Brain Imaging and Support Vector Machines for Brain Computer Interface

Khachab, Maha; Kaakour, S.; Mokbel, Chafic

doi:10.1109/isbi.2007.357031

Cited by 4 publications

(5 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Classical feature extraction algorithms try to explore the time-frequency characteristics of the EEG signals (Wavelet, Filter banks) or to statistically analyze the EEG signals (PCA, ICA) in order to reduce relevant features that would help to decode the intended action. An approach proposed by (Khachab et al, 2007) consists of using a brain imaging algorithm to deduce the electrical activity on a grid defined on the cortical surface. These activations are considered to form a feature vector.…”

Section: Discussionmentioning

confidence: 99%

“…As discussed previously, the electrical activity on the cortex surface also reflects the movement to be executed. Therefore, it has been proposed (Khachab et al, 2007) to use brain imaging techniques in order to extract reliable features for the decoding process of a BCI system. The underlying idea is to consider a grid representing the cortical surface and compute the electrical activity in every point of this grid.…”

Section: Eeg-based Brain Imaging Featuresmentioning

confidence: 99%

“…More detail is provided in (Cristianini & Taylor, 2000). In (Khachab et al, 2007) several kernel functions have been used and compared.…”

Section: Support Vector Machines (Svm)mentioning

confidence: 99%

See 2 more Smart Citations

Brain Imaging and Machine Learning for Brain-Computer Interface

Khachab¹,

Mokbel²,

Kaakour³

et al. 2010

Biomedical Imaging

Self Cite

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 99%

Section: Eeg-based Brain Imaging Featuresmentioning

confidence: 99%

See 1 more Smart Citation

Brain Imaging and Machine Learning for Brain-Computer Interface

Khachab¹,

Mokbel²,

Kaakour³

et al. 2010

Biomedical Imaging

Self Cite

View full text Add to dashboard Cite

“…is called kernel function and it can be any function which fulfills Mercer's condition. The kernel function computes a dot product in some high dimensional euclidean space H which needs not necessarily known [42][43].…”

Section: Support-vector-machines (Svms)mentioning

confidence: 99%

A BCI System for Physically Challenged Based on Alpha Dynamics of EEG Signals

Khare

2023

Preprint

View full text Add to dashboard Cite

Millions of people around the world suffer from paralysis. People having paralysis need new devices with sophisticated technologies to help them for comfortable live. Most existing brain computer interfaces (BCIs) are designed to control a wheelchair using eye movement, joy stick and chin movement have a many limitation and inconvenient to use. In this study, a brain computer interface based on alpha dynamics of EEG has been attempted to provide an alternative channel for severely disabled for communicating with the external world. In the present study, people who are not able to move their body part can also control the wheelchair just by thought process Wheelchair EEG data was mapped in alpha frequency band. Wavelet packet transform was used for feature extraction. Performance of three classifiers was compared for classification of five imagery tasks. Radial basis neural network was observed most suitable for classification EEG signal. EEG based offline system for wheelchair control has been designed and implemented on laboratory developed prototype.

show abstract

“…The choice of a SVM-based classifier and a RBF kernel function relies on previous works that considered this configuration (Shoker et al, 2005;Guler and Ubeyli, 2007;Khachab et al, 2007). Furthermore, a SVM classifier has improved the accuracy in 13% when compared to LDA (Linear Discriminant Analysis) and 16.3% when compared to NN (Neural Networks), using the same features (Nicolaou et al, 2008).…”

Section: Classifier: Svmmentioning

confidence: 99%

Evaluation of PSD Components and Aar Parameters as Input Features for a SVM Classifier Applied to a Robotic Wheelchair

Ferreira¹,

Bastos-Filho²,

Sarcinelli-Filho³

et al. 2009

Proceedings of the International Conference on Biomedical Electronics and Devices

View full text Add to dashboard Cite

Two distinct signal features suitable to be used as input to a Support-Vector Machine (SVM) classifier in an application involving hands motor imagery and the correspondent EEG signal are evaluated in this paper. Such features are the Power Spectral Density (PSD) components and the Adaptive Autoregressive (AAR) parameters. Different classification times (CT) and time intervals are evaluated, for the AAR-based and the PSD-based features, respectively. The best result (an accuracy of 97.1%) is obtained when using PSD components, while the AAR parameters generated an accuracy of 94.3%. The results also demonstrate that it is possible to use only two EEG channels (bipolar configuration around C 3 and C 4), discarding the bipolar configuration around C z. The algorithms were tested with a proprietary EEG data set involving 4 individuals and with a data set provided by the University of Graz (Austria) as well. The resulting classification system is now being implemented in a Brain-Computer Interface (BCI) used to guide a robotic wheelchair.

show abstract

Brain Imaging and Support Vector Machines for Brain Computer Interface

Cited by 4 publications

References 9 publications

Brain Imaging and Machine Learning for Brain-Computer Interface

Brain Imaging and Machine Learning for Brain-Computer Interface

A BCI System for Physically Challenged Based on Alpha Dynamics of EEG Signals

Evaluation of PSD Components and Aar Parameters as Input Features for a SVM Classifier Applied to a Robotic Wheelchair

Contact Info

Product

Resources

About