NFL Employs AI to Forecast Injuries, Boost Player Wellness

The National Football League has partnered with Amazon Web Services to develop an injury prediction tool, aiming to enhance player health and manage risks. This initiative reflects the league's investment in technology to keep players on the field, especially given the hundreds of millions of dollars teams commit to their rosters each season.

Credit: NFL

Julie Souza, global head of sports at AWS, stated, "Fans want their favourite players on the field. The team owners certainly want those players on the field. The athletes themselves want to be on the field." Souza added that efforts to improve player health represent a "noble endeavour."

The "Digital Athlete" tool aggregates video and data from all 32 teams, encompassing training, practice, and games. This comprehensive information helps teams monitor player workload, assess injury risk, and track league-wide trends and benchmarks.

Credit: NFL

All teams have utilised the Digital Athlete portal for three seasons, with medical staff praising its benefits. They describe it as a "one-stop shop" for injury information that was previously unavailable from a single source.

Tyler Williams, vice president of health and performance for the Minnesota Vikings, explained that the tool provides more information for asking better questions and making effective interventions. He summarised sports science as "How can we measure and assess to make ourselves more effective and efficient."

Digital Athlete gathers data through sensors in shoulder pads, cameras, and optical tracking during practice and games. This process is similar to how NextGen stats collects information but on a far larger scale.

While NextGen Stats generates approximately 500 million data points in an entire season, Digital Athlete accumulates that volume weekly. This immense data necessitates machine learning and artificial intelligence to extract meaningful insights.

Souza emphasised, "The sheer volume of data means somebody can’t be sitting there with a clipboard or Excel figuring that out." She noted it is "absolutely a job for high-performance computers, machine learning, artificial intelligence, all of those things."

A key strength of Digital Athlete is its capacity to consolidate data from over 1,500 players across all 32 teams. This provides training staff and coaches with superior insights into injury susceptibility and effective mitigation strategies.

Teams have used the tool to plan training camp schedules and determine appropriate weekly workloads during the season. It helps identify players or position groups that might need reduced work to prevent soft-tissue injuries.

Williams noted that the data often reinforces existing beliefs but occasionally highlights issues he might have missed. Empirical evidence aids in persuading players to take days off or convincing coaches to adjust practice loads.

The model tracks decelerations, accelerations, total on-field workload, and changes of direction to inform staff about a player's exertion levels. While the NFL has reported a reduction in overall injuries since the tool's introduction, Williams acknowledges multiple contributing factors.

Williams stated, "Everybody is always going to want the smoking gun that if we do A and pair it with B, we will get C." He stressed that while nobody is preventing injuries, the goal is to mitigate risk through comprehensive measurement and assessment.

The NFL has also leveraged data from these systems to model the impact of rule changes. Digital Athlete simulated 10,000 seasons to project how the new kickoff rule, introduced last season, would affect injuries.

The data has also guided the league in identifying optimal helmets for concussion protection. Dawn Aponte, an NFL executive, indicated that the data informed changes to quarterback helmets, adding more padding at the back.

Aponte reported that the NFL recorded its lowest number of concussions last year since tracking began. She attributed this to the ability to analyse data, leading to improved equipment, better-performing helmets, and enhanced manufacturing based on impact types.

Initially, Aponte observed scepticism from some, who questioned if this was "junk science." However, she now sees widespread acceptance from coaches, medical staff, and players, viewing it as an "additive tool."

She explained that the tool complements coaching decisions, providing data-driven insights. "When you’re actually able to point them to certain things and say, hey, this is what we’re seeing," Aponte stated, "When they start losing players, particularly in training camp, they pay more attention."

---

• The NFL collaborated with Amazon Web Services to develop the "Digital Athlete" AI tool for injury prediction and player health management.

• The tool collects extensive player data from training, practice, and games across all 32 teams, providing insights into workload and injury risk.

• Medical staff use the "one-stop shop" portal to inform practice schedules and player workload adjustments, aiming to mitigate injury risks.

Source: AP NEWS