Respiration processor

1. A system for determining a respiration rate of a patient using an acoustic signal received from an acoustic sensor adapted to monitor the patient, the system comprising:

an input configured to receive the acoustic signal from the acoustic sensor; and

one or more hardware processors configured to:

generate a time domain envelope signal from the acoustic signal,

individually tag each local extremum of a plurality of local extrema in the time domain envelope signal as potentially corresponding to an inspiration or an expiration of the patient,

determine to reject at least one local extremum of the plurality of local extrema as not corresponding to the inspiration or the expiration,

determine in real time a respiratory rate from the plurality of local extrema without using the at least one local extremum that was determined to be rejected, and

output the respiratory rate.

2. The system of claim 1 , wherein the one or more hardware processors are configured to determine to reject a first local extremum of the plurality of local extrema according to a time difference between the first local extremum and a second local extremum of the plurality of local extrema, the at least one local extremum that was determined to be rejected comprising the first local extremum.

3. The system of claim 2 , wherein the first local extremum and the second local extremum are consecutively-occurring local extrema in the time domain envelope signal.

4. The system of claim 2 , wherein the one or more hardware processors are configured to determine to reject the first local extremum further according to a magnitude difference between a first magnitude of the first local extremum and a second magnitude of the second local extremum.

5. The system of claim 2 , wherein the one or more hardware processors are configured to:

determine to reject the first local extremum from a comparison of the time difference and a threshold, and

adjust the threshold.

6. The system of claim 5 , wherein the one or more hardware processors are configured to:

determine that the patient is a neonate, and

adjust the threshold responsive to determining that the patient is the neonate.

7. The system of claim 1 , wherein the one or more hardware processors are configured to determine to reject a first local extremum of the plurality of local extrema according to a change in a magnitude of the time domain envelope signal between the first local extremum and a second local extremum of the plurality of local extrema, the at least one local extremum that was determined to be rejected comprising the first local extremum.

8. The system of claim 1 , wherein the one or more hardware processors are configured to determine to reject a first local extremum of the plurality of local extrema because the first local extremum is at a beginning or an end of a processing unit, the processing unit corresponding to a set of the plurality of local extrema that the one or more hardware processors are configured to process as a batch, the at least one local extremum that was determined to be rejected comprising the first local extremum.

9. The system of claim 1 , wherein the one or more hardware processors are configured to determine to reject a first local extremum of the plurality of local extrema according to a width of the time domain envelope signal around the first local extremum, the at least one local extremum that was determined to be rejected comprising the first local extremum.

10. The system of claim 1 , wherein the one or more hardware processors are configured to:

determine a statistical parameter from processing a set of the plurality of local extrema as a batch, the set of the plurality of local extrema not comprising a first local extremum of the plurality of local extrema, and

determine to reject the first local extremum from a deviation of a characteristic of the first local extremum relative to the statistical parameter, the at least one local extremum that was determined to be rejected comprising the first local extremum.

11. The system of claim 1 , wherein the one or more hardware processors are configured to:

filter the time domain envelope signal with a filter to smooth the time domain envelope signal, the filter having a cutoff frequency, and

adjust the cutoff frequency.

12. The system of claim 11 , wherein the one or more hardware processors are configured to adjust the cutoff frequency responsive to the respiratory rate.

13. The system of claim 1 , wherein the one or more hardware processors are configured to individually tag each local extremum of the plurality of local extrema by identifying any local extremum of the plurality of local extrema in the time domain envelope signal that are between a lower threshold and an upper threshold, the lower threshold corresponding to a noise floor.

14. The system of claim 1 , further comprising:

the acoustic sensor; or

a display configured to present the respiratory rate.

15. A method for determining a respiration rate of a patient using an acoustic signal received from an acoustic sensor monitoring the patient, the method comprising:

detecting, using the acoustic sensor, the acoustic signal;

generating, by one or more hardware processors, a time domain envelope signal from the acoustic signal;

individually tagging, by the one or more hardware processors, each local extremum of a plurality of local extrema in the time domain envelope signal as potentially corresponding to an inspiration or an expiration of the patient;

determining, by the one or more hardware processors, to reject at least one local extremum of the plurality of local extrema as not corresponding to the inspiration or the expiration;

determining, by the one or more hardware processors, in real time a respiratory rate from the plurality of local extrema without using the at least one local extremum that was determined to be rejected; and

presenting the respiratory rate on a display.

16. The method of claim 15 , wherein said determining to reject the at least one local extremum comprises determining to reject a first local extremum of the plurality of local extrema according to a time difference between the first local extremum and a second local extremum of the plurality of local extrema, the first local extremum and the second local extremum being consecutively-occurring local extrema in the time domain envelope signal, the at least one local extremum that was determined to be rejected comprising the first local extremum.

17. The method of claim 16 , wherein said determining to reject the at least one local extremum comprises determining to reject the first local extremum from a comparison of the time difference and a threshold, and further comprising:

determining that the patient is a neonate; and

adjusting the threshold responsive to determining that the patient is the neonate.

18. The method of claim 15 , wherein said determining to reject the at least one local extremum comprises determining to reject a first local extremum of the plurality of local extrema according to a change in a magnitude of the time domain envelope signal between the first local extremum and a second local extremum of the plurality of local extrema, the first local extremum and the second local extremum being consecutively-occurring local extrema in the time domain envelope signal, the at least one local extremum that was determined to be rejected comprising the first local extremum.

19. The method of claim 15 , further comprising:

filtering the time domain envelope signal with a filter to smooth the time domain envelope signal, the filter having a cutoff frequency; and

adjusting the cutoff frequency responsive to the respiratory rate.

20. The method of claim 15 , wherein said tagging comprises identifying any local extremum of the plurality of local extrema in the time domain envelope signal that are between a lower threshold and an upper threshold.