How does the NFL take care of player health? With the help of AI

The National Football League (NFL) has found an interesting way to monitor player health and optimize performance – through the use of the Digital Athlete tool, an AI-powered system developed in collaboration with Amazon Web Services (AWS).

Injuries are not uncommon in football and can lead to significant financial losses, forcing teams to regularly review possible ways to ensure effective management of their players. The NFL has been taking steps to help teams and medical staff make better decisions about player health.

In recent years, the NFL has partnered with AWS to develop a tool that “represents the next generation of player health and safety for the league.”

Digital Athlete uses AI and machine learning to give a full overview of each player’s health, performance, and workload. To achieve this, it uses real game-day data, as well as information gathered from video footage filmed during games and practices.

Using tracking devices embedded in players’ equipment, AWS can effectively combine that data with available information about other metrics, such as weather and play type.

The tool then uses it to simulate millions of NFL games and in-game scenarios, aiming to highlight which players would be at the most risk of injury, offering insights that will support injury prevention and recovery.

AWS uses a set of 38 cameras to capture 3D player movements (5k video at 60 frames per second) and uploads it to the cloud for future analysis. Digital Athlete can then analyze each player’s position at any given time, providing teams with a detailed understanding of how each injury occurred – and even if fatigue could’ve been partly to blame.

According to AWS, Digital Athlete leverages everything it knows to run infinite game simulations, reconstructing every possible play and analyzing how and when an injury occurred. This data feeds into Digital Athlete’s Risk Mitigation Modeling feature, which can then determine how much training an athlete can ideally undertake while staying safe.

There is another component of Digital Athlete that’s still in development, called 3D Pose Estimation. It can analyze the exact movements that led to a certain injury.

“During each week of games, the Digital Athlete processes about 6.8 million video frames and documents around 100 million locations and positions of players on the field. And during team practices, the platform processes about 15,000 miles of player tracking data per week, equating to more than 500 million data points (10 Hz location data) per week,” AWS says.

The technology is currently used across all 32 NFL clubs, gathering data from over 1,500 players.