Movement-Tracking System Collects Health and Behavioral Data

When a signal rebounds, it creates a map, sectioned into vertical and horizontal frames, that indicates where people are in a three-dimensional (3D) space. The vertical frames capture height and build, while the horizontal frames determine general location. As individuals move about, the system analyzes the RF frames to generate short trajectories, called tracklets.

To train the system and tag identities, all users first wear low-powered accelerometer sensors, which are used to label the reflected RF signals as per their respective identities via an algorithm that correlates acceleration features with motion features. When users walk, for instance, the acceleration oscillates, but becomes a flat line when they stop. When the best match between acceleration data and tracklets is met, the tracklet is labeled with the matching user's identity. The sensors do not need charging, and, after training, the individuals don't need to wear them again.

The researchers then tested Marko in six locations: two assisted living facilities, three apartments inhabited by couples, and one townhouse with four residents. The study demonstrated the system's ability to distinguish individuals based solely on RF wireless signals. In one assisted living facility, the researchers monitored a patient with dementia who would often become agitated. By matching her increased pacing with the visitor log, they determined the patient was agitated more during the days following family visits. The study was presented at the annual Human Factors in Computing Systems conference, held during May 2019, in Glasgow (United Kingdom).

“We live in a sea of wireless signals, and the way we move and walk around changes these reflections. We developed the system that listens to those reflections ... to better understand people's behavior and health,” said lead author PhD student Chen-Yu Hsu, who added that “video is more invasive. Using radio signals to do all this work strikes a good balance between getting some level of helpful information, but not making people feel uncomfortable.”

“With respect to imaging through cameras, it offers a less data-rich and more targeted model of collecting information, which is very welcome from the user privacy perspective,” commented Professor Cecilia Mascolo, PhD, of the department of computer science and technology at Cambridge University (United Kingdom). “The data collected, however, is still very rich, and the paper evaluation shows accuracy which can enable a number of very useful applications, for example in elderly care, medical adherence monitoring, or even hospital care.”

Related Links:
Massachusetts Institute of Technology