University of Delaware researchers have created an innovative AI model with 95% accuracy in predicting athletes’ injury risk post-concussion, having implications far beyond sports.
Athletes returning to play after a concussion face an elevated risk of injury. A novel artificial intelligence model developed by researchers at the University of Delaware aims to change that by accurately predicting the likelihood of lower-extremity injuries with 95% accuracy.
Published in the journal Sports Medicine, this study marks a significant advancement in sports medicine, addressing a complex issue that has long baffled medical professionals.
Athletes are susceptible to post-concussion injuries, such as sprains, strains, broken bones or torn ACLs.
“This is due to brain changes we see post-concussion,” Thomas Buckley, a professor of kinesiology and applied physiology in UD’s College of Health Sciences, said in a news release.
These subtle yet impactful changes affect skills like balance, cognition and reaction times.
“Even a minuscule difference in balance, reaction time, or cognitive processing of what’s happening around you can make the difference between getting hurt and not,” Buckley added.
AI: A New Frontier in Injury Risk Assessment
To develop this AI model, Buckley joined forces with an interdisciplinary team from UD. Austin Brockmeier, an assistant professor of electrical and computer engineering, and César Claros, a doctoral student, played crucial roles in this collaboration, along with Wei Qian, an associate professor of statistics in the College of Agriculture and Natural Resources, and Melissa Anderson, a former KAAP doctoral fellow and now an assistant professor at Ohio University.
The AI model analyzes over 100 variables, including sports and medical histories, concussion type and cognitive data before and after the concussion.
“Every athlete is unique, especially across various sports,” Brockmeier added. “Tracking an athlete’s performance over time, rather than relying on absolute values, helps identify disturbances, deviations, or deficits that, when compared to their baseline, may signal an increased risk of injury.”
Interestingly, the AI model’s accuracy remains high even without accounting for the specific sport played by the athlete, suggesting that individual characteristics significantly contribute to injury risks.
“We tested a version of the model that doesn’t have access to the athlete’s sport, and it still accurately predicted injury risk,” added Brockmeier.
From Research to Real-World Applications
The research shows that the risk of musculoskeletal injury post-concussion persists and may even increase over time as athletes unconsciously adapt to minor neurological deficits.
“Common sense would suggest that injuries would occur early in an athlete’s return to play, but that’s simply not true,” Buckley added. “Our research shows that the risk of future injury increases over time as athletes compensate and adapt to small deficits they may not even be aware of.”
The next steps involve collaborating with UD Athletics’ strength and conditioning staff to design real-time interventions aimed at mitigating these risks. Dan Watson, the deputy athletic director of competitive excellence and campus recreation, sees enormous potential in the AI model.
“In sport performance, we have two goals: improve the athlete’s abilities in their sport and to keep them on the field,” Watson added.
Beyond Sports: AI’s Broader Implications
The model’s applications extend far beyond athletics. Brockmeier envisions its use in predicting fall risks for patients with conditions like Parkinson’s disease. Claros is exploring its potential in aging research, especially regarding cognitive impairments.
“We want to use brain measurements to investigate whether baseline lifestyle measurements such as weight, BMI and smoking history are predictive of future mild cognitive impairment or Alzheimer’s disease,” added Claros.
The University of Delaware’s AI model opens new horizons for injury prevention and could revolutionize not only sports safety but also medical protocols for various at-risk populations, making it a significant breakthrough in contemporary health research.
Source: University of Delaware