Brailsports, a spin-off of imec, the University of Antwerp, and Ghent University, is developing a platform to help athletes and their coaches optimize training. The platform uses AI models to accurately estimate the athlete’s fitness and fatigue levels, allowing them to adjust training schedules and volumes to improve performance and reduce injury risk. Brailsports is working closely with the Lotto Dstny cycling team in the development of this platform.
The right training volume is crucial for athletes: both overtraining and undertraining can weaken fitness levels. To find the right training balance, personal coaches often base training schedules on the lactate test – the classic exercise test that measures the concentration of lactic acid in the athlete’s blood as they gradually increase their effort until reaching maximum exertion. Similarly, popular fitness apps use the results of exercise tests as a basis for estimating metabolism and determining appropriate training zones.
But the use of such performance tests poses two key issues:
- Firstly, they are highly invasive, requiring athletes to push themselves to the limit.
- Secondly, the results provide only a snapshot in time. If an athlete falls ill, it their training regimen may become too demanding.
Conversely, if an athlete improves faster than expected, their workouts may be too easy due to outdated threshold estimates.
A better assessment of athletes’ fatigue levels
As an alternative to the lactate test, Brailsports has developed a model that can more accurately assess an individual athlete’s fatigue levels. The model uses both internal (heart rate) and external (wattage, tempo) parameters to estimate the impact of each workout on the athlete’s lactate thresholds. By analyzing this data, the Brailsports model provides a more precise evaluation of the athlete’s fatigue.
While many current fitness apps and wearables already provide “training load” insights, these are typically based on generalized data from scientific studies, such as fatigue lasting for an average of 7 days and fitness improvements lasting 42 days. However, these generic estimates fail to account for individual differences – an elite athlete will typically recover faster than a recreational cyclist. To address this, Brailsports uses personalized models that analyze a large volume of data points, allowing them to more accurately estimate post-training fatigue levels for each user.
Building on expertise and scientific insights
The spin-off’s algorithmic model builds on the AI expertise developed within imec. It combines this technical knowledge with insights from sports physiology, training theory, and coaching. The co-founding team includes Erika Lutin (CEO of Brailsports), Bart Nonneman (sports coaching & AI, imec), Tim Verdonck (statistics & AI, University of Antwerp), Jan Boone (physiology & training, Ghent University), and Steven Latré (fellow AI at imec).
Co-founder Tim Verdonck (University of Antwerp/imec) notes: ‘Brailsports won’t replace the coach any time soon, but it will make their work even more effective. By smoothly extracting insights from the data and automating part of the work, schedules can be continuously adjusted based on individual data’.
Co-founder Professor Jan Boone (Ghent University): ‘By supplementing the classic lactate test with a data-driven approach, we are taking a step forward in terms of performance.’
Co-founder and CEO Erika Lutin: ‘For now, we’re focusing mainly on coaches of professional cycling teams, but the same data-driven approach is also going to have great benefits for recreational runners and cyclists, who often also set themselves quite ambitious goals and don’t want to get injured.’
Visit the Brailsports website for more info at brailsports.com