Researchers at MIT's Computer Science at Artificial Intelligence Lab (CSAIL) published a new paper this week in which they showed that they can uniquely identify a person they've monitored before. The device, on which the university's professor Dina Katabi has been working for several years, can be used to measure a person's vital signs, including heart rate and breathing. The analysis is made using wireless signals, and thus no physical contact is needed.
The system called "RF-ReID", which stands for radio-frequency re-identification, can be used to continuously monitor up to 40 people in a group living setting. As it detects heart rate and breathing, RF-ReID could be deployed in retirement homes, which will let the caregivers keep track of vital signs of those admitted.
It works by taking the collective pulse of the larger community and monitors wider trends about members' day-to-day health. If anomalies such as difficulty in breathing or reduced heart rates are found, care workers would be alerted. They can then separate those individuals from the rest of the group, and conduct COVID-19 tests on them. Katabi said:
“The new invention allows us to identify measurements from the same person, but without collecting identifying private information about them, including their appearance.”
While other approaches, namely installation of cameras and wearable devices have been tried for some time now, Katabi says these approaches are impractical and don't guarantee privacy to users. RF-ReID offers other uses too, such as early identification of Parkinson's patients who have a distinctive walking style. Until now, the system has been deployed in 19 different homes, and results concluded that after training it on at least 20 people, a new person could be re-identified in less than 10 seconds of physical activity.