· · · Repository hosted by TU Delft Library

This dissemination, use and exploitation plan (DUEP) is a dynamic document, which will be continuously updated throughout BRAIN's duration. It serves the following purposes: -To document the overall strategy for the dissemination and exploitation of the knowledge gained in BRAIN -To document the exploitation plans of each partner -To be a repository of the history of presentations and publications that result from BRAIN work -To generally disseminate information on BRAIN and its progress, in such a way that other projectsin the area can utilize BRAIN's results. This document comprises three sections, namely exploitable knowledge, knowledge dissemination, and publishable results. The section on exploitable knowledge lists the project results that are classified as knowledge and have a potential for product or service development. The section on knowledge dissemination summarizes the major activities (workshops, conferences, etc.) in which BRAIN's results have been presented. The last section provides a summary of each exploitable result that BRAIN has generated.

Research in Electroencephalogram (EEG) based Brain-Computer Interfaces (BCIs) has been considerably expanding during the last few years. Such an expansion owes to a large extent to the multidisciplinary and challenging nature of BCI research. Signal processing undoubtedly constitutes an essential component of a BCI system since from the EEG acquisition to the translation of brain activity into meaningful commands, multivariate signal processing algorithms are intensively applied. In this tutorial, the basic BCI concepts, EEG monitoring, BCI operation, the electrophysiological sources of BCI control, future directions, and ambitions are introduced. The main BCI types, namely motor imagery (ERD/ERS), steady state visual evoked potentials (SSVEP), and P300 based BCIs are presented along with practical application examples.The EEG processing for BCI applications is then described in depth. The multivariate nature of the EEG combined with the neuroscience knowledge on hemispheric brain specialization are advantageously taken into account to derive spatial filters (i.e. acrossthe EEG electrodes) to analyze the patterns resulting from motor imagery, visual evoked potentials, and the P300 paradigm.

Brain-computer interfaces (BCIs) have been emerging during the last twenty years, as plausible alternatives for offering communication and control to physically challenged people. Yet, the idea of achieving control by simply thinking is appealing to a wider range of users. Availability, lower cost, and convenience make the electroencepha-log ram (EEG) the preferred choice in brain monitoring for BCI. Most of current BCI implementations rely on motor imagery as a neuro-physiological mechanism to achieve command and control. Limb movement imagination (e.g. left/right hand, foot) is particularly suitable forBCI because its corresponding brain activity, localized in the primary motor cortex, shares various features with actual movement.The EEG patterns resulting from motor imagery can be characterized in termsof spatial and frequency-band localization which are subject dependent. A technique termed common-spatial-patterns (CSP) permits to identify such space-frequency signatures in an automated way. Thistutorial document which is based on a tutorial presented at the IEEE/BiOCAS (Intelligent Biomedical Systems) conference in 2008, surveys the non-invasive BCI research field and focuses on motor imageryby presenting the details of the EEG signal processing and by illustrating them with results on six subjects performing three sorts ofmotor imagery, namely right hand movement, left hand movement, and footmovement.

The EU-project BRAIN aims at developing practical, user-friendly Brain-computer interfaces (BCI) to be used as communication restoration devices for the physically challenged. Current BCIs are laboratory prototypes and require expert assistance for successful operation. The description of Task 3.1 (quoted below) regards SSVEP based BCI.

The dissemination and exploitation strategies of BRAIN are reportedin this document. Dissemination activities included maintaining theproject website http://www.brain-project.org/, demonstrating BRAIN’sBCI system at large events such as Hannover fair 2010 and CeBIT 2011, drawing media attention on BRAIN developments, and writing publications for both scientific journal and specialized conferences. Theexploitation activities mainly focused on setting up a plan for commercial exploitation of main BRAIN achievements. A total of six mainachievements were identified in BRAIN. 1) The electrodes that use water instead of conductive gel, 2) the use of high flicker frequencyto evoke SSVEP and phase modulation, 3) the flexible visual stimulation appliance, 4) the wizard software platform to easily customizeBCI operation to a particular user, 5) the intuitive GUI, and 6) theopen software platform OpenBCI. BRAIN received significant attention during the public demonstrations at the Hannover Fair 2010 and CeBIT 2011 as witnessed by the large media coverage reported in the press. These events did not only served the purpose of demonstratingthe BCI system but also made it possible to recruit a large number of volunteers, out of the booth visitors, to test the system. This allowed BRAIN to publish important results linking demographics factors with the BCI operation. BRAIN resulted in a large number of scientific publications. Because of rapid dissemination and the possibility to interact with the research community, the results were primarily disseminated through attendance to specialized conferences and workshops. BRAIN was present at major BCI events and other events organized by EU projects. After considering the options for commercialexploitation of BRAIN results, it was decided to focus on the application on communication restoration and to target, at the first instance, the research market. BRAIN has unique features that stem fromthe main achievements in the project. The envisioned product is a phase modulated HF-SSVEP based BCI that uses a flexible visual stimulation appliance and records EEG with electrodes that use water instead of conductive gel. The software platform of this system will be based on openBCI and integrate the wizard and intuitive GUI platforms.From a benchmark against a competing system such as the g.tec’s intendiX, it appears that the BRAIN system can offer higher performanceand higher user-friendliness while being comparable in terms of price. The application and target market are such that the most appropriate entity to lead the commercialization of this system is TMSi. TMSi’s existing sales channels will be utilized to bring this systemto the market. Prospective launching of the system can happen in 2013. BRAIN was part of large cluster of EU-projects dealing with brain and neural computer interfaces. Through regular participation in various events organized by this cluster, BRAIN could provide valuable input to the effort on BCI standardization and to the organizationof a BNCI roadmap.

Brain-Computer Interfaces (BCIs) enable people to control appliances without involving the normal output pathways of peripheral nervesand muscles. A particularly promising type of BCI is based on the Steady-State Visual Evoked Potential (SSVEP). Users can selectcommands by focusing their attention on repetitive visual stimuli(RVSi) that change one of their properties (e.g. color or pattern) with a certain frequency. These properties as well as the devicethe RVSi are rendered on, can greatly affect the performance, applicability, comfort and safety of the BCI. Despite this fact, stimulation properties have received fairly little attention in the BCI literature to this date. Furthermore, a heavy emphasis is placedon BCI performance to the detriment of other important factors suchas comfort and safety. The research reported in this document aimsat studying the effects of stimulation properties on performance aswell as comfort of SSVEP-based BCIs. Research was performed in bothoffline and online settings, using a custom made high-performance BCI. Comfort was measured using a custom questionnaire. A largevariability across subjects was found, but the results confirm that stimulation properties have a considerable impact on performance and comfort of SSVEP based BCIs. In general, a large difference between stimulation states is beneficial for BCI performance, but detrimen-tal to user comfort. A couple of configurations were found that provide a good compromise between comfort and performance.

A brain computer interface (or BCI) is a communication or control system which does not need any neuromuscular activity to produce a message or action. In this project, an SSVEP based BCI is used. This means that when the user concentrates on a LED, flickering with a certain frequency, the steady state visual evoked potential that is caused by this flickering and measured with EEG electrode Oz, will control the system. The amplitude of the SSVEP is expected to bedifferent in each subject. This inter subject difference motivatesfor BCI calibration, since a higher SSVEP amplitude implies a better performance of the BCI. The inter subject difference is investigatedin this project. Further, for calibration, it is convenient to use ashort (<10 min) calibration sequence, with which the SSVEP amplitudeis predicted. System identification techniques are used in this pro-cess. In this project, linear and nonlinear approaches are used to predict the SSVEP. The reliability of this prediction is investigated.

This document describes the High Frequency (HF) Steady-State Visual Evoked Potential (SSVEP) based Brain Computer Interface (BCI) developed at Philips Research Europe (PRE). The interface is based on the fact that the oscillatory visual stimuli can elicit oscillatory brain activity at the same frequency or at that of higher harmonics. This signal can be detected by analyzing human electroencephalographic (EEG) recordings. This HF SSVEP BCI system is especially geared towards using high frequencies (above 30Hz) and is utilizing spatial filtering of the EEG signals in the occipital region to extract the dominant frequency component. Our HF SSVEP BCI system supports both calibration procedure and normal operation where brain signals are used for control and/or communication. The calibration is done for aparticular user and prior to the operation. The system setup includes three main components: an electronic device for EEG signal acquisition, a processing unit for EEG signal processing, and an electronic devicefor rendering oscillatory stimuli. The HF SSVEP BCI is designed with the possibility to interface with arbitrary devices or software application. This document reports on the software platform and the operation of the HF SSVEP BCI. The algorithmic details can be found in [1].

The goal of dry electrode development activity within the WP2 is tobuild a dry electrode prototype for brain wave sensing that is comfortable for the user and provides sufficient signal quality. The electrodes are to be utilized in BCI applications, namely Steady-StateVisually Evoked Potential (SSVEP),Event Related Synchronization andDe-synchronization (ERD/ERS),and P300 based BCIs. Due to the statusof the dry electrode technology and our non-encouraging results onthe evaluation of the contactless dry sensors we re-focused our efforts in developing an EEG system using dry electrodes that have galvanic contact to the human scalp. The first goal we set is to reliably detect the alpha brain rhythm (brain waves in the range from 8 to12Hz) as these brain waves are the most prominent ones in the EEG spectrum.The outcome of the evaluation presented here is that the signal quality of dry electrodes is sufficient to reliably measure alpha brainactivity. This is con-firmed through user studies with the medicallycertified amplifier - Mobi from TMSi. However, due to the skin contact problems and high input impedance, reported in the deliverable,robustness of dry-electrode (in combination with the amplifier) hasto be further improved. In particular, the dry electrode design andamplifier front end have to be further optimized.The robustness of the dry electrode-amplifier combination has to befurther improved to achieve a stable signal and reliable performancewhen measuring brain waves of people with long and thick hair. Thefollowing directions for further developments are envisioned:- Optimization of dry electrode design, focusing on- electrode material, i.e., using “bio-approved” materials such asgold and silver/silver-chloride used in this deliverable that willenable low impedance to the skin- number of pins to achieve good contact and increase the comfort- Optimization of amplifier front-end to cope with variation in input impedance- Optimize existing amplifier technology for usage with the developed dry electrodes.

Focusing of attention on a repetitive visual stimulation (RVS) at aconstant frequency, elicits the so called steady-state visual evokedpotential (SSVEP). This effect can be advantageously utilized in brain-computer interfaces (BCIs). SSVEP based BCIs can offer higher bitrates and require shorter training time as compared to other BCI modalities. Detection of the SSVEP from the EEG can be facilitated through spatial filtering (linear combination of the signals recorded at several electrodes). Literature offers several options to performthis. In this paper we propose a taxonomy to categorize these methods and we extensively evaluate them using 22 stimulation frequencies.We suggest improvements to existing methods to increase the SSVEP detection performance. We also consider practical aspects in the discussion of results.

Brain-computer interfaces (BCI) based on Steady State Visual EvokedPotential (SSVEP) can provide higher information transfer rate thanother BCI modalities. For the sake of safety and comfort, the frequency of the repetitive visual stimulus (RVS) necessary to elicit an SSVEP, should be higher than 30 Hz. However, in the frequency rangeabove 30 Hz, only a limited number of frequencies can elicit sufficiently strong SSVEPs for BCI purposes. Consequently, the conventionalapproach, consisting in presenting various repetitive visual stimuli having different frequency each, is not practical for SSVEP basedBCI functioning. Indeed this would bring low communication bitrates.In order to increase the number of possible repetitive visual stimuli, we consider modulating the phase of the stimulus instead of thefrequency. Thus, several stimuli, sharing the same frequency, but with different phase can be presented to the user. The approach presented in this document, to detect the phase of the stimulus is termedphase synchrony. It consists in using as feature, to identify a subject's focus of attention, the phase difference between the SSVEP and the stimulus. The phase is extracted through the Hilbert transformapplied on an univariate signal resulting from spatially filteringthe electroencephalogram.

Brain-computer interfaces (BCI) based on Steady State Visual EvokedPotential (SSVEP) can provide higher information transfer rate thanother BCI modalities. For the sake of safety and comfort, the frequency of the repetitive visual stimulus (RVS) necessary to elicit an SSVEP, should be higher than 30 Hz. However, in the frequency rangeabove 30 Hz, only a limited number of frequencies can elicit sufficiently strong SSVEPs for BCI purposes. Consequently, the conventionalapproach, consisting in presenting various repetitive visual stimuli having different frequency each, is not practical for SSVEP basedBCI functioning. Indeed this would bring low communication bitrates.In order to increase the number of possible repetitive visual stimuli, we consider modulating the phase of the stimulus instead of thefrequency. Thus, several stimuli, sharing the same frequency, but with different phase can be presented to the user. The approach presented in this document, to detect the phase of the stimulus is termedphase synchrony. It consists in using as feature, to identify a subject's focus of attention, the phase difference between the SSVEP and the stimulus. The phase is extracted through the Hilbert transformapplied on an univariate signal resulting from spatially filteringthe electroencephalogram. We have conducted experiments with seven subjects to estimate the information transfer rate that can be achieved using the phase synchrony analysis method.

Repetitive visual stimulation elicits specific brain responses knownas steady state visual evoked potentials (SSVEP). The SSVEP manifests as oscillatory components at the stimulation frequency or harmonics in brain signals such as the electroencephalogram (EEG) or magnetoencephalogram. Analysis of the dynamics of the SSVEP permits to characterize the neurophysiological basis of visual processing. Classical SSVEP analysis is restricted to the study of the EEG power at thestimulation frequency. In this paper, we focus on the dynamics of the alpha peak frequency under flicker stimulation. The alpha peak frequency is person specific and plays an important role on mental load. High resolution time-frequency methods are necessary to preciselyidentify the alpha peak frequency. We utilize therefore matching pursuit methods using stochastic dictionaries. We show that the alphapeak frequency decreases during flicker stimulation. We argue that this phenomenon can partially explain the relative fast habituation of the SSVEP, i.e.~the strength of the SSVEP decreases after few seconds of continuous flicker stimulation.

Brain-computer interfaces (BCI) based on Steady State Visual Evoked Potential (SSVEP) can provide higher information transfer rate and require shorter calibration than BCIs based on other modalities. For safety and comfort, the frequency of the repetitive visual stimuli seliciting the SSVEP should be higher than 30 Hz. However, in such frequency range, only a limited number of frequencies can elicit sufficiently strong SSVEPs for BCI purposes. Thus, the conventional approach, consisting in presenting various repetitive visual stimuli at different frequencies, is not feasible for high frequencies. Indeed this would bring low communication bitrates. To increase the number of possible repetitive visual stimuli, we consider modulating the phase of the stimulus instead of the frequency. In this paper, we present an approach to reliably detect the stimulus phase from the recorded SSVEP.

A relative high power in the alpha band (8 - 13Hz) at a particular EEG position is believed to indicate an active inhibitory state of the corresponding cortical site. According to this hypothesis an active inhibitory state corresponds to a lower engagement of the cortical site and is accompanied by a higher engagement of other sites (possibly neighboring ones). In this project the influence of a repetitive visual stimulation on the alpha band was investigated. A special attention was paid to the difference between the individual alpha frequency and alpha power before and after the stimulation. Data was analyzed by using the matching pursuit time-frequency distribution that provides better resolution in time and frequency than STFT or Wigner-Ville transforms.

Recent technological advances in the field of skin electrodes and on-body sensors indicate a possibility of having an alternative to the traditionally used conductive gel electrodes for measuring electrical signals of the brain (electroencephalogram, EEG). This paper evaluates whether water-based and dry contact electrode solutions can replace the gel ones. The quality of the obtained signal by three headsets, each using 8 electrodes of a different type, is estimated onthe steady state visual evoked potential (SSVEP) brain-computer interface (BCI) use case. The stimuli frequencies in the low (12 to 21Hz) and high (28 to 40Hz) frequency domain were used. Six people, that had different hair length and type, participated in the experiment. SSVEP response in terms of power spectra across different electrodes is compared and the impact of noise on temporal characteristics ofthe response is discussed. For people with shorter hair style the performance of water-based and dry electrodes comes close to the gelones in the optimal setting. On average, the classification accuracy of 0.63 for dry and 0.88 for water-based electrodes is achieved, compared to the 0.96 obtained for gel electrodes. The theoretical maximum of the average information transfer rate across participants was 23bpm for dry, 38bpm for water-based and 67bpm for gel electrodes. Furthermore, the convenience level of all three setups was seen as comparable. These results demonstrate that, having optimized headset and electrode design for dry and water-based electrodes for people with different hair length and type, dry and water-based electrode scan replace gel ones in BCIs and Neurofeedback applications where lower communication speed is acceptable.

Steady-State Visual Evoked Potential (SSVEP) is an oscillatory electrical response appearing in the electroencephalogram (EEG) in response to flicker stimulation. The SSVEP manifests more prominently in electrodes located near the visual cortex and has oscillatory components at the stimulation frequency and/or harmonics. The phase and amplitude of the SSVEP are sensitive to stimulus parameters such as frequency, modu-lation depth, and spatial frequency. Research related to SSVEP and the human visual system has mainly focused on brain computer interfaces (BCI) applications, cognitive and memory performance, pathophysiology of diseases. Some other research has been focus inthe influence of light properties (i.e. colour and size of stimuli)on brain activity. Sensitivity to flicker can be studied from a perception viewpoint. By presenting flickering light to an observer itis possible to find the frequency and modulation depth that is required to detect that the light is flickering rather than steady. Thesensitivity varies between people and change due to factors like age, concentration and fatigue, while for the stimulus; colour, background an intensity of the light have an influence on the perception. Also, we can find various studies on the response of human visual per-ception to flicker. However, very little research has been investigated about the SSVEP sensitivity base on the influence of the frequency and modulation depth of flicker. In this work, we present ZEERunit, a criterion to measure the intensity of SSVEP. This unit allows us to quantify the oscillatory response of the brain to flicker. With the ZEER unit we are able to make the estimation of the SSVEP sensitivity curve according to the frequency and modulation depth of flicker. We implemented an experiment to acquire the visual perception and SSVEP response to the flicker stimulation with goal of characterizing the link between visual perception of flicker and the corresponding SSVEP response. An experiment was conducted where 25 flicker stimuli with different properties were presented to 12 voluntary participants. The flicker stimuli result from the combina-tion of five different frequencies and five different Modulation Depths (MDs).The MDs were selected around the values defined by a perception curve which defines the relation between perception and MD for a given flicker frequency. In the study, the EEG and the visual perception data from each participant were collected. The EEG data was pre-processing by peak-filtering, subsampling, artifact correction and averaging the result. Then we applied statistical analysis on the distribution of the samples before and during stimulation. The results were spatially analysed all over the scalp and we used 2 different methodsfor the estimation of SSVEP sensitivity and flicker perception curve: absolute modulation depth and psychometric method. The results of the estimated curves indicate that visual perception of flicker and the SSVEP sensitivity are not aligned. We can start to see entrainment at stimuli frequencies below ~35 Hz only for flicker that are perceived by the observer. On the contrary, for higher stimuli frequencies it is possible to elicit oscillatory responses below the visual perception of flicker.

A brain computer interface (BCI) is a communication or control system which does not need any neuromuscular activity to produce the message or action. Therefore it can be a communication solution for those people with severe neuromuscular disorders as well as for healthy people in practical applications in entertainment, and communication restoration. SSVEPs are oscillatory responses elicited by a flickering light stimulus that can be recorded in the electroencephalogram(EEG). It is a repetitive evoked potential whose constituent discrete frequency components remain constant in amplitude and phase over aninfinitive long time period. The SSVEP response consists of a number of sinusoids with frequencies given by the stimulus flickering frequency and a number of harmonics. This study is based on brain computer interfaces based on steady-state-visual-evoked-potentials. Thebrain regions where the response is produced has been demonstrated along this work to be stable for a certain frequency and so on spatial filters that discriminate the regions responsible of the SSVEP response have been consider a good solution. We present an approach to automatically obtain the optimum spatial filter to detect the SSVEP at a given stimulation frequency using short signal segments where the stimulation is presented at intermittent periods interspersed with breaks. The best harmonic combination is calculated and asan output all the necessary tools for a suitable detection, thresholdand detection filters.

In order to guide our actions, the human brain relies most on the information that is received from the visual sense. However, the visual input the brain receives through the eyes is often subject to noise and ambiguity. Consider for example a person driving a car throughthe fog. In this situation, the visual information that the brain receives is severely degraded. It is obvious that the brain needs a way to resolve such ambiguities in order to be able to provide the reliable visual information necessary to guide our actions. The neuralmechanisms that are involved in resolving ambiguity in the visual input have been the subject of many studies for over a century. In recent years, many studies have used psychophysics and neuroimaging toinvestigate how and where the brain processes and resolves ambiguous visual information. In this study we used a structure-from-motionstimulus to investigate the neural responses during the viewing of an ambiguous stimulus. To investigate these neural responses underlying ambiguous visual perception we used a novel approach. Intracranial electroencephalography is a relatively new method that provides several advantages over more commonly used methods such as scalp EEGor magnetic resonance imaging. The electrodes that are used to record the electrical activity of the brain are implanted directly on thesurface of the brain. This allows for more accurate measurements ofbrain activity becausethere is no interference from the skull and skin which cause problems in scalp EEG.Using intracranial EEG, we identified two separate locations in the left hemisphere of one partici-pant that showed a significant difference in alpha band power during a switch in perception when viewing an unambiguous stimulus compared to viewing an ambiguous stimulus. This difference could indicatethat the neuronal populations at theselocations respond preferentially to bottom-up visual information and not to top-down visual information. A subsequent analysis of the time-frequency spectrum revealedno significant differences but clearly showed a decrease in power in the lower frequencies before the participant responded to a changein the stimulus. This decrease in power is likely related to the preparation of the motor response. The results of this study show thatintracranial EEG is useful new method that can provide an improvement over both scalp EEG and MRI.

Brain computer interfaces (BCI) that use the steady-state-visual-evoked-potential (SSVEP) as neural source, offer two main advantages over other types of BCIs: shorter calibration times and higher information transfer rates. SSVEPs elicited by high frequency (larger than30 Hz) repetitive visual stimulation are less prone to cause visualfatigue, safer, and more comfortable for the user. However in the high frequency range there is a practical limitation because only fewfrequencies can elicit sufficiently strong SSVEPs for BCI purposes.We bypass this limitation by using only one stimulation frequency and different phases. To detect the phase from the recorded SSVEP, weuse spatial filtering combined to phase synchrony analysis. We developed an online BCI implementation which was tested on six subjects and resulted on an average accuracy of 95.5% and an average bit rateof 34 bits-per-minute. Our approach has the advantage of entailing only minimal visual annoyance for the user.