Stimulus tagging using aperiodic visual stimulation in a VEP-based BCI

From AIRWiki
Jump to: navigation, search
Stimulus tagging using aperiodic visual stimulation in a VEP-based BCI
Coordinator: MatteoMatteucci (matteo.matteucci@polimi.it)
Tutor: MatteoMatteucci (matteo.matteucci@polimi.it)
Collaborator:
Students: DavideCastellone (kastel88@email.it), GiuseppeBroccio (giuse.bro@email.it)
Research Area: BioSignal Analysis
Research Topic: Brain-Computer Interface
Start: 2010/09/01
End: 2010/07/30
Status: Closed
Level: Bs
Type: Thesis


Part 1: project profile

Project name

Project short description

Visual-evoked potentials (VEPs) are a possible way to drive the a Brain-Computer Interface (BCI). This projects aims at maximizing the discrimination between different stimuli by using numerical codes derived from techniques of digital telecommunications. This subset of VEP-BCI is usually called code-based VEP, or c-VEP.

Dates

Start date: 2009/10/01

People involved

Project head(s)

prof. Matteo Matteucci

Students currently working on the project

Giuseppe Broccio

Davide Castellone

Laboratory work and risk analysis

Several EEG sessions will be recorded. Data acquisition is performed at the IIT-Lab. All risks are those related to the use of video terminals and EEG devices.

WARNING Some people may experience epileptic seizures when viewing flashing lights or patterns. Even people who have never had any seizures may nonetheless have an undetected epileptic condition.

  • 2010-04-12: First data acquisition session
  • 2010-04-16: Second and third session
  • 2010-05-26: Fourth and fifth session
  • 2010-06-07: Sixth and seventh session
  • 2010-06-11: Eigth session (coloured flashers)

The protocol briefly explained

EEG is recorded at points Fz, Cz, Pz, O1 and O2 while the patient fixates a particular symbol on the flashing pattern (30 to 90 seconds). This is repeated several times with different symbols, frame rates and sequence length. Any recording can be used as a training for the classifier. Classification is done by comparing two different recordings (provided that they were recorded with the same flashing pattern and frame rate).

Part 2: project description

What this project means to demonstrate is the possibility to extend the SSVEP paradigm to multiple visual signals (i.e. blinking lights) using a numeric encoding, in a way similar to what has already been attempted with noise-tagging.

Libraries used

Roadmap

  1. Video output
    1. Specification - done
    2. Design - done
    3. Implementation - done
    4. Testing - done
    5. Deployment - done
  2. c-VEP
    1. Specification/Design/Implementation - done
    2. Data Acquisition - in progress
    3. Data Analysis - in progress
    4. Conclusions


Bibliography

  • J.R. Wolpaw et al., Brain-computer interfaces for communication and control
  • Erich E. Sutter., The brain response interface: communication through visually-induced electrical brain responses
  • J. Farquhar, J. Blankespoor, R. Velk, P. Desain, Towards a noise tagging auditory BCI-Paradigm, in 4th International BCI Workshop and Training Course 2008