Watch out fitness fakers, shaking your iPhone for a few extra fitness points won’t work for much longer. Researchers at Northwestern University are developing an activity tracker that can discern when you’re cheating. The new technology will work with iPhone apps and fitness wristbands that count your steps as you walk. It's designed to help health care providers and insurance companies accurately reward healthy behavior or monitor patients, though it’ll presumably help folks trying to fulfill personal goals and stick to daily quotas as well.
While most data tracking programs are able to detect deceptive activity with 38 percent accuracy, the new method has increased accuracy to 84 percent. The improved tracker is “trained” on both deceptive and normal activity, and can not only identify common cheating behaviors, but can learn to recognize new ones.
In the study, recently published in PLOS ONE, 14 volunteers were asked to try to trick their phone trackers. They employed a variety of strategies, from sitting on a chair and shaking their phones, to putting their phones in their pockets and moving their torso or legs to mimic walking. The new tracking system was able to accurately identify each deception.
"As health care providers and insurance companies rely more on activity trackers, there is an imminent need to make these systems smarter against deceptive behavior," said lead study author Sohrab Saeb, a postdoctoral fellow at the Center for Behavioral Intervention Technologies at Northwestern University Feinberg School of Medicine. "We’ve shown how to train systems to make sure data is authentic."
But, according to Saeb, there is still one way to trick the system: “If someone attaches an activity tracker to a dog, the system can’t recognize that."