How AI and machine learning could prevent injury and optimize performance in the world of professional sport

The days of sports professionals pushing their limits and hoping for the best — and risking injury or exhaustion in the process — could be behind us. Artificial intelligence (AI) and machine learning programs are increasingly relied upon to predict sports injuries and enhance player performance, the Wall Street Journal reports.

Many sports injuries can be predicted — with the right data. Injury can seriously impact an athlete’s physical and mental well-being, as well as their career potential, and is even more common among amateur athletes and young people than among professionals. But many injuries could be prevented — up to a third of injuries in professional football arise from overuse and could therefore be predicted by algorithms, according to Nature.

The use of AI in sports is seeing rapid global market growth: Valued at an estimated USD 1.4 bn in 2020, the market is expected to reach some USD 19.2 bn by 2030, according to a recent Allied Market Research report.

So how do the programs work? By collecting lots of data. Positional and biometric data gathering is creating hundreds of new metrics to optimize decision-making processes in sports, says Deloitte Insights. “There are athletes that are treating their body like a business, and they’ve started to leverage data and information to better manage themselves,” the WSJ quotes Kitman Labs founder and CEO Stephen Smith as saying. “We will see way more athletes playing far longer and playing at the highest level far longer as well.”

Tracking indicators can measure how players (or balls, or other objects) move around a field or court, using metrics including speed, acceleration, jump height, and lateral motion.

Biometric indicators — including pulse rate, blood oxygen levels, sweat rate and sleep rhythms — can measure players’ physical exertion, where they’re susceptible to injury, and their rest needs.

Some sports scientists also factor in contextual data — ranging from a player’s mood, to how much water they’ve drunk or how far they have traveled in a given period, their body mass index, and previous injuries.

Companies already operating in the market include nutrition coaching app FoodVisor, which uses machine learning to recognize over 1.2k types of food and estimate its quantity and nutritional value on a plate; AI system Zone7, which crunches data from tens of thousands of athletes to analyze changes in body movement and make injury risk forecasts and training suggestions; Skillcorner, which tracks football players’ movement and performance; and injury risk analytics provider Kitman Labs.

Are these programs being used in MENA? Not yet, as far as we can tell. MENA has seen the growth of data analytics in areas including digital fan engagement by the UAE Jiu-Jitsu Federation (UAEJJF), performance monitoring of fan services in the region by LaLiga, and the rise of Egypt’s own football data analytics startup Arqam FC. Programs that predict sports injuries don’t appear to be in use here yet, though.

All the potential applications are still to be explored: “In the last five or ten years, teams have been capturing huge volumes of data,” data scientist Derek McHugh is quoted by Nature as saying. “There’s a gap between how much data we’ve been capturing and what’s actually been done with it,” he says. “We’re trying to fill that gap using machine learning.”

But data privacy concerns could put the brakes on widespread adoption: Questions about who actually owns player data — including sensitive health information — remain unanswered. Some players are reluctant to have their data collected, citing privacy concerns and the fear that it could impact contract negotiations. And while certain professional sports leagues have put some frameworks in place to partially safeguard player data, the new tech is developing quickly, outpacing attempts to regulate data use, notes Deloitte.