Ahmadi, M, Farrokhi Nia, A, Michalka, SW, Sumich, AL ORCID: https://orcid.org/0000-0003-4333-8442, Wuensche, B and Billinghurst, M, 2023. Comparing performance of dry and gel EEG electrodes in VR using MI paradigms. In: VRST 2023: Proceedings of the 29th ACM Symposium on Virtual Reality Software and Technology. New York: ACM. ISBN 9798400703287
Preview |
Text
1820582_Sumich.pdf - Published version Download (574kB) | Preview |
Abstract
Brain–computer interfaces (BCIs) are an emerging technology with numerous applications. Electroencephalogram (EEG) motor imagery (MI) is among the most common BCI paradigms and has been used extensively in healthcare applications such as post-stroke rehabilitation. Using a Virtual Reality (VR) game, Push Me, we con-ducted a pilot study to compare MI accuracy with Gel or active-dry EEG electrodes. The motivation was to (1) investigate the MI paradigm in a VR environment and (2) compare MI accuracy using active dry and gel electrodes with different Machine Learning (ML)classifications (SVM, KNN and RF). The results indicate that while gel-based electrodes, in combination with SVM, achieved the high-est accuracy, dry electrode EEG caps achieved similar outcomes, especially with SVM and KNN models.
Item Type: | Chapter in book |
---|---|
Description: | Paper presented at VRST 2023: 29th ACM Symposium on Virtual Reality Software and Technology, Christchurch, New Zealand, 9-11 October 2023. Article no. 46 |
Creators: | Ahmadi, M., Farrokhi Nia, A., Michalka, S.W., Sumich, A.L., Wuensche, B. and Billinghurst, M. |
Publisher: | ACM |
Place of Publication: | New York |
Date: | 9 October 2023 |
ISBN: | 9798400703287 |
Identifiers: | Number Type 10.1145/3611659.3617436 DOI 1820582 Other |
Divisions: | Schools > School of Social Sciences |
Record created by: | Jonathan Gallacher |
Date Added: | 23 Oct 2023 09:43 |
Last Modified: | 23 Oct 2023 09:47 |
URI: | https://irep.ntu.ac.uk/id/eprint/50072 |
Actions (login required)
Edit View |
Statistics
Views
Views per month over past year
Downloads
Downloads per month over past year