Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Brain-computer interface for physically impaired people
Brain-Computer interface (BCI) is a promising field of research that can change life as we know it, staring from the healthcare, home devices and armed force to video gaming. BCI provides a new communication channel between human and computers using brain signals to perform certain control actions....
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Published: |
AUC Knowledge Fountain
2016
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| Summary: | Brain-Computer interface (BCI) is a promising field of research that can change life as we know it, staring from the healthcare, home devices and armed force to video gaming. BCI provides a new communication channel between human and computers using brain signals to perform certain control actions. BCI systems can help physically impaired patients to increase their degree of independence, and give hope to ‘Locked-In Syndrome’ (LIS) Patients and Amyotrophic Lateral Sclerosis ALS to communicate again. The aim of this thesis is to develop a Generic BCI system that uses Blink and Wink as control signals. This system avoids problems like long training periods, the risks of flashing lights and using many electrodes and sophistication of hybrid systems. Blinks and Winks are signals generated by the eye lid muscles and it is considered as an artifact in the brain signals. In this study we propose the use of this artifact signals to recognize human intention. The reason behind choosing blink and wink signals is because its features can be distinguishable from the normal brain activities which allow the system to easily detect it. The Blink and Wink based BCI system uses oddball paradigm technique to facilitate the use for more interactive commands. The system is tested with P300 speller and smart home control panel. For P300 Speller test the system achieved 87.5% accuracy and for the smart home control panel test the system achieved 92.5% accuracy. The test shows that the developed system is generic and efficient as no offline training for calibration was done by any of the subjects while the system achieved better accuracy compared to other BCI systems. |
|---|