MIT 'human radar' device tracks movement, gait speed of Parkinson's patients at home

Radar technology is used to track vehicles, measure driving speeds, forecast weather changes and, now, to remotely monitor the severity and progression of Parkinson’s disease.

Researchers from MIT and the University of Rochester have built a wireless device that uses radio signals to pick up on the movements of Parkinson’s patients throughout their own homes, creating what the research team has termed “human radar.”

The new technology is described in a study that was published this month in the journal Science Translational Medicine. In it, the research team concluded that their system for “continuous, objective, sensitive and passive assessment” could successfully be used not only to shape more effective treatment plans for Parkinson’s patients but also to improve clinical trials of potential new therapies for the disease.

The device is approximately the size of a Wi-Fi router but emits signals with only a fraction of the power of a router. The low-power radio waves pass through solid objects in a patient’s home, like walls and furniture, but bounce off humans due to the water content of the body.

Because radio waves always travel at the same speed, the device can calculate a person’s walking speed by analyzing how long it takes for the waves to bounce back to it. It’s equipped with machine learning algorithms that can pinpoint which of those reflected waves came from a specific patient and then use those readings to calculate gait speed.

According to the researchers, tracking at-home movements can give doctors a better idea of the severity and progression of an individual’s Parkinson’s disease compared to in-clinic assessments because of how difficult it may be for patients to go in for regular check-ups and how unreliable those one-off snapshots of their condition may be.

In contrast, in the study, the device was used to monitor the gait of 50 individuals—about two-thirds of them with Parkinson’s, and the rest an age-matched control group without the disease—for a full year. With a year’s worth of readings, the researchers were able to eliminate outliers and variations in gait speed and home in specifically on changes caused by Parkinson’s.

“Monitoring the patient continuously as they move around the room enabled us to get really good measurements of their gait speed. And with so much data, we were able to perform aggregation that allowed us to see very small differences,” Guo Zhang, a Ph.D. student at MIT and co-lead author of the study, said to MIT News.

Their findings included evidence showing that gait speed declined about twice as quickly for those with Parkinson’s compared to those in the control group. They were also able to definitively link daily fluctuations in speed with medication response—a previously “cumbersome” and largely subjective task, since it typically relies on self-reported assessments from patients, according to Yingcheng Liu, another MIT grad student and co-lead author of the study.

“By being able to have a device in the home that can monitor a patient and tell the doctor remotely about the progression of the disease and the patient’s medication response so they can attend to the patient even if the patient can’t come to the clinic—now they have real, reliable information—that actually goes a long way toward improving equity and access,” said senior author Dina Katabi, Ph.D., a professor at MIT and principal investigator in its Computer Science & Artificial Intelligence Laboratory.