Detecting and comparing the onset of self-paced and cue-based finger movements from EEG signals
2015 (English)In: 7th Annual International IEEE Conference on Computer Science and Electronic Engineering (CEEC), Colchester, UK: IEEE conference proceedings, 2015, Vol. 7, 157-160 p.Conference paper (Refereed)
We asked four subjects to perform the task of pressing a taster button with their thumbs, while their EEG recordings were obtained, in order to determine the probability of the subjects' intention to make the movement in comparison to the idle state. Humans usually spontaneously decide when to initiate movements to complete daily-life tasks, but sometimes our movements can also be externally triggered. Thus, the subjects first performed motor tasks at the instants defined by the animation shown on the screen and second, the subjects performed self-initiated movements. In this paper, we study if there is a difference in the classification results and coherence measures of EEG signals in these two paradigms. We used the Support Vector Machine (SVM) classifier on features extracted by applying Burg's algorithm to EEG signals, which arose as a solution with high accuracy.
Place, publisher, year, edition, pages
Colchester, UK: IEEE conference proceedings, 2015. Vol. 7, 157-160 p.
Brain-Computer Interface, EEG, Support Vector Machines, Neural Signal Processing
Medical and Health Sciences Natural Sciences Engineering and Technology
IdentifiersURN: urn:nbn:se:kth:diva-182093DOI: 10.1109/CEEC.2015.7332717ISI: 000377258900029ScopusID: 2-s2.0-84963641796OAI: oai:DiVA.org:kth-182093DiVA: diva2:903158
7th Annual International IEEE Conference on Computer Science and Electronic Engineering (CEEC)
QC 201603072016-02-142016-02-142016-07-06Bibliographically approved