Sleep apnea diagnosis system and method of generating information using non-obtrusive audio analysis

Patent metadata

Field	Value
Publication number	US-9833189-B2
Application number	US-201314647475-A
Country	US
Kind code	B2
Filing date	Dec 16, 2013
Priority date	Dec 17, 2012
Publication date	Dec 5, 2017
Grant date	Dec 5, 2017

Abstract

An electronic apparatus includes an array of microphones for detecting audible sounds generated by a patient and for generating audio information representing the detected audible sounds, a first beamformer having a first adaptability speed and configured to generate first audio information and first noise information from the audio information, a second beamformer having a second adaptability speed which is slower than the first adaptability speed, the second adaptive beamformer configured to generate second audio information and second noise information from the audio information, an audio classification unit for generating audio classification information based on the first audio information, a head movement detection unit for generating head movement information based on at least one of the second audio information, the first noise information, and the second noise information, and a diagnosis unit for determining a sleep apnea diagnosis based on the audio classification information and the head movement information.

First claim

The invention claimed is:

1. An electronic apparatus comprising:
an array of microphones for detecting audible sounds generated by a patient and for generating audio information representing the detected audible sounds generated by the patient;
a first beamformer structured to form a first beam and having a first adaptability speed, wherein the first beamformer is configured to generate first audio information and first noise information from the audio information, wherein the first audio information is comprised of audible sounds from within the first beam and the first noise information is comprised of audible sounds from outside the first beam;
a second beamformer structured to form a second beam and having a second adaptability speed which is slower than the first adaptability speed, wherein the second adaptive beamformer is configured to generate second audio information and second noise information from the audio information, wherein the second audio information is comprised of audible sounds from within the second beam and the second noise information is comprised of audible sounds from outside second beam;
an audio classification unit for generating audio classification information based on the first audio information;
a head movement detection unit for generating head movement information based on changes in the second audio information or differences between the first noise information and the second noise information; and
a diagnosis unit for determining a sleep apnea diagnosis based on the audio classification information and the head movement information.

2. The electronic apparatus of claim 1 , wherein the first adaptability speed is fast enough to track normal head movements of the patient and the second adaptability speed is too slow to track the normal head movements of the patient.

3. The electronic apparatus of claim 1 , wherein the first beamformer and the second beamformer are one of filter and sum type beamformers and delay and sum type beamformers.

4. The electronic apparatus of claim 1 , wherein the head movement detection unit comprises a first head movement detection unit and a second head movement detection unit and the head movement information comprises first head movement information and second head movement information, wherein the first head movement detection unit is configured to calculate the first head movement information based on the second audio information and the second head movement detection unit is configured to calculate the second head movement information based on the first noise information and the second noise information.

5. The electronic apparatus of claim 4 , wherein the first head movement detection unit is an envelope detection unit for calculating the first head movement information based on variations in the second audio information.

6. The electronic apparatus of claim 4 , wherein the second head movement detection unit is structured to calculate the second head movement information based on differences between the first noise information and the second noise information.

7. The electronic apparatus of claim 1 , wherein the array of microphones are located within about 10 cm or less of each other.

8. A method of diagnoding sleep apnea, the method comprising:
detecting audible sounds with an array of microphones, wherein the audible sounds are generated by a patient, and generating audio information representing the detected audible sounds generated by the patient;
processing the audio information with a first beamformer having a first beam and a first adaptability speed and generating first audio information and first noise information from the audio information, wherein the first audible information is comprised of audible sounds from within the first beam and the first noise information is comprised of audible sounds from outside the first beam;
processing the audio information with a second beamformer having a second beam and a second adaptability speed which is slower than the first adaptability speed and generating second audio information and second noise information from the audio information, wherein the second audio information is comprised of audible sounds from within the second beam and the second noise information is comprised of audible sounds from outside the second beam;
generating audio classification information based on the first audio information;
generating head movement information based on changes in the second audio information or differences between the first noise information and the second noise information; and
determining a sleep apnea diagnosis based on the audio classification information and the head movement information.

9. The method of claim 8 , wherein the first adaptability speed is fast enough to track normal head movements of the patient and the second adaptability speed is too slow to track the normal head movements of the patient.

10. The method of claim 8 , wherein the first beamformer and the second beamformer are one of filter and sum type beamformers and delay and sum type beamformers.

11. The method of claim 8 , wherein the generating head movement information comprises generating first head movement information based on the second audio information and generating second head movement information based on the first noise information and the second noise information.

12. The method of claim 11 , wherein the generating first head movement information comprises performing envelope analysis on the second audio information to determine variations in the second audio information.

13. The method of claim 11 , wherein the generating second head movement information comprises performing filter analysis to determine differences between the first noise information and the second noise information.

14. The method of claim 11 , wherein the array of microphones are located within about 10 cm or less of each other.

15. A non-transitory computer readable medium storing one or more programs, including instructions, which when executed by a computer, causes the computer to perform the method as claimed in claim 8 .

Assignees

• Koninklijke Philips Nv

Inventors

• Triki Mahd

Classifications

• A61B2562/04

  Arrangements of multiple sensors of the same type · CPC title

• A61B5/7207

  of noise induced by motion artifacts · CPC title

• A61B7/003

  Detecting lung or respiration noise · CPC title

• H04R1/406

  microphones · CPC title

• H04R3/005

  for combining the signals of two or more microphones (specially adapted for hearing aids H04R25/407) · CPC title