Laura Bowen, Doctoral Researcher in the School of Sport, Exercise and Rehabilitation Sciences describes her PhD thesis in three minutes.
Title: Laura Bowen: University of Birmingham Three Minute Thesis Final (2013)
Duration: 3.26 mins
Speaker Names (if given): S1 Laura Bowen, Doctoral Researcher, School of Sport, Exercise and Rehabilitation Sciences
S1 Earlier this month, Gareth Bale moved from Tottenham to Real Madrid for a record-breaking £85 million, after settling for a measly £300,000 a week wage. This means, if he were to get injured resulting in time out, which is on average 18 days in football, he'd get paid £770,00 without even kicking a ball. So, injury prevention is clearly important. Can we predict these expensive and sometimes career-ending injuries?
Of course in some cases, injuries are the result of random dirty tackles, but in others, it's a combination of fatigue, physical weakness and overuse. These injuries have the potential to be predicted, and the goal of my PhD is to create a model to achieve this. We can firstly monitor a player's activity using the GPS devices they wear in both training sessions and matches. From this, we can establish relationships between this data and the injury.
As you can see here for example, this is one player's activity during one training session. The same player, this is his activity analysed over a month. One way of identifying injury risk is to track the relationship between muscle stress and high-intensity movement. Generally, the more high-intensity movements there are, the greater the muscle stress.
So if you went for a run, there'd be more stress on your muscles than if you went for a leisurely stroll. That same principal applies here. The yellow line is muscle stress and the purple line is high-intensity movement. As you can see, they are interdependent, and injury risk is highlighted when there is a discrepancy between the two. So when muscle stress continues to rise despite a fall in high-intensity movements, in this case, the player should have taken a couple of days to recover, but instead, he continued to play and train for the next six days until he was sidelined with a muscle strain.
So the information is all there. The pattern recognition process is the challenge, analysing relationships and reoccurrences in a mass of data, both in one player's history and in the squad as a whole. In this case, I identified muscle fatigue from over-training as the major factor in the resulting injury.
But, does it always work? Well in the club I'm currently at, there have been five muscle injuries so far this season. Of these, and based on the model, I have managed to identify injury risk between three days and two weeks prior to injury in all cases. The model itself is not perfect, but prevention is cheaper than the cure, because I could work on research into this for the next 77 years, and still not have earned as much as Gareth Bale will during one injury. Thank you.