Brain–Computer Interfaces - Index of

Detecting Mental States by Machine Learning
Techniques: The Berlin Brain–Computer
Interface
Benjamin Blankertz, Michael Tangermann, Carmen Vidaurre,
Thorsten Dickhaus, Claudia Sannelli, Florin Popescu, Siamac Fazli,
Márton Danóczy, Gabriel Curio, and Klaus-Robert Müller
1 Introduction
The Berlin Brain-Computer Interface (BBCI) uses a machine learning approach
to extract user-specific patterns from high-dimensional EEG-features optimized for
revealing the user’s mental state. Classical BCI applications are brain actuated tools
for patients such as prostheses (see Section 4.1) or mental text entry systems ([1]
and see [2–5] for an overview on BCI). In these applications, the BBCI uses natural
motor skills of the users and specifically tailored pattern recognition algorithms for
detecting the user’s intent. But beyond rehabilitation, there is a wide range of possible
applications in which BCI technology is used to monitor other mental states,
often even covert ones (see also [6] in the fMRI realm). While this field is still
largely unexplored, two examples from our studies are exemplified in Sections 4.3
and 4.4.
1.1 The Machine Learning Approach
The advent of machine learning (ML) in the field of BCI has led to significant
advances in real-time EEG analysis. While early EEG-BCI efforts required neurofeedback
training on the part of the user that lasted on the order of days, in ML-based
systems it suffices to collect examples of EEG signals in a so-called calibration during
which the user is cued to perform repeatedly any one of a small set of mental
tasks. This data is used to adapt the system to the specific brain signals of each
user (machine training). This step of adaption seems to be instrumental for effective
BCI performance due to a large inter-subject variability with respect to the brain
signals [7]. After this preparation step, which is very short compared to the subject
B. Blankertz (B)
Machine Learning Laboratory, Berlin Institute of Technology, Berlin, Germany; Fraunhofer
FIRST (IDA), Berlin,Germany
e-mail: blanker@cs.tu-berlin.de
B. Graimann et al. (eds.), Brain–Computer Interfaces, The Frontiers Collection,
DOI 10.1007/978-3-642-02091-9_7, C○ Springer-Verlag Berlin Heidelberg 2010
113