Imagine a stadium where ultra-high-resolution video feeds and camera-carrying drones track how individual players’ joints flex during a game, how high they jump or fast they run—and, using AI, precisely identify athletes’ risk of injury in real time.
Coaches and elite athletes are betting on new technologies that combine artificial intelligence with video to predict injuries before they happen and provide highly tailored prescriptions for workouts and practice drills to reduce the risk of getting hurt. In coming years, computer-vision technologies similar to those used in facial-recognition systems at airport checkpoints will take such analysis to a new level, making the wearable sensors in wide use by athletes today unnecessary, sports-analytics experts predict.
Soccer has been among the biggest adopters of AI-driven data analytics as teams look for any kind of edge in the global sport. But some individual sports are also beginning to use these technologies. At the 2022 Winter Olympics in Beijing, ten U.S. figure skaters used a system called 4D Motion, developed by New Jersey-based firm 4D Motion Sports, to help track fatigue that can be the result of taking too many jumps in practice, says Lindsay Slater, sports sciences manager for U.S. Figure Skating and an assistant professor of physical therapy at the University of Illinois Chicago. Skaters strapped a small device to the hip and then reviewed the movement data with their coach when practice was done.
“We’ve actually gotten the algorithm to the point where we can really define the takeoff and landing of a jump, and we can estimate that the stresses at the hip and the trunk are quite high,” Dr. Slater says. “Over the course of the day, we found that the athletes have reduced angular velocity, reduced jump height, they’re cheating more jumps, which is where those chronic and overuse injuries tend to happen.” She says U.S. Figure Skating is assessing the 4D system in a pilot project before expanding its use to more of its athletes.
Also presenting challenges are data-privacy issues and the question whether individual athletes should be compensated when teams collect their information to feed AI algorithms.
The U.S. currently has no regulations that prohibit companies from capturing and using player training data, according to Adam Solander, a Washington, D.C., attorney who represents several major sports teams and data-analytics firms. He notes the White House is developing recommendations on rules governing artificial intelligence and the use of private data.
Those regulations will need to strike a balance in order to allow potentially important technologies to help people, while still taking privacy rights of individuals into consideration, Mr. Solander says.
This data revolution will mean that some overuse injuries may be greatly reduced in the future, says Stephen Smith, CEO and founder of Kitman Labs, a data firm working in several pro sports leagues with offices in Silicon Valley and Dublin. “There are athletes that are treating their body like a business, and they’ve started to leverage data and information to better manage themselves,” he says. “We will see way more athletes playing far longer and playing at the highest level far longer as well.”
While offering prospects for keeping players healthy, this new frontier of AI and sports also raises difficult questions about who will own this valuable information—the individual athletes or team managers and coaches who benefit from that data. Privacy concerns loom as well.
A baseball app called Mustard is among those that already employ computer vision. Videos recorded and submitted by users are compared to a database of professional pitchers’ moves, guiding the app to suggest prescriptive drills aimed to help throw more efficiently. Mustard, which comes in a version that is free to download, is designed to help aspiring ballplayers improve their performance, as well as avoiding the kind of repetitive motions that can cause long-term pain and injury, according to CEO and co-founder Rocky Collis.
Computer vision is also making inroads in apps for other sports, like golf, and promises to have relevance for amateurs as well as pros in the future. In wider use now are algorithms using a form of AI known as machine learning that crunches statistical data from sensors and can analyze changes in body position or movement that could indicate fatigue, weaknesses or a potential injury. Liverpool Football Club in the U.K. says it reduced the number of injuries to its players by a third over last season after adopting an AI-based data-analytics program from the company Zone7. The information is used to tailor prescriptions for training and suggest optimal time to rest.
Algorithms still have many hurdles to overcome in predicting the risk of an injury. For one, it’s difficult to collect long-term data from athletes who jump from team to team every few years. Also, data collected by sensors can vary slightly depending on the manufacturer of the device, while visual data has an advantage of being collected remotely, without the worry that a sensor might fail, analytics experts say.
Sports-data firm SkillCorner collects broadcast television video from soccer leagues around the world and runs it through an algorithm that tracks individual players’ location and speed. The data is used to derive performance metrics that aid player scouting and recruitment.
For now, one sports-data firm that has adopted computer vision is using it not to predict injuries, but to predict the next superstar. Paris-based SkillCorner collects broadcast television video from 45 soccer leagues around the world and runs it through an algorithm that tracks individual players’ location and speed, says Paul Neilson, the company’s general manager.
You must be logged in to post a comment.