NFL Incorporates Machine Learning, AI Technology to Prevent Player Injuries

In December 2019, the National Football League (NFL) announced a partnership with Amazon Web Services (AWS) to apply the data collected from the league’s Next Gen Stats technology to artificial intelligence and machine learning technologies to prevent player injuries during games. The new program will be used to develop what the NFL calls the "Digital Athlete” platform, a simulation of an NFL player that will be used to model scenarios in the game environment without any risk to athletes.

The announcement comes in an era of increasing injury incidents; According to the NFL’s latest injury data reports, an average of 6-7 injuries occurred per game in the 2018 season. The NFL plans to employ the new technology to specifically address factors such as game rules, equipment, and rehabilitation and recovery strategies. Key players are hoping the partnership will enable the NFL to predict player injuries and prevent them before they happen.

Using Data for Injury Prevention

In October 2019, Thomas Insights interviewed Zebra Technologies Corporation, a key player in the partnership and the NFL’s ongoing Next Gen Stats program. Working in tandem with the NFL, Zebra Technologies embedded RFID chips in players’ gear, footballs, and throughout the stadium to transmit real-time location, speed, and acceleration for every play run during games and practices.

In the 2019 NFL season, Zebra installed 1,700 tags on player equipment and 20,000 tags on footballs, bringing their technology to every team in the NFL.

In the interview, Zebra’s Regional Manager of Client Services and Operation Adam Petrus explained the RFID chips’ ability to help coaches in the injury rehabilitation process.

"[Coaches] can track their players’ performance during practice… and it’s a great tool for if a player is injured. You’re able to utilize their pre-injury data as a baseline [so] you can really monitor their rehabilitation to ensure that they’re fully game-ready,” Petrus explained. "Let’s say they had a broken or sprained ankle, and they’re not hitting those same times. Well, then you know that foot just isn’t quite 100% yet, so [you can plan to] continue to rehab, work with the medical department, and… get him back on the field.”

Petrus also mentioned that the stats can be applied during the game to track player fatigue and adjust plays accordingly.

Collaborating to Reduce Concussions, Other Serious Injuries

In the press release, the NFL highlighted the importance of the program in studying – and hopefully preventing – one of the most dangerous injuries to its athletes: concussions.

"AWS and the NFL will also collaborate to develop computer vision models using Amazon SageMaker, Amazon SageMaker Ground Truth, and Amazon Rekognition to advance the ability to detect concussions and identify the forces that cause them,” the released noted. "Any techniques developed to detect and prevent concussion may also be capable of being extended to reduce a wide range of injuries. Any reductions, especially in concussions and lower extremity injuries, could have a potentially profound effect on player safety.”

However, The Verge spoke to Zachary Binney, an epidemiologist and consultant for Major League Baseball and college sports teams on injury prevention, who is skeptical of the results of the partnership until he sees results.

"Predicting injuries is challenging because there are so many factors that could contribute to a possible injury, from an athlete’s physical characteristics on a particular day to slight divots on a field. One athlete might have five attributes that research shows puts them at risk for an injury and still not get hurt, but another might look perfectly fine and tear a ligament the next day,” Binney told the outlet. “It is just an incredibly difficult problem.”

The Verge also notes that the program could gain better insight into injuries by adding capabilities to collect data on how hard body parts are being hit. However, despite the Digital Athlete partnership’s room for improvement, if proven effective, the research could stand to revolutionize injury prevention for all athletes, even those outside of the NFL.