A Plug&Play P300 BCI Using Information Geometry
This work addresses the challenge of improving calibration efficiency and generalization in P300-based brain-computer interfaces, which is incremental but impactful for users needing more accessible and adaptive BCI systems.
The paper tackles the problem of classifying Event Related Potentials (ERPs) in brain-computer interfaces by extending Riemannian geometry methods, previously limited to SMR-based BCIs, through a new covariance matrix estimation. The result is a method that increases performance, reduces calibration data needs, and generalizes well across sessions and subjects, as demonstrated on a P300-based game with an online adaptive implementation allowing high accuracy without calibration.
This paper presents a new classification methods for Event Related Potentials (ERP) based on an Information geometry framework. Through a new estimation of covariance matrices, this work extend the use of Riemannian geometry, which was previously limited to SMR-based BCI, to the problem of classification of ERPs. As compared to the state-of-the-art, this new method increases performance, reduces the number of data needed for the calibration and features good generalisation across sessions and subjects. This method is illustrated on data recorded with the P300-based game brain invaders. Finally, an online and adaptive implementation is described, where the BCI is initialized with generic parameters derived from a database and continuously adapt to the individual, allowing the user to play the game without any calibration while keeping a high accuracy.