A group of Brock University graduate students took top spot for the first time in a biomechanics challenge hosted by 3D motion capture equipment provider Movella.
This year, 81 teams from 28 countries were asked to create a coaching model for cross-country skiers by analyzing data collected from eight athletes who completed laps at varying degrees of difficulty while
wearing Movella’s XSens sensor-equipped suits. Teams were given two weeks to analyze the data and
prepare a presentation to share with judges.
Brock’s team included four Master of Science in Applied Health Sciences students focused on
Kinesiology: Alex MacNeil (BKin ’22), Tamar Kritzer (BKin ’22), Noah Polesky and Alexis Napper (BKin
’22). As in previous years, Assistant Professor of Kinesiology Shawn Beaudette offered the experiential
education opportunity to students in his Advanced Biomechanics Research Methods class.
The team began their analysis by creating an automated system to rank skiers’ performances and used principal component analysis (PCA) to simplify the complex data so they could better understand and
interpret it.
“PCA helps reveal patterns and relationships that might not be apparent from the raw data. We could see
the variations in the performances between a beginner and advanced athlete,” said Kritzer.
Using a machine learning model, the team was able to identify seven main cross-country skiing techniques from 6,000 movement segments and associate them with different athletic skill levels and power outputs.
Presenting the information to judges not necessarily familiar with biomechanics was a great experience in knowledge mobilization and translation said Polesky.
“Our team can talk all day about PCA, but if we can’t tell an athlete or a coach how it’s going to make them a better skier, then they won’t really care,” he said. “We had to figure out how to make this tool relevant.”
To better illustrate the differences between beginner and elite skiers to judges, the team created a
skeletal animation that shows how each level of skier moves and their power output. Elite skiers are
represented by the blue line, black is average and red is beginner.
More important than the visual presentation was how the animation can be used as a tool for coaching.
Data collected from an athlete wearing the XSens motion capture equipment can be analyzed and
compared to the animation to determine similarities to an elite skier and identify areas of improvement.
What’s even more powerful, says Polesky, is that data can be collected and analyzed several more times
from the same skier to help determine if their coach’s training methods are effective in improving athletic performance.
MacNeil adds that the technology allows coaches to move beyond their “coaching eye,” which describes
the observational skills needed to evaluate and correct athletic performance.
“It takes the whole guessing game away from coaching,” he said.
In recognition of placing first, Brock’s team received a full-body 3D motion capture suit, a free 12-month
software licence and a €5,000 grant to further develop the project they began for the challenge.
They also came away from the competition with new experiences applying academic knowledge and tools
to a practical situation.
“I got to play around with coding tools and software I can use to build my thesis project,” said Napper.
As first-place winners of the competition, Brock’s team will have an additional opportunity to
demonstrate their tool’s relevance and gain practical experience. Over the next four months, they will be
working closely with Movella and the University of Trento in Italy to fully develop the project. Data
collection will be expanded beyond the eight skiers used in the competition, and results will be used to
help train athletes for the 2023 winter Olympics.