Smario: a toolbox for brain-computer interfacing analysis and design

In this paper we introduce Smario, a MATLAB open source toolbox for the analysis of BCI signals and implementation of translation algorithms for BCI systems. The Smario functions have been created based on the design of EEGLAB, they are accessible through the graphic user interface but they can also be run and edited using MATLAB syntax. Smario reads BCI2000 files in DAT and MAT formats, and documentation is available to implement functions to read other formats. This toolbox incorporates two feature selection methods based on discriminative measures; r² and SEPCOR. Using the graphical capabilities of EEGLAB, these feature selection modules provide 2-D and 3-D topographic maps of scalp data fields for selected features and graphs of the extracted features for selected channels. Smario´s modularized design allows the user to create a translation algorithm using existing feature extraction, selection and classification modules. These modules are easily configurable and interchangeable, to provide the user with means to compare different translation algorithms. Furthermore, the implementation of new modules is possible and guidelines for this purpose have been included in the documentation. Available modules are Common Average Reference (CAR), MEM filter, Autoregressive (AR) analysis, Fast Fourier Transform (FFT), Discrete Wavelet Transform (DWT), Bayesian Classifier, Linear Discriminant Analysis (LDA), Support Vector Machine (SVM), SEPCOR and r². Smario is available for download at http://www.hst.aau.dk/bci/Smariovf/Smindex.html where documentation and sample files can also be found.