Abstract

Classification of electroencephalogram (EEG) signal is important in mental decoding for brain-computer interfaces (BCI). We introduced a feature extraction approach based on frequency domain analysis to improve the classification performance on different mental tasks using single-channel EEG. This biologically inspired method extracts the most discriminative spectral features from power spectral densities (PSDs) of the EEG signals. We applied our method on a dataset of six subjects who performed five different imagination tasks: (i) resting state, (ii) mental arithmetic, (iii) imagination of left hand movement, (iv) imagination of right hand movement, and (v) imagination of letter “A.” Pairwise and multiclass classifications were performed in single EEG channel using Linear Discriminant Analysis and Support Vector Machines. Our method produced results (mean classification accuracy of 83.06% for binary classification and 91.85% for multiclassification) that are on par with the state-of-the-art methods, using single-channel EEG with low computational cost. Among all task pairs, mental arithmetic versus letter imagination yielded the best result (mean classification accuracy of 90.29%), indicating that this task pair could be the most suitable pair for a binary class BCI. This study contributes to the development of single-channel BCI, as well as finding the best task pair for user defined applications.

Highlights

  • The idea of people being able to control their brain rhythm by performing specific mental tasks constitutes the main research focus on electroencephalogram (EEG) based mental control tasks, which gave birth to the brain-computer interface (BCI) [1]

  • We examined the plot of sensitivity versus 1-specificity, namely, receiver operating characteristics (ROC) curve and the area under the curve (AUC) [47] for evaluating the reliability of classification procedure

  • A new feature extraction method for EEG signals based on biologically inspired frequency domain

Read more

Summary

Introduction

The idea of people being able to control their brain rhythm by performing specific mental tasks constitutes the main research focus on electroencephalogram (EEG) based mental control tasks, which gave birth to the brain-computer interface (BCI) [1]. Multichannel EEG recording reduces the portability of daily use BCI and constitutes the main drawback for end users [19]. To address these problems, many methods have been proposed in the literature including electrode reduction algorithms and feature extraction methods based on a few electrodes [16, 19,20,21,22,23,24]. Many studies tested feature extraction methods and classification algorithms only on BCI competition datasets [16, 19, 21,22,23,24]. For a real BCI application, it is advantageous to conduct BCI studies including data recording, instead of only using BCI competition datasets [25, 26]

Objectives
Methods
Results
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call