CMMM 2013 - Machine Learning for Feature Extraction and Classification in EEG-Based Brain-Computer Interfaces

Over the years, especially during the last decade, the research of brain-computer interface (BCI) technology has attracted increasing interdisciplinary interest from diverse fields such as neuroscience, biomedical engineering, and machine learning. A BCI provides a platform through which a biological brain and a computer can communicate with each other. It can serve as a communication and control channel for people with severe motor disabilities. There are many other potential applications, such as alarming paroxysmal episodes for patients with neuropathic disorders (e.g., predicting epileptic seizures), manipulating delicate equipment in dangerous or inconvenient environments, augmenting human cognitive and behavioral performance, as a new approach for entertainment (e.g., brain-controlled game machines), or for biometric authentication systems.
Electroencephalogram (EEG) signals are measurements of the electrical activity of the populations of neurons in the brain cortex, using electrodes mounted on the scalp. Currently EEG is the most popular way to measure brain activity in current BCI research due to its low cost and relative portability. A BCI adopting EEG signals as the information carrier is usually referred to as an EEG-based BCI. A general EEG-based BCI consists of four basic components, which are EEG-signal acquisition, feature extraction, pattern classification, and device control. EEG signals are characterized by high temporal resolution, relatively poor spatial resolution, and various types of noise and artifacts. As such, Digital Signal Processing and Machine Learning techniques have played an important role in the successful operation of BCIs, especially in the process of feature extraction and pattern classification?ttwo core components of a BCI.
The main focus of this special issue will be on recent developments in feature extraction and classification methods for EEG-based BCIs, with the aim of promoting the latest research in BCIs. New Machine Learning methods applicable to BCIs or successful combinations of existing Machine Learning techniques with BCIs are both highly preferred. Potential topics include, but are not limited to:
Feature extraction
Feature selection
Support vector machines
Gaussian processes
Kernel methods
Ensemble learning
Sequential learning models
Semisupervised learning
Active learning
Transfer learning
Before submission authors should carefully read over the journal’s Author Guidelines, which are located at http://www.hindawi.com/journals/cmmm/guidelines/. Prospective authors should submit an electronic copy of their complete manuscript through the journal Manuscript Tracking System at http://mts.hindawi.com/submit/journals/cmmm/bci/ according to the following timetable:
Manuscript Due Friday, 19 July 2013
First Round of Reviews Friday, 11 October 2013
Publication Date Friday, 6 December 2013
Lead Guest Editor
Shiliang Sun, Department of Computer Science and Technology, East China Normal University, Shanghai, China
Guest Editors
Yijun Wang, Swartz Center for Computational Neuroscience, University of California San Diego, San Diego, CA, USA
Tom Diethe, Machine Learning and Perception Group, Microsoft Research Cambridge, Cambridge, UK