We’re all going to die eventually – but what if you knew when you were in danger of dropping dead just because of the way you walked? A new study shows that wrist-worn motion sensor measurements can be used to predict one’s mortality risk up to five years later. In one of the largest validations of wearable technology to date, the research opens up the possibility of one day using the motion detection system in smartphones to monitor patients’ health status without the need for in-person visits to the doctor’s office.
The study published Thursday in the journal PLOS Digital Health, was carried out using data from over 100,000 Brits from the massive UK Biobank project, which began collecting health and biometric information from participants in 2006 and will continue for another 14 years. Using a week of wrist sensor data, researchers at the University of Illinois at Urbana-Champaign developed a model that breaks down a person’s acceleration and distance traveled into six-minute chunks. According to study author Bruce Schatz, a University of Illinois computer science researcher, the scientists chose this duration to mimic the six-minute walk test: a measurement of heart and lung function commonly taken during a doctor’s appointment that prompts participants to walk at a normal pace for six minutes and compares the total distance covered with benchmarks based on age.
The test is “a very good external measure of what’s going on internally” and could easily be replicated using the accelerometer that’s in a wrist sensor or a cheap phone, Schatz told The Daily Beast. “I know for sure that these types of models will work with cheap phones.”
The researchers’ model’s predictions of future deaths were 72 percent correct at one year and 73 percent at five years – a similar level of accuracy found in a study published last year that analyzed the same dataset but used hours. instead of minutes of data. This new study, Schatz argued, is a more promising demonstration of passive surveillance technologies like phone and wrist sensors because his team’s model requires less data and offers a high level of user privacy.
“If you record all the data, it’s true that people have characteristic gaits and you can tell who the person is. But it’s entirely possible to be part of the signal that’s good enough to measure vital signs but completely obscures who the person is,” he said.
“I know for a fact that these types of models work with cheap phones.”
— Bruce Schatz, University of Illinois
Still, using everyday technology to passively monitor patients could pose problems when users cannot provide continuous informed consent, situations that could be complicated by degenerative diseases or lack of technological literacy. These ethical questions, Schatz said, are still speculative but deserve coordinated consideration by scientists as research advances.
While the sensors used in the study were nearly identical to those found in basic cellphones and smartphones, future work should validate this model in a large sample when users carry phones in their pockets rather than wearing sensors on their wrists. Downloading an app that measures your everyday health could be a convenient and painless way to keep people healthier for longer.
“If you want to improve the overall health of the entire population, a project like this is really important,” said Schatz.