Classifications of motor imagery tasks in brain computer interface using Euclidean distance

Abstract

In this paper we propose and discuss a new classification method of motor imagery tasks based on patterns and Euclidean distance. The proposed method is simple, fast, but considerably sensitive with respect to the selected features/frequencies for classification. Choosing a predefined number of features leads to results similar to GTEC/BCI2000 while an optimal selection gives improved results but still requires additional investigation.