On the temporal behavior of EEG recorded during real finger movement

Authors:
Bashar Awwad;Shiekh Hasan
Affiliations:
School of Medical Sciences, University of Aberdeen, Aberdeen, UK;School of Medical Sciences, University of Aberdeen, Aberdeen, UK
Venue:
MLDM'11 Proceedings of the 7th international conference on Machine learning and data mining in pattern recognition
Year:
2011

Citing 7
Cited 0

Hidden Markov models for online classification of single trial EEG data

Pattern Recognition Letters
Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data

ICML '01 Proceedings of the Eighteenth International Conference on Machine Learning
Shallow parsing with conditional random fields

NAACL '03 Proceedings of the 2003 Conference of the North American Chapter of the Association for Computational Linguistics on Human Language Technology - Volume 1
Pattern Recognition and Machine Learning (Information Science and Statistics)

Pattern Recognition and Machine Learning (Information Science and Statistics)
Discrimination of motor imagery-induced EEG patterns in patients with complete spinal cord injury

Computational Intelligence and Neuroscience - Neuromath: advanced methods for the estimation of human brain activity and connectivity
2009 Special Issue: Time Domain Parameters as a feature for EEG-based Brain-Computer Interfaces

Neural Networks
A discriminative model corresponding to hierarchical HMMs

IDEAL'07 Proceedings of the 8th international conference on Intelligent data engineering and automated learning

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper studies the temporal behavior of EEG data during self-paced real finger movement using Hidden Markov Models and Conditional Random Fields and proposes novel temporal classification methods for movement classification versus idle state. Results are compared to those from Linear Discriminant Analysis based classification. It is demonstrated that using the temporal information in the classification model itself can significantly improve the performance of self-paced EEG classification. The proposed methods are tested on 15 subjects, achieving between 57% and 88% cross validation accuracy, with an average 6% improvement in classification accuracy.