Smart sensors will make sure you lift with your knees bent and back straight

Devices like the Apple Watch may be able to remind you to stand up every so often and prompt you to get some exercise, but they’re not going to save your back from the perils of incorrectly lifting heavy objects. For that you’ll need a smart insole sensor, developed by researchers at the University of Quebec at Chicoutimi in Canada. Its smart algorithm promises to make sure you’re lifting with proper form more effectively than any health and safety video training video can do.

“We developed a smart tool that allows the detection of awkward postures, leading to the appearance of work-related musculoskeletal disorders,” researcher Eya Barkallah told Digital Trends. “The tool is composed of a set of sensors: A helmet containing an accelerometer, and an insole including four force sensors that enable us to calculate the displacement of center of pressure. This latter represents the point of application of the sum of all forces acting between the human body and the ground on which he’s standing. The detection is performed by classifying a posture into one of six predefined classes of postures using artificial neural networks.”

In tests, the device was able to correctly classify wearers’ posture with 95 percent accuracy. While the idea of permanently wearing a helmet may not be practical in all settings, there are some where the system could be extremely helpful. “Some industrial plant such as the aluminum casting industry have very hazardous environments where this system could be used,” Professor Martin Otis, who also worked on the project, told us. “We also have an interest in military and sport applications. For example, it’s possible to measure performance in a specific sport and detect the progression of some movements. We could also monitor isolated workers in forests.”

Going forward, Otis said that the team hopes to convince employee unions and boards of directors that the technology could be a useful tool for them to adopt. They are also looking for a company to partner with that’s interested in evaluating and helping improve the technology for different environments.

A paper describing the project was recently published in the journal Sensors.