Using sensors and ML to prevent warfighter injury
StorySeptember 09, 2022
A team of researchers at the Johns Hopkins Applied Physics Laboratory (APL – Laurel, Maryland) is developing a system to monitor physical fatigue and possible injury in soldiers in near-real time using body-worn sensors and machine learning (ML) algorithms.
Such a data-driven system, say the scientists, could prevent musculoskeletal (MSK) and other bodily injuries. Physical fatigue – a natural consequence of performing strenuous activity, particularly under heavy and irregular loads and often under less-than-optimal conditions – is often the first sign that these injuries might occur later.
Data from the U.S. Army Public Health Center reveal that MSK injuries among active-duty soldiers cause over 10 million limited-duty days (LDD) each year and account for more than 70% of the medically nondeployable population. MSK injuries and their follow-on effects are a leading cause for medical disability and discharge from service.
Biomechanical engineer Mike Vignos, an instructor at Johns Hopkins, is leading the project – now in its second year – to create ML algorithms that use data collected by people wearing sensors to reliably identify and quantify the severity of physical fatigue. The long-term goal: To predict the risk of MSK injury in near-real time and identify those at high risk for injury before they get hurt.
One of the most significant challenges to creating such ML algorithms is acquiring the data to train them. While previous research has seen an association between fatigue and MSK injury, this relationship has not been quantified, due in part to the fact that recreating harsh conditions and injuries in a laboratory is infeasible, even dangerous.
Vignos and his team are drawing on other studies undertaken at APL to produce computational models of the human body, as well as a compilation of data that correlates injuries with the environments in which they occur. As for the actual physical data-gathering from the exerted individuals. Vignos says that his team is still trying to identify the optimal sensor suite, one that could gather the necessary data without proving too cumbersome for soldiers who may be carrying heavy equipment or must maneuver in difficult conditions.
“Right now we’re still in the ‘over-sensorized’ phase, since we’re in the laboratory where we can afford to put five or six different sensors on people. But in the field there won’t be a scenario where that’s a realistic option,” Vignos says. “We’re still working out what that sweet spot is, that minimally viable sensor setup of ideally one or maybe two wearable sensors that will allow us to assess fatigue without contributing to the problem we’re trying to solve. We’re also exploring the option of doing this completely with remote sensing, which would be ideal because it removes the need for warfighters to wear sensors.”
During 2021, the team created an algorithm that can identify fatigue in simple binary terms: tired or not tired. This year’s effort – as the team collects data during exertion tests – is attempting to create a numerical score that can be used to quantify fatigue with more precision. (Figure 1.)
[Figure 1 | Team member Bryndan Lindsey (right) records qualitative data during pilot tests of the sensor suite. In the background, an avatar can be seen that is designed to replicate the real-time movement of the test subject. Photo: Johns Hopkins APL/Ed Whitman.]
“Our goal is to create a score that is generalizable across all these drivers of physical fatigue and that enables us to reliably predict and prevent MSK injuries that might otherwise take soldiers out of the line of duty,” Vignos says.
Biomedical engineer Kathleen Perrino, who manages an APL research portfolio dedicated to predictive health, says she believes that injury prevention informed by a fatigue score could take a number of forms. “Once we’re armed with an objective score for physical fatigue, there are a variety of ways we could apply it to improve the health of our warfighters,” Perrino explains. “We might change the armor that they wear, or the way their gear is distributed across their bodies. We might combine the score with knowledge about how a person’s genetic predispositions affect their biomechanics, and so individualize how people are trained and equipped. There are any number of ways it could be applied, which is exciting, as well as challenging.”
The APL, a scientific and engineering research and development division of Johns Hopkins University, has served as a technical resource since World War II for the U.S. Department of Defense (DoD), NASA, and other government agencies. Over that time, the lab has developed numerous systems and technologies in the areas of air and missile defense, naval warfare, cybersecurity, and space. The work currently being done on bodily injury and fatigue is aligned with APL’s portfolio focused on predictive health and human-performance modeling.
