With more than 10 million people around the globe affected by Parkinson’s disease, which is now considered the fastest-growing neurological disease, clinicians could use all the help they can get in improving methodology for tracking its severity and progression. As it stands, the motor skill and cognitive function tests they commonly administer to get a handle on a Parkinson’s patient’s condition are too prone to being rendered inaccurate by outside factors. This includes physician distance as a barrier to proper treatment, and studies point to some 40% of these patients having never been seen by a neurologist or disease specialist, in many cases because of difficulty traveling or living too far away.
Researchers from MIT and elsewhere have recently stepped up with a convenient in-home device capable of monitoring a patient’s movement behaviors, which are indicative of severity, progression, and medication response. This Wi-Fi router-sized device aggregates data passively via radio signals reflected off an individual’s body during everyday movement around the house without requiring incorporation as a wearable or any extra behavioral modification from its subject. Even breathing patterns during sleep can be picked up by the device.
A year-long, in-home study comprised of 50 participants found that clinicians can leverage the data with machine-learning algorithms to accurately track Parkinson’s progression and medication response, surpassing what is possible through periodic in-clinic evaluations. The passively amassed data could include more than 200,000 gait speed measurements and other substantial, medically useful metrics.