Rejecting noise in a signal
Reducing noise in a signal is provided. A first filter includes a selectively definable passband for filtering a first signal used in determining at least one patient parameter. A characteristic analyzer detects a characteristic of a second signal and generates a variability measurement value using a series of data values within the second signal over a previously occurring window of time. A filter controller coupled to the characteristic analyzer uses the variability measurement value to define a characteristic of the passband for the first filter and selectively tunes the passband of the first filter according to the defined characteristic. Related apparatus, systems, techniques, and articles are also described.
1. A patient monitoring device configured to reduce noise in a measured pulse rate signal, comprising:
a pulse rate sensor operating in a red to infrared electromagnetic spectrum configured to generate a pulse rate signal in response to pulse rate measurements;
an independently measured heart rate sensor operating outside the red to infrared electromagnetic spectrum configured to generate a heart rate signal in response to heart rate measurements;
a characteristic analyzer configured to determine a measured heart rate variability based on the heart rate measurements;
a filter controller configured to dynamically tune an adjustable filter in order to generate a reduced-noise pulse rate signal based on the determined heart rate variability;
a parameter processor providing a pulse rate output signal based on the reduced-noise pulse rate signal; and
wherein the characteristic analyzer is further configured to determine a predicted heart rate variability, determine a guard band based of the measured heart rate variability and the predicted heart rate variability, determine a frequency passband for the pulse rate signal based on the heartrate variability and the guard band, and dynamically tune the adjustable filter based on the frequency passband.
2. The patient monitoring device according to claim 1 , wherein the heart rate measurements comprises electrocardiogram (ECG) measurements.
3. The patient monitoring device according to claim 1 , wherein the characteristic analyzer is further configured to continually determine the measured heart rate variability and continually tune the adjustable filter.
4. The patient monitoring device according to claim 1 , wherein the characteristic analyzer continually determines the heart rate variability over successive time intervals and continually adjusts the adjustable filter.
5. The patient monitoring device according to claim 1 , wherein the frequency passband comprises at least one of (a) a center frequency; (b) a width of a frequency envelope; (c) lower and upper cutoff frequencies; and (d) a shape of the frequency envelope.
6. The patient monitoring device according to claim 1 , wherein the pulse measurements are first pulse rate measurements, further configured for receiving second pulse rate measurements in the red to infrared electromagnetic spectrum; and
generating the pulse rate signal based on the first and second pulse rate measurements.
7. The patient monitoring device according to claim 6 , further configured for:
computing a blood oxygen saturation level (SpO2) based on measurements the first and second pulse rate measurements; and
providing a blood oxygen output based on the blood oxygen saturation level (SpO2).
8. The patient monitoring device according to claim 1 , further configured for:
determining a signal quality index (SQI) by calculating a pulse rate variability (PRV) over a predetermined period and comparing the pulse rate variability to the measured heart rate variability; and
providing a signal quality output based on the signal quality index (SQI).
9. The patient monitoring device according to claim 1 , a wherein the parameter processor is further configured to:
determine a signal quality index (SQI) by calculating a pulse rate variability (PRV) and comparing the pulse rate variability to the measured heart rate variability; and
provide a signal quality output reflecting the signal quality index (SQI).
10. A method for reducing noise in a measured pulse rate signal, comprising:
receiving pulse rate measurements in a red to infrared electromagnetic spectrum;
generating a pulse rate signal based on the pulse rate measurements;
receiving heart rate measurements, independent of the pulse rate measurements, outside the red to infrared electromagnetic spectrum;
generating a heart rate signal based on the heart rate measurements;
determining a measured heart rate variability based on the heart rate signal;
determining a predicted heart rate variability;
computing a guard band based the measured heart rate variability and the predicted heart rate variability;
determining a frequency passband for the pulse rate signal based on the heart rate variability and the guard band;
dynamically tuning an adjustable filter based on the frequency passband;
determining a reduced-noise pulse rate signal based on the pulse rate signal and the adjustable filter; and
providing a pulse rate output reflecting the reduced-noise pulse rate signal.
11. The method of claim 10 , wherein the heart rate measurements comprises electrocardiogram (ECG) measurements.
12. The method of claim 10 , further comprising continually determining the measured heart rate variability and continually tuning the adjustable filter.
13. The method of claim 10 , wherein defining the frequency passband comprises determining at least one of (a) a center frequency; (b) a width of a frequency envelope; (c) lower and upper cutoff frequencies; and (d) a shape of the frequency envelope.
14. The method of claim 10 , further comprising dynamically tuning the adjustable filter using at least one of (a) a time domain measurement technique; (b) a frequency domain measurement technique; (c) a joint time-frequency domain measurement technique; and (d) a nonlinear dynamic measurement technique.
15. The method of claim 10 , wherein the pulse rate measurements are first pulse rate measurements, further comprising:
receiving second pulse rate measurements in the red to infrared electromagnetic spectrum; and
generating the pulse rate signal based on the first and second pulse rate measurements.
16. The method of claim 15 , further comprising:
computing a blood oxygen saturation level (SpO2) based on measurements the first and second pulse rate measurements; and
providing a blood oxygen output based on the blood oxygen saturation level (SpO2).
17. The method of claim 10 , further comprising:
determining a signal quality index (SQI) by calculating a pulse rate variability (PRV) over a predetermined period and comparing the pulse rate variability to the measured heart rate variability; and
providing a signal quality output based on the signal quality index (SQI).