2020 Youth Olympic Games: a testing ground for new technologies
Almost 2,000 athletes aged 18 and under have descended on Lausanne and the surrounding area for the Youth Olympic Games. EPFL, in partnership with the University of Lausanne (UNIL) and Lausanne University Hospital (CHUV), has used this unique opportunity to test a series of technologies developed in the lab before bringing them to the general public.
EPFL boasts a longstanding track record in pushing the boundaries of sports innovation. The school has contributed to the 2020 YOG education program, using the games as a testing ground for new technologies developed in conjunction with the UNIL-EPFL Sport and Health Center (CSS) and CHUV. Pascal Vuilliomenet, a project manager at EPFL’s Vice Presidency for Innovation, is intimately familiar with the 30 or so laboratories carrying out sports-related research and development work. It’s his job to create synergies. “The 2020 YOG is a major event,” says Vuilliomenet. “It’s a chance to get everyone working together on applied projects, to strengthen existing collaborative relationships – and to start new ones.”
«The 2020 YOG is a major event. It’s a chance to get everyone working together on applied projects, to strengthen existing collaborative relationships – and to start new ones.»
Health and injury prevention in sports
For EPFL, UNIL and CHUV, the Lausanne 2020 YOG have been a unique opportunity to showcase “Health for Performance,” an unprecedented and ambitious educational project that combines technology with sports science and injury prevention techniques to promote the benefits of good health for athletic performance.
Athletes have had a chance to create their own avatar, or digital double. After completing a series of tests, their data have been fed into the system. The results have been collated in an app, developed by EPFL startup Katapult, where athletes have been able to see their personal musculoskeletal and motor profile, highlighting their strengths and weaknesses.
The project is a joint effort between researchers and the CSS. “It’s a virtuous circle,” explains Vuilliomenet. “The CSS, which works directly with athletes, tells us what products and services it wants to develop, then we find labs capable of meeting that demand. The CSS tests the prototypes and reports back to us. And in the process, we generate a rich vein of data that can be used in other areas of research.”
The idea behind Health for Performance is to bring about a paradigm shift in sports coaching and training. Whereas in the past the emphasis was on strength, speed and endurance, the project stresses how moving in the right way can help athletes reach their full potential. “These days, everyone wants to trail race, that’s what’s hot,” says Stéphane Maeder, head of the CSS. “And people want to run faster, go further, or lift heavier weights. But what happens? They get injured! We want to change mindsets and place the emphasis on quality. Athletes should stop thinking about how far they can push themselves, and focus instead on how they’re moving their bodies.” Athletes competing in the 2020 YOG will have a chance to use some of the assessment tools developed by Maeder and his team. These same tools will gradually be rolled out at the CSS.
Sports health laboratories and startups
At the next YOG, competitors will have a chance to try out the avatar they create using the Katapult app, which will be enhanced by various technologies developed by EPFL, UNIL, CHUV and other Smart Move partners. The app will produce ever more detailed results. For instance, athletes will gain insights into how sleep and diet affect their health and performance, and even check their overall state of health by looking at biomarkers in their sweat. The technologies will include contributions from several EPFL labs and startups.
Identifying and tracking individuals
Alexandre Alahi, a tenure-track assistant professor at EPFL’s Visual Intelligence for Transportation Laboratory, works on the implications of human behavior for self-driving cars. He has developed an app that uses a smartphone camera to detect a person’s movements, posture, body position and direction of gaze, including in the middle of a crowd. The system can identify and continue tracking people even if they temporarily leave the camera’s field of view because it recognizes their behavior, posture, movement and activity profile. The technology works by building a unique, 17-point model of a person’ skeleton. “The system could give coaches precise, detailed, real-time insights into athletes’ movements and postures,” says Prof. Alahi.
The Computer Vision Laboratory, led by Professor Pascal Fua, has developed a system that uses one or more cameras to analyze an athlete’s movements. Using algorithms, the system reconstructs the position of the athlete’s body in space and calculates the trajectory of the joints. Prof. Fua and his team are developing versions of the technology that can operate in increasingly complex environments.
The lab’s next challenge, working with Swatch Group’s sports watch arm Swiss Timing, is to track competitive divers as they execute their dives. “Our system should help judges score dives more objectively, and improve training,” explains Prof. Fua. “The idea is to overlay dozens of recorded dives, check for consistency of movement, and perhaps even suggest improvements.”
Eating well to perform well
How can marathon runners keep their blood sugar levels stable? How can sprinters gain explosive strength? In theory, science should be able to address questions like these. But the sheer number of variables in play means that, in practice, precise answers are more elusive. “An apple’s origin, variety, whether or not it’s organic – all these factors affect different people in different ways,” explains Chloé Allémann, a scientific assistant who leads the Food and You project at EPFL’s Digital Epidemiology Lab.
Diet and health are closely related. Likewise, the foods athletes eat impact their performance. “Many health issues can be addressed with a balanced diet and regular exercise,” says Allémann. “In terms of sports, we could develop custom diets for individual athletes.”
From her diet and lifestyle research – spanning aspects such as exercise, sleep, everyday habits, overall health, gut microbiota and blood sugar – Allémann understands that we all react differently to the foods we eat. “Blood sugar levels vary from one person to the next,” she explains. “Eating a banana before exercising is beneficial for some people but not for others.”
Wearable biochemical sensors
EPFL spinoff Xsensio – working with the Nanoelectronic Devices Laboratory, from which it originated – has developed on-chip sensors that measure biomarkers in sweat. The chip, which measures barely 5 mm2 and is worn on a smart patch stuck to the skin, sends real-time, continuous readings to a smartphone app or other external system. “The sensors can measure variations in electrolyte, metabolite or protein concentrations in an athlete’s sweat,” says Esmeralda Megally, Xsensio co-founder and CEO. “Electrolyte changes, for instance, tell us how dehydrated the wearer is, or how well they’re acclimatizing to heat. Similarly, lactate levels give us insights into muscle activity, while cortisol is an indicator of stress. The data should also be of interest to medical practitioners.”