Eye interference reduction in electroencephalogram recordings using a radial basic function

Abstract

The electroencephalogram (EEG) signal is the manifestation of brain activity recorded as changes in electrical potentials at multiple locations over the scalp and it can be distorted by many other sources of electrical activity, called eye artefacts. It is important to remove these artefact signals before analysing the EEG signal, to obtain accurate information about brain activity and avoid mistakes in its interpretation. To deal with this problem, the present study proposes an artificial neural network, as a filter to remove ocular artefacts. In the proposed method, the number of radial basis function (RBF) neurons and input output space clustering are adaptively determined. Furthermore, the structure of the system and the parameters of the corresponding RBF units are trained automatically and relatively fast adaptation is attained. By the least-square error estimator techniques, the proposed system is suitable for real EEG applications. The proposed system improves results yielded by conventional techniques of ocular reduction, such as principal component analysis, support vector machines and independent component analysis systems. As a consequence, the algorithm could serve as an effective framework to reduce substantially eye interference in EEG recordings.