Optimized code-books, for enhancing the performance of cVEP Brain Computer Interfaces
Keywords
Loading...
Authors
Issue Date
2022-02-15
Language
en
Document type
Journal Title
Journal ISSN
Volume Title
Publisher
Title
ISSN
Volume
Issue
Startpage
Endpage
DOI
Abstract
Visually evoked potentials (VEPs) can be measured in the EEG as response
to a visual stimulus. Commonly, VEPs are displayed by averaging
multiple responses to a certain stimulus or a classifier is trained to identify
the response to a certain stimulus.
Brain-computer interfaces (BCIs) based on code-modulated visual evoked
potentials (cVEPs) provide high information transfer rates, which make
them a promising alternative communication tool.
One of the important components of a BCI is the set of codes that are
flashed as a stimulus, eliciting specific responses. Codes that give rise to
more differentiable responses, can increase the performance of a BCI significantly.
Creating a codebook is not trivial as various chosen mathematical
and statistical properties do not necessarily transform to the responses.
One way to create a custom codebook is to grow it, taking into account
the responses it produces at each step. This guarantees the preservation
of chosen differentiating properties, when the codes elicit responses during
their use. In this paper, this method of codebook designing was investigated.
In addition to this, the influence of the use of prior probabilities (by incorporation)
to further guide this generative process, was investigated.
An objective metric to simulate BCI trials was constructed and utilized
to compared the quality of codebooks generated using the aforementioned
procedure, against the codebooks that are common-practice for use in BCIs.
Description
Citation
Supervisor
Faculty
Faculteit der Sociale Wetenschappen