Extraction of one or more discrete heart sounds from heart sound information

• US 20050043643A1
• Filed: 07/16/2004
• Published: 02/24/2005
• Est. Priority Date: 07/18/2003
• Status: Active Grant

• Current Assignee: Board of Trustees of The University of Illinois
• Original Assignee: Board of Trustees of The University of Illinois
• Inventors: Priemer, Roland
• Granted Patent: US 7,351,207 B2
• Time in Patent Office: Days
• US Class Current: 600/528
• CPC Class Codes: A61B 2562/0204 Acoustic sensors
• CPC Class Codes: A61B 2562/046 in a matrix array
• CPC Class Codes: A61B 7/02 Stethoscopes acoustic details thereof G10K11/00
• CPC Class Codes: A61B 7/023 for introduction into the body, e.g. into the oesophagus oesophageal ECG probes A61B5/285
• 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

Abstract

A heart sound analyzer component of an apparatus in one example extracts from composite heart sound information one or more discrete heart sounds of one or more corresponding distinct heart sound sources.

Claims

1. An apparatus, comprising:

a heart sound analyzer component that extracts from composite heart sound information one or more discrete heart sounds of one or more corresponding distinct heart sound sources.

2. The apparatus of claim 1, wherein the heart sound analyzer component employs independent component analysis to extract the one or more discrete heart sounds from the composite heart sound information.

3. The apparatus of claim 2, wherein a heart comprises a plurality of distinct heart sound sources, wherein the plurality of distinct heart sound sources comprise the one or more distinct heart sound sources;

wherein the heart sound analyzer component runs the independent component analysis in a neural network construct that employs a plurality of nodes in number at least equal to a number of the plurality of distinct heart sound sources to extract the one or more discrete heart sounds from the composite heart sound information.

4. The apparatus of claim 1, wherein the one or more discrete heart sounds comprise a first discrete heart sound of a first distinct sound source of the one or more distinct heart sound sources;

wherein the one or more discrete heart sounds comprise a second discrete heart sound of a second distinct sound source of the one or more distinct heart sound sources;

wherein the first discrete heart sound and the second discrete heart sound occur contemporaneously, wherein the first discrete heart sound and the second discrete heart sound are mixed in the composite heart sound information;

wherein the heart sound analyzer component employs independent component analysis of the composite heart sound information to create statistically independent outputs of one or more of the first discrete heart sound and the second discrete heart sound.

5. The apparatus of claim 1, wherein the heart sound analyzer component extracts from the composite heart sound information the one or more discrete heart sounds that comprise one or more of:

a mitral valve sound;

a tricuspid valve sound;

an aortic valve sound;

a pulmonary valve sound;

a noise sound component;

a low frequency sound; and

an abnormal heart sound.

6. The apparatus of claim 1, further comprising:

a stethoscope sensor head that is locatable over a heart of a patient;

wherein the heart comprises a plurality of distinct heart sound sources, wherein the plurality of distinct heart sound sources comprise the one or more distinct heart sound sources, wherein the stethoscope sensor head comprises a plurality of microphones in number at least equal to a number of the plurality of distinct heart sound sources;

wherein the plurality of microphones obtain and pass the composite heart sound information to the heart sound analyzer component.

7. The apparatus of claim 1, further comprising:

an esophageal stethoscope sensor head that may pass down an esophagus of a patient;

wherein a heart of the patient comprises a plurality of distinct heart sound sources, wherein the plurality of distinct heart sound sources comprise the one or more distinct heart sound sources, wherein the esophageal stethoscope sensor head comprises a flexible tube that contains a plurality of microphones in number at least equal to a number of the plurality of distinct heart sound sources;

wherein the plurality of microphones obtain and pass the composite heart sound information to the heart sound analyzer component.

8. The apparatus of claim 1, wherein the one or more discrete heart sounds comprise an abnormal heart sound, wherein the abnormal heart sound overlaps with one or more other sounds in the composite heart sound information;

wherein the abnormal heart sound is associated with a dysfunction of a heart of a patient, wherein the heart sound analyzer component extracts the abnormal heart sound from the composite heart sound information and outputs the abnormal heart sound to a doctor for diagnosis of the dysfunction.

9. The apparatus of claim 1, wherein the heart sound analyzer comprises a plurality of data buffers that allow real-time heart sound analysis.

10. The apparatus of claim 9, wherein the heart sound analyzer component runs independent component analysis on a first portion of the composite heart sound information from a first data buffer of the plurality of data buffers contemporaneously with steering a second portion of the composite heart sound information into a second data buffer of the plurality of data buffers;

wherein upon completion of the independent component analysis on the first portion of the composite heart sound information, the heart sound analyzer runs independent component analysis on the second portion of the composite heart sound information from the second data buffer.

11. The apparatus of claim 1, wherein the heart sound analyzer component runs data whitening software on the composite heart sound information to reduce a noise sound component of the composite heart sound information;

wherein after running the data whitening software, the heart sound analyzer component employs dimensionality reduction software to determine one or more eigen values of the composite heart sound information, wherein the heart sound analyzer reduces a number of the one or more eigen values until the one or more eigen values are equal in number to a number of the one or more discrete heart sounds;

wherein after running the dimensionality reduction software, the heart sound analyzer component employs blind source separation software to extract the one or more discrete heart sounds from the composite heart sound information.

12. The apparatus of claim 11, wherein after extraction of the one or more discrete heart sounds from the composite heart sound information, the heart sound analyzer component runs digital filtering software and/or channel permutation software to spectrally modify, spectrally shift, and/or amplify one or more of the one or more discrete heart sounds.

13. The apparatus of claim 12, wherein the heart sound analyzer component runs the digital filtering software and/or the channel permutation software to identify the one or more corresponding distinct heart sound sources associated with the one or more discrete heart sounds.

14. An apparatus, comprising:

a plurality of microphones of a stethoscope sensor head that are employable to capture a plurality of composite heart sounds that allow extraction of one or more discrete heart sounds from the plurality of composite heart sounds.

15. The apparatus of claim 14, wherein a heart comprises a plurality of distinct heart sound sources, wherein the stethoscope sensor head comprises the plurality of microphones in number at least equal to a number of the plurality of distinct heart sound sources to allow the extraction of one or more discrete heart sounds from the plurality of composite heart sounds;

wherein the plurality of microphones are sufficiently spaced to capture separate composite heart sounds of the plurality of composite heart sounds.

16. The apparatus of claim 14, wherein a heart comprises a plurality of distinct heart sound sources, wherein the plurality of microphones are arranged on the stethoscope sensor head in a geometry compatible with a geometry of the plurality of distinct heart sound sources;

wherein the plurality of distinct heart sound sources comprise a plurality of heart valves, wherein the plurality of microphones are arranged on the stethoscope sensor head to be near the plurality of heart valves to capture strong audio signals from the plurality of heart valves.

17. The apparatus of claim 14, wherein the stethoscope sensor head comprises an esophageal stethoscope sensor head that may pass down an esophagus of a patient;

wherein a heart of the patient comprises a plurality of distinct heart sound sources, wherein the plurality of distinct heart sound sources comprise the one or more distinct heart sound sources, wherein the esophageal stethoscope sensor head comprises a flexible tube that contains a plurality of microphones in number at least equal to a number of the plurality of distinct heart sound sources;

wherein the plurality of microphones obtain and pass the composite heart sound information to a heart sound analyzer component for extraction of the one or more discrete heart sounds from the plurality of composite heart sounds.

18. A method, comprising the steps of:

obtaining composite heart sound information from a heart; and

extracting from the composite heart sound information one or more discrete heart sounds of one or more corresponding distinct heart sound sources.

19. The method of claim 18, wherein the heart comprises a plurality of distinct heart sound sources, wherein the plurality of distinct heart sound sources comprise the one or more distinct heart sound sources;

wherein the step of obtaining the composite heart sound information from the heart comprises the step of;

capturing the composite heart sound information with a stethoscope sensor head that is locatable over a heart of a patient and comprises a plurality of microphones in number at least equal to a number of the plurality of distinct heart sound sources;

passing the composite heart sound information to the heart sound analyzer component for extraction of the one or more discrete heart sounds from the composite heart sound information.

20. The method of claim 18, wherein a heart comprises a plurality of distinct heart sound sources, wherein the plurality of distinct heart sound sources comprise the one or more distinct heart sound sources;

wherein the one or more discrete heart sounds comprise a first discrete heart sound of a first distinct sound source of the plurality of distinct heart sound sources, wherein the one or more discrete heart sounds comprise a second discrete heart sound of a second distinct sound source of the plurality of distinct heart sound sources, wherein the first discrete heart sound and the second discrete heart sound occur contemporaneously and are mixed in the composite heart sound information;

wherein the step of extracting from the composite heart sound information the one or more discrete heart sounds of the one or more corresponding distinct heart sound sources comprises the step of;

running independent component analysis in a neural network construct that employs a plurality of nodes in number at least equal to a number of the plurality of distinct heart sound sources to extract one or more of the first heart sound and/or the second heart sound from the composite heart sound information.