Player safety is of paramount importance in contact sports, and as such the NFL has partnered with AWS to transform how its teams seek to avoid injuries to their players.
The development of the ‘Digital Athlete’ is enabling teams in the league to better predict and ultimately prevent player injuries by creating a virtual representation of an NFL player using AI and machine learning.
Running simulations to warn players at high risk of injury
The Digital Athlete is an injury prediction tool that collects data from sources such as the Next Generation Stats System (NGS), which provides data on each player’s location, speed, and acceleration at all times during a game. Elsewhere, data is also gathered from video footage of games and practice sessions, as well as sensors located around the stadiums which record performance metrics from devices fitted to the players’ equipment.
By combining these data points with other variables such as weather, equipment, and the type of play (such as a run or a punt), the Digital Athlete’s algorithms can run millions of simulations of specific scenarios during an NFL game to warn teams if a player is at a high risk of an injury. The insights generated can then be used to create personalised training, injury prevention, and recovery plans.
Determining the root cause of an injury
The footage captured by the 38 cameras positioned around each stadium is uploaded to the AWS cloud to be analysed by a computer vision (CV) algorithm, which locates the core and extremities of every player on the field simultaneously. From this, a 3D virtual skeleton of each player is created, and the precise position of a player’s joints and movements during each play is tracked.
This enables clubs to draw insights into what may have caused an injury. For example, if analysis of the movements of a player’s virtual skeleton indicates that their gait changed during a game, the injury suffered may be attributed to fatigue.
How well do you really know your competitors?
Access the most comprehensive Company Profiles on the market, powered by GlobalData. Save hours of research. Gain competitive edge.
Thank you!
Your download email will arrive shortly
Not ready to buy yet? Download a free sample
We are confident about the unique quality of our Company Profiles. However, we want you to make the most beneficial decision for your business, so we offer a free sample that you can download by submitting the below form
By GlobalDataNFL teams benefit from a vast dataset to reconstruct injury conditions
All 32 NFL clubs have had access to the Digital Athlete team portal since the 2023-24 season. This provides them with insights drawn from a large dataset, with around 6.8 million video frames processed by the Digital Athlete each game week, as well as approximately 15,000 miles of player tracking data from team practices per week.
The simulations that can be run from this also incorporate data on specific players, equipment, and environment (such as the weather or the playing surface), to reconstruct the occurrence of an injury and understand what may have contributed to it.
Further development of the technology will also improve the 3D player visualisations, increasing the depth of analysis and facilitating a greater understanding of the root causes of injuries to players.
