A smartwatch may ID Parkinson’s up to 7 years before symptoms seen
AI used accelerometry data to distinguish Parkinson’s patients, healthy people
Using a smartwatch to track how fast people can move may help identify who’s at risk for Parkinson’s disease up to seven years before symptoms become evident and doctors make a diagnosis, a study suggests.
Wearing the smartwatch for a week was enough to feed an artificial intelligence (AI)-powered machine learning model enough data to predict those who would develop the disease.
“We have shown here that a single week of data captured can predict events up to seven years in the future. With these results we could develop a valuable screening tool to aid in the early detection of Parkinson’s,” Cynthia Sandor, PhD, who led the study at Cardiff’s UK Dementia Research Institute, said in a news release. “This has implications both for research, in improving recruitment into clinical trials, and in clinical practice, in allowing patients to access treatments at an earlier stage, in future when such treatments become available.”
The study, “Wearable movement-tracking data identify Parkinson’s disease years before clinical diagnosis,” was published in Nature Medicine by Sandor’s team and researchers at the Neuroscience and Mental Health Innovation Institute at Cardiff University.
A feature of Parkinson’s is the death and dysfunction of dopamine-producing nerve cells (neurons). Without enough of the signaling molecule, or neurotransmitter, motor symptoms such as tremor, abnormally slow movements, and muscle rigidity begin to appear.
Using AI to read a smartwatch’s accelerometry data to predict Parkinson’s
It can take a long time before these symptoms manifest. This is commonly called a prodromal phase. When symptoms do appear, many brain cells have already died and the damage can be irreversible. As such, being able to diagnose Parkinson’s early is important.
“For most people with Parkinson’s disease, by the time they start to experience symptoms, many of the affected brain cells have already been lost. This means that diagnosing the condition early is challenging,” said Kathryn Peall, MD, PhD, who co-directs the Neuroscience and Mental Health Innovation Institute.
The researchers drew on data from 103,712 people registered with the UK Biobank database. Between 2013-2016, participants wore a medical-grade smartwatch that used accelerometry to measure speed of movement continuously over seven days. Accelerometry refers to the use of an accelerometer sensor, a small electronic component that can detect changes in motion and orientation, to measure and track a device’s acceleration and movement.
The participants were separated into two groups. One group had already been diagnosed with Parkinson’s, while the other received the diagnosis up to seven years after their smartwatch data were collected. They were compared to a group of healthy people of the same age and sex.
Smart watch data is easily accessible and low cost. As of 2020, around 30 per cent of the U.K. population wear smart watches. By using this type of data, we would potentially be able to identify individuals in the very early stages of Parkinson’s disease within the general population.
A machine learning model trained to use the accelerometry data performed better than other methods at distinguishing people with diagnosed Parkinson’s or prodromal disease from healthy people. The other methods included genetic information, lifestyle factors, blood tests, and early symptoms of disease.
Because their use is widespread, smartwatches are being used to track and record health data. For people with Parkinson’s, this could mean being able to monitor both motor and nonmotor symptoms, including drops in blood pressure.
“Smart watch data is easily accessible and low cost. As of 2020, around 30 per cent of the U.K. population wear smart watches. By using this type of data, we would potentially be able to identify individuals in the very early stages of Parkinson’s disease within the general population,” Sandor said.
“Accelerometry is a potentially important, low-cost screening tool for determining people at risk of developing Parkinson’s disease and identifying participants for clinical trials of neuroprotective treatments,” the researchers wrote.
About the Author
