System and methods for video-based monitoring of vital signs
The present invention relates to the field of medical monitoring, and in particular non-contact, video-based monitoring of pulse rate, respiration rate, motion, and oxygen saturation. Systems and methods are described for capturing images of a patient, producing intensity signals from the images, filtering those signals to focus on a physiologic component, and measuring a vital sign from the filtered signals.
1. A video-based method of measuring a patient's vital sign, comprising:
receiving a video signal from a video camera, the video signal having a field of view;
extracting from the video signal two or more time-varying color signals in the field of view;
identifying a frequency content of each time-varying color signal;
combining at least two time-varying color signals having a shared frequency content;
calculating a statistic of the combined signal, wherein the statistic comprises an amplitude, a variability, a skew, or a signal to noise ratio;
measuring a vital sign from the combined signal; and
determining a weight for the vital sign, the weight being based on the statistic, and adding the vital sign to a running average based on the weight.
2. The method of claim 1 , wherein combining comprises averaging the at least two time-varying color signals.
3. The method of claim 1 , wherein the vital sign comprises heart rate.
4. The method of claim 3 , wherein measuring the heart rate comprises accumulating frequency peaks from the combined signal, selecting a median frequency, and updating a running average heart rate from the selected frequency.
5. The method of claim 1 , further comprising applying a weight to the combined signal prior to measuring the vital sign from the combined signal, the weight being based on the statistic.
6. The method of claim 1 , wherein extracting the time-varying color signal comprises extracting two time-varying color signals from a region within the field of view, and wherein measuring the vital sign comprises measuring oxygen saturation from the two time-varying color signals.
7. The method of claim 1 , wherein each time-varying color signal extracted from the video signal is from a distinct region and wherein combining at least two time-varying color signals having a shared frequency content comprises combining at least two time-varying color signals having a shared frequency content that from non-adjacent regions.
8. A video-based method of measuring a patient's vital sign, comprising:
generating a video signal from a video camera having a field of view;
extracting from the video signal a plurality of time-varying intensity signals;
combining at least two time-varying intensity signals from a first region of the field of view to produce a first combined intensity signal;
measuring a first vital sign of the patient from the first combined intensity signal;
combining at least two time-varying intensity signals from a second region of the field of view to produce a second combined intensity signal; and
measuring a second vital sign of the patient from the second combined intensity signal.
9. The method of claim 8 , wherein the first region is selected based on intensity signals comprising a first modulation at a first shared frequency, and the second region is selected based on intensity signals comprising a second modulation at a second shared frequency, the first shared frequency being different from the second shared frequency.
10. The method of claim 8 , wherein the first vital sign is heart rate and the second vital sign is respiration rate.
11. The method of claim 8 , further comprising displaying the image signal on a display, highlighting on the display the first region for the first vital sign, and highlighting on the display the second region for the second vital sign.
12. The method of claim 9 , wherein the first modulation at the first shared frequency is above an amplitude threshold.
13. The method of claim 8 , wherein extracting the plurality of intensity signal comprises extracting at least one red intensity signal comprising a time-varying average of red values, at least one green intensity signal comprising a time-varying average of green values, and at least one blue intensity signal comprising a time-varying average of blue values.
14. A camera-based method of measuring a patient's vital sign, comprising:
receiving an image signal from a camera having a field of view;
receiving a user input identifying a region in the image signal;
measuring a first vital sign of the patient, comprising:
from the region, selecting a first set of pixels that exhibit a first modulation at a first shared frequency, wherein the first selected pixels are non-contiguous; and
measuring the first vital sign of the patient from the selected first pixels;
displaying a number derived from the first measured vital sign; and
triggering an alarm based on the displayed number and stored alarm conditions, wherein the stored alarm conditions omit a sensor-off or sensor-disconnect alarm.
15. The method of claim 14 , further comprising:
measuring and outputting a second vital sign of the patient, comprising:
from the region, selecting a second set of pixels that exhibit a second modulation at a second shared frequency different from the first shared frequency, wherein the second selected pixels are non-contiguous; and
measuring the second vital sign of the patient from the selected second pixels.
16. The method of claim 14 , wherein the first vital sign comprises heart rate and the second vital sign comprises respiration rate.
17. The method of claim 14 , further comprising receiving a second user input moving the region to a new location, and subsequently measuring the first vital sign from the new location.
18. The method of claim 14 , wherein the alarm is a system-level alarm, a physiologic alarm, or both.
19. A video-based method of measuring a patient's vital sign, comprising:
receiving a video signal from a video camera, the video signal having a field of view that includes exposed skin of a patient;
extracting from the video signal a time-varying amplitude signal for each of a plurality of regions, each region including exposed skin of the patient;
identifying a frequency content of each amplitude signal;
selecting two or more non-adjacent regions that have a shared frequency content comprising a modulation at a shared frequency, wherein each of the two or more non-adjacent regions satisfy a quality criterion, the quality criterion comprising a signal to noise ratio that satisfies a threshold, or a skew value that satisfies a threshold;
combining the amplitude signals of the selected regions;
measuring a vital sign from the combined signal; and
outputting the vital sign for further processing or display.
20. The method of claim 19 , further comprising removing an amplitude signal from the combined signal upon failure of the removed amplitude signal to satisfy the criterion.