Self-paced Brain computer interface (SBCI) systems allow individuals with motor disabilities to use their brain signals to control devices, whenever they wish. These systems are required to identify the user’s “intentional control (IC)” commands and they must remain inactive during all periods in which users do not intend control (called “no control (NC)” periods). This dissertation addresses three issues related to the design of SBCI systems: 1) their presently high false positive (FP) rates, 2) the presence of artifacts and 3) the identification of a suitable evaluation metric. To improve the performance of SBCI systems, the following are proposed: 1) a method for the automatic user-customization of a 2-state SBCI system, 2) a two-stage feature reduction method for selecting wavelet coefficients extracted from movement-related potentials (MRP), 3) an SBCI system that classifies features extracted from three neurological phenomena: MRPs, changes in the power of the Mu and Beta rhythms; 4) a novel method that effectively combines methods developed in 2) and 3 ) and 5) generalizing the system developed in 3) for detecting a right index finger flexion to detecting the right hand extension. Results of these studies using actual movements show an average true positive (TP) rate of 56.2% at the FP rate of 0.14% for the finger flexion study and an average TP rate of 33.4% at the FP rate of 0.12% for the hand extension study. These FP results are significantly lower than those achieved in other SBCI systems, where FP rates vary between 1-10%. We also conduct a comprehensive survey of the BCI literature. We demonstrate that many BCI papers do not properly deal with artifacts. We show that the proposed BCI achieves a good performance of TP=51.8% and FP=0.4% in the presence of eye movement artifacts. Further tests of the performance of the proposed system in a pseudo-online environment, shows an average TP rate =48.8% at the FP rate of 0.8%. Finally, we propose a framework for choosing a suitable evaluation metric for SBCI systems. This framework shows that Kappa coefficient is more suitable than other metrics in evaluating the performance during the model selection procedure. / Applied Science, Faculty of / Electrical and Computer Engineering, Department of / Graduate

Les interfaces cerveau-machine (ICM) utilisent les signaux émis par le cerveau pour contrôler des machines ainsi que des appareils (claviers, voitures, neuro-prothèses). Après plusieurs décennies de développement, les techniques de ICM modernes montrent une maturité relative par rapport aux dernières décennies et reçoivent de plus en plus d'attention dans les applications grand public du monde réel, en particulier dans le domaine des interactions homme-machine pour personnes en bonne santé, par exemple les neuro-jeux. L'objectif de cette thèse est de développer un modèle d'ICM et des algorithmes de traitement de signaux EEG pour relever ces défis, donc conduire à une ICM non-invasive, portable et facile à utiliser, exploitant des rythmes EEG pour les applications grand public (non médicales). Pour atteindre cet objectif, un examen de l'état de l'art (prototypes existants et produits commerciaux, configurations expérimentales, algorithmes) a d'abord été effectué pour acquérir une bonne compréhension de ce domaine. Les contributions de cette thèse comprennent : 1) un paradigme ICM hybride avec peu d'électrodes, 2) la réduction de la dimensionnalité pour l'ICM multi-canal (avec un nombre élevé d'électrodes), 3) la réduction et la sélection de canal, 4) l'amélioration de la classification pour l'ICM avec des électrodes prédéterminées. Les résultats expérimentaux montrent que les méthodes proposées dans cette thèse peuvent améliorer les performances de classification et/ou augmenter l'efficacité du système (par exemple, réduire le temps d'apprentissage, réduire le coût du matériel), de manière à contribuer à des ICM pour des applications générales. / Brain-computer interfaces (BCIs) use signals from the brain to control machines and devices (keyboards , cars, neuro- prostheses) . After several decades of development, modern BCI techniques show a relative maturity compared to the past decades and receive more and more attention in real-world general public applications, in particular in the domain of BCI-based human-computer interactions for healthy people, such as neuro-games. The aim of this thesis is to develop an experimental setup and signal processing algorithms for non-invasive, portable and easy-to-use BCI systems for large public (non-medical) applications. To achieve this goal, a review of the state of the art (existing prototypes and commercial products, experimental setup, algorithms) is first performed to get a full scope and a good understanding in this field. The main contributions of this thesis include: 1) a hybrid BCI paradigm with a few electrodes , 2) dimensionality reduction for multi-channel BCI (with a high number of electrodes ), 3) reduction and selection channel , 4) improved classification for BCI with a few predetermined electrodes. The experimental results show that the methods proposed in this thesis can improve classification performance and / or increase the efficiency of the system ( for example, reduce the learning time, reduce the cost of equipment ) , so as to contribute to BCI for the general applications.

No description available.

No description available.

No description available.

People who suffer from amyotrophic lateral sclerosis (ALS) eventually lose all voluntary muscle control. In the late stages of the disease, traditional augmentative and alternative communication (AAC) devices fail to provide adequate levels of communication. Brain-computer interface (BCI) technology has provided effective communication after all other AAC devices have failed. Nonetheless, EEG-based BCI devices may also fail for people with late-stage ALS due to loss of voluntary eye movement. Specifically, some people may suffer from random eye movement (nystagmus) and/or drooping of the eyelids (ptosis). Presently, it is unclear in the literature whether BCI operation requires voluntary control of eye movement. The current study attempts to simulate involuntary random eye movement in able-bodied individuals employing the P300-based BCI. To simulate involuntary random eye movement, the stimuli shift in the X and Y dimensions. Stimulus movement ‘Jitter’ occurs between each stimulus presentation in increments of 1-5 pixels (Jitter 1), 5-10 pixels (Jitter 2), 10-15 pixels-(Jitter 3), or a no movement control condition. Data collected from a previous study using 22 participants compared the control condition to Jitter 1 and Jitter 2 indicated higher accuracy for control and Jitter 1 than Jitter 2. No significant differences were found in accuracy, selections per minute, or bitrate. Waveform analysis indicated significantly higher P300 amplitude for the control condition and Jitter 1 than Jitter 2. Preference survey scores showed a preference for Jitter 1 as compared to control and Jitter 2. This finding was unexpected and may be due to the slight movement of Jitter 1 forcing participants to be vigilant, but not distracted. Based on our finding in this study, the current study examines the amount of pixel movement that could lead to reductions in performance. Participants completed a control condition and the three levels of Jitter in a counter-balanced design. Preliminary data for the current study was collected from 15 participants. No significant differences were observed between the three conditions in measures of BCI accuracy, selections per minute, and bitrate. Furthermore, preference survey scores indicated no significant difference in condition preference. Based on the findings of the first study, as well as the data collected so far in the current study, it appears that random movement does not have a significant impact on the ability of healthy participants to operate the BCI system. This could indicate that individuals with random eye movement should be able to operate the system with high rates of accuracy.

Brain-Computer Interfaces (BCIs) that rely on motor imagery currently allow subjects to control quad-copters, robotic arms, and computer cursors. Recent advancements have been made possible because of breakthroughs in fields such as electrical engineering, computer science, and neuroscience. Currently, most real-time BCIs use hand-crafted feature extractors, feature selectors, and classification algorithms. In this work, we explore the different classification algorithms currently used in electroencephalographic (EEG) signal classification and assess their performance on intracranial EEG (iEEG) data. We first discuss the motor imagery task employed using iEEG signals and find features that clearly distinguish between different classes. Second, we compare the different state-of-the-art classifiers used in EEG BCIs in terms of their error rate, computational requirements, and feature interpret-ability. Next, we show the effectiveness of these classifiers in iEEG BCIs and last, show that our new classification algorithm that is designed to use spatial, spectral, and temporal information reaches performance comparable to other state-of-the-art classifiers while also allowing increased feature interpret-ability. / Master of Science / Brain-Computer Interfaces (BCIs) as the name suggests allows individuals to interact with computers using electrical activity captured from different regions of the brain. These devices have been shown to allows subjects to control a number of devices such as quad-copters, robotic arms, and computer cursors. Applications such as these obtain electrical signals from the brain using electrodes either placed non-invasively on the scalp (also known as an electroencephalographic signal, EEG) or invasively on the surface of the brain (Electrocorticographic signal, ECoG). Before a participant can effectively communicate with the computer, the computer is calibrated to recognize different signals by collecting data from the subject and learning to distinguish them using a classification algorithm. In this work, we were interested in analyzing the effectiveness of using signals obtained from deep brain structures by using electrodes place invasively (also known as intracranial EEG, iEEG). We collected iEEG data during a two hand movement task and manually analyzed the data to find regions of the brain that are most effective in allowing us to distinguish signals during movements. We later showed that this task could be automated by using classification algorithms that are borrowed from electroencephalographic (EEG) signal experiments.

There are hundreds of thousands of people who could benefit from a Brain-Computer Interface. However, not all are willing to undergo surgery, so an EEG is the prime candidate for use as a BCI. The features of Event-Related Desynchronization and Synchronization could be used for a switch and have been in the past. A new method of feature selection was proposed to optimize classification of active motor movement vs a non-active idle state. The previous method had pre-selected which frequency and electrode to use as electrode C3 at the 20Hz bin. The new method used SPSS statistical software to determine the most significant frequency and electrode combination. This improved method found increased accuracy in classifying cases as either active or idle states. Future directions could be using multiple features for classification and BCI control, or exploiting the difference between ERD and ERS, though for either of these a more advanced algorithm would be required.

Thesis advisor: Robert Signorile / It has long been known that as neurons fire within the brain they produce measurable electrical activity. Electroencephalography (EEG) is the measurement and recording of these electrical signals using sensors arrayed across the scalp. Though there is copious research in using EEG technology in the fields of neuroscience and cognitive psychology, it is only recently that the possibility of utilizing EEG measurements as inputs in the control of computers has emerged. The idea of Brain-Computer Interfaces (BCIs) which allow the control of devices using brain signals evolved from the realm of science fiction to simple devices that currently exist. BCIs naturally present themselves to many extremely useful applications including prosthetic devices, restoring or aiding in communication and hearing, military applications, video gaming and virtual reality, and robotic control, and have the possibility of significantly improving the quality of life of many disabled individuals. However, current BCIs suffer from many problems including inaccuracies, delays between thought, detection, and action, exorbitant costs, and invasive surgeries. The purpose of this research is to examine the Emotiv EPOC© System as a cost-effective gateway to non-invasive portable EEG measurements and utilize it to build a thought-based BCI to control the Parallax Scribbler® robot. This research furthers the analysis of the current pros and cons of EEG technology as it pertains to BCIs and offers a glimpse of the future potential capabilities of BCI systems. / Thesis (BA) — Boston College, 2010. / Submitted to: Boston College. College of Arts and Sciences. / Discipline: Computer Science Honors Program. / Discipline: Computer Science.

Identifying finger and wrist flexion based actions using single channel surface electromyogram have a number of rehabilitation, defence and human computer interface applications. These applications are currently infeasible because of unreliability in classification of sEMG when the level of muscle contraction is low and when there are multiple active muscles. The presence of noise and cross-talk from closely located and simultaneously active muscles is exaggerated when muscles are weakly active such as during maintained wrist and finger flexion. It has been established in literature that surface electromyogram (sEMG) and other such biosignals are fractal signals. Some researchers have determined that fractal dimension (FD) is related to strength of muscle contraction. On careful analysis of fractal properties of sEMG, this research work has established that FD is related to the muscle size and complexity and not to the strength of muscle contraction. The work has also identified a novel feature, maximum fractal length (MFL) of the signal, as a good measure of strength of contraction of the muscle. From the analysis, it is observed that while at high level of contraction, root mean square (RMS) is an indicator of strength of contraction of the muscle, this relationship is not very strong when the muscle contraction is less than 50% maximum voluntary contraction. This work has established that MFL is a more reliable measure of strength of contraction compared to RMS, especially at low levels of contraction. This research work reports the use of fractal properties of sEMG to identify the small changes in strength of muscle contraction and the location of the active muscles. It is observed that fractal dimension (FD) of the signal is related with the properties of the muscle while maximum fractal length (MFL) is related to the strength of contraction of the associated muscle. The results show that classifying MFL and FD of a single channel sEMG from the forearm it is possible to accurately identify a set of finger and wrist flexion based actions even when the muscle activity is very weak. It is proposed that such a system could be used to control a prosthetic hand or for human computer interface.