Brain Computer Interface Signals Research Articles

Low cost implementation of a Motor Imagery experiment with BCI system and its use in neurorehabilitation.

The application of rehabilitation programs based on videogames with brain-computer interfaces (BCI) allows to provide feedback to the user with the expectation of stimulate the brain plasticity that will restore the motor control. The use of specific mental strategies such as Motor Imagery (MI) in neuroscientific experiments with BCI systems often requires the acquisition of sophisticated interfaces and specialized software for execution, which usually have a high implementation costs. We present a combination of low-cost hardware and open-source software for the implementation of videogame based on virtual reality with MI and its potential use as neurotherapy for stroke patients. Three machine learning algorithms for the BCI signals classification are shown: LDA (Linear Discriminant Analysis) and two Support Vector Machines (SVM) in order to determine which task of MI is being performed by the user in a particular moment of the experiment. All classification algorithms was evaluated in 8 healthy subjects, the average accuracy of the best classifier was 96.7%, which shows that it is possible to carry out serious neuroscientific experiments with MI using low-cost BCI systems and achieve comparable accuracies with more sophisticated and expensive devices.

Evaluation of EEG Features in Decoding Individual Finger Movements from One Hand

With the advancements in modern signal processing techniques, the field of brain-computer interface (BCI) is progressing fast towards noninvasiveness. One challenge still impeding these developments is the limited number of features, especially movement-related features, available to generate control signals for noninvasive BCIs. A few recent studies investigated several movement-related features, such as spectral features in electrocorticography (ECoG) data obtained through a spectral principal component analysis (PCA) and direct use of EEG temporal data, and demonstrated the decoding of individual fingers. The present paper evaluated multiple movement-related features under the same task, that is, discriminating individual fingers from one hand using noninvasive EEG. The present results demonstrate the existence of a broadband feature in EEG to discriminate individual fingers, which has only been identified previously in ECoG. It further shows that multiple spectral features obtained from the spectral PCA yield an average decoding accuracy of 45.2%, which is significantly higher than the guess level (P < 0.05) and other features investigated (P < 0.05), including EEG spectral power changes in alpha and beta bands and EEG temporal data. The decoding of individual fingers using noninvasive EEG is promising to improve number of features for control, which can facilitate the development of noninvasive BCI applications with rich complexity.

A New Approach to Separate Haemodynamic Signals for Brain-Computer Interface Using Independent Component Analysis and Least Squares

Brain-computer interface (BCI) is one technology that allows a user to communicate with external devices through detecting brain activity. As a promising noninvasive technique, functional near-infrared spectroscopy (fNIRS) has recently earned increasing attention in BCI studies. However, in practice fNIRS measurements can suffer from significant physiological interference, for example, arising from cardiac contraction, breathing, and blood pressure fluctuations, thereby severely limiting the utility of the method. Here, we apply the multidistance fNIRS method, with short-distance and long-distance optode pairs, and we propose the combination of independent component analysis (ICA) and least squares (LS) with the fNIRS recordings to reduce the interference. The short-distance fNIRS measurement is treated as the virtual channel and the long-distance fNIRS measurement is treated as the measurement channel. Least squares is used to optimize the reconstruction value for brain activity signal. Monte Carlo simulations of photon propagation through a five-layered slab model of a human adult head were implemented to evaluate our methodology. The results demonstrate that the ICA method can separate the brain signal and interference; the further application of least squares can significantly recover haemodynamic signals contaminated by physiological interference from the fNIRS-evoked brain activity data.

A Haemodynamic Brain–Computer Interface Based on Real-Time Classification of near Infrared Spectroscopy Signals during Motor Imagery and Mental Arithmetic

Over the past decade, an increasing number of studies have investigated near infrared (NIR) spectroscopy for signal acquisition in brain–computer interface (BCI) systems. However, although a BCI relies on classifying brain signals in real-time, the majority of previous studies did not perform real-time NIR spectroscopy signal classification but derived knowledge about the feasibility of NIR spectroscopy for BCI purposes from offline analyses. The present study investigates whether NIR spectroscopy signals evoked by two different mental tasks (i.e. motor imagery and mental arithmetic) can be classified in real-time in order to control a NIR-BCI application. Furthermore, since this is the first study that attempts to distinguish between the haemodynamic responses to these two tasks, we aimed to investigate whether this task-combination is feasible for controlling a NIR-BCI. Twelve healthy participants were asked to control a moving ball on a computer screen by performing motor imagery and mental arithmetic tasks. The real-time classification of their task-specific NIR spectroscopy signals yielded accuracy rates ranging from 45% up to 93%. Offline analyses across all participants showed that both tasks evoked different haemodynamic responses in prefrontal and sensorimotor cortex areas. On the one hand, these results demonstrate the considerable potential of NIR spectroscopy for BCI signal acquisition and the feasibility of the applied mental tasks for NIR-BCI control. On the other hand, since the classification accuracy showed an unsatisfactory stability across measurement sessions, we conclude that further investigations and progress in methodological issues are needed and we discuss further steps that have to be taken until it is conceivable to implement a real-time capable NIR-BCI that works with sufficient accuracy across a large group of individuals.

P300-based brain–computer interface for environmental control: an asynchronous approach

Brain–computer interface (BCI) systems allow people with severe motor disabilities to communicate and interact with the external world. The P300 potential is one of the most used control signals for EEG-based BCIs. Classic P300-based BCIs work in a synchronous mode; the synchronous control assumes that the user is constantly attending to the stimulation, and the number of stimulation sequences is fixed a priori. This issue is an obstacle for the use of these systems in everyday life; users will be engaged in a continuous control state, their distractions will cause misclassification and the speed of selection will not take into account users' current psychophysical condition. An efficient BCI system should be able to understand the user's intentions from the ongoing EEG instead. Also, it has to refrain from making a selection when the user is engaged in a different activity and it should increase or decrease its speed of selection depending on the current user's state. We addressed these issues by introducing an asynchronous BCI and tested its capabilities for effective environmental monitoring, involving 11 volunteers in three recording sessions. Results show that this BCI system can increase the bit rate during control periods while the system is proved to be very efficient in avoiding false negatives when the users are engaged in other tasks.

Investigation of Non-stationarity in Brain Activity via Robust Principal Component Analysis

Event Abstract Back to Event Investigation of Non-stationarity in Brain Activity via Robust Principal Component Analysis Motoaki Kawanabe1, 2, Javier Pascual1 and Carmen Vidaurre1* 1 Berlin Institute of Technology, Machine Learning Laboratory, Germany 2 Fraunhofer FIRST, IDA, Germany Investigating non-stationarities is an important topic and a difficult problem in neuroscience. In this paper we propose a method that allows, visualizing the spatial pattern of the most prominent variability in the neuroscientific data (EEG in this case). A key feature of this procedure is that allows the study of the relevant brain activity by attenuating the influence of outlying signals. In this case we apply the method in .brain-computer interface (BCI) data. BCIs are systems that translate the users intent, coded by a small set of mental tasks, into control actions such as computer applications or prostheses. Fluctuations in BCI data may be caused by change of task involvement, fatigue etc., or by artifacts such as swallowing, blinking or yawning. Recently, Krauledat [1] applied principal component analysis (PCA) to session-wise average covariance matrices and discussed session-to-session variability of BCI signals. Inspired by the previous work, we investigate non-stationarity between single-trial covariance matrices by a novel robust PCA proposed in this research.Technically, our robust PCA is developed upon the theory of robust statistics. As in our previous work on the robust CSP [2], we replace the Euclidean error measure with an “outlier-insensitive” one for defining the desired low-dimensional approximations of observed data. From this robust objective function, we derive an iterative algorithm which can detect outlying trials automatically and which makes the PCA calculation robust by reducing their weights. The outputs of our procedure are eigen matrices, which indicate prominent changes in covariance matrices across trials. We can usually get further insight by showing the first eigen vectors of the eigen matrices as scalp maps (see Fig. 1).The data of the analysis were recorded in a one-day session from 1 healthy BCI-novice user. The subject was sitting in a comfortable chair with arms lying relaxed on armrests. Brain activity was recorded from the scalp with multi-channel EEG amplifiers using 119 Ag/AgCl electrodes in an extended 10-20 system sampled at 1000 Hz with a band-pass from 0.05 to 200 Hz. The subject performed a ‘calibration measurement’ in which every 8s one of three different visual cues (arrows pointing left, right, down) indicating which motor imagery to perform: left/right hand or foot. Three runs with 25 trials of each motor condition were recorded. Here we analyzed the data of two classes. Figure 1. Comparison of robust PCA and PCA to analyze the spatial patterns of variability in brain activity in BCI. The figure illustrates the first principal component of the direction of main variation for each class. This component is wrongly estimated using conventional PCA. We can see a strong influence of lateral electrodes in the motor area. However, robust PCA reveals that this direction is mainly caused by outlying trials. Our robust method is a powerful tool to study variations in brain activity and can be applied not only to data where difference in rhythmical power is to be expected but also to other activity such as evoked potentials. Figure 1

High gamma mapping using EEG

High gamma (HG) power changes during motor activity, especially at frequencies above 70 Hz, play an important role in functional cortical mapping and as control signals for BCI (brain–computer interface) applications. Most studies of HG activity have used ECoG (electrocorticography) which provides high-quality spatially localized signals, but is an invasive method. Recent studies have shown that non-invasive modalities such as EEG and MEG can also detect task-related HG power changes. We show here that a 27 channel EEG (electroencephalography) montage provides high-quality spatially localized signals non-invasively for HG frequencies ranging from 83 to 101 Hz. We used a generic head model, a weighted minimum norm least squares (MNLS) inverse method, and a self-paced finger movement paradigm. The use of an inverse method enables us to map the EEG onto a generic cortex model. We find the HG activity during the task to be well localized in the contralateral motor area. We find HG power increases prior to finger movement, with average latencies of 462 ms and 82 ms before EMG (electromyogram) onset. We also find significant phase-locking between contra- and ipsilateral motor areas over a similar HG frequency range; here the synchronization onset precedes the EMG by 400 ms. We also compare our results to ECoG data from a similar paradigm and find EEG mapping and ECoG in good agreement. Our findings demonstrate that mapped EEG provides information on two important parameters for functional mapping and BCI which are usually only found in HG of ECoG signals: spatially localized power increases and bihemispheric phase-locking.

Brain–computer interfaces (BCIs): Detection instead of classification

Many studies over the past two decades have shown that people can use brain signals to convey their intent to a computer through brain–computer interfaces (BCIs). These devices operate by recording signals from the brain and translating these signals into device commands. They can be used by people who are severely paralyzed to communicate without any use of muscle activity. One of the major impediments in translating this novel technology into clinical applications is the current requirement for preliminary analyses to identify the brain signal features best suited for communication. This paper introduces and validates signal detection, which does not require such analysis procedures, as a new concept in BCI signal processing. This detection concept is realized with Gaussian mixture models (GMMs) that are used to model resting brain activity so that any change in relevant brain signals can be detected. It is implemented in a package called SIGFRIED (SIGnal modeling For Real-time Identification and Event Detection). The results indicate that SIGFRIED produces results that are within the range of those achieved using a common analysis strategy that requires preliminary identification of signal features. They indicate that such laborious analysis procedures could be replaced by merely recording brain signals during rest. In summary, this paper demonstrates how SIGFRIED could be used to overcome one of the present impediments to translation of laboratory BCI demonstrations into clinically practical applications.

Towards a robust BCI: error potentials and online learning

Recent advances in the field of brain-computer interfaces (BCIs) have shown that BCIs have the potential to provide a powerful new channel of communication, completely independent of muscular and nervous systems. However, while there have been successful laboratory demonstrations, there are still issues that need to be addressed before BCIs can be used by nonexperts outside the laboratory. At IDIAP Research Institute, we have been investigating several areas that we believe will allow us to improve the robustness, flexibility, and reliability of BCIs. One area is recognition of cognitive error states, that is, identifying errors through the brain's reaction to mistakes. The production of these error potentials (ErrP) in reaction to an error made by the user is well established. We have extended this work by identifying a similar but distinct ErrP that is generated in response to an error made by the interface, (a misinterpretation of a command that the user has given). This ErrP can be satisfactorily identified in single trials and can be demonstrated to improve the theoretical performance of a BCI. A second area of research is online adaptation of the classifier. BCI signals change over time, both between sessions and within a single session, due to a number of factors. This means that a classifier trained on data from a previous session will probably not be optimal for a new session. In this paper, we present preliminary results from our investigations into supervised online learning that can be applied in the initial training phase. We also discuss the future direction of this research, including the combination of these two currently separate issues to create a potentially very powerful BCI.

BCI meeting 2005-workshop on BCI signal processing: feature extraction and translation

This paper describes the outcome of discussions held during the Third International BCI Meeting at a workshop charged with reviewing and evaluating the current state of and issues relevant to brain-computer interface (BCI) feature extraction and translation. The issues discussed include a taxonomy of methods and applications, time-frequency spatial analysis, optimization schemes, the role of insight in analysis, adaptation, and methods for quantifying BCI feedback.