Researchers in Chicago have come up with a way to train wearable devices to spot fake activity. Shaking your phone while lounging on the sofa may currently pass as a brisk walk, but the new system can apparently differentiate with much greater precision between what’s real activity, and what isn’t. Although the study was carried out with smartphones, it can also be applied to fitness trackers and smartwatches.
“As health care providers and insurance companies rely more on activity trackers, there is an imminent need to make these systems smarter against deceptive behaviour,” said lead study author Sohrad Saeb, a postdoctoral fellow at the Center for Behavioral Intervention Technologies at Northwestern University Feinberg School of Medicine.
In the study, subjects used various strategies to trick their devices. They shook the phone and swung their hands while sitting to pretend they were walking. They also pretended to sit by not moving their arms while walking.
Results shows that the new algoritm is capable of identifying genuine activity 85% of the time, more than double the 38% accuracy of current devices. .
“Very few studies have tried to make activity tracking recognition robust against cheating,” said senior author Konrad Kording, a research scientist at RIC and an associate professor in physical medicine and rehabilitation at Feinberg.
“This technology could have broad implications for companies that make activity trackers and insurance companies alike as they seek to more reliably record movement.”
The days of being able to trick your fitness tracker about how much you exercise each week could soon be over.
Nevertheless, Saed went on to say that the system isn’t perfect just yet.
“If someone attaches an activity tracker to a dog, the system can’t recognize that.”