Vital sign monitors have become the cornerstone of a wide variety of commercial and personal safety applications, ranging from baby monitoring to elderly care. Many such monitors exist in the today market. In each case, the patient – or caregiver – is required to activate the monitoring process by connecting a sensor to generate and read the relevant data. These sensors sometimes need to stay in place for an extended period, creating issues when dealing with sleeping adults or even children who tend to remove foreign objects that cause discomfort. To avoid these issues there is a need for a system that has remote sensing capabilities, which is not available. Until now.
This blog post introduces a simple yet efficient, real-time method for measuring a person’s breathing sinus rhythm using only remote sensors. This work started as a research project in Essence’s CTO office with the problem to be solved being very well defined: enable the tracking and monitoring of a person’s (elderly, baby, etc.) vital signs without requiring any action or cooperation on their part.
Though the problem may be easy enough to define, the solution is quite complex. The first challenge our team considered was the need for remote sensing capabilities that wouldn’t interfere with the person’s day to day life. We found the solution in radar technology. A radar emits electro-magnetic pulses, kind of like a cellphone. These pulses are then reflected back from the target, which can be a human, a beating heart or a moving breathing torso. These pulses create a harmonic pattern. Using signal processing algorithms we have managed to track this harmonic pattern and measure the breathing and heartrate of multiple people in a room at a range of up to 8 meters.
To augment these results, using multiple sensors is desired to increase the reliability of the readings. In particular, a camera is one such sensor with a proven additional value. Therefore, we have also developed algorithms that use streaming video to detect points of interest on the subject’s body, and the corresponding optical flow is estimated and tracked using the well-known Lucas-Kanade algorithm on a frame by frame basis. These detected points may also vary in their location in the image (correlated with the subject’s breathing rhythm) and are then tracked using a real time iterative tracker. Finally, the readings are stitched together to create a refined sinus estimation of the breathing rhythm.
We performed extensive studies with both infants and adults, yielding a maximal error of 1 BPM between the calculated results and the true breathing rate of the subjects. Indeed, we received very promising and very accurate results to build on as we proceed with the work, keeping our top goal in mind: Developing devices for a Better life made Possible.
Nir is a researcher with Essence Group CTO’s office, responsible for bringing to life future technologies for the company. Nir’s research interests are statistical signal processing, namely, estimation, detection theory and machine learning applied to problems in computer vision and radar remote sensing. Nir is also a Ph.D. student in Ben-Gurion University, researching the problem of radar remote sensing, classification and recognition of miniature drones.
