Wagemans J, De Leeuw A, Catteeuw P, et al. Development of an algorithm-based approach using neuromuscular test results to indicate an increased risk for non-contact lower limb injuries in elite football players.


The full article can be found here.


Tell us more about yourself and the author team.

The supervisor of our team is Prof. Dr Dirk Vissers, a full professor in physiotherapy and rehabilitation sciences who teaches exercise physiology at the University of Antwerp. Dr Arie-Willem De Leeuw, a postdoctoral researcher in sports data science, is the brains behind the curtain, with his vast knowledge about artificial intelligence in sports. Next up, we have Dr Peter Catteeuw, exercise scientist and head of performance at the current champion of Belgium and Belgian Cup winner Royal Antwerp Football Club. He is the man who initiated this entire project with an abstract idea. Lastly, Jente Wagemans is a physiotherapist and a doctoral researcher passionate about sports-medical sciences. Our complementary backgrounds make us a complete research team. This enables us to think outside the box and combine ideas to render innovative concepts.


What is the story behind your study?

The initial idea came from Peter and his colleague at the club. They currently use an experience-based algorithm that substantiates their return-to-sports decision-making. “What if we could make these decisions more evidence-based, with a scientifically data-driven algorithm?” The greatest football club in Antwerp teamed up with the University of Antwerp, and intrigued and curious as we were – and still are – we got to work.


In your own words, what did you find?

Our main finding is that we explored player-related data to identify a predictor, or a combination of predictors, for an increased risk of non-contact injuries of the lower limbs. This data-driven approach looks beyond the usual suspects and can bring new insights into injury prevention. If more data becomes available, the model can be trained to improve. This would enable the development of an interface for coaches and medical staff with early warning signs to adapt training programs.


What was the main challenge you faced in your study?

How to handle missing data and work with a dataset influenced by a professional competition, i.e. transfers of players, busy schedule. If collecting data in one professional football team is already proven challenging, collecting big data at a national, European or global level might be an important barrier to overcome.


If there is one take-home message from your study, what would that be?

The sun is coming from behind the dark clouds of preventable injuries: we have developed a proof-of-concept showing that neuromuscular test data can predict an increased risk of non-contact lower limb injury. This might pave the way towards a more data-driven, personalized football training and coaching approach.

(Visited 95 times, 1 visits today)