Authors:
Anett Seeland
1
;
Laura Manca
2
;
Frank Kirchner
3
and
Elsa Andrea Kirchner
3
Affiliations:
1
Robotics Innovation Center (RIC) and German Research Center for Artificial Intelligence (DFKI GmbH), Germany
;
2
University of Bremen, Germany
;
3
Robotics Innovation Center (RIC), German Research Center for Artificial Intelligence (DFKI GmbH) and University of Bremen, Germany
Keyword(s):
Movement Prediction, ERD/ERS, MRCP, Brain-computer Interface, BCI.
Related
Ontology
Subjects/Areas/Topics:
Animation and Simulation
;
Applications and Services
;
Artificial Intelligence
;
Biomedical Engineering
;
Biomedical Signal Processing
;
Computer Vision, Visualization and Computer Graphics
;
Data Manipulation
;
Health Engineering and Technology Applications
;
Human-Computer Interaction
;
Medical Image Detection, Acquisition, Analysis and Processing
;
Methodologies and Methods
;
Motion Control
;
Neurocomputing
;
Neurotechnology, Electronics and Informatics
;
Pattern Recognition
;
Physiological Computing Systems
;
Sensor Networks
;
Soft Computing
Abstract:
In brain-computer interfaces (BCIs) based on electroencephalography (EEG), two distinct types of EEG patterns
related to movement have been used for detecting the brain’s preparation for voluntary movements: a)
event-related patterns in the time domain named movement related cortical potentials (MRCPs) and b) patterns
in the frequency domain named event-related desynchronization/synchronization (ERD/ERS). The applicability
of those patterns in BCIs is often evaluated by the classification performance. To this end, the known
spatio-temporal differences in EEG activity can be of interest, since they might influence the classification
performance of the two different patterns. In this paper, we compared the classification performance based
on ERD/ERS and MRCP while varying the time point of prediction as well as the used electrode sites. Empirical
results were obtained from eight subjects performing voluntary right arm movements. Results show:
a) classification based on MRCP is superior
compared to ERD/ERS close to the movement onset whereas the
opposite results farther away from the movement onset, b) the performance maximum of MRCP is located at
central electrodes whereas it is at fronto-central electrodes for ERD/ERS. In summary, the results contribute to
a better insight into the spatial and temporal differences between ERD/ERS and MRCP in terms of prediction
performance.
(More)