A new study published in the journal Frontiers in Bioengineering and Biotechnology describes how the analysis tool — known as the Engine for Automatic Biomechanical Evaluation — works. It can be used outside of a laboratory setting and doesn’t require subjects to wear physical body markers.
SwRI researchers teamed up with the Air Force Special Warfare Training Wing at Joint Base San Antonio to test the tool. Researchers videotaped trainees performing certain functional movement assessments and processed the footage using algorithms to analyze full-body movement patterns.
“There are established markers that might be indicative of injury risk: pigeon toes, whether your knees buckle inward or outward when you move, et cetera. But the problem is that only a handful of people actually exhibit those markers,” said Lance Frazer, a senior research engineer with Southwest Research Institute. “We actually tracked the way the entire body moves in tandem, which not too many people to our knowledge have done before.”
Frazer said the tool attempts to capture what he calls a “biomechanical fingerprint.”
“Everyone does certain movements a little bit differently," he explained. "Sometimes all of those minor differences work collectively in a good way. But sometimes they work collectively in a bad way.”
The researchers screened 156 Air Force trainees. Their model correctly predicted 70% of the injuries that occurred during an eight-week training course. SwRI is now gearing up for a much larger study of the tool.
“In an ideal world, you would have trainees do some functional movement assessments in front of our cameras before they start military training,” Frazer said. “We would pick up on those that are predisposed to injury risk. Then, at that point, it would be up to … the military to design an intervention.”
