Limits...
Temporal and spatial features of single-trial EEG for brain-computer interface.

Zhao Q, Zhang L - Comput Intell Neurosci (2007)

Bottom Line: Brain-computer interface (BCI) systems create a novel communication channel from the brain to an output device bypassing conventional motor output pathways of nerves and muscles.Combining IRA features with the CSP method, we obtain the optimal spatial and temporal features with which we achieve the best classification rate.The high classification rate indicates that the proposed method is promising for an EEG-based brain-computer interface.

View Article: PubMed Central - PubMed

Affiliation: Department of Computer Science and Engineering, Shanghai Jiao Tong University, Shanghai 200240, China. qbzhao@sjtu.edu.cn

ABSTRACT
Brain-computer interface (BCI) systems create a novel communication channel from the brain to an output device bypassing conventional motor output pathways of nerves and muscles. Modern BCI technology is essentially based on techniques for the classification of single-trial brain signals. With respect to the topographic patterns of brain rhythm modulations, the common spatial patterns (CSPs) algorithm has been proven to be very useful to produce subject-specific and discriminative spatial filters; but it didn't consider temporal structures of event-related potentials which may be very important for single-trial EEG classification. In this paper, we propose a new framework of feature extraction for classification of hand movement imagery EEG. Computer simulations on real experimental data indicate that independent residual analysis (IRA) method can provide efficient temporal features. Combining IRA features with the CSP method, we obtain the optimal spatial and temporal features with which we achieve the best classification rate. The high classification rate indicates that the proposed method is promising for an EEG-based brain-computer interface.

No MeSH data available.


(a) The joint distribution of four features withmaximal mutual information between features and events types during left-handmovement imagery. (b) The joint distribution of four features with maximalmutual information between features and events types during right-hand movementimagery.
© Copyright Policy - open-access
Related In: Results  -  Collection


getmorefigures.php?uid=PMC2267213&req=5

fig12: (a) The joint distribution of four features withmaximal mutual information between features and events types during left-handmovement imagery. (b) The joint distribution of four features with maximalmutual information between features and events types during right-hand movementimagery.

Mentions: Here the entropy H(s) can be computed in the process of computingthe mutual information with output class , so there is little change in computational load withrespect to MIFS. The variable β gives flexibility tothe algorithm as in MIFS. If we set β zero, the proposedalgorithm chooses features in the order of the mutual information with theoutput. As β grows, it deselectsthe redundant features more efficiently. In general, we can set β = 1in compliance with(31). For allthe experiments to be discussed later, we set it to 1. The estimation of mutualinformation (MI) between each feature and event labels are showed in Figure 11.Based on the algorithm, we obtain a subset of relevant features, which possessthe larger MI of all the features, for the classification procedure. Figure 12shows joint distribution of four features with maximal mutual information.


Temporal and spatial features of single-trial EEG for brain-computer interface.

Zhao Q, Zhang L - Comput Intell Neurosci (2007)

(a) The joint distribution of four features withmaximal mutual information between features and events types during left-handmovement imagery. (b) The joint distribution of four features with maximalmutual information between features and events types during right-hand movementimagery.
© Copyright Policy - open-access
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC2267213&req=5

fig12: (a) The joint distribution of four features withmaximal mutual information between features and events types during left-handmovement imagery. (b) The joint distribution of four features with maximalmutual information between features and events types during right-hand movementimagery.
Mentions: Here the entropy H(s) can be computed in the process of computingthe mutual information with output class , so there is little change in computational load withrespect to MIFS. The variable β gives flexibility tothe algorithm as in MIFS. If we set β zero, the proposedalgorithm chooses features in the order of the mutual information with theoutput. As β grows, it deselectsthe redundant features more efficiently. In general, we can set β = 1in compliance with(31). For allthe experiments to be discussed later, we set it to 1. The estimation of mutualinformation (MI) between each feature and event labels are showed in Figure 11.Based on the algorithm, we obtain a subset of relevant features, which possessthe larger MI of all the features, for the classification procedure. Figure 12shows joint distribution of four features with maximal mutual information.

Bottom Line: Brain-computer interface (BCI) systems create a novel communication channel from the brain to an output device bypassing conventional motor output pathways of nerves and muscles.Combining IRA features with the CSP method, we obtain the optimal spatial and temporal features with which we achieve the best classification rate.The high classification rate indicates that the proposed method is promising for an EEG-based brain-computer interface.

View Article: PubMed Central - PubMed

Affiliation: Department of Computer Science and Engineering, Shanghai Jiao Tong University, Shanghai 200240, China. qbzhao@sjtu.edu.cn

ABSTRACT
Brain-computer interface (BCI) systems create a novel communication channel from the brain to an output device bypassing conventional motor output pathways of nerves and muscles. Modern BCI technology is essentially based on techniques for the classification of single-trial brain signals. With respect to the topographic patterns of brain rhythm modulations, the common spatial patterns (CSPs) algorithm has been proven to be very useful to produce subject-specific and discriminative spatial filters; but it didn't consider temporal structures of event-related potentials which may be very important for single-trial EEG classification. In this paper, we propose a new framework of feature extraction for classification of hand movement imagery EEG. Computer simulations on real experimental data indicate that independent residual analysis (IRA) method can provide efficient temporal features. Combining IRA features with the CSP method, we obtain the optimal spatial and temporal features with which we achieve the best classification rate. The high classification rate indicates that the proposed method is promising for an EEG-based brain-computer interface.

No MeSH data available.