Work

Reducing Calibration Time for the P300 Brain-Computer Interface Speller

Year: 2010
Type: conference-paper
Abstract: With a Brain-Computer Interface (BCI), it is nowadays possible to achieve a direct pathway between the brain and computers thanks to the analysis of some particular brain activities. The detection of even-related potentials, like the P300, provides a way to create BCIs. The generation of the P300 wave is achieved with the oddball paradigm, which allows detecting targets selected by the user on a screen. The P300-Speller is based on this principle. The detection of the P300 requires efficient signal processing and machine learning techniques. Thus, a calibration step is needed for training the models. However, the duration of this calibration can be a drawback. We propose to evaluate the optimal number of characters that should be spelt in order to provide a working system with a minimum calibration duration. The evaluation has been tested on data recorded on 20 healthy subjects. It is possible to spell only seven symbols during the calibration to reach an initialized system with an average accuracy of at least 80%.
Source: HAL
Institution: GIPSA-Lab
Language: French
Open Access status: green
OpenAlex ID: W2610760675
FWCI: 1.009
Cites: 19
Cited by: 3
Related to: 10