Please use this identifier to cite or link to this item:
https://hdl.handle.net/2440/81606
Citations | ||
Scopus | Web of Science® | Altmetric |
---|---|---|
?
|
?
|
Type: | Conference paper |
Title: | Investigation of the trade-off between time window length, classifier update rate and classification accuracy for restorative brain-computer interfaces |
Author: | Darvishi, S. Ridding, M. Abbott, D. Baumert, M. |
Citation: | Proceedings of the 35th Annual International Conference of the IEEE EMBS, 2013: pp.1567-1570 |
Publisher: | IEEE |
Publisher Place: | USA |
Issue Date: | 2013 |
Series/Report no.: | IEEE Engineering in Medicine and Biology Society Conference Proceedings |
ISBN: | 9781457702167 |
ISSN: | 1557-170X 2694-0604 |
Conference Name: | Annual International Conference of the IEEE Engineering in Medicine and Biology Society (35th : 2013 : Osaka, Japan) |
Statement of Responsibility: | Sam Darvishi, Michael C. Ridding, Derek Abbott, Mathias Baumert |
Abstract: | Recently, the application of restorative brain-computer interfaces (BCIs) has received significant interest in many BCI labs. However, there are a number of challenges, that need to be tackled to achieve efficient performance of such systems. For instance, any restorative BCI needs an optimum trade-off between time window length, classification accuracy and classifier update rate. In this study, we have investigated possible solutions to these problems by using a dataset provided by the University of Graz, Austria. We have used a continuous wavelet transform and the Student t-test for feature extraction and a support vector machine (SVM) for classification. We find that improved results, for restorative BCIs for rehabilitation, may be achieved by using a 750 milliseconds time window with an average classification accuracy of 67% that updates every 32 milliseconds. |
Rights: | ©2013 IEEE |
DOI: | 10.1109/EMBC.2013.6609813 |
Description (link): | http://embc2013.embs.org/ |
Published version: | http://dx.doi.org/10.1109/embc.2013.6609813 |
Appears in Collections: | Aurora harvest Electrical and Electronic Engineering publications |
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.