Context-aware adaptive spelling in motor imagery BCI
- Submitting institution
-
The University of Essex
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 1317
- Type
- D - Journal article
- DOI
-
10.1088/1741-2560/13/3/036018
- Title of journal
- Journal of Neural Engineering
- Article number
- 036018
- First page
- 036018
- Volume
- 13
- Issue
- 3
- ISSN
- 1741-2552
- Open access status
- Deposit exception
- Month of publication
- May
- Year of publication
- 2016
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- Yes
- Number of additional authors
-
2
- Research group(s)
-
B - Brain Computer Interfaces and Neural Engineering (BCI-NE)
- Citation count
- 17
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- EEG signals change over time even for the same task. This is a significant barrier to BCI system performance. This foundational work, published in a high-impact journal and cited by world-leading BCI and machine-learning experts, demonstrated an unsupervised ML system in an adaptive BCI speller which could retrain itself in real-time using contextual assistance thus overcoming the significant problem of non-stationary brain signals. This system demonstrated performance close to that of a supervised system despite requiring absolutely no manual data labelling. The method's applicability was verified via closed-loop experiments with real participants and outperformed state-of-the-art algorithms.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -