Wearable sensors can detect Parkinson’s disease almost nine years before clinical diagnosis by recording how a person turns while walking. This conclusion comes from researchers at the Tubingen University Hospital. And their research findings were published in the journal Annals of Neurology.

The study used ten years of monitoring data from over a thousand volunteers aged 50 and older. Participants wore a compact sensor on their waist that measured the parameters of turning while walking through a corridor, including angle, duration, and speed. Researchers paid special attention to angular velocity and how quickly a person turned at the fastest point of the movement.

The study showed that a decrease in angular turning speed begins about 8.8 years before the clinical onset of Parkinson’s disease. This sign appears much earlier than other motor symptoms of the disease, such as characteristic tremor, rigidity, or gait disorders.

To verify the reliability of the detected link, researchers used a machine learning model. It accounted for age, gender, and turning parameters, and was able to predict the development of Parkinson’s disease with an AUC of 80.5%. This level is considered high for long-term prediction, especially when dealing with subtle changes.

“This research opens a vital window for early intervention,” said Associate Professor Brook Galna from Murdoch University’s School of Allied Health and co-author of this study.

 “By detecting changes in turning speed through wearable sensors, in combination with other early signs of Parkinson’s, we can identify individuals at risk long before symptoms become clinically apparent,” he said.

“Earlier detection of people at risk of developing Parkinson’s will speed the discovery and testing of neuroprotective treatments designed to slow disease progression and keep people living independently for longer,” the aforesaid researchers concluded.