Decoding Parkinson’s Patients Movements In The Real World

Scientists have traditionally studied how the brain controls movement by asking patients to perform structured tasks while connected to multiple sensors in a lab. While these studies have provided important insights, these experiments do not fully capture how the brain functions during everyday activities, be it walking to the kitchen for a snack or strolling through a park. For people living with Parkinson’s disease, this gap between laboratory research and real-world behavior has limited efforts to improve gait symptoms outside of the clinic. 

Now, researchers from UC San Francisco have taken an important step toward closing that gap by successfully moving the laboratory into the living room. In a new study published February 13 in Science Advances, the team demonstrated that brain activity recorded from fully implanted devices while patients are at home can be used to reliably determine whether a person is walking or not. 

By analyzing synchronized neural and movement data collected during more than 80 hours of unsupervised daily activity, researchers identified individualized patterns of brain activity associated with walking. These neural signatures allowed an implanted deep brain stimulation (DBS) device to classify movement states using signals generated during natural, at-home activities.

“This is the first demonstration that a fully implanted device can be used to detect a specific movement state in humans during real-world activity,” said senior study author Doris Wang, MD, PhD, a neurosurgeon and associate professor of Neurological Surgery at UCSF. “Our findings show that it is possible to identify meaningful neural signals outside the laboratory, which is an important step toward more personalized and responsive neuromodulation therapies.”

Gait impairment is one of the most disabling symptoms of Parkinson’s disease. Patients often experience short, shuffling steps, difficulty initiating movement, and instability during turning. These changes increase fall risk and can significantly affect independence and quality of life. Current DBS therapy delivers continuous stimulation, but symptoms such as walking difficulty can fluctuate throughout the day and often do not respond to DBS settings that treat tremor, slowness, or stiffness.

In this early feasibility study, four participants with Parkinson’s disease were implanted with a bidirectional investigational DBS system that recorded neural activity from movement-related brain regions, including the motor cortex and globus pallidus. Wearable sensors provided simultaneous movement measurements, allowing researchers to match brain signals with periods of walking and other activities. The results showed that walking could be distinguished from non-walking states based on neural signal alone, with patterns that varied across individuals.

“We identified personalized neural biomarkers associated with gait and demonstrated that these signals can be used for real-time movement state classification within the constraints of an implanted device,” said Wang. “This establishes a framework for future adaptive DBS systems that could adjust stimulation in response to a patient’s activity state.”

The authors emphasize that the study was designed to demonstrate feasibility rather than clinical efficacy. The sample size was small, and additional studies will be required to determine whether movement-state detection can improve clinical outcomes. The research team is now planning future trials to evaluate whether stimulation settings optimized for walking can be dynamically applied using these neural biomarkers.

“By enabling the study of brain activity during natural behavior, the approach may ultimately expand the reach of brain–computer interfaces and adaptive neuromodulation beyond controlled laboratory environments and into everyday life,” said Wang.