Master Thesis MSTR-3393

BibliographyHassib, Mariam: Mental Task Classification Using Single-Electrode Brain Computer Interfaces.
University of Stuttgart, Faculty of Computer Science, Electrical Engineering, and Information Technology, Master Thesis No. 3393 (2012).
133 pages, english.
CR-SchemaH.5.2 (Information Interfaces and Presentation User Interfaces)
Abstract

In the recent years, the field of Human-Computer Interaction (HCI) has greatly evolved to involve new and exciting interaction paradigms that allow users to interact with their environment and with technology in a more intuitive and ergonomic way. These interaction paradigms include voice, touch, virtual reality, and more recently, the brain.

A brain-computer interface (BCI) is a an interface system allowing users to control devices without using the normal output pathways of peripherals, instead, by using neural activity generated in the brain. BCIs have a huge potential in a multitude of fields, all the way from providing users with severe motor disabilities with means for interaction with the external world, to entertainment, gaming, user state monitoring, and self-tracking systems. The potentials of BCI have sparked the interest of researchers, gaming markets and healthcare providers more and more in the recent years. The is due to the emergence of new commercial lightweight, low cost Electroencephalograph (EEG) equipment that made it possible to create more portable and usable BCI systems and expanded their fields of application.

This Master thesis aims to explore the state of the art commercial BCI as well as the uses and challenges related to them. Commercially available EEG equipment, namely the Neurosky Brainband and Neurosky Mindset, will be investigated. User tests will be carried out to investigate whether such equipment with low accuracy and low cost can be used to recognize various mental activities. This would be performed by first collecting a dataset of brain signals during performing a set of mental tasks, which is one of the contributions of this project, followed by applying a set of signal processing algorithms, then exploring various classification techniques to classify the collected signals

Full text and
other links
PDF (4877211 Bytes)
Department(s)University of Stuttgart, Institute for Natural Language Processing
Superviser(s)Alireza Sahami
Entry dateNovember 26, 2012
   Publ. Computer Science