System and device for non-invasive drink detection
A wearable device is provided. The wearable device includes at least one motion sensor operable to detect one or more motion signals, a processor coupled to the at least one motion sensor, one or more biological sensors coupled to the processor and operable to detect one or more biological indicators of a user, and a memory configured to store instructions executable by the processor. The instructions, when executed, are operable to: obtain at least one of the one or more biological indicators of the user; correlate the at least one biological indicators of the user with the detected one or more motion signals; and determine that a drink event is detected based on the correlation between the detected one or more motion signals and the at least one biological indicators.
1. A method comprising:
obtaining, from at least one motion sensor, one or more motion signals;
obtaining, from one or more biological sensors, one or more biological indicators of a user, the one or more biological indicators including a surge in a heart rate;
correlating the one or more biological indicators of the user with the one or more motion signals;
determining that a drink event is detected based on the correlation between the one or more motion signals and the one or more biological indicators, wherein the drink event ends upon a return, based on the one or more motion sensor signals, to a position prior to the drink event being detected; and
estimating a volume of the drink event using an area under a curve of the surge in the heart rate over a baseline heart rate.
2. The method of claim 1 , further comprising pre-processing the one or more motion signals using dynamic time warping.
3. The method of claim 1 , wherein the at least one motion sensor includes an inertial motion unit, an accelerometer, a magnetometer, and/or a gyroscope.
4. The method of claim 1 , further comprising processing the one or more motion signals by a machine learning classification technique.
5. The method of claim 4 , wherein the machine learning classification technique includes k-nearest neighbor classifier and/or a neural network classifier.
6. The method of claim 1 , wherein the one or more biological sensors includes a thermometer, and the one or more biological indicators includes tissue temperature and ambient temperature.
7. The method of claim 6 , wherein the one or more biological indicators includes a decrease in the tissue temperature and/or the decrease in the tissue temperature with respect to the ambient temperature.
8. The method of claim 1 , wherein the one or more biological sensors includes a respiration detector, and the one or more biological indicators includes a pause in breathing.
9. The method of claim 1 , wherein the one or more biological sensors includes a near-infrared spectrometer, and the one or more biological indicators includes an increase in tissue perfusion and/or blood volume.
10. The method of claim 1 , wherein the one or more biological sensors includes a photoplethysmography monitor, and the one or more biological indicators includes a blood oxygen saturation, the heart rate, a heart rate variation, a blood pressure, and/or a respiration rate.
11. The method of claim 1 , wherein the one or more biological sensors includes a bioimpedance monitor, and the one or more biological indicators includes tissue hydration variation.
12. The method of claim 1 , wherein the one or more biological sensors includes an atmospheric pressure sensor, an altitude sensor, a relative humidity sensor, a microphone, a clock, an event marker, and/or a camera.
13. The method of claim 12 , wherein the one or more biological indicators includes atmospheric pressure, altitude, humidity, sounds associated with eating and/or drinking, a time, span of the time, images and/or videos associated with the drinking.
14. The method of claim 1 , further comprising:
obtaining, from one or more additional sensors, one or more additional indicators;
correlating the one or more additional indicators with the one or more biological indicators of the user and the one or more motion signals;
determining that the drink event is detected based on the correlation between the one or more additional indicators, the one or more motion signals, and the one or more biological indicators.
15. The method of claim 14 , wherein the one or more additional sensors includes a global position system component, and the one or more additional indicators includes a physical location.
16. The method of claim 15 , wherein the physical location includes a restaurant and/or a bar.
17. The method of claim 15 , wherein the physical location is determined by a calendar, events and/or postings on one or more social media applications, ambient light, UV exposure, humidity, altitude, images, videos, and/or previously visited locations.