Thomas Navin Lal

Support vector channel selection in BCI

Designing a brain computer interface (BCI) system one can choose from a variety of features that may be useful for classifying brain activity during a mental task. For the special case of classifying electroencephalogram (EEG) signals we propose the usage of the state of the art feature selection algorithms Recursive Feature Elimination and Zero-Norm Optimization which are based on the training of support vector machines (SVM) . These algorithms can provide more accurate solutions than standard filter methods for feature selection . We adapt the methods for the purpose of selecting EEG channels. For a motor imagery paradigm we show that the number of used channels can be reduced significantly without increasing the classification error. The resulting best channels agree well with the expected underlying cortical activity patterns during the mental tasks. Furthermore we show how time dependent task specific information can be visualized.

Classifying EEG and ECoG signals without subject training for fast BCI implementation: comparison of nonparalyzed and completely paralyzed subjects

We summarize results from a series of related studies that aim to develop a motor-imagery-based brain-computer interface using a single recording session of electroencephalogram (EEG) or electrocorticogram (ECoG) signals for each subject. We apply the same experimental and analytical methods to 11 nonparalysed subjects (eight EEG, three ECoG), and to five paralyzed subjects (four EEG, one ECoG) who had been unable to communicate for some time. While it was relatively easy to obtain classifiable signals quickly from most of the nonparalyzed subjects, it proved impossible to classify the signals obtained from the paralyzed patients by the same methods. This highlights the fact that though certain BCI paradigms may work well with healthy subjects, this does not necessarily indicate success with the target user group. We outline possible reasons for this failure to transfer.

Robust EEG channel selection across subjects for brain-computer interfaces

Most EEG-based brain-computer interface (BCI) paradigms come along with specific electrode positions, for example, for a visual-based BCI, electrode positions close to the primary visual cortex are used. For new BCI paradigms it is usually not known where task relevant activity can be measured from the scalp. For individual subjects, Lal et al. in 2004 showed that recording positions can be found without the use of prior knowledge about the paradigm used. However it remains unclear to what extent their method of recursive channel elimination (RCE) can be generalized across subjects. In this paper we transfer channel rankings from a group of subjects to a new subject. For motor imagery tasks the results are promising, although cross-subject channel selection does not quite achieve the performance of channel selection on data of single subjects. Although the RCE method was not provided with prior knowledge about the mental task, channels that are well known to be important (from a physiological point of view) were consistently selected whereas task-irrelevant channels were reliably disregarded.

Classifying event-related desynchronization in EEG, ECoG and MEG signals

We employed three different brain signal recording methods to perform Brain-Computer Interface studies on untrained subjects. In all cases, we aim to develop a system that could be used for fast, reliable preliminary screening in clinical BCI application, and we are interested in knowing how long screening sessions need to be. Good performance could be achieved, on average, after the first 200 trials in EEG, 75–100 trials in MEG, or 25–50 trials in ECoG. We compare the performance of Independent Component Analysis and the Common Spatial Pattern algorithm in each of the three sensor types, finding that spatial filtering does not help in MEG, helps a little in ECoG, and improves performance a great deal in EEG. In all cases the unsupervised ICA algorithm performed at least as well as the supervised CSP algorithm, which can suffer from poor generalization performance due to overfitting, particularly in ECoG and MEG.

Voluntary brain regulation and communication with electrocorticogram signals.

Brain-computer interfaces (BCIs) can be used for communication in writing without muscular activity or for learning to control seizures by voluntary regulation of brain signals such as the electroencephalogram (EEG). Three of five patients with epilepsy were able to spell their names with electrocorticogram (ECoG) signals derived from motor-related areas within only one or two training sessions. Imagery of finger or tongue movements was classified with support-vector classification of autoregressive coefficients derived from the ECoG signals. After training of the classifier, binary classification responses were used to select letters from a computer-generated menu. Offline analysis showed increased theta activity in the unsuccessful patients, whereas the successful patients exhibited dominant sensorimotor rhythms that they could control. The high spatial resolution and increased signal-to-noise ratio in ECoG signals, combined with short training periods, may offer an alternative for communication in complete paralysis, locked-in syndrome, and motor restoration.

Hilbertian Metrics on Probability Measures and Their Application in SVM’s

In this article we investigate the field of Hilbertian metrics on probability measures. Since they are very versatile and can therefore be applied in various problems they are of great interest in kernel methods. Quit recently Topsoe and Fuglede introduced a family of Hilbertian metrics on probability measures. We give basic properties of the Hilbertian metrics of this family and other used metrics in the literature. Then we propose an extension of the considered metrics which incorporates structural information of the probability space into the Hilbertian metric. Finally we compare all proposed metrics in an image and text classification problem using histogram data.

Combining a Filter Method with SVMs

Our goal for the competition was to evaluate the usefulness of simple machine learning techniques. We decided to use the Fisher criterion (see Chapter 2) as a feature selection method and Support Vector Machines (see Chapter 1) for the classification part. Here we explain how we chose the regularization parameter C of the SVM, how we determined the kernel parameter σ and how we estimated the number of features used for each data set. All analyzes were carried out on the training sets of the competition data. We choose the data set Arcene as an example to explain the approach step by step.

Learning and Prediction of the Nonlinear Dynamics of Biological Neurons with Support Vector Machines

Based on biological data we examine the ability of Support Vector Machines (SVMs) with gaussian kernels to learn and predict the nonlinear dynamics of single biological neurons. We show that SVMs for regression learn the dynamics of the pyloric dilator neuron of the australian crayfish, and we determine the optimal SVM parameters with regard to the test error. Compared to conventional RBF networks, SVMs learned faster and performed a better iterated one-step-ahead prediction with regard to training and test error. From a biological point of view SVMs are especially better in predicting the most important part of the dynamics, where the membranpotential is driven by superimposed synaptic inputs to the threshold for the oscillatory peak.

Towards Learning Path Planning for Solving Complex Robot Tasks

For solving complex robot tasks it is necessary to incorporate path planning methods that are able to operate within different high-dimensional configuration spaces containing an unknown number of obstacles. Based on Advanced A*-algorithm (AA*) using expansion matrices instead of a simple expansion logic we propose a further improvement of AA* enabling the capability to learn directly from sample planning tasks. This is done by inserting weights into the expansion matrix which are modified according to a special learning rule. For an examplary planning task we show that Adaptive AA* learns movement vectors which allow larger movements than the initial ones into well-defined directions of the configuration space. Compared to standard approaches planning times are clearly reduced.