Turning gaming EEG peripherals into trainable brain computer interfaces

Companies such as NeuroSky and Emotiv Systems are selling non-medical EEG devices for human computer interaction. These devices are significantly more affordable than their medical counterparts, and are mainly used to measure levels of engagement, focus, relaxation and stress. This information is sought after for marketing research and games. However, these EEG devices have the potential to enable users to interact with their surrounding environment using thoughts only, without activating any muscles. In this paper, we present preliminary results that demonstrate that despite reduced voltage and time sensitivity compared to medical-grade EEG systems, the quality of the signals of the Emotiv EPOC neuroheadset is sufficiently good in allowing discrimina tion between imaging events. We collected streams of EEG raw data and trained different types of classifiers to discriminate between three states (rest and two imaging events). We achieved a generalisation error of less than 2% for two types of non-linear classifiers.

Single-trial event-related potential analysis for brain-computer interfaces

Abstract. Different types of mental activity are utilised as an input in Brain-Computer Interface (BCI) systems. One such activity type is based on Event-Related Potentials (ERPs). The characteristics of ERPs are not visible in single-trials, thus averaging over a number of trials is necessary before the signals become usable. An improvement in ERP-based BCI operation and system usability could be obtained if the use of single-trial ERP data was possible. The method of Independent Component Analysis (ICA) can be utilised to separate single-trial recordings of ERP data into components that correspond to ERP characteristics, background electroencephalogram (EEG) activity and other components with non- cerebral origin. Choice of specific components and their use to reconstruct “denoised” single-trial data could improve the signal quality, thus allowing the successful use of single-trial data without the need for averaging. This paper assesses single-trial ERP signals reconstructed using a selection of estimated components from the application of ICA on the raw ERP data. Signal improvement is measured using Contrast-To-Noise measures. It was found that such analysis improves the signal quality in all single-trials.

Brain-Computer Interfaces for augmented communication: Asynchronous and adaptive algorithms and evaluation with end users

This thesis aimed at addressing some of the issues that, at the state of the art, avoid the P300-based brain computer interface (BCI) systems to move from research laboratories to end users’ home. An innovative asynchronous classifier has been defined and validated. It relies on the introduction of a set of thresholds in the classifier, and such thresholds have been assessed considering the distributions of score values relating to target, non-target stimuli and epochs of voluntary no-control. With the asynchronous classifier, a P300-based BCI system can adapt its speed to the current state of the user and can automatically suspend the control when the user diverts his attention from the stimulation interface. Since EEG signals are non-stationary and show inherent variability, in order to make long-term use of BCI possible, it is important to track changes in ongoing EEG activity and to adapt BCI model parameters accordingly. To this aim, the asynchronous classifier has been subsequently improved by introducing a self-calibration algorithm for the continuous and unsupervised recalibration of the subjective control parameters. Finally an index for the online monitoring of the EEG quality has been defined and validated in order to detect potential problems and system failures. This thesis ends with the description of a translational work involving end users (people with amyotrophic lateral sclerosis-ALS). Focusing on the concepts of the user centered design approach, the phases relating to the design, the development and the validation of an innovative assistive device have been described. The proposed assistive technology (AT) has been specifically designed to meet the needs of people with ALS during the different phases of the disease (i.e. the degree of motor abilities impairment). Indeed, the AT can be accessed with several input devices either conventional (mouse, touchscreen) or alterative (switches, headtracker) up to a P300-based BCI.

Development of Brain-Computer Interfaces using Evolvable Hardware

Brain-Computer Interfaces are usually tackled from a medical point of view, correlating observed phenomena to physical facts known about the brain. Existing methods of classification lie in the application of deterministic algorithms and depend on certain degree of knowledge about the underlying phenomena so as to process data. In this demo, different architectures for an evolvable hardware classifier implemented on an FPGA are proposed, in line with the objective of generalizing evolutionary algorithms regardless of the application.

Brain-Computer Interfaces: Desarrollo de un sistema de identificación de estados mentales alfa

En el mundo actual las aplicaciones basadas en sistemas biométricos, es decir, aquellas que miden las señales eléctricas de nuestro organismo, están creciendo a un gran ritmo. Todos estos sistemas incorporan sensores biomédicos, que ayudan a los usuarios a controlar mejor diferentes aspectos de la rutina diaria, como podría ser llevar un seguimiento detallado de una rutina deportiva, o de la calidad de los alimentos que ingerimos. Entre estos sistemas biométricos, los que se basan en la interpretación de las señales cerebrales, mediante ensayos de electroencefalografía o EEG están cogiendo cada vez más fuerza para el futuro, aunque están todavía en una situación bastante incipiente, debido a la elevada complejidad del cerebro humano, muy desconocido para los científicos hasta el siglo XXI. Por estas razones, los dispositivos que utilizan la interfaz cerebro-máquina, también conocida como BCI (Brain Computer Interface), están cogiendo cada vez más popularidad. El funcionamiento de un sistema BCI consiste en la captación de las ondas cerebrales de un sujeto para después procesarlas e intentar obtener una representación de una acción o de un pensamiento del individuo. Estos pensamientos, correctamente interpretados, son posteriormente usados para llevar a cabo una acción. Ejemplos de aplicación de sistemas BCI podrían ser mover el motor de una silla de ruedas eléctrica cuando el sujeto realice, por ejemplo, la acción de cerrar un puño, o abrir la cerradura de tu propia casa usando un patrón cerebral propio. Los sistemas de procesamiento de datos están evolucionando muy rápido con el paso del tiempo. Los principales motivos son la alta velocidad de procesamiento y el bajo consumo energético de las FPGAs (Field Programmable Gate Array). Además, las FPGAs cuentan con una arquitectura reconfigurable, lo que las hace más versátiles y potentes que otras unidades de procesamiento como las CPUs o las GPUs.En el CEI (Centro de Electrónica Industrial), donde se lleva a cabo este TFG, se dispone de experiencia en el diseño de sistemas reconfigurables en FPGAs. Este TFG es el segundo de una línea de proyectos en la cual se busca obtener un sistema capaz de procesar correctamente señales cerebrales, para llegar a un patrón común que nos permita actuar en consecuencia. Más concretamente, se busca detectar cuando una persona está quedándose dormida a través de la captación de unas ondas cerebrales, conocidas como ondas alfa, cuya frecuencia está acotada entre los 8 y los 13 Hz. Estas ondas, que aparecen cuando cerramos los ojos y dejamos la mente en blanco, representan un estado de relajación mental. Por tanto, este proyecto comienza como inicio de un sistema global de BCI, el cual servirá como primera toma de contacto con el procesamiento de las ondas cerebrales, para el posterior uso de hardware reconfigurable sobre el cual se implementarán los algoritmos evolutivos. Por ello se vuelve necesario desarrollar un sistema de procesamiento de datos en una FPGA. Estos datos se procesan siguiendo la metodología de procesamiento digital de señales, y en este caso se realiza un análisis de la frecuencia utilizando la transformada rápida de Fourier, o FFT. Una vez desarrollado el sistema de procesamiento de los datos, se integra con otro sistema que se encarga de captar los datos recogidos por un ADC (Analog to Digital Converter), conocido como ADS1299. Este ADC está especialmente diseñado para captar potenciales del cerebro humano. De esta forma, el sistema final capta los datos mediante el ADS1299, y los envía a la FPGA que se encarga de procesarlos. La interpretación es realizada por los usuarios que analizan posteriormente los datos procesados. Para el desarrollo del sistema de procesamiento de los datos, se dispone primariamente de dos plataformas de estudio, a partir de las cuales se captarán los datos para después realizar el procesamiento: 1. La primera consiste en una herramienta comercial desarrollada y distribuida por OpenBCI, proyecto que se dedica a la venta de hardware para la realización de EEG, así como otros ensayos. Esta herramienta está formada por un microprocesador, un módulo de memoria SD para el almacenamiento de datos, y un módulo de comunicación inalámbrica que transmite los datos por Bluetooth. Además cuenta con el mencionado ADC ADS1299. Esta plataforma ofrece una interfaz gráfica que sirve para realizar la investigación previa al diseño del sistema de procesamiento, al permitir tener una primera toma de contacto con el sistema. 2. La segunda plataforma consiste en un kit de evaluación para el ADS1299, desde la cual se pueden acceder a los diferentes puertos de control a través de los pines de comunicación del ADC. Esta plataforma se conectará con la FPGA en el sistema integrado. Para entender cómo funcionan las ondas más simples del cerebro, así como saber cuáles son los requisitos mínimos en el análisis de ondas EEG se realizaron diferentes consultas con el Dr Ceferino Maestu, neurofisiólogo del Centro de Tecnología Biomédica (CTB) de la UPM. Él se encargó de introducirnos en los distintos procedimientos en el análisis de ondas en electroencefalogramas, así como la forma en que se deben de colocar los electrodos en el cráneo. Para terminar con la investigación previa, se realiza en MATLAB un primer modelo de procesamiento de los datos. Una característica muy importante de las ondas cerebrales es la aleatoriedad de las mismas, de forma que el análisis en el dominio del tiempo se vuelve muy complejo. Por ello, el paso más importante en el procesamiento de los datos es el paso del dominio temporal al dominio de la frecuencia, mediante la aplicación de la transformada rápida de Fourier o FFT (Fast Fourier Transform), donde se pueden analizar con mayor precisión los datos recogidos. El modelo desarrollado en MATLAB se utiliza para obtener los primeros resultados del sistema de procesamiento, el cual sigue los siguientes pasos. 1. Se captan los datos desde los electrodos y se escriben en una tabla de datos. 2. Se leen los datos de la tabla. 3. Se elige el tamaño temporal de la muestra a procesar. 4. Se aplica una ventana para evitar las discontinuidades al principio y al final del bloque analizado. 5. Se completa la muestra a convertir con con zero-padding en el dominio del tiempo. 6. Se aplica la FFT al bloque analizado con ventana y zero-padding. 7. Los resultados se llevan a una gráfica para ser analizados. Llegados a este punto, se observa que la captación de ondas alfas resulta muy viable. Aunque es cierto que se presentan ciertos problemas a la hora de interpretar los datos debido a la baja resolución temporal de la plataforma de OpenBCI, este es un problema que se soluciona en el modelo desarrollado, al permitir el kit de evaluación (sistema de captación de datos) actuar sobre la velocidad de captación de los datos, es decir la frecuencia de muestreo, lo que afectará directamente a esta precisión. Una vez llevado a cabo el primer procesamiento y su posterior análisis de los resultados obtenidos, se procede a realizar un modelo en Hardware que siga los mismos pasos que el desarrollado en MATLAB, en la medida que esto sea útil y viable. Para ello se utiliza el programa XPS (Xilinx Platform Studio) contenido en la herramienta EDK (Embedded Development Kit), que nos permite diseñar un sistema embebido. Este sistema cuenta con: Un microprocesador de tipo soft-core llamado MicroBlaze, que se encarga de gestionar y controlar todo el sistema; Un bloque FFT que se encarga de realizar la transformada rápida Fourier; Cuatro bloques de memoria BRAM, donde se almacenan los datos de entrada y salida del bloque FFT y un multiplicador para aplicar la ventana a los datos de entrada al bloque FFT; Un bus PLB, que consiste en un bus de control que se encarga de comunicar el MicroBlaze con los diferentes elementos del sistema. Tras el diseño Hardware se procede al diseño Software utilizando la herramienta SDK(Software Development Kit).También en esta etapa se integra el sistema de captación de datos, el cual se controla mayoritariamente desde el MicroBlaze. Por tanto, desde este entorno se programa el MicroBlaze para gestionar el Hardware que se ha generado. A través del Software se gestiona la comunicación entre ambos sistemas, el de captación y el de procesamiento de los datos. También se realiza la carga de los datos de la ventana a aplicar en la memoria correspondiente. En las primeras etapas de desarrollo del sistema, se comienza con el testeo del bloque FFT, para poder comprobar el funcionamiento del mismo en Hardware. Para este primer ensayo, se carga en la BRAM los datos de entrada al bloque FFT y en otra BRAM los datos de la ventana aplicada. Los datos procesados saldrán a dos BRAM, una para almacenar los valores reales de la transformada y otra para los imaginarios. Tras comprobar el correcto funcionamiento del bloque FFT, se integra junto al sistema de adquisición de datos. Posteriormente se procede a realizar un ensayo de EEG real, para captar ondas alfa. Por otro lado, y para validar el uso de las FPGAs como unidades ideales de procesamiento, se realiza una medición del tiempo que tarda el bloque FFT en realizar la transformada. Este tiempo se compara con el tiempo que tarda MATLAB en realizar la misma transformada a los mismos datos. Esto significa que el sistema desarrollado en Hardware realiza la transformada rápida de Fourier 27 veces más rápido que lo que tarda MATLAB, por lo que se puede ver aquí la gran ventaja competitiva del Hardware en lo que a tiempos de ejecución se refiere. En lo que al aspecto didáctico se refiere, este TFG engloba diferentes campos. En el campo de la electrónica:  Se han mejorado los conocimientos en MATLAB, así como diferentes herramientas que ofrece como FDATool (Filter Design Analysis Tool).  Se han adquirido conocimientos de técnicas de procesado de señal, y en particular, de análisis espectral.  Se han mejorado los conocimientos en VHDL, así como su uso en el entorno ISE de Xilinx.  Se han reforzado los conocimientos en C mediante la programación del MicroBlaze para el control del sistema.  Se ha aprendido a crear sistemas embebidos usando el entorno de desarrollo de Xilinx usando la herramienta EDK (Embedded Development Kit). En el campo de la neurología, se ha aprendido a realizar ensayos EEG, así como a analizar e interpretar los resultados mostrados en el mismo. En cuanto al impacto social, los sistemas BCI afectan a muchos sectores, donde destaca el volumen de personas con discapacidades físicas, para los cuales, este sistema implica una oportunidad de aumentar su autonomía en el día a día. También otro sector importante es el sector de la investigación médica, donde los sistemas BCIs son aplicables en muchas aplicaciones como, por ejemplo, la detección y estudio de enfermedades cognitivas.

Non-invasive Detection of Cortical Control Signals for Brain-Computer Interfaces

Thesis (Ph.D.)--University of Washington, 2016-05

Comparing interaction techniques for serious games through brain-computer interfaces:a user perception evaluation study

This paper examines the application of commercial and non-invasive electroencephalography (EEG)-based brain-computer (BCIs) interfaces with serious games. Two different EEG-based BCI devices were used to fully control the same serious game. The first device (NeuroSky MindSet) uses only a single dry electrode and requires no calibration. The second device (Emotiv EPOC) uses 14 wet sensors requiring additional training of a classifier. User testing was performed on both devices with sixty-two participants measuring the player experience as well as key aspects of serious games, primarily learnability, satisfaction, performance and effort. Recorded feedback indicates that the current state of BCIs can be used in the future as alternative game interfaces after familiarisation and in some cases calibration. Comparative analysis showed significant differences between the two devices. The first device provides more satisfaction to the players whereas the second device is more effective in terms of adaptation and interaction with the serious game.

Motor imagery induced EEG patterns in spinal cord injury patients and their impact on Brain-Computer Interface accuracy

OBJECTIVE: Assimilating the diagnosis complete spinal cord injury (SCI) takes time and is not easy, as patients know that there is no 'cure' at the present time. Brain-computer interfaces (BCIs) can facilitate daily living. However, inter-subject variability demands measurements with potential user groups and an understanding of how they differ to healthy users BCIs are more commonly tested with. Thus, a three-class motor imagery (MI) screening (left hand, right hand, feet) was performed with a group of 10 able-bodied and 16 complete spinal-cord-injured people (paraplegics, tetraplegics) with the objective of determining what differences were present between the user groups and how they would impact upon the ability of these user groups to interact with a BCI. APPROACH: Electrophysiological differences between patient groups and healthy users are measured in terms of sensorimotor rhythm deflections from baseline during MI, electroencephalogram microstate scalp maps and strengths of inter-channel phase synchronization. Additionally, using a common spatial pattern algorithm and a linear discriminant analysis classifier, the classification accuracy was calculated and compared between groups. MAIN RESULTS: It is seen that both patient groups (tetraplegic and paraplegic) have some significant differences in event-related desynchronization strengths, exhibit significant increases in synchronization and reach significantly lower accuracies (mean (M) = 66.1%) than the group of healthy subjects (M = 85.1%). SIGNIFICANCE: The results demonstrate significant differences in electrophysiological correlates of motor control between healthy individuals and those individuals who stand to benefit most from BCI technology (individuals with SCI). They highlight the difficulty in directly translating results from healthy subjects to participants with SCI and the challenges that, therefore, arise in providing BCIs to such individuals.

An optimized ERP Brain-computer interface based on facial expression changes

OBJECTIVE: Interferences from spatially adjacent non-target stimuli are known to evoke event-related potentials (ERPs) during non-target flashes and, therefore, lead to false positives. This phenomenon was commonly seen in visual attention-based brain-computer interfaces (BCIs) using conspicuous stimuli and is known to adversely affect the performance of BCI systems. Although users try to focus on the target stimulus, they cannot help but be affected by conspicuous changes of the stimuli (such as flashes or presenting images) which were adjacent to the target stimulus. Furthermore, subjects have reported that conspicuous stimuli made them tired and annoyed. In view of this, the aim of this study was to reduce adjacent interference, annoyance and fatigue using a new stimulus presentation pattern based upon facial expression changes. Our goal was not to design a new pattern which could evoke larger ERPs than the face pattern, but to design a new pattern which could reduce adjacent interference, annoyance and fatigue, and evoke ERPs as good as those observed during the face pattern. APPROACH: Positive facial expressions could be changed to negative facial expressions by minor changes to the original facial image. Although the changes are minor, the contrast is big enough to evoke strong ERPs. In this paper, a facial expression change pattern between positive and negative facial expressions was used to attempt to minimize interference effects. This was compared against two different conditions, a shuffled pattern containing the same shapes and colours as the facial expression change pattern, but without the semantic content associated with a change in expression, and a face versus no face pattern. Comparisons were made in terms of classification accuracy and information transfer rate as well as user supplied subjective measures. MAIN RESULTS: The results showed that interferences from adjacent stimuli, annoyance and the fatigue experienced by the subjects could be reduced significantly (p < 0.05) by using the facial expression change patterns in comparison with the face pattern. The offline results show that the classification accuracy of the facial expression change pattern was significantly better than that of the shuffled pattern (p < 0.05) and the face pattern (p < 0.05). SIGNIFICANCE: The facial expression change pattern presented in this paper reduced interference from adjacent stimuli and decreased the fatigue and annoyance experienced by BCI users significantly (p < 0.05) compared to the face pattern.

Facilitating effects of transcranial direct current stimulation on motor imagery brain-computer interface with robotic feedback for stroke rehabilitation

OBJECTIVE: To investigate the efficacy and effects of transcranial direct current stimulation (tDCS) on motor imagery brain-computer interface (MI-BCI) with robotic feedback for stroke rehabilitation. DESIGN: A sham-controlled, randomized controlled trial. SETTING: Patients recruited through a hospital stroke rehabilitation program. PARTICIPANTS: Subjects (N=19) who incurred a stroke 0.8 to 4.3 years prior, with moderate to severe upper extremity functional impairment, and passed BCI screening. INTERVENTIONS: Ten sessions of 20 minutes of tDCS or sham before 1 hour of MI-BCI with robotic feedback upper limb stroke rehabilitation for 2 weeks. Each rehabilitation session comprised 8 minutes of evaluation and 1 hour of therapy. MAIN OUTCOME MEASURES: Upper extremity Fugl-Meyer Motor Assessment (FMMA) scores measured end-intervention at week 2 and follow-up at week 4, online BCI accuracies from the evaluation part, and laterality coefficients of the electroencephalogram (EEG) from the therapy part of the 10 rehabilitation sessions. RESULTS: FMMA score improved in both groups at week 4, but no intergroup differences were found at any time points. Online accuracies of the evaluation part from the tDCS group were significantly higher than those from the sham group. The EEG laterality coefficients from the therapy part of the tDCS group were significantly higher than those of the sham group. CONCLUSIONS: The results suggest a role for tDCS in facilitating motor imagery in stroke.

Mental states monitoring through passive Brain-Computer Interface systems

The monitoring of cognitive functions aims at gaining information about the current cognitive state of the user by decoding brain signals. In recent years, this approach allowed to acquire valuable information about the cognitive aspects regarding the interaction of humans with external world. From this consideration, researchers started to consider passive application of brain–computer interface (BCI) in order to provide a novel input modality for technical systems solely based on brain activity. The objective of this thesis is to demonstrate how the passive Brain Computer Interfaces (BCIs) applications can be used to assess the mental states of the users, in order to improve the human machine interaction. Two main studies has been proposed. The first one allows to investigate whatever the Event Related Potentials (ERPs) morphological variations can be used to predict the users’ mental states (e.g. attentional resources, mental workload) during different reactive BCI tasks (e.g. P300-based BCIs), and if these information can predict the subjects’ performance in performing the tasks. In the second study, a passive BCI system able to online estimate the mental workload of the user by relying on the combination of the EEG and the ECG biosignals has been proposed. The latter study has been performed by simulating an operative scenario, in which the occurrence of errors or lack of performance could have significant consequences. The results showed that the proposed system is able to estimate online the mental workload of the subjects discriminating three different difficulty level of the tasks ensuring a high reliability.

Adaptive Brain-Computer Interface Systems For Communication in People with Severe Neuromuscular Disabilities

Brain-computer interfaces (BCI) have the potential to restore communication or control abilities in individuals with severe neuromuscular limitations, such as those with amyotrophic lateral sclerosis (ALS). The role of a BCI is to extract and decode relevant information that conveys a user's intent directly from brain electro-physiological signals and translate this information into executable commands to control external devices. However, the BCI decision-making process is error-prone due to noisy electro-physiological data, representing the classic problem of efficiently transmitting and receiving information via a noisy communication channel.

This research focuses on P300-based BCIs which rely predominantly on event-related potentials (ERP) that are elicited as a function of a user's uncertainty regarding stimulus events, in either an acoustic or a visual oddball recognition task. The P300-based BCI system enables users to communicate messages from a set of choices by selecting a target character or icon that conveys a desired intent or action. P300-based BCIs have been widely researched as a communication alternative, especially in individuals with ALS who represent a target BCI user population. For the P300-based BCI, repeated data measurements are required to enhance the low signal-to-noise ratio of the elicited ERPs embedded in electroencephalography (EEG) data, in order to improve the accuracy of the target character estimation process. As a result, BCIs have relatively slower speeds when compared to other commercial assistive communication devices, and this limits BCI adoption by their target user population. The goal of this research is to develop algorithms that take into account the physical limitations of the target BCI population to improve the efficiency of ERP-based spellers for real-world communication.

In this work, it is hypothesised that building adaptive capabilities into the BCI framework can potentially give the BCI system the flexibility to improve performance by adjusting system parameters in response to changing user inputs. The research in this work addresses three potential areas for improvement within the P300 speller framework: information optimisation, target character estimation and error correction. The visual interface and its operation control the method by which the ERPs are elicited through the presentation of stimulus events. The parameters of the stimulus presentation paradigm can be modified to modulate and enhance the elicited ERPs. A new stimulus presentation paradigm is developed in order to maximise the information content that is presented to the user by tuning stimulus paradigm parameters to positively affect performance. Internally, the BCI system determines the amount of data to collect and the method by which these data are processed to estimate the user's target character. Algorithms that exploit language information are developed to enhance the target character estimation process and to correct erroneous BCI selections. In addition, a new model-based method to predict BCI performance is developed, an approach which is independent of stimulus presentation paradigm and accounts for dynamic data collection. The studies presented in this work provide evidence that the proposed methods for incorporating adaptive strategies in the three areas have the potential to significantly improve BCI communication rates, and the proposed method for predicting BCI performance provides a reliable means to pre-assess BCI performance without extensive online testing.

Use of brain computer interface to drive : preliminary results

This paper reports on the implementation of a non-invasive electroencephalography-based brain-computer interface to control functions of a car in a driving simulator. The system is comprised of a Cleveland Medical Devices BioRadio 150 physiological signal recorder, a MATLAB-based BCI and an OKTAL SCANeR advanced driving experience simulator. The system utilizes steady-state visual-evoked potentials for the BCI paradigm, elicited by frequency-modulated high-power LEDs and recorded with the electrode placement of Oz-Fz with Fz as ground. A three-class online brain-computer interface was developed and interfaced with an advanced driving simulator to control functions of the car, including acceleration and steering. The findings are mainly exploratory but provide an indication of the feasibility and challenges of brain-controlled on-road cars for the future, in addition to a safe, simulated BCI driving environment to use as a foundation for research into overcoming these challenges.