EEG Feature Selection for BCI Based on Motor Imaginary Task

AbstractThe greatest problem met when a Brain Computer Interface (BCI) based on electroencephalographic (EEG) signals is to be created is a huge dimensionality of EEG feature space and at the same time very limited number of possible observations. The first is a result of a huge amount of data which can be recorded during the single trial, the latter - the result of individuality of EEG signals, which can significantly differ in different frequency bands determined for different subjects. These two reasons force the brain researches to reduce the huge EEG feature space to only some features, those which allow to build a BCI of a satisfactory accuracy. The paper presents the comparison of two methods of feature selection - blind source separation (BSS) method and method using interpretable features. The comparison was carried out with the data set recorded during EEG session with a subject whose task was to imagine movements of right and left hand.

Download Full-text

Genetic Algorithms for Feature Selection for Brain–Computer Interface

International Journal of Pattern Recognition and Artificial Intelligence ◽

10.1142/s0218001415590089 ◽

2015 ◽

Vol 29 (05) ◽

pp. 1559008 ◽

Cited By ~ 22

Author(s):

Izabela Rejer

Keyword(s):

Genetic Algorithms ◽

Feature Selection ◽

Brain Computer Interface ◽

Computer Interface ◽

Eeg Signals ◽

Crucial Problem ◽

Left And Right ◽

Selection For ◽

First Time

The crucial problem that has to be solved when designing an effective brain–computer interface (BCI) is: how to reduce the huge space of features extracted from raw electroencephalography (EEG) signals. One of the strategies for feature selection that is often applied by BCI researchers is based on genetic algorithms (GAs). The two types of GAs that are most commonly used in BCI research are the classic algorithm and the Culling algorithm. This paper presents both algorithms and their application for selecting features crucial for the correct classification of EEG signals recorded during imagery movements of the left and right hand. The results returned by both algorithms are compared to those returned by an algorithm with aggressive mutation and an algorithm with melting individuals, both of which have been proposed by the author of this paper. While the aggressive mutation algorithm has been published previously, the melting individuals algorithm is presented here for the first time.

Download Full-text

EEG Classification for BCI Based on CSP and SVM-GA

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.459.228 ◽

2013 ◽

Vol 459 ◽

pp. 228-231 ◽

Cited By ~ 4

Author(s):

Hao Yang ◽

Song Wu

Keyword(s):

Classification Performance ◽

Computer Interface ◽

Support Vector ◽

Eeg Signals ◽

Common Spatial Pattern ◽

Data Set ◽

Linear Discriminant ◽

Brain Signals ◽

Electroencephalogram Eeg ◽

The Brain

Electroencephalogram (EEG) is generally used in Brain-Computer Interface (BCI) applications to measure the brain signals. However, the multichannel EEG signals characterized by unrelated and redundant features will deteriorate the classification accuracy. This paper presents a method based on common spatial pattern (CSP) for feature extraction and support vector machine with genetic algorithm (SVM-GA) as a classifier, the GA is used to optimize the kernel parameters setting. The proposed algorithm is performed on data set Iva of BCI Competition III. Results show that the proposed method outperforms the conventional linear discriminant analysis (LDA) in average classification performance.

Download Full-text

NON-INVASIVE BCI METHOD: EEG - ELECTROENCEPHALOGRAPHY

International Conference on Technics Technologies and Education ◽

10.15547/ictte.2019.02.095 ◽

2019 ◽

pp. 139-145

Author(s):

Selma Büyükgöze

Keyword(s):

Brain Computer Interface ◽

Computer Interface ◽

Electrode Placement ◽

Eeg Signals ◽

Non Invasive ◽

Brain Signals ◽

Brain States ◽

The Brain

Brain Computer Interface consists of hardware and software that convert brain signals into action. It changes the nerves, muscles, and movements they produce with electro-physiological signs. The BCI cannot read the brain and decipher the thought in general. The BCI can only identify and classify specific patterns of activity in ongoing brain signals associated with specific tasks or events. EEG is the most commonly used non-invasive BCI method as it can be obtained easily compared to other methods. In this study; It will be given how EEG signals are obtained from the scalp, with which waves these frequencies are named and in which brain states these waves occur. 10-20 electrode placement plan for EEG to be placed on the scalp will be shown.

Download Full-text

Feature selection for classification in Steady state visually evoked potentials (SSVEP)-based brain-computer interfaces with genetic algorithm

Bio-Algorithms and Med-Systems ◽

10.1515/bams-2020-0013 ◽

2020 ◽

Vol 16 (2) ◽

Author(s):

Stanisław Karkosz ◽

Marcin Jukiewicz

Keyword(s):

Genetic Algorithm ◽

Feature Selection ◽

Steady State ◽

Brain Computer Interface ◽

Morphological Features ◽

Computer Interface ◽

Visually Evoked Potentials ◽

Computer Interfaces ◽

Subject Variability ◽

Selection For

AbstractObjectivesOptimization of Brain-Computer Interface by detecting the minimal number of morphological features of signal that maximize accuracy.MethodsSystem of signal processing and morphological features extractor was designed, then the genetic algorithm was used to select such characteristics that maximize the accuracy of the signal’s frequency recognition in offline Brain-Computer Interface (BCI).ResultsThe designed system provides higher accuracy results than a previously developed system that uses the same preprocessing methods, however, different results were achieved for various subjects.ConclusionsIt is possible to enhance the previously developed BCI by combining it with morphological features extraction, however, it’s performance is dependent on subject variability.

Download Full-text

Multi-way feature selection for ECoG-based Brain-Computer Interface

Expert Systems with Applications ◽

10.1016/j.eswa.2018.06.054 ◽

2018 ◽

Vol 114 ◽

pp. 402-413 ◽

Cited By ~ 1

Author(s):

Anastasia Motrenko ◽

Vadim Strijov

Keyword(s):

Feature Selection ◽

Brain Computer Interface ◽

Computer Interface ◽

Selection For

Download Full-text

Genetic-based feature selection for efficient motion imaging of a brain–computer interface framework

Journal of Neural Engineering ◽

10.1088/1741-2552/aad567 ◽

2018 ◽

Vol 15 (5) ◽

pp. 056020 ◽

Cited By ~ 10

Author(s):

Hongli Chang ◽

Jimin Yang

Keyword(s):

Feature Selection ◽

Brain Computer Interface ◽

Computer Interface ◽

Selection For ◽

Motion Imaging

Download Full-text

Classification Algorithms for EEG-Based Brain-Computer Interface

Advances in Medical Technologies and Clinical Practice - Advanced Classification Techniques for Healthcare Analysis ◽

10.4018/978-1-5225-7796-6.ch003 ◽

2019 ◽

pp. 52-73

Author(s):

Sravanth Kumar Ramakuri ◽

Chinmay Chakraboirty ◽

Anudeep Peddi ◽

Bharat Gupta

Keyword(s):

Brain Activity ◽

Biomedical Signal Processing ◽

Computer Interface ◽

Eeg Signals ◽

Large Sets ◽

Eeg Data ◽

Important Measurement ◽

The Brain

In recent years, a vast research is concentrated towards the development of electroencephalography (EEG)-based human-computer interface in order to enhance the quality of life for medical as well as nonmedical applications. The EEG is an important measurement of brain activity and has great potential in helping in the diagnosis and treatment of mental and brain neuro-degenerative diseases and abnormalities. In this chapter, the authors discuss the classification of EEG signals as a key issue in biomedical research for identification and evaluation of the brain activity. Identification of various types of EEG signals is a complicated problem, requiring the analysis of large sets of EEG data. Representative features from a large dataset play an important role in classifying EEG signals in the field of biomedical signal processing. So, to reduce the above problem, this research uses three methods to classify through feature extraction and classification schemes.

Download Full-text