Funding: (IM)2 Swiss National Center of Competence in Research
Duration: July 2002 -- June 2004
Partners: CSEM (CH), University of Geneva, Functional Brain Mapping Lab (CH), University of Geneva, Computer Vision on Multimedia Lab (CH)
Contact Persons (IDIAP): Samy Bengio, Jose Millan

The main problem we would like to tackle in this white paper is to investigate the possibility of classifying spontaneous brain activity based on either reconstructed brain activity maps, or directly from EEG recordings. In this way, the user should be able to think about a particular, previously learnt symbol, and the system should classify that thought according to a predefined alphabet of possible symbols. A symbol could be a particular geometric shape, a song, a mental image of a scenery, an action, a number, etc.

The project will focus on the design of the entire processing chain, from sensors to classification and interpretation, with an overall performance goal of enhancing the number of correct bits transmitted per second (bit rate).