Discrimination of task-related eeg signals using diffusion adaptation and S-transform coherency

Author

Eftaxias, Konstantinos ; Sanei, Saeid

Author_Institution

Fac. of Eng. & Phys. Sci., Univ. of Surrey, Guildford, UK

fYear

2014

fDate

21-24 Sept. 2014

Firstpage

1

Lastpage

6

Abstract

This work presents a novel approach for discriminating complex mental and motor tasks using diffusion adaptation and brain connectivity measures. In particular, in this paper, we use a S-transform based measure to estimate the connectivity on single-trial basis and diffusion Kalman filtering to train a model that can classify different tasks. The superiority of the method is proven when compared with solutions that don´t rely on cooperation.

Keywords

Kalman filters; electroencephalography; medical signal processing; transforms; S-transform based measure; S-transform coherency; brain connectivity measures; complex mental tasks; connectivity estimation; diffusion Kalman filtering; diffusion adaptation; motor tasks; single-trial basis; task-related EEG signals; Adaptation models; Atmospheric measurements; Brain modeling; Electroencephalography; Frequency measurement; Kalman filters; Vectors; Diffusion adaptation; S-transform; brain connectivity; diffusion Kalman filtering;

fLanguage

English

Publisher

ieee

Conference_Titel

Machine Learning for Signal Processing (MLSP), 2014 IEEE International Workshop on

Conference_Location

Reims

Type

conf

DOI

10.1109/MLSP.2014.6958868

Filename

6958868