Method for generating output data

• US 7,761,302 B2
• Filed: 06/07/2005
• Issued: 07/20/2010
• Est. Priority Date: 06/03/2005
• Status: Active Grant

• Current Assignee: South Manchester University Hospital N H S Trust
• Original Assignee: South Manchester University Hospital N H S Trust
• Inventors: Smith, Jaclyn Ann, McGuinness, Kevin, Woodcock, Ashley Arthur
• Primary Examiner(s): Vo; Huyen X.
• Application Number:
  US11/146,905
• Publication Number:
  US 20060277037A1
• Time in Patent Office: 1,869 Days
• Field of Search: 704/206, 704/207, 704/209, 704/243, 704/244, 704/231, 704/236, 704/245, 704/250, 704/270, 704/256, 600529-542
• US Class Current: 704/270
• CPC Class Codes: A61B 7/003 Detecting lung or respiration noise
• CPC Class Codes: G10L 17/26 Recognition of special voice characteristics, e.g. for use in lie detectors; Recognition of animal voices

Abstract

A method for generating output data identifying a class of a predetermined plurality of classes. The method comprises receiving data representing an acoustic signal; determining an amplitude of at least a first predetermined frequency component of said acoustic signal; and comparing the or each amplitude with a respective primary threshold; and generating output data identifying one of said classes to which said acoustic signal should be allocated, based upon said comparison.

Claims

1. A calibration method for calibrating an apparatus comprising a process to differentiate acoustic signals attributable to cough from other acoustic signals, the method comprising:

receiving, by the processor, first data representing a first acoustic signal representing at least one cough;

receiving, by the processor, second data representing a second acoustic signal representing an acoustic event other than cough; and

processing, by the processor, said first and second data to generate configuration data for said apparatus, said configuration data defining at least one rule which when applied differentiates acoustic signals attributable to cough from other acoustic signals, wherein said rule is defined by one or more selected frequencies each having an associated amplitude and the one or more selected frequencies are each selected from a plurality of predetermined frequencies such that the amplitude of said first and second acoustic signals at said selected frequencies has a difference greater than the difference between the amplitude of said first and second acoustic signals at all others of said predetermined frequencies.

2. The method according to claim 1, wherein the amplitude associated with each selected frequency is defined to be centrally placed between the amplitude of said first acoustic signal, and said amplitude of said second acoustic signal at the respective selected frequency.

3. The method according to claim 1, further comprising:

filtering at least one of said first and second data to remove vibrations below a predetermined amplitude.

4. The method according to claim 1, further comprising:

filtering the received data representing at least one cough by removing data that represents vibrations with a frequency that exceeds an upper cough cut off frequency.

5. The method according to claim 1, further comprising:

filtering at least one of said first and second data to remove frequency components above a predetermined frequency threshold.

6. The method according to claim 1, further comprising:

filtering at least one of said first and second data to remove frequency components below a predetermined frequency threshold.

7. A device storing processor executable program instructions to perform a method of calibrating an apparatus to differentiate acoustic signals attributable to cough from other acoustic signals according to claim 1.

8. A computer apparatus comprising:

a program memory containing processor readable instructions; and

a processor for reading and executing the instructions contained in the program memory;

wherein said processor readable instructions comprise instructions controlling the processor to carry out the method of claim 1.

9. A calibration method for calibrating an apparatus comprising a processor to differentiate acoustic signals attributable to cough from other acoustic signals, the method comprising:

receiving, by the processor, first data representing a first acoustic signal representing at least one cough;

receiving, by the processor, second data representing a second acoustic signal representing an acoustic event other than cough; and

processing, by the processor, said first data and second data together to generate configuration data for said apparatus, said configuration data defining at least one single rule which when applied to a received acoustic signal generates data indicating whether said received acoustic signal is attributable to cough or is attributable to an acoustic event other than cough, wherein the at least one rule is defined by at least one selected frequency and a respective threshold amplitude associated with each of the one or more selected frequencies, and the at least one rule is configured to be applied to said received acoustic signal by a comparison of the amplitude of the at least one selected frequency component of the received signal with the threshold amplitude associated with that frequency.

10. A calibration method according to claim 9, wherein the second acoustic signal represents speech sounds emanating from said subject.