Comparison of EEG Pattern Classification Methods for Brain-Computer Interfaces

Percept-related EEG classification using machine learning approach and features of functional brain connectivity

Machine learning is a promising approach for electroencephalographic (EEG) trials classification. Its efficiency is largely determined by the feature extraction and selection techniques reducing the dimensionality of input data. Dimensionality reduction is usually implemented via the mathematical approaches (e.g., principal component analysis, linear discriminant analysis, etc.) regardless of the origin of analyzed data. We hypothesize that since EEG features are determined by certain neurophysiological processes, they should have distinctive characteristics in spatiotemporal domain. If so, it is possible to specify the set of EEG principal features based on the prior knowledge about underlying neurophysiological processes. To test this hypothesis, we consider the classification of EEG trials related to the perception of ambiguous visual stimuli. We observe that EEG features, underlying the different ambiguous stimuli interpretations, are defined by the network properties of neuronal activity. Having analyzed functional neural interactions, we specify the brain area in which neural network architecture exhibits differences for different classes of EEG trials. We optimize the feedforward multilayer perceptron and develop a strategy for the training set selection to maximize the classification accuracy, being 85% when all channels are used. The revealed localization of the percept-related features allows about 95% accuracy, when the number of channels is reduced up to 90%. Obtained results can be used for classification of EEG trials associated with more complex cognitive tasks. Taking into account that cognitive activity is subserved by a distributed functional cortical network, its topological properties have to be considered when selecting optimal features for EEG trial classification.

Classifying the Perceptual Interpretations of a Bistable Image Using EEG and Artificial Neural Networks

In order to classify different human brain states related to visual perception of ambiguous images, we use an artificial neural network (ANN) to analyze multichannel EEG. The classifier built on the basis of a multilayer perceptron achieves up to 95% accuracy in classifying EEG patterns corresponding to two different interpretations of the Necker cube. The important feature of our classifier is that trained on one subject it can be used for the classification of EEG traces of other subjects. This result suggests the existence of common features in the EEG structure associated with distinct interpretations of bistable objects. We firmly believe that the significance of our results is not limited to visual perception of the Necker cube images; the proposed experimental approach and developed computational technique based on ANN can also be applied to study and classify different brain states using neurophysiological data recordings. This may give new directions for future research in the field of cognitive and pathological brain activity, and for the development of brain-computer interfaces.

A P300-based cognitive assessment battery

BACKGROUND

It is well established that some patients who are diagnosed as being in a vegetative state or a minimally conscious state show reliable signs of volition that may only be detected by measuring neural responses. A pertinent question is whether these patients are capable of higher cognitive processes.

METHODS

Here, we develop a series of EEG paradigms that probe several core aspects of cognition at the bedside without the need for motor responses and explore the sensitivity of this approach in a group of healthy controls.

RESULTS

Using analysis of ERPs alone, this method can determine with high reliability whether individual participants are able to attend a stimulus stream, maintain items in working memory, or solve complex grammatical reasoning problems.

CONCLUSION

We suggest that this approach could form the basis of a brain-based battery for assessing higher cognition in patients with severe motor impairments or disorders of consciousness.

Comparison of EEG-features and classification methods for motor imagery in patients with disorders of consciousness

Current research aims at identifying voluntary brain activation in patients who are behaviorally diagnosed as being unconscious, but are able to perform commands by modulating their brain activity patterns. This involves machine learning techniques and feature extraction methods such as applied in brain computer interfaces. In this study, we try to answer the question if features/classification methods which show advantages in healthy participants are also accurate when applied to data of patients with disorders of consciousness. A sample of healthy participants (N = 22), patients in a minimally conscious state (MCS; N = 5), and with unresponsive wakefulness syndrome (UWS; N = 9) was examined with a motor imagery task which involved imagery of moving both hands and an instruction to hold both hands firm. We extracted a set of 20 features from the electroencephalogram and used linear discriminant analysis, k-nearest neighbor classification, and support vector machines (SVM) as classification methods. In healthy participants, the best classification accuracies were seen with coherences (mean = .79; range = .53−.94) and power spectra (mean = .69; range = .40−.85). The coherence patterns in healthy participants did not match the expectation of central modulated -rhythm. Instead, coherence involved mainly frontal regions. In healthy participants, the best classification tool was SVM. Five patients had at least one feature-classifier outcome with p0.05 (none of which were coherence or power spectra), though none remained significant after false-discovery rate correction for multiple comparisons. The present work suggests the use of coherences in patients with disorders of consciousness because they show high reliability among healthy subjects and patient groups. However, feature extraction and classification is a challenging task in unresponsive patients because there is no ground truth to validate the results.

Toward a model-based predictive controller design in brain-computer interfaces

A first step in designing a robust and optimal model-based predictive controller (MPC) for brain-computer interface (BCI) applications is presented in this article. An MPC has the potential to achieve improved BCI performance compared to the performance achieved by current ad hoc, nonmodel-based filter applications. The parameters in designing the controller were extracted as model-based features from motor imagery task-related human scalp electroencephalography. Although the parameters can be generated from any model-linear or non-linear, we here adopted a simple autoregressive model that has well-established applications in BCI task discriminations. It was shown that the parameters generated for the controller design can as well be used for motor imagery task discriminations with performance (with 8-23% task discrimination errors) comparable to the discrimination performance of the commonly used features such as frequency specific band powers and the AR model parameters directly used. An optimal MPC has significant implications for high performance BCI applications.

Optimization of electrode channels in Brain Computer Interfaces

What is the optimal number of electrodes one can use in discrimination of tasks for a Brain Computer Interface (BCI)? To address this question, the number and location of scalp electrodes in the acquisition of human electroencephalography (EEG) and discrimination of motor imagery tasks were optimized by using a systematic optimization approach. The systematic analysis results in the most reliable procedure in electrode optimization as well as a validating means for the other feature selection techniques. We acquired human scalp EEG in response to cue-based motor imagery tasks. We employed a systematic analysis by using all possible combinations of the channels and calculating task discrimination errors for each of these combinations by using linear discriminant analysis (LDA) for feature classification. Channel combination that resulted in the smallest discrimination error was selected as the optimum number of channels to be used in BCI applications. Results from the systematic analysis were compared with another feature selection algorithm: forward stepwise feature selection combined with LDA feature classification. Our results demonstrate the usefulness of the fully optimized technique for a reliable selection of scalp electrodes in BCI applications.

Model-based responses and features in Brain Computer Interfaces

Novel model based features are introduced in the discrimination of motor imagery tasks using human scalp electroencephalography (EEG) towards the development of Brain Computer Interfaces (BCI). We have acquired human scalp EEG under open-loop and feedback conditions in response to cue-based motor imagery tasks. EEG signals, transformed into frequency specific bands such as mu, beta and movement related potentials, were used for feature extraction with the aim to discriminate tasks. Data were classified using features such as power spectrum and model-based parameters. Two different feature selection methods: stepwise and principal component analysis (PCA), were combined with linear discriminant analysis (LDA). Different training/validation criteria were applied for classification of task related features. Results show that the scalp EEG correlate of the imagery tasks of hands/toes/tongue movements under open-loop conditions and left/right hand movements under feedback conditions, can be well discriminated with classification errors below 20%. Model based techniques, which resulted in classification errors in the range of 2%-30%, have the potential to use advanced control systems theory in the development of BCI to achieve improved performance compared to the performance achieved by currently applied proportional control or filter algorithms.