Notes from the Cybathlon BCI series 2019

OPINION

October 2019

Ricardo Chavarriaga

Development of brain-computer interface technologies is a long term endeavour that faces multiple challenges. The transition from the controlled conditions of research laboratories to real-life is paved with uncertainties about multiple factors that affect human and machine performance. The Cybathlon tries to address these difficulties by setting up an event in which people with disabilities can compete on several disciplines where they perform day-to-day activities using state-of-the-art assistive technologies [1].

The Cybathlon BCI series 2019 is an opportunity for multiple research and development groups to test their systems in a competition setup. It took place at the TU Graz campus on September 2019 in the frame of the Graz BCI conference. There, 6 different teams competed in an EEG-controlled racing video-game. Representatives from Austria, France, Italy, Thailand and UK worked together for several months with end-users, people with motor disabilities, to develop their technical systems, and train both pilot and machine to perform well in the race.

The Cybathlon BCI series provides a great platform to test the robustness of the interfaces [2]. The competition framework requires BCI pilots to overcome the stressful situation of performing in front of an audience and being in direct competition with others teams. In addition, the event does not allow for performance evaluation over multiple days pushing teams to look for ways to reduce performance variability [3].

Figure 1. Teams and pilots prepare for one of the qualifying races. On the screen, Wilfried Panatier, pilot of team NITRO, is being set with the EEG electrodes cap by his team members.

After the event, the Team NITRO (INRIA, France) declared that “this was our first participation to a Cybathlon event, for the scientists as well as for the pilots. We were excited by the BCI competition and the suspense of the races (…) we also experienced how difficult it is to travel and participate to an international meeting when you are physically hindered. We learned a lot from the competition and are motivated to prepare further for the next event.”

Pascal Prietl, pilot of the local team MIRAGE 91 (TU Graz, Austria) told us how he “started training in April; at first only twice a month and as the event approached it became 2-3 times a week. The event was very interesting, exciting and cool! And with our 2nd place we are very satisfied. The team did a great job!”. The team leader, Reinmar Kobler expressed “I am very proud of the success of our team – especially our pilot Pascal. He managed to focus on the mental tasks and ignore all distractors. I think playing the game together and coaching him for this particular situation was critical for our success.”

In turn, Francesco Bettella, pilot of the WHI Team (U Padova, Italy) recounted that he also started the training in April, a few times per week. Adding that, “Although I was in good control since the beginning, during the race the main difficulty was to stay relaxed and focused on the task. After such an amazing experience in Graz, I am ready to start again the training in preparation of the Cybathlon 2020 in Zurich”. The team captain, Luca Tonin explained that “the biggest challenge was to make our BCI system robust enough to work in such a stressful and competitive environment. The Cybathlon BCI Series in Graz gave us the opportunity to evaluate our system in a real situation, probably even more demanding than normal daily activities. Once again, the idea of strongly promoting the learning interactions between user and machine above all seems to greatly do the job.”

Figure 2. Left, Team WHI. From left to right: Luca Tonin, Francesco Betella, Gloria Beraldo and Stefano Tortora. Right, Team Mirage 91. Pilot Pascal Prietl at the center surrounded by his team mates. Picture ©: Maximilian Wührer / TU Graz / ETH Zurich.

Francesco and his team WHI grabbed the 1st place in the final, followed by teams MIRAGE 91 (TU Graz, Austria) and NeuroCONCISE (Ulster University, UK). However, the real winner of this kind of events is the whole BCI research community as it allows a valuable tool to advnce the field and uncover new ways for bringing these technologies closer to achieving their potential to improve the quality of life of people with motor disabilities.

Figure 3. Final podium at the Cybathlon BCI series 2019. In the front row, from left to right, we can see the team pilots Pascal Prietl (Team MIRAGE 91, Austria), Francesco Betella (Team WHI, Italy) and Owen Collumb (Team NeuroCONCISE, UK). Picture ©: Maximilian Wührer / TU Graz / ETH Zurich.

Figure 3. Final podium at the Cybathlon BCI series 2019. In the front row, from left to right, we can see the team pilots Pascal Prietl (Team MIRAGE 91, Austria), Francesco Betella (Team WHI, Italy) and Owen Collumb (Team NeuroCONCISE, UK). Picture ©: Maximilian Wührer / TU Graz / ETH Zurich.

References

[1] Novak, D., Sigrist, R., Gerig, N.J., Wyss, D., Bauer, R., Götz, U., Riener, R., 2018. Benchmarking brain-computer interfaces outside the laboratory: The Cybathlon 2016. Front. Neurosci. 11, 1–14. DOI: 10.3389/fnins.2017.00756

[2] Ienca, M., Kressig, R.W., Jotterand, F., Elger, B., 2017. Proactive Ethical Design for Neuroengineering, Assistive and Rehabilitation Technologies: the Cybathlon Lesson. J. Neuroeng. Rehabil. 14, 115. DOI: 10.1186/s12984-017-0325-z

[3] Perdikis, S., Tonin, L., Saeedi, S., Schneider, C., Millán, J. del R., 2018. The Cybathlon BCI race: Successful longitudinal mutual learning with two tetraplegic users. PLOS Biol. 16, e2003787. DOI: 10.1371/journal.pbio.2003787

Author’s Biography

Passionate of responsible development and social implications of technology, Ricardo Chavarriaga has more than 12 years of experience in brain-machine interaction, computational neuroscience, and artificial intelligence. His work focuses on the conception of neurotechnologies that allow symbiotic interaction between human and intelligent machines.

Ricardo has a degree on Electronics Engineering from the Pontificia Universidad Javeriana in Cali, Colombia and a PhD in Computational Neuroscience from the Ecole Polytechnique Fédérale de Lausanne (EPFL) In Switzerland. He is the editor of the IEEE Brain eNewsletter and chair of the IEEE Standards Association working group on Neurotechnologies for Brain-Machine Interfacing. He is affiliated with EPFL and is an executive in residence at the Geneva Center for Security Policy (GCSP).