A new international study suggests that wearable technology could detect Parkinson’s disease up to nine years before a clinical diagnosis by monitoring how people turn while walking.
The research, conducted by five institutions including the University Hospital of Kiel and Murdoch University, followed 1,051 participants over the age of 50 for ten years. Participants wore a single sensor on their lower back, which measured turning movements such as angle, speed, and duration along a 20-metre hallway.
Data collected at the University Hospital Tübingen in Germany revealed that slower peak angular velocity—how quickly someone turns at their fastest point—was linked to a higher risk of developing Parkinson’s disease. Turning speeds began to decline roughly 8.8 years before diagnosis, making it one of the earliest detectable motor signs of the condition.
Researchers validated their findings using a machine learning model that incorporated age, sex, and peak angular velocity to predict which participants would develop Parkinson’s. The model achieved an area under the curve (AUC) of 80.5%, indicating strong predictive accuracy.
“This research opens a vital window for early intervention,” said Associate Professor Brook Galna from Murdoch University’s School of Allied Health.
“By detecting changes in turning speed through wearable sensors, alongside other early signs of Parkinson’s, we can identify individuals at risk long before symptoms appear,” he added.
Early detection, experts say, could accelerate the development of neuroprotective treatments aimed at slowing disease progression and helping patients maintain independence longer.
Related topic：
