Breathing disorder identification, characterization and diagnosis methods, devices and systems

• US 9,801,590 B2
• Filed: 05/17/2012
• Issued: 10/31/2017
• Est. Priority Date: 05/17/2011
• Status: Active Grant

• Current Assignee: University Health Network
• Original Assignee: University Health Network
• Inventors: Alshaer, Hisham, Fernie, Geoffrey Roy, Bradley, T. Douglas, Akl, Ahmad Ziad
• Primary Examiner(s): Stout, Michael C
• Assistant Examiner(s): Kolderman, Nicholas E
• Application Number:
  US14/117,963
• Publication Number:
  US 20140188006A1
• Time in Patent Office: 1,993 Days
• CPC Class Codes: A61B 5/4818 Sleep apnoea
• CPC Class Codes: A61B 5/6803 Head-worn items, e.g. helmets, masks, headphones or goggles
• 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 5/7475 User input or interface means, e.g. keyboard, pointing device, joystick
• CPC Class Codes: A61B 7/003 Detecting lung or respiration noise

Abstract

Disclosed herein are breathing disorder identification, characterization and diagnosis methods, devices and systems. In general the disclosed methods, devices and systems may rely on the characterization of breath sound amplitudes, periodic breath sounds and/or aperiodic breath sounds to characterize a breathing disorder as obstructive (e.g. obstructive sleep apnea—OSA) or non-obstructive (e.g. central sleep apnea—CSA).

Claims

1. A computed-implemented method, automatically implemented by one or more processors of a computing device, for automatically characterizing recorded breath sounds as indicative of one of obstructive and central sleep apnea, the method comprising:

identifying by the at least one process a segment of the recorded breath sounds encompassing at least one of an apnea and a hypopnea, said segment comprising a falling edge and a rising edge separated by a low-amplitude segment associated with said apnea or hypopnea;

automatically evaluating by the at least one processor a breath sound amplitude profile pattern of said identified segment, wherein said pattern is at least partially defined by said falling edge and said rising edge, against one or more preset sound amplitude profile pattern criteria set to distinguish between segments indicative of obstructive sleep apnea and central sleep apnea;

wherein said evaluating comprises evaluating a similarity between said rising edge and said falling edge of said identified segment, and

wherein said one or more criteria comprise a similarity threshold set to distinguish obstructive from central apnea events;

classifying by the at least one processor said identified segment as being indicative of one of obstructive and central sleep apnea based on a result of said evaluating step; and

outputting an indication as to one of obstructive and central sleep apnea based on said classifying.

2. The method of claim 1, said profile pattern comprising a fall/rise pattern characterized as either a crescendo-decrescendo pattern indicative of central sleep apnea or a gradual fall-abrupt rise pattern indicative of obstructive sleep apnea.

3. The method of claim 1, said criteria comprising a similarity index calculated by the at least one processor by comparing said falling edge and said rising edge using a Dynamic Time Warping (DTW) subroutine.

4. The method of claim 1 further comprising repeating the steps for multiple segments.

5. The method of claim 1, further comprising the steps of:

automatically identifying by the at least one processor inspiration and expiration phases of the recorded breath sounds;

isolating by the at least one processor expiration breath sounds from said recorded sounds; and

extracting by the at least one processor said breath sound amplitude profile from said isolated expiration sounds, whereby said segment is representative of expiration sounds only.

6. The method of claim 1, said step of identifying said segment further comprising:

scanning by the at least one processor said extracted breath sound amplitude profile to identify a prospect event segment;

evaluating by the at least one processor characteristics of said prospect event segment for consistency with at least one of;

one or more preset apnea-specific criteria, and one or more preset hypopnea-specific criteria distinct from said apnea-specific criteria; and

identifying by the at least one processor said prospect event as encompassing an apnea or a hypopnea upon it satisfying each of said one or more apnea-specific criteria or each of said one or more preset hypopnea-specific criteria.

7. The method of claim 1, further comprising:

extracting by the at least one processor one or more spectral characteristics associated with said identified segment; and

evaluating by the at least one processor said extracted spectral characteristics against one or more corresponding classification criteria predetermined to distinguish obstructive breathing events from non-obstructive breathing events;

said classifying step comprising combining by the at least one processor amplitude-related and spectrum-related evaluation results to classify said identified segment.

8. The method of claim 7, said one or more spectral characteristics associated with at least a respective one of periodic breath sounds and aperiodic breath sounds manifested in a period encompassing said identified segment.

9. The method of claim 1, further comprising the step of acquiring breath sounds via a face mask comprising a microphone disposed, upon a candidate wearing the mask during sleep, at a distance above a nose and mouth area of the candidate'"'"'s face.

10. A non-transitory computer-readable medium comprising statements and instructions stored thereon for implementation by one or more processors of a computing system to automatically characterize recorded breath sounds as indicative of one of obstructive and central sleep apnea, that cause the computing system to perform the following operations:

identify a segment of the recorded breath sounds encompassing at least one of an apnea and a hypopnea, said segment comprising a falling edge and a rising edge separated by a low-amplitude segment associated with said apnea or hypopnea;

automatically evaluate a breath sound amplitude profile pattern of said identified segment, wherein said pattern is at least partially defined by said falling edge and said rising edge, against one or more preset sound amplitude profile pattern criteria set to distinguish between segments indicative of obstructive sleep apnea and central sleep apnea;

wherein said evaluating comprises evaluating a similarity between said rising edge and said falling edge of said identified segment, and wherein said one or more criteria comprise a similarity threshold set to distinguish obstructive from central apnea events;

classify said identified segment as being indicative of one of obstructive and central sleep apnea based on a result of said evaluating step; and

output an indication as to one of obstructive and central sleep apnea based on said classified segment.

11. The non-transitory computer-readable medium of claim 10, said profile pattern comprising a fall/rise pattern characterized as either a crescendo-decrescendo pattern indicative of central sleep apnea or a gradual fall-abrupt rise pattern indicative of obstructive sleep apnea.

12. A system for automatically characterizing recorded breath sounds as indicative of one of obstructive and central sleep apnea, the system comprising:

one or more processors;

memory storing instructions, the instructions comprising instructions that, when executed by the one or more processors, cause the processors to;

identify a segment of the recorded breath sounds encompassing at least one of an apnea and a hypopnea;

automatically evaluate a breath sound amplitude profile pattern of said identified segment, wherein said pattern is at least partially defined by said falling edge and said rising edge, against one or more preset sound amplitude profile pattern criteria set to distinguish between segments indicative of obstructive sleep apnea and central sleep apnea;

wherein said evaluating comprises evaluating a similarity between said rising edge and said falling edge of said identified segment, and wherein said one or more criteria comprise a similarity threshold set to distinguish obstructive from central apnea events;

classify said identified segment as being indicative of one of obstructive and central sleep apnea based on a result of said evaluating step; and

output an indication as to one of obstructive and central sleep apnea based on said classified segment.

13. The system of claim 12, further comprising a face mask having a microphone mounted thereon and reproducibly disposable, upon the candidate wearing the mask during sleep, at a distance above a nose and mouth area of the candidate so to intercept and capture airflow sounds emanating therefrom for processing.

14. The system of claim 13, said mask further comprising a removable data storage medium operatively coupled to said microphone for storing recorded breath sounds thereon prior to processing.