Reading detection based on electroencephalogram processing

  • Authors:
  • Inês Oliveira;Ovidiu Grigore;Nuno Guimarães

  • Affiliations:
  • LASIGE, FCUL, University of Lisbon, Lisbon, Portugal;LASIGE, FCUL, University of Lisbon, Lisbon, Portugal;LASIGE, FCUL, University of Lisbon, Lisbon, Portugal

  • Venue:
  • ICCOMP'09 Proceedings of the WSEAES 13th international conference on Computers
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper describes a study regarding the detection of silent visual reading and non reading mental activities through electroencephalogram (EEG) processing. Our work is in the context of human computer interaction research field, and we pretend to integrate EEG signals in applications to assist and analyze reading tasks. The need of users to be constantly and tightly coupled with the applications is being highly stimulated by the design of universally-accessible interactive system, where the use of biomedical signals has become an emerging area. The work focuses on building reliable capture and preprocessing procedures, extracting relevant features and testing simple learning algorithms. The detection process uses left hemisphere EEG signals, which is referred to as being the relevant brain area for this type of tasks. The signals were processed to extract the power spectrum density of delta, theta and alpha rhythms, known frequencies of this type of signals. We also present two real time demonstration applications,