MIT engineers recently unveiled a radar-like device that can test people for signs of Parkinson’s disease while they sleep. Now, they say the same system can monitor Parkinson’s patients differently by tracking how they walk.
“We know very little about the brain and its diseases,” said Dina Katabi, a professor of electrical engineering at MIT who led the project. “My goal is to develop non-invasive tools to provide new insights into brain function and its diseases.”
The system developed by Katabi and her colleagues Yingcheng Liu and Guo Zhang uses a radio transmitter-receiver that can be installed in a person’s home. The device broadcasts a far less powerful signal than a typical home Wi-Fi router. The signal can pass through walls, but is reflected by water in the body. The receiver picks up the reflected signal like a radar system and constantly monitors human movement.
Last month, Katabi said the technology could be used to detect changes in sleepers’ breathing patterns, an early sign of Parkinson’s disease. Her team used the same hardware and different software to track the ongoing condition of Parkinson’s patients as they walked around the house. People with Parkinson’s disease tend to walk slower than healthy people of the same age.
According to a paper published Wednesday in Science Translational Medicine, Katabi and her colleagues tested the system on 50 people over a year, 34 of whom had Parkinson’s disease. The receiver collected 200,000 observations. By analyzing the data, they could distinguish between subjects with the disease and those without.
What’s more, they were able to measure the effectiveness of levodopa, a drug designed to control symptoms of the disease. For example, a patient’s walking speed improved for a period of time after taking the drug, then worsened as the effect wore off.
The system will reduce the need for doctor visits as doctors can monitor patients in real-time at home. The additional data could also help doctors prescribe optimal doses of Parkinson’s drugs. And because it collects data so quickly, the system could greatly speed up testing of new Parkinson’s drugs.
How soon can such a system be deployed? Katabi said it does not require FDA approval because the system only collects data.
“The device doesn’t make medical decisions, only measurements,” she said. “In principle, it should be possible soon.”
Hiawatha Bray can be reached at firstname.lastname@example.org him on twitter @GlobeTechLab.