Detecting subjects with disordered breathing

1. A sound analysis apparatus for detecting a presence of multiple subjects in an environment with sleep disordered breathing, the apparatus comprising:

an acoustic sensor configured to acquire an acoustic signal in the environment;

one or more processors configured to:

determine a plurality of acoustic signal components from the acquired acoustic signal, the plurality of acoustic signal components differing from one another in respect of at least one signal characteristic or property;

determine a plurality of signal envelopes or energy signals based on the acoustic signal components, at least one signal envelope being derived from each of the acoustic signal components of the plurality of acoustic signal components; and

analyze the determined plurality of signal envelopes or energy signals to detect whether there are one or more subjects in the environment with sleep disordered breathing, wherein analyzing comprises identifying (i) a number of repetition patterns in the determined plurality of signal envelopes or energy signals and (ii) a repetition interval for each of the identified repetition patterns, and further analyzing the number of repetition patterns and the repetition intervals to subdivide a time axis of a whole duration from a start to an end of a sound measurement of the acquired acoustic signal and provide an overview of sleep disordered breathing to be acquired for each subject, wherein the number of repetition patterns identified in each subdivided interval is indicative of a number of different subjects in the environment with sleep disordered breathing; and

a user interface configured to render information relating to sleep disordered breathing based on the detected one or more subjects in the environment with sleep disordered breathing, wherein the rendered information includes an identification of the detected one or more subjects and one or more characteristic of the detected sleep disordered breathing of a respective detected and identified subject.

2. An apparatus as claimed in claim 1 , wherein the plurality of acoustic signal components is for any one or more of:

different frequency ranges;

different Mel-frequency cepstral coefficients;

different acoustic levels;

different temporal characters; and

different spectral signatures.

3. An apparatus as claimed in claim 1 , wherein the one or more processors are configured to:

compare the determined plurality of signal envelopes or energy signals;

combine signal envelopes or energy signals that are similar; and

identify the repetition patterns in the combined signal envelopes or energy signals.

4. An apparatus as claimed in claim 1 , wherein the one or more processors are configured to:

analyze the determined plurality of signal envelopes or energy signals using independent component analysis, principal component analysis, multivariate singular spectrum analysis, or clustering algorithm analysis to determine a secondary plurality of signal envelopes or energy signals; and

identify the repetition patterns in the secondary determined plurality of signal envelopes or energy signals.

5. An apparatus as claimed in claim 1 , wherein the one or more processors are configured to:

compare the identified repetition intervals and associated repetition patterns to reference data that is typical for sleep disordered breathing to extract the identified repetition intervals and associated repetition patterns that relate to sleep disordered breathing.

6. An apparatus as claimed in claim 5 , wherein the reference data typical for sleep disordered breathing comprises breathing rate ranges typical for sleep disordered breathing.

7. An apparatus as claimed in claim 1 , wherein the one or more processors are configured to:

analyze the identified repetition intervals and associated repetition patterns to segment the acquired acoustic signal into parts; and

for each part of the acquired acoustic signal:

identify a number of substantially different breathing patterns in the part; and

assign a label to the part, wherein the label is indicative of the identified number of substantially different breathing patterns in the part.

8. An apparatus as claimed in claim 7 , wherein the one or more processors are configured to:

identify a part of the acquired acoustic signal corresponding to a disordered breathing episode for one subject; and

analyze at least one characteristic of the identified part of the acquired acoustic signal to identify the subject.

9. An apparatus as claimed in claim 8 , wherein the one or more processors are configured to analyze the at least one characteristic of the identified part of the acquired acoustic signal by:

matching the at least one characteristic of the identified part of the acquired acoustic signal to at least one corresponding characteristic stored in a memory with an identity of a subject to which the at least one corresponding characteristic relates; and

identifying the subject as the subject to which the at least one corresponding characteristic relates.

10. An apparatus as claimed in claim 8 , wherein the one or more processors are configured to:

analyze the identified part of the acquired acoustic signal to determine information on the sleep disordered breathing of the identified subject.

11. An apparatus as claimed in claim 1 , wherein the sleep disordered breathing comprises any one or more of snoring, asthma, stridor, and apnea.

12. A method of operating a sound analysis apparatus for detecting a presence of multiple subjects in an environment with sleep disordered breathing, the method comprising:

acquiring an acoustic signal, via an acoustic sensor, in an environment;

determining, via one or more processors, a plurality of acoustic signal components from the acquired acoustic signal, the plurality of acoustic signal components differing from one another in respect of at least one signal characteristic or property;

determining, via the one or more processors, a plurality of signal envelopes or energy signals based on the acoustic signal components, where at least one signal envelope being derived from each of the acoustic signal components of the plurality of acoustic signal components;

analyzing, via the one or more processors, the determined plurality of signal envelopes or energy signals to detect whether there are one or more subjects in the environment with sleep disordered breathing, wherein the analyzing comprises identifying (i) a number of repetition patterns in the determined plurality of signal envelopes or energy signals and (ii) a repetition interval for each of the identified repetition patterns, and further comprises analyzing the number of repetition patterns and the repetition intervals to subdivide a time axis of a whole duration from a start to an end of a sound measurement of the acquired acoustic signal and provide an overview of sleep disordered breathing to be acquired for each subject, wherein the number of repetition patterns identified in each subdivided interval is indicative of a number of different subjects in the environment with sleep disordered breathing; and

rendering, via a user interface, information relating to sleep disordered breathing based on the detected one or more subjects in the environment with sleep disordered breathing, wherein the rendered information includes an identification of the detected one or more subjects and one or more characteristic of the detected sleep disordered breathing of a respective detected and identified subject.

13. A non-transitory computer readable medium having computer readable code embodied therein, the computer readable code being configured such that, on execution by a suitable computer or processor, the computer or processor is caused to perform the method as claimed in claim 12 .