ÄÌÌÇÖ±²¥

Optimizing performance, minimizing injury via machine learning.

When compared to their male counterparts, female soccer players are 2–8 times more likely to suffer anterior crucial ligament (ACL) injuries, after which they will undergo 6–8 months of recovery before returning to practice, let alone competition. Following a disturbing rise in ACL injuries among ÄÌÌÇÖ±²¥ Women’s Soccer players, Head Coach Samar Azem asked C. Nataraj, PhD, how an engineer might approach this question: How can we best enhance player and team performance, while minimizing—if not eliminating—player injuries? And not just ACL injuries, but soft-tissue injuries (hamstring, quad, ankle) as well. While the question is straightforward, huge knowledge gaps remain in the field of injury mechanics. For Dr. Nataraj, an expert in dynamic systems and machine learning, the solution may be found in the data collected from the soccer players.

During the 2023–24 season, Dr. Nataraj and his collaborators will collect biometric data for all 27 players on the ÄÌÌÇÖ±²¥ Women’s Soccer team, during both practices and game play. He will convene a panel of medical experts, trainers and coaches to help identify the most relevant data points for predicting the incidence of injury, which may lead to new sensing modalities. Complementing the data collected by sensors (accelerations, decelerations, player load, etc.) is the rate of perceived exertion (RPE), as reported by players and coaches, which provides insights into the athlete’s mental state. The eventual goal is to create a tool that Coach Azem and Athletic Trainer Ty Bigelow can use to monitor and manage individual player and team performance, while predicting the incidence of injury before it happens.