The future of gait analysis
At last year’s World Track and Field Championship in London, scientists from Leeds Beckett University deployed 49 highspeed cameras around the stadium and along the marathon course to record the movements of the best runners in the world. The result was billed as “the largest biomechanics study in the sport’s history,” and it revealed some surprising insights.
For one thing, more than 70 per cent of the 1 48 male and female marathoners landed heel-first with each st ride–a muchmaligned ga it pattern that some critics der ideas in efficient. But apparently many of the best runners in the world, including the top four finishers in the men’s race, are heel-strike rs. There were also some surprisingly quirky gaits on display: women’s 10,000m champion Almaz Ayana, for example, had an asymmetric stride with one leg reaching 20 centimetres farther than the other.
As fascinating as this data is, there are two drawbacks. One is that few of us will ever have a team of scientists poring over highspeed camera data of our strides. The other is that, even for the lucky runners who get this treatment, it’s not always clear what to do with the resulting data. Should Ayana, who already holds the 10,000m world record, try to make her stride more symmetric? Or should the rest of us try to emulate her awkward-but-fast gait?
One possible vision for the future comes from researchers at the University of Calgary, who deployed a simple, wearable accelerometer on runners, with the results analyzed by a computer using “machine learning” to detect which gait traits are most important. The study, which was published in the Journal of Sports Sciences, showed that the computer could tell the difference between “recreational” and “competitive” runners (as classified based on their race times) with better than 80 per cent accuracy.
Understanding t he gait differences between recreational and competitive runners is useful because the latter group tend to be more injury-proof, especially in the knee and hip areas. The main differences between the two groups showed up in their consistency: for competitive runners, the forces and accelerations and lower-leg movements in one stride are almost identical to the next stride. Inexperienced runners, in contrast, tend to have more stride-tostride variation. And that, lead researcher Christian Clermont explains, offers a simple target for future wearable devices to start tracking: don’t worry if you’re heel striking, as long as it’s consistent.