System and method of determining breathing rates from oscillometric data
A system and method for determining the breathing rate of a patient using only an oscillometric device as the physiological sensor. Here, the oscillometric device is mounted on a patient's limb, and oscillometric pulse waveforms are obtained as the device's cuff deflates, thus obtaining pulse wave signals and artifact signals over multiple patient breaths. A computer processor analyzes these signals, and removes artifacts according to various algorithms. The resulting signal can be viewed as containing both an amplitude modulated envelope of pulse waves (AM signals) and a frequency modulated sequence of pulses at various time intervals (FM signals). The main harmonics of the AM and FM signals both contain breathing rate data, and system accuracy can be improved by comparing the AM harmonics with the FM harmonics. The final breathing rate data, often a function of the AM and FM harmonics, is output or stored in memory.
1. A method of automatically determining a breathing rate of a patient, said method comprising:
obtaining pulse waveforms from an oscillometric device mounted on a limb of said patient, said pulse waveforms thus being oscillometric type pulse waveforms;
analyzing said pulse waveforms, using at least one processor, and determining artifact-free regions of said pulse waveforms, thus obtaining edited pulse waveforms;
analyzing said edited pulse waveforms, using said at least one processor, and determining AM envelope signals and FM between-pulse-time signals of said edited pulse waveforms;
analyzing said AM envelope signals and said FM between-pulse-time signals using said at least one processor, and determining an AM envelope main harmonic of said AM envelope signals and an FM between-pulse-time main harmonic of said FM between-pulse-time signals;
in response to said AM envelope main harmonic and said FM between-pulse-time main harmonic being within a predetermined limit of each other, using said at least one processor to calculate a value from a function comprising said AM envelope main harmonic and said FM between-pulse-time main harmonic;
and record or output said value as said breathing rate of said patient.
2. The method of claim 1 , wherein said oscillometric device further comprises a tri-axial accelerometer-gyroscope device comprising a tri-axial accelerometer and/or a tri-axial gyroscope sensor, said tri-axial accelerometer-gyroscope device reports limb movement of said oscillometric device to said at least one processor, and said at least one processor further uses said limb movement to determine at least some of said artifact-free regions of said pulse waveforms.
3. The method of claim 2 , wherein said at least some of said artifact-free regions of said pulse waveforms are determined by obtaining oscillometric cuff deflation signals, and analyzing said cuff deflation signals by any of:
a) analyzing areas where neighboring pulses exhibit below average cross-correlation;
b) obtaining any of tri-axial gyroscope signal or tri-axial accelerometer signals from said tri-axial accelerometer-gyroscope device and automatically deweighting those cuff deflation signals obtained during a time that said tri-axial accelerometer-gyroscope device detects motion above a preset threshold; and
c) analyzing said AM envelope signals of said edited pulse waveforms, and automatically deweighting pulse waveform data associated with envelope outliers above a preset threshold.
4. The method of claim 1 , wherein said at least one processor determines said AM envelope signals by determining an oscillometric envelope of pulse peak amplitudes, and determining a time varying amplitude of said AM envelope signals.
5. The method of claim 1 , wherein said at least one processor further determines said FM between-pulse-time signals by computing time differences between peak indices of said pulse waveforms, and using these time differences to compute instantaneous pulse rates of said patient;
and wherein said at least one processor further uses said instantaneous pulse rates to determine said FM between-pulse-time signals.
6. The method of claim 1 , wherein at least one processor determines said AM envelope main harmonic by determining a power spectral density of the main harmonics of an oscillometric envelope of said pulse waveforms; and wherein said at least one processor determines said FM between-pulse-time main harmonic by determining a power spectral density of an instantaneous pulse rate signal that is based on individual pulse positions of the pulse waveforms with respect to each other in time.
7. The method of claim 1 , wherein said breathing rate is a weighted combination of said AM envelope main harmonic and said FM between-pulse-time main harmonic.
8. The method of claim 1 , wherein said limb of said patient comprises a wrist of said patient, and said oscillometric device is mounted on said patient's wrist.
9. The method of claim 8 , wherein said oscillometric device comprises said at least one processor, and said oscillometric device further comprises a display configured to display said breathing rate of said patient.
10. A system for automatically determining a breathing rate of a patient, said system comprising:
an oscillometric device configured to be mounted on a limb of said patient, said oscillometric device comprising at least one processor, memory, pressure cuff, and pressure cuff sensor, said device configured to obtain pulse waveforms, said pulse waveforms thus being oscillometric type pulse waveforms;
said at least one processor configured to analyze said pulse waveforms, and determine artifact-free regions of said pulse waveforms, thus obtaining edited pulse waveforms;
said at least one processor further configured to analyze said edited pulse waveforms, and determine AM envelope signals and FM between-pulse-time signals of said edited pulse waveforms;
said at least one processor further configured to analyze said AM envelope signals and said FM between-pulse-time signals, and determine an AM envelope main harmonic of said AM envelope signals and an FM between-pulse-time main harmonic of said FM between-pulse-time signals;
said at least one processor configured to determine when said AM envelope main harmonic and said FM between-pulse-time main harmonic are within a predetermined limit of each other, and when within the predetermined limit of each other, to calculate a value from a function comprising said AM envelope main harmonic and said FM between-pulse-time main harmonic, and to record or output said value as said breathing rate of said patient.
11. The system of claim 10 , wherein said oscillometric device further comprises a tri-axial accelerometer-gyroscope device comprising a tri-axial accelerometer and/or a tri-axial gyroscope sensor, said tri-axial accelerometer gyroscope device further configured to report limb movement of said oscillometric device to said at least one processor, and said at least one processor further configured to use said limb movement to determine at least some of said artifact-free regions of said pulse waveforms.
12. The system of claim 11 , wherein said at least one processor is configured to determine said at least some of said artifact-free regions of said pulse waveforms by obtaining oscillometric cuff deflation signals, and analyzing said cuff deflation signals by any of:
a) analyzing areas where neighboring pulses exhibit below average cross-correlation;
b) obtaining any of tri-axial gyroscope signal or tri-axial accelerometer signals from said tri-axial accelerometer-gyroscope device and automatically deweighting those cuff deflation signals obtained during a time that said tri-axial accelerometer-gyroscope device detects motion above a preset threshold; and
c) analyzing said AM envelope signals of said edited pulse waveforms, and automatically deweighting pulse waveform data associated with envelope outliers above a preset threshold.
13. The system of claim 10 , wherein said at least one processor is further configured to determine said AM envelope signals by determining an oscillometric envelope of pulse peak amplitudes, and determine a time varying amplitude of said AM envelope signals.
14. The system of claim 10 , wherein said at least one processor is further configured to determine said FM between-pulse-time signals by computing time differences between peak indices of said pulse waveforms, and using these time differences to compute instantaneous pulse rates of said patient;
and wherein said at least one processor further uses said instantaneous pulse rates to determine said FM between-pulse-time signals.
15. The system of claim 10 , wherein said at least one processor is configured to determine said AM envelope main harmonic by determine a power spectral density of the main harmonics of an oscillometric envelope of said pulse waveforms; and wherein said at least one processor is configured to determine said FM between-pulse-time main harmonic by determining a power spectral density of an instantaneous pulse rate signal that is based on individual pulse positions of the pulse waveforms with respect to each other in time.
16. The system of claim 10 , wherein said breathing rate is a weighted combination of said AM envelope main harmonic and said FM between-pulse-time main harmonic.
17. The system of claim 10 , wherein said limb of said patient comprises a wrist of said patient, and said oscillometric device comprises a combined chassis and cuff device configured to be mounted on said patient's wrist.
18. The system of claim 17 , wherein said oscillometric device further comprises any of a display configured to display said breathing rate of said patient, or a wireless transceiver configured to output said breathing rate of said patient.