Motor imagery classification in Brain computer interface (BCI) based on EEG signal by using machine learning technique

N E Md Isa,M S Razalli,M Z Ilyas,A Amir

doi:10.11591/eei.v8i1.1402

Abstract

This paper focuses on classification of motor imagery in Brain Computer Interface (BCI) by using classifiers from machine learning technique. The BCI system consists of two main steps which are feature extraction and classification. The Fast Fourier Transform (FFT) features is extracted from the electroencephalography (EEG) signals to transform the signals into frequency domain. Due to the high dimensionality of data resulting from the feature extraction stage, the Linear Discriminant Analysis (LDA) is used to minimize the number of dimension by finding the feature subspace that optimizes class separability. Five classifiers: Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Naïve Bayes, Decision Tree and Logistic Regression are used in the study. The performance was tested by using Dataset 1 from BCI Competition IV which consists of imaginary hand and foot movement EEG data. As a result, SVM, Logistic Regression and Naïve Bayes classifier achieved the highest accuracy with 89.09% in AUC measurement.

Highlights

There are many techniques to get the signals from the brain, such as electroencephalogram (EEG), Magnetic resonance imaging (MRI) and computer tomography (CT)
We study five machine learning algorithm: Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Naïve Bayes, Decision Tree and Logistic Regression
In AUC measurement, SVM, Logistic Regression, Naïve Bayes give the highest accuracy with 72.20% for subject b and 89.09% for subject g

Summary

Introduction

There are many techniques to get the signals from the brain, such as electroencephalogram (EEG), Magnetic resonance imaging (MRI) and computer tomography (CT). EEG is one of the most commonly used in BCI to records electrical activity and brain waves by placing the electrode on the scalp. EEG capable in capturing brain information processing quickly with high temporal resolution, but it has low spatial resolution and high noise level which make it challenging to extract useful information from EEG signals for BCI application [5]. Motor imagery is a popular paradigm in EEG-based BCI system [6]. Machine learning technique is commonly used in this classification process as it has the ability to model high-dimensional datasets [8]. Machine learning is the technique which can be briefly defined as enabling computers make successful predictions using past experiences [9]

Objectives

Methods

Results

Conclusion