by Dr. Reinhold Scherer
Deputy Head of the Institute for Knowledge Discovery in Graz (Austria)
The 6th International Brain-Computer Interface Conference took place from September 16 – 19, 2014 at the Graz University of Technology, Graz, Austria. This year’s conference theme was “The Future of Brain-Computer Interaction: Basics, Shortcomings, Users”. About two hundred researchers from 35 nations participated in the event and made it a huge success.
It was a pleasure for the Graz-BCI team to host the conference for the 6th time. The first day of the conference focused on networking. Participants had the opportunity to interact with representatives of European projects, the industry and end-user groups, and to discuss issues and possible solutions related to BCI. Manfred Halver from the Austrian Research Promotion Agency was present and available to answer relevant questions concerning the Horizon 2020 EU Framework Programme for Research and Innovation. Days 2 to 4 were focused on research and on sharing of scientific ideas. Brain-Computer Interfacing is a very interdisciplinary research area. Therefore, the organizers aimed at providing a balanced mix of novel research results in several areas, which included, among others, engineering, signal processing, machine learning and pattern recognition, basic neuroscience, psychology, and work in end-users with different medical conditions. By mixing talks and poster from all these areas and using a single-track structure, it was possible to create a varied and exciting program for the attendees. Other key success factors were the keynote speakers Scott Makeig (Towards Robust, Pervasive Brain-Computer Interfaces) and Nick F. Ramsey (Sweet spots for BCI implants: the matter of finding and interfacing).
As usual, talented and emerging young scientists that pursue a PhD degree have been awarded prizes that were sponsored by Brain Products GmbH. The awards consisting of a check of Euro 500 along with a trophy and certificate were presented in two categories: Best Talk and Best Poster.
The “Best Poster Award” winner was Helena Erlbeck from the University of Würzburg, Germany. The title of the winning poster was “Unifeature vs. multifeature oddball and magnitude of deviance effects on the mismatch negativity”. The mismatch negativity (MMN) is frequently used to investigate cognitive functioning in individuals with disorder of consciousness. The attention span of such individuals is generally short and as such monitoring paradigms have to be both of short duration and allow for reliable detection of the MMNs. Multifeature oddball paradigms comprising multiple deviants may represent a valuable solution to this challenge. Typical paradigms comprise one standard and one rare deviant, thus focusing on only one feature of typically auditory stimuli. Multifeature oddball paradigms present several deviants in various dimensions. Helena and colleagues showed that duration deviants elicited larger MMNs in the multifeature oddball, and that MMN amplitudes increased with increasing deviant minus-standard distance in terms of the target stimulus feature. In a nutshell, duration oddballs elicit reliable MMNs in multifeature paradigms and may therefore represent a valuable tool in clinical settings.
The “Best Talk Award” winner was Catharina Zich from the Neuropsychology Lab at the University of Oldenburg, Germany. She gave an inspirational talk with the title “Real-time artifact correction enables EEG-based feedback inside the fMRI scanner”. Combining the advantages of the electroencephalogram (EEG) – high temporal resolution – with the advantages of functional magnetic resonance imaging (fMRI) – high spatial resolution – would allow gaining new insights of brain functioning. The impact of MRI scanner artifacts on EEG signal quality presents a challenge. Catharina and colleagues combined both technologies with the aim to investigate brain patterns involved in EEG-BCI operation in more details. The authors showed that hand motor imagery induces event-related desynchronization (ERD) over contralateral areas during continuous MRI scanning and that the ERD inside and outside the MRI scanner strongly correlate. This was achieved by reduction of MRI related artifacts such as gradient artifacts and ballistocardiogram artifacts. Offline and online artifact corrections methods were presented and evaluated in N=22 able bodied individuals. This is a proof-of-principle that meaningful EEG-based BCI feedback is possible in a MRI scanner.
The proceedings of the conference are freely available and can be downloaded at http://bci.tugraz.at.
©Brain Products GmbH 2014