Ulster University Logo

Ulster Institutional Repository

Creating a nonparametric brain-computer interface with neural time-series prediction preprocessing

Biomedical Sciences Research Institute Computer Science Research Institute Environmental Sciences Research Institute Nanotechnology & Advanced Materials Research Institute

Coyle, DH, Prasad, G and McGinnity, TM (2006) Creating a nonparametric brain-computer interface with neural time-series prediction preprocessing. In: the 28th International IEEE Engineering in Medicine and Biology Conference, New York, USA. UNSPECIFIED. 4 pp. [Conference contribution]

Full text not available from this repository.

URL: http://ieeexplore.ieee.org/Xplore/login.jsp?url=http%3A%2F%2Fieeexplore.ieee.org%2Fiel5%2F4028925%2F4461641%2F04462222.pdf%3Farnumber%3D4462222&authDecision=-203

Abstract

The issue of subject-specific parameter selection in an electroencephalogram (EEG)-based brain-computer interface (BCI) is tackled in this paper. Hjorth- and Barlow-based feature extraction procedures (FEPs) are investigated along with linear discriminant analysis (LDA) for classification. These are well-known nonparametric FEPs but their simplicity prevents them from matching the performance of more complex FEPs. Neural time-series prediction preprocessing (NTSPP), which has been shown to enhance the separability of both time- and frequency-based features, is used to improve the applicability of these FEPs. NTSPP uses a number of prediction modules (PMs) to perform m-step ahead prediction of EEG time-series recorded whilst subjects perform left/right motor imagery-based mental tasks. Depending on the PMs, the NTSPP framework normally requires subject-specific parameters to be predefined. In this work each PM is a self-organizing fuzzy neural network (SOFNN). The SOFNN has a self-organizing structure and augmented nonlinear approximation capabilities however, a number of parameters must be defined prior to training. This is problematic therefore the practicality of a general set of parameters, previously selected via a sensitivity analysis (SA), is analyzed. The results indicate that a general set of NTSPP parameters may provide the best results and therefore a fully nonparametric BCI may be realizable.

Item Type:Conference contribution (Paper)
Faculties and Schools:Faculty of Computing & Engineering
Faculty of Computing & Engineering > School of Computing and Intelligent Systems
Research Institutes and Groups:Computer Science Research Institute
Computer Science Research Institute > Intelligent Systems Research Centre
ID Code:8794
Deposited By:Dr Damien Coyle
Deposited On:01 Feb 2010 12:01
Last Modified:01 Feb 2010 12:01

Repository Staff Only: item control page