91av

Computers can measure your pulse remotely by bouncing ultrasound waves off your body and analysing the reflections to detect the tiny chest movements caused by your heart pumping. The system could one day operate on home smart speaker devices.

Researchers at the University of the Chinese Academy of Sciences in Beijing and the University of Massachusetts Amherst used a laptop connected to a speaker to send out chirps of ultrasound at a frequency of 48 kilohertz. They then used a standard microphone to record the signals that were reflected by objects in the room, including people.

They were able to measure changes in the reflected audio to map any movement within the room as a bat does with echolocation. The technology is sensitive enough to detect a person’s chest expanding and contracting by approximately 5 millimetres during breathing, from which it was possible to calculate their respiration rate.

It was also possible to detect a pattern of much smaller chest movements – of about 0.5 mm – caused by the heartbeat. This helped the researchers measure the pulse, and the measurement was found to have a median error of less than one beat per minute. The technique works even if the subject is moving.

Jie Xiong at the University of Massachusetts Amherst says that signal processing was required to tell heart rate from ultrasound noise in the room and from the subject’s breathing.

He believes that the software can easily be run on any commercially available smart speaker on the market and could be used to remotely monitor the breathing and heart rate of vulnerable individuals at home without the need for specialist hardware. It could also act as an automatic alarm if someone had a fall.

The researchers were unable to test their system on a commercial smart speaker because manufacturers control that software that runs on them. But the quality of speakers and microphones on most of these devices is higher than those the team used, which suggests smart speakers could measure heartbeats even more accurately.

Several flaws need to be addressed before the software is ready for a public launch. Wearing bulky clothes can make it more difficult to detect a heartbeat, and when a person is moving around a room quickly there is additional noise in the signal that can make it harder to detect smaller chest movements.

The system can also detect and distinguish between multiple heartbeats when more than one person is in a room, but differentiating them becomes more difficult if two people are very close to one another.