Filtering Techniques for Channel Selection in Motor Imagery EEG Applications: A Survey

Baig, Muhammad, Aslam, Nauman and Shum, Hubert P. H. (2020) Filtering Techniques for Channel Selection in Motor Imagery EEG Applications: A Survey. Artificial intelligence review, 53 (2). pp. 1207-1232. ISSN 0269-2821

Preview	Text (Final published version) Baig2020_Article_FilteringTechniquesForChannelS.pdf Available under License Creative Commons Attribution 4.0. Download (599kB) | Preview
Preview	Text (Early online version) Baig2019_Article_FilteringTechniquesForChannelS.pdf - Published Version Available under License Creative Commons Attribution 4.0. Download (609kB) | Preview
Preview	Text AIR2019.pdf - Accepted Version Download (1MB) | Preview

Abstract

Brain Computer Interface (BCI) Systems are used in a wide range of applications such as communication, neuro-prosthetic and environmental control for disabled persons using robots and manipulators. A typical BCI system uses different types of inputs; however, Electroencephalography (EEG) signals are most widely used due to their non-invasive EEG electrodes, portability, and cost efficiency. The signals generated by the brain while performing or imagining a motor related task (Motor Imagery (MI)) signals are one of the important inputs for BCI applications. EEG data is usually recorded from more than 100 locations across the brain, so efficient channel selection algorithms are of great importance to identify optimal channels related to a particular application. The main purpose of applying channel selection is to reduce computational complexity while analysing EEG signals, improve classication accuracy by reducing over-tting, and decrease setup time. Different channel selection evaluation algorithms such as ltering, wrapper, and hybrid methods have been used for extracting optimal channel subsets by using predened criteria. After extensively reviewing the literature in the eld of EEG channel selection, we can conclude that channel selection algorithms provide a possibility to work with fewer channels without aecting the classication accuracy. In some cases, channel selection increases the system performance by removing the noisy channels. The research in the literature shows that the same performance can be achieved using a smaller channel set, with 10-30 channels in most cases. In this paper, we present a survey of recent development in ltering channel selection techniques along with their feature extraction and classication methods for MI-based EEG applications.

Actions (login required)

View Item

Downloads