Physiological monitoring devices and methods for noise reduction in physiological signals based on subject activity type
Methods and apparatus for monitoring a subject are described. A monitoring device configured to be attached to a body of a subject includes a sensor that is configured to detect and/or measure physiological information from the subject and at least one motion sensor configured to detect and/or measure subject motion information. The physiological sensor and motion sensor are in communication with a processor that is configured to receive and analyze signals produced by the physiological sensor and motion sensor. The processor is configured to process motion sensor signals to identify an activity characteristic of the subject. Once an activity characteristic is identified, the processor is configured to select a noise reference in response to identification of the activity characteristic of the subject, and then process physiological sensor signals using the noise reference to generate an output signal having reduced noise relative to the physiological sensor signal, to produce physiological information about the subject.
1. A method of monitoring a subject via a monitoring device, wherein the monitoring device includes a physiological sensor and a plurality of motion sensors, and wherein the physiological sensor and the plurality of motion sensors are in communication with at least one processor, the method comprising:
processing, via the at least one processor, a motion sensor signal to identify an activity characteristic of the subject;
selecting one of the plurality of motion sensors as a noise reference via the at least one processor in response to based on identification of the activity characteristic; and
processing, via the at least one processor, a physiological sensor signal received from the physiological sensor using the noise reference to generate an output signal having reduced noise relative to the physiological sensor signal.
2. The method of claim 1 , wherein the plurality of motion sensors comprises first and second motion sensors, wherein the first and second motion sensors are different types of sensors, and wherein selecting the one of the plurality of motion sensors as the noise reference comprises:
selecting, from among the first and second motion sensors, the first motion sensor as the noise reference based on identification of the activity characteristic as periodic motion; or
selecting, from among the first and second motion sensors, the second motion sensor as the noise reference based on identification of the activity characteristic as non-periodic motion.
3. The method of claim 2 , wherein the physiological sensor comprises an optical sensor, and wherein the second motion sensor is configured to provide a signal response that more closely corresponds to motion-related optical noise present in the physiological sensor signal than the first motion sensor.
4. The method of claim 2 , wherein processing the physiological sensor signal using the noise reference comprises:
processing the physiological sensor signal using a frequency-domain-based algorithm or circuit responsive to identification of the activity characteristic as periodic motion; or
processing the physiological sensor signal using a time-domain-based algorithm or circuit responsive to identification of the activity characteristic as non-periodic motion.
5. The method of claim 2 , wherein the physiological sensor is a photoplethysmography (PPG) sensor, wherein the first motion sensor comprises an inertial sensor, and wherein the second motion sensor comprises an optomechanical sensor.
6. The method of claim 5 , wherein the optomechanical sensor is configured to modulate light generated by an optical emitter in response to body motion, and to direct the light upon a pathway which is substantially unaffected by blood flow.
7. The method of claim 1 , wherein processing the physiological sensor signal using the noise reference further comprises:
in response to identification of the activity characteristic as periodic motion, processing the physiological sensor signal using a first algorithm or circuit; or
in response to identification of the activity characteristic as non-periodic motion, processing the physiological sensor signal using a second algorithm or circuit that is different from the first algorithm or circuit.
8. The method of claim 7 , wherein the first algorithm or circuit comprises frequency-domain processing and/or filtering, and wherein the second algorithm or circuit comprises time-domain processing and/or filtering.
9. The method of claim 8 , wherein the first and second algorithms or circuits utilize a same sensor or sensors of the plurality of motion sensors.
10. The method of claim 8 , wherein the first algorithm or circuit is configured to perform spectral subtraction, and wherein the second algorithm or circuit is configured to perform adaptive filtering.
11. The method of claim 1 , wherein body motion information contained in the noise reference is greater than blood flow information contained therein.
12. The method of claim 1 , wherein the noise reference comprises information associated with motion of the subject and is substantially free of information associated with blood flow of the subject.
13. The method of claim 1 , wherein the physiological information includes one or more of the following: subject heart rate, subject respiration rate, subject RR-interval (RRi), subject heart rate variability (HRV), subject blood pressure, subject blood analyte levels, subject cardiovascular properties.
14. The method of claim 1 , wherein the activity characteristic comprises a first activity characteristic or a second activity characteristic that is different than the first activity characteristic, wherein the plurality of motion sensors comprises first and second motion sensors, wherein the first and second motion sensors are different types of sensors, and wherein selecting the one of the plurality of motion sensors as the noise reference comprises:
selecting, from among the first and second motion sensors, the first motion sensor as the noise reference based on identification of the first activity characteristic; or
selecting, from among the first and second motion sensors, the second motion sensor as the noise reference based on identification of the second activity characteristic.
15. The method of claim 14 , wherein the first motion sensor comprises an inertial sensor, and wherein the second motion sensor comprises an optomechanical sensor.
16. A monitoring device configured to be attached to a bodily location of a subject, the monitoring device comprising:
a physiological sensor configured to detect and/or measure physiological data from the subject and generate a physiological sensor signal representative thereof;
a plurality of motion sensors, each configured to detect and/or measure subject motion data and generate a respective motion sensor signal representative thereof; and
at least one processor coupled to the physiological sensor and the plurality of motion sensors, wherein the at least one processor is configured to perform operations comprising:
processing the respective motion sensor signal received from one of the plurality of motion sensors to identify an activity characteristic of the subject;
selecting one of the plurality of motion sensors as a noise reference based on identification of the activity characteristic; and
processing the physiological sensor signal received from the physiological sensor using the noise reference to generate an output signal having reduced noise relative to the physiological sensor signal.
17. The monitoring device of claim 16 , wherein the plurality of motion sensors comprises first and second motion sensors, wherein the first and second motion sensors are different types of sensors, and wherein selecting the one of the plurality of motion sensors as the noise reference comprises:
selecting, from among the first and second motion sensors, the first motion sensor as the noise reference based on identification of the activity characteristic as periodic motion; or
selecting, from among the first and second motion sensors, the second motion sensor as the noise reference based on identification of the activity characteristic as non-periodic motion.
18. The monitoring device of claim 17 , wherein the physiological sensor comprises an optical sensor, and wherein a signal response of the second motion sensor more closely corresponds to motion-related optical noise present in the physiological sensor signal than a signal response of the first motion sensor.
19. The monitoring device of claim 18 , wherein processing the physiological sensor signal using the noise reference comprises:
processing the physiological sensor signal using a frequency-domain-based algorithm or circuit responsive to identification of the activity characteristic as periodic motion; or
processing the physiological sensor signal using a time-domain-based algorithm or circuit responsive to identification of the activity characteristic as non-periodic motion.
20. The monitoring device of claim 19 , wherein the physiological sensor is a photoplethysmography (PPG) sensor, wherein the first motion sensor comprises an inertial sensor, and wherein the second motion sensor comprises an optomechanical sensor.
21. The monitoring device of claim 20 , wherein the optomechanical sensor is configured to modulate light generated by an optical emitter in response to body motion, and to direct the light upon a pathway which is substantially unaffected by blood flow.
22. A computer program product comprising a non-transitory computer readable storage medium having computer readable program code embodied therein, which, when executed by at least one processor, causes the at least one processor to perform operations comprising:
processing a motion sensor signal to identify an activity characteristic of the subject;
selecting one of a plurality of motion sensors as a noise reference based on identification of the activity characteristic; and
processing a physiological sensor signal received from the physiological sensor using the noise reference to generate an output signal having reduced noise relative to the physiological sensor signal.