

52. The system of claim 47, wherein the operation of calculating said frequency spectrum of said time data segment comprises transforming said time data segment with a Fast Fourier Transform.

53. The system of claim 52, wherein the operation generating said probability indicator indicative of the likelihood that the patient has cardiovascular disease further comprises:

a. forming a sample set from said time data segment and portions of time data segments adjacent thereto, wherein said sample set overlaps said time data segment and associated time data segments adjacent thereto; and

b. suppressing the contribution of time samples at the beginning or end of said sample set with either a Blackman or Kaiser Window windowing function, wherein said frequency spectrum is calculated with said Fast Fourier Transform by transforming said sample set after operation of said Blackman or Kaiser Window windowing function thereupon.

54. The system of claim 47, wherein the operation generating said probability indicator indicative of the likelihood that the patient has cardiovascular disease further comprises for each time data segment of said plurality of associated time data segments:

a. calculating a mean time data power of said high-pass filtered heart cycle data of said time data segment; and

b. comparing said mean time data power with a mean time data power threshold for each time data segment of said plurality of associated time data segments, wherein for each said spectral component of said plurality of associated spectral components, if said normalized spectrum value exceeds said bruit power detection threshold AND if said mean time data power exceeds said mean time data power threshold AND said skew is less than said skew threshold, then said frequency associated with said spectral component, said time associated with said time data segment, said skew, and said normalized spectrum value are associated with said bruit candidate associated with said cardiovascular disease.

55. A physical computer-readable medium containing computer program instructions for a method comprising:

a. capturing cardiovascular sound signals of a patient at a point of care of the patient;

b. forwarding to a location remote from the point of care the captured cardiovascular sound signals; and

c. receiving from the location remote from the point of care a probability indicator indicative of the likelihood that the patient has cardiovascular disease, the probability indicator being generated based on the captured cardiovascular sound signals that were forwarded to the location remote from the point of care, wherein the operation of generating said probability indicator indicative of the likelihood that the patient has cardiovascular disease comprises;

i. parsing said captured cardiovascular sound signals so as to generate a corresponding set of parsed heart cycle signals representative of said captured cardiovascular sound signals relative to an associated heart cycle;

ii. generating high-pass filtered heart cycle data by high-pass filtering said parsed heart cycle signals so as to substantially retain frequency components above 300 Hz and so as to substantially attenuate frequency components below 200 Hz;

iii. segmenting said high-pass filtered heart cycle data so as to generate a plurality of associated time data segments;

iv. for each time data segment of said plurality of associated time data segments;

a) calculating a frequency spectrum of said time data segment, wherein said frequency spectrum comprises a plurality of spectrum values for each of a corresponding plurality of associated spectral components;

b) calculating a normalized frequency spectrum of said time data segment, wherein for each spectral component of said corresponding plurality of associated spectral components, each of said plurality of spectrum values is normalized with respect to a corresponding associated noise floor, so as to generate a corresponding plurality of normalized spectrum values;

c) calculating a skew of said normalized frequency spectrum, wherein said skew is responsive to a second moment of said normalized spectrum values of said normalized frequency spectrum;

d) comparing said skew with a skew threshold; and

e) for each said spectral component of said plurality of associated spectral components;

i) comparing a normalized spectrum value of said corresponding plurality of normalized spectrum values with a bruit power detection threshold; and

ii) if said normalized spectrum value exceeds said bruit power detection threshold AND said skew is less than said skew threshold, then a frequency associated with said spectral component, a time associated with said time data segment, said skew, and said normalized spectrum value are associated with a bruit candidate associated with said cardiovascular disease; and

v. identifying whether or not said bruit candidate is likely associated with said cardiovascular disease, wherein said probability indicator is responsive to whether or not said bruit candidate is identified as being a bruit likely associated with said cardiovascular disease, and if so, said probability indicator is responsive to at least one characteristic of said bruit candidate.

56. The physical computer-readable medium of claim 55, wherein said probability indicator is responsive to a probability of at least one repetitive bruit.

57. The physical computer-readable medium of claim 55, said method further comprising receiving from the location remote from the point of care background noise signals that were previously captured simultaneously with said cardiovascular sound signals at said point of care of the patient, and processing at said location remote from said point of care the captured background noise signals, wherein said probability indicator is generated responsive to both said captured cardiovascular sound signals and said captured background noise signals.

58. The physical computer-readable medium of claim 57, wherein the operation of identifying whether or not said bruit candidate is likely associated with said cardiovascular disease comprises:

a. determining whether or not said bruit candidate is also detected within said captured background noise signals; and

b. if said bruit candidate is detected within said captured background noise signals, then treating said bruit candidate as not being a bruit likely associated with said cardiovascular disease.

59. The physical computer-readable medium of claim 58, wherein the operation of identifying whether or not said bruit candidate is likely associated with said cardiovascular disease comprises:

a. determining whether or not said bruit candidate is relatively close in time and frequency to a different bruit candidate that had been previously treated as not being a bruit likely associated with said cardiovascular disease; and

b. if said bruit candidate is relatively close in time and frequency to said different bruit candidate that had been previously treated as not being a bruit likely associated with said cardiovascular disease, then treating said bruit candidate as not being a bruit likely associated with said cardiovascular disease.

60. The physical computer-readable medium of claim 55, wherein the operation of calculating said frequency spectrum of said time data segment comprises transforming said time data segment with a Fast Fourier Transform.

61. The physical computer-readable medium of claim 60, said method further comprising:

a. forming a sample set from said time data segment and portions of time data segments adjacent thereto, wherein said sample set overlaps said time data segment and associated time data segments adjacent thereto; and

b. suppressing the contribution of time samples at the beginning or end of said sample set with either a Blackman or Kaiser Window windowing function, wherein said frequency spectrum is calculated with said Fast Fourier Transform by transforming said sample set after operation of said Blackman or Kaiser Window windowing function thereupon.

62. The physical computer-readable medium of claim 55, said method further comprising for each time data segment of said plurality of associated time data segments:

a. calculating a mean time data power of said high-pass filtered heart cycle data of said time data segment; and

b. comparing said mean time data power with a mean time data power threshold for each time data segment of said plurality of associated time data segments, wherein for each said spectral component of said plurality of associated spectral components, if said normalized spectrum value exceeds said bruit power detection threshold AND if said mean time data power exceeds said mean time data power threshold AND said skew is less than said skew threshold, then said frequency associated with said spectral component, said time associated with said time data segment, said skew, and said normalized spectrum value are associated with said bruit candidate associated with said cardiovascular disease.