Real-time detection of P300 brain events: brain-computer interfaces for EEG-based communication aids
| aut.embargo | No | en_NZ |
| aut.thirdpc.contains | No | en_NZ |
| aut.thirdpc.permission | No | en_NZ |
| aut.thirdpc.removed | No | en_NZ |
| dc.contributor.advisor | Kasabov, Nik (Nikola) | |
| dc.contributor.advisor | Schliebs, Stefan | |
| dc.contributor.advisor | Walker, Charles | |
| dc.contributor.author | Alkhater, Rehab | |
| dc.date.accessioned | 2012-11-12T01:20:41Z | |
| dc.date.available | 2012-11-12T01:20:41Z | |
| dc.date.copyright | 2012 | |
| dc.date.created | 2012 | |
| dc.date.issued | 2012 | |
| dc.date.updated | 2012-11-11T21:39:47Z | |
| dc.description.abstract | This thesis aims to design a real-time EEG-based communication aid using brain-computer interface (BCI) technologies. The study evaluates the feasibility of using the Emotive headset as an affordable EEG input system that is suitable for daily usage under realistic conditions. A further objective of this research is to increase the spelling speed of the P300 Speller. Multiple-screen verbal and graphical versions of the spelling paradigm are introduced to increase the number of letters that can be spelled in a particular time period. The experiments were conducted using the OpenViBE platform on six participants. The xDAWN spatial filter was used to detect the activated area of the brain while the LDA and the SVM were employed to classify the data into target and non-target samples. In terms of Emotiv feasibility, this system has evidenced its capability to detect the P300 brain waves used as the control signals for the P300 BCI. The obtained accuracies are comparable to those presented in other studies in which expensive medical EEG recording systems were utilized. The users’ performance with the verbal and graphical versions of the speller is similar to the performance obtained when using the typical alphanumerical speller, although with higher spelling speed. Accordingly, the use of these new versions is highly recommended. The results show significant differences between individual users’ performance. The shape of their brain activity pattern recorded within 500 ms of the visual stimulation, which is used as a control signal, as well as other factors were considered. For most participants involved in this study, the target signals are remarkably distinguishable from the non-target ones; however, a case of BCI illiteracy is identified. To summarise, the interface performance is affected positively by higher amplitude of P300 brain waves and users’ motivation; however, it is affected negatively by loss of attention, motor movements and mental fatigue. | en_NZ |
| dc.identifier.uri | https://hdl.handle.net/10292/4707 | |
| dc.language.iso | en | en_NZ |
| dc.publisher | Auckland University of Technology | |
| dc.rights.accessrights | OpenAccess | |
| dc.subject | Brain-computer interfaces | en_NZ |
| dc.subject | P300 | en_NZ |
| dc.subject | Real-time | en_NZ |
| dc.subject | EEG | en_NZ |
| dc.subject | Communication aids | en_NZ |
| dc.title | Real-time detection of P300 brain events: brain-computer interfaces for EEG-based communication aids | en_NZ |
| dc.type | Thesis | |
| thesis.degree.discipline | ||
| thesis.degree.grantor | Auckland University of Technology | |
| thesis.degree.level | Masters Theses | |
| thesis.degree.name | Master of Creative Technologies | en_NZ |