A team at MIT is using widely available low-res Webcam imaging to measure the human pulse, and is now working on expanding readings to respiration and blood pressure.
Elizabeth Armstrong Moore
Elizabeth Armstrong Moore is based in Portland, Oregon, and has written for Wired, The Christian Science Monitor, and public radio. Her semi-obscure hobbies include climbing, billiards, board games that take up a lot of space, and piano.
How'd you like to check your pulse, respiration, and blood pressure as you brush your teeth in the mirror each morning? A PhD candidate at the Harvard-MIT Division of Health Sciences and Technology is working to make this a reality in the near future.
The system works by measuring the slightest variations in brightness produced by blood flow through blood vessels in the face. Poh used public-domain software to identify facial positions in any given image and break that information into separate red, green, and blue portions of the video images.
To deal with both movement in front of the lens as well as different ambient light, Poh adapted a method known as ICA (Independent Component Analysis)--a signal-processing technique originally developed to extract a single voice from a room of conversations--to find the pulse signal amid all the video noise.
The pulse results turned out to be pretty reliable when compared with measurements taken by an FDA-approved monitoring device. Poh's readings were within three beats per minute, even when test subjects were moving in front of the camera, and even with up to three people standing in view of the camera at the same time.
I can see the future, and it feels like Star Trek. Family teeth-brushing time will soon include checking in on pulse rates and blood pressures. But Poh says he also sees practical applications for online telemedicine screenings, or for ongoing monitoring of, say, elderly relatives from afar.
Fokko Wieringa, senior scientist at TNO Science & Industry in the Netherlands, published a paper describing a similar pulse-detection system in 2005, but he tells MIT News:
The exciting thing about this new method is that they identify a fixed region on the face and track it... The achieved gain in signal quality allows them to use a simple and cheap camera, even on moderately moving persons. These combined features are very original. My hands are itching to exchange ideas and cooperate with these young researchers; it's exciting stuff.
Poh may have a long road ahead of him before he's able to get blood pressure and blood-oxygen measurements from the same video images. But he's optimistic, because blood oxygen sensors already use optical detection, the difference being that they rely on dedicated light sources as opposed to ambient lighting.
"It's not going to be easy," he says, "but it theoretically should be possible."