Method and system of an acoustic scene analyzer for body sounds

• US 8,475,396 B2
• Filed: 02/10/2012
• Issued: 07/02/2013
• Est. Priority Date: 02/11/2011
• Status: Active Grant

• Current Assignee: AventuSoft, LLC
• Original Assignee: AventuSoft, LLC
• Inventors: Jones, Kevin, Srivastava, Garima
• Primary Examiner(s): Gedeon, Brian T
• Application Number:
  US13/370,542
• Publication Number:
  US 20120209132A1
• Time in Patent Office: 508 Days
• Field of Search: 600/528, 600/586, 381/67, 381/94.1
• US Class Current: 600/586
• 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: G01S 5/18 using ultrasonic, sonic, or infrasonic waves
• CPC Class Codes: G10L 21/028 using properties of sound source
• CPC Class Codes: G10L 21/10 Transforming into visible information
• CPC Class Codes: G10L 25/66 for extracting parameters related to health condition detecting or measuring for diagnostic purposes A61B5/00

Abstract

A system and method for visualizing auditory scene analysis by way of a portable device is provided. In one embodiment, the method steps include capturing multiple sounds from a sensor array of microphones connected to the portable device, performing auditory scene analysis on detected body sounds in accordance with a psychoacoustic representation of body organ functions, and rendering to a display of the portable device a visualization of auditory scene auscultation of the body sounds, including user input functionality for separated sound source tracks, sound source identifier tracks, and sound source location trajectories.

Claims

1. A portable device for visualizing auditory scene auscultation of body sounds, the portable device comprising:

A visual display with user input;

a sensor array of microphones to capture body sounds;

a processor coupled to the visual display and the sensor array configured to;

perform auditory scene analysis in accordance with a psychoacoustic interpretation of the captured body sounds;

render to the display a visualization of auditory scene auscultation of the body sounds, including;

separated sound source tracks;

sound source identifier tracks; and

sound source location trajectories,

a speaker coupled to the processor to play the separated sound source tracks responsive to user input; and

a power supply to provide power to electronic components of the portable device.

2. The portable device of claim 1, wherein the sensor array captures from body sounds both vibrations from infrasound and acoustically transmitted audible sound.

3. The portable device of claim 1, where the processor programmed in coupled sequence:

a first software module to extract features from the auditory scene analysis based on psychoacoustic characteristics of the body sounds;

a second software module to identify unique sources of the body sounds from the features; and

a third software module to estimate masking patterns of each identified sound source based on sound generation characteristics of the sound sources.

4. The portable device of claim 3, further comprising:

a memory that stores frequency masking characteristics of multiple body organs,

wherein the processor evaluates the frequency masking characteristics of the multiple body organs for separating identified sound sources.

5. The portable device of claim 4, where the memory stores masking patterns of regular and irregular:

heartbeat conditions, respiratory conditions, and gastrointestinal conditions.

6. The portable device of claim 2, further comprising in coupled sequence:

a fourth software module to perform sound source separation based on the identified unique sound sources, masking patterns, and extracted features.

7. The portable device of claim 2, further comprising in coupled sequence:

a fifth software module to localize sound sources and generate a location for identified sound sources; and

a sixth software module to track the location of the identified sound sources.

8. A method for visualizing auditory scene auscultation of body sounds by way of a portable device, the method comprising the steps of:

capturing body sounds from a sensor array detecting both vibrations from infrasound and acoustically transmitted audible sound connected to the portable device;

performing auditory scene analysis on the captured body sounds in accordance with a psychoacoustic representation of body organ functions;

rendering to a display of the portable device a visualization of auditory scene auscultation of the body sounds, including;

separated sound source tracks, sound source identifier tracks; and

sound source location trajectories.

9. The method of claim 8, comprising sound source tracking by:

short-term monitoring of location cues to discriminate distances between sources; and

long-term monitoring of acoustic cues for resolving features associated with sound source identities,

wherein the location cues and acoustic cues are selectively stored and retrieved in a history within a memory of the portable device according to a coincidence detection model.

10. The method of claim 8, comprising calculating inter-aural time differences between body sounds captured at the microphone for place-coding and localizing the sound source according to the inter-aural time and level differences.

11. The method of claim 8, comprising calculating a biological representation of loudness, frequency, timbre and pitch for analyzing abnormalities occurring in identified sound source sites.

12. The method of claim 8, comprising separating a composite signal of the body sounds into sound components dues to individual body structures and body activities.

13. The method of claim 8, analyzing abnormalities occurring in specific source sites and reporting a location of the abnormality.

14. The method of claim 8, performing amplitude compression and frequency compression of the captured body sound in accordance with the psychoacoustic mapping procedure to compensate for the biological representation of loudness, pitch and timbre of human hearing.

15. The method of claim 8, comprising translating inaudible acoustic signals in a masked frequency region of an identified body organ sound spectrum to frequencies in an audible frequency region according to a psychoacoustic frequency mapping.

16. The method of claim 8, comprising generating a two-dimensional representation of locations of various heart sounds, arterial and ventricular septal defects, and systolic or diastolic murmurs.

17. A portable device for visualizing auscultation of body sounds, the portable device comprising:

a visual display with user input;

a sensor array of microphones to capture body sounds comprising vibrations from infrasound and acoustically transmitted audible sound;

a processor coupled to the visual display and the sensor array programmed with sequential order of processing;

a first software module to extract features from an auditory scene analysis based on psychoacoustic characteristics of the body sounds;

a second software module to identify unique sources of the body sounds from the features;

a third software module to estimate masking patterns of each identified sound source based on sound generation characteristics of the sound sources;

a fourth software module to perform sound source separation based on the identified unique sound sources, masking patterns, and extracted features;

a fifth software module to localize sound sources and generate a location for identified sound sources; and

a sixth software module to track the location of the identified sound sources,

wherein the processor determines if the sensor array is in non-body contact or body contact and compensates accordingly during psychoacoustic interpretation of the captured body sounds, and renders to the display a visualization of auditory scene auscultation of the body sounds.

18. The portable device of claim 17, further comprising a speaker or headset coupled to the processor to play the separated sound source tracks responsive to user input.

19. The portable device of claim 17, wherein the processor includes a

software module for two organizational processes:

segmentation to decompose the auditory scene into time-frequency segments; and

grouping to combine the time-frequency segments from a same source into a single perceptual stream.

20. The portable device of claim 19, wherein with the grouping:

simultaneous organization integrates time-frequency segments that overlap in time, and

sequential organization groups segments across time.