Researchers at the Massachusetts Institute of Technology (MIT) are building a device that uses machine learning, and is similar to a Wi-Fi router, to tract breathing, heart rate, sleep, gait, just by sitting in one spot.

The device can help people living with conditions like Parkinson's, Alzheimer's, depression, and pulmonary diseases and enable their physicians to wirelessly monitor their health. According to MIT professor Dina Katabi, the novel device will be able to replace the array of expensive, bulky, uncomfortable gear we currently need to get clinical data about the body.

It transmits a low-power wireless signal throughout a space the size of a one- or two-bedroom apartment (even through walls), and the signal reflects off people's bodies. The device then uses machine learning to analyse those reflected signals and extract physiological data. 

The device takes advantage of the fact that every time we move -- even if it's just a teeny, tiny bit, such as when we breathe -- we change the electromagnetic field surrounding us, Katabi said, at the recently held MIT Technology Review's EmTech conference in Cambridge, Massachusetts.

It can accurately monitor sleep, including individual sleep stages, in a person's own bed, with no changes to the way they sleep or what they wear. Because the device would be installed in a home, it could track the resident over time, too, which could be useful for watching sleep-disrupting conditions like Alzheimer's or depression, she said. Importantly, the data is collected only about specific traits and only with a person's consent. In addition, it is encrypted and is limited to certain designated recipients, Katabi said.