Tutorial on Noninvasive Electroencephalogram-based Brain-Computer Interfaces


Organizer: João Luís Garcia Rosa

Department of Computer Science

University of Sao Paulo - Brazil

Abstract: Brain-Computer Interfaces (BCI) is a form of communication that enables individuals unable to perform movements to connect to external assistive devices using the electroencephalogram (EEG) or other brain signals. Noninvasive BCIs capture changes in blood flow or fluctuations in electric and magnetic fields caused by the activity of large populations of neurons. The EEG, a non-invasive technique, measures the electrical activity of the brain in different locations of the head, typically using electrodes placed on the scalp. With the proper removal of artifacts, signal processing and machine learning, human EEG carries enough information about the intention of planning and execution. Brain models based on neurodynamics seek to understand and represent the reasons why neurons are excitable cells. The microscopic electric current of each neuron adds with the currents from other neurons, which causes a difference in macroscopic electric potential, measured by EEG, which records the patterns of populations of neurons mesoscopic activity. That is, a good neural model must reproduce the dynamics of neurons, taking into account the dynamic properties of populations of neurons, in addition to the electrophysiological properties of individual neurons. The objectives of the proposed tutorial are to show how the understanding of electrical activity of the brain, measured noninvasively by EEG, can provide a way to allow communication without muscle movements. The intention is, from the study of the neurodynamic behavior of the brain, to investigate ways and propose models that enable the noninvasive brain-computer interfaces. In recent decades, the EEG-based BCIs have attracted the attention of researchers in the fields of neuroscience, neural engineering and clinical rehabilitation. The plan is to use the data obtained through BCI to analyze the pre-motor movements, changes in the brain that occur before there is actually a movement, and apply them to a proper handling of prosthetic devices.

Outline of the tutorial: Neurodynamics concepts, Hodgkin-Huxley model, attractors, EEG, BCI, neuron populations, mesoscopic approach, a biologically inspired neural network brain model based on a mesoscopic approach (Freeman K-sets).

Tutorial Topic: Computational Neuroscience

Rationale: A noninvasive tool to understand the brain without surgery in order to build useful computational models, like brain-computer interfaces.

Relevance for IJCNN: Computational brain models and biologically inspired neural networks are subjects of interest to IJCNN audience

Presentation Slides: [To be included]

Bio of the organizer: João Luís G. Rosa is a Professor Doctor at the Department of Computer Science in the Institute of Mathematics and Computer Sciences (ICMC) - University of São Paulo (USP), in São Carlos, Brazil. He is with the Bio-inspired Computing Laboratory (BioCom). His research interests include computational neurodynamics, brain-computer interfaces, and biologically plausible neural networks. About his academic experience, he has taught graduate level courses on computer science, disciplines Artificial Neural Networks, and Theory of Computation, and undergraduate level courses for computer engineering and computer science, disciplines Programming Languages, Artificial Intelligence, Algorithms, and Theory of Computation and Formal Languages, at University of São Paulo at São Carlos, São Paulo, Brazil. He has been interested on neural networks since 1990, and since 1998 he has contributed to the field of Biologically Plausible Artificial Neural Networks with published papers and supervision of undergraduate and graduate students. Since 2009, he has been a reviewer for the ACM Computing Reviews. He is also a reviewer for several periodicals and conferences. He has published two books (on Artificial Intelligence Fundamentals and on Formal Languages and Automata, both in Portuguese), two book chapters (the last one, in 2013, on Biologically Plausible Artificial Neural Networks), and papers in journals and conference proceedings.

Disclaimer: The opinions expressed in this web page and the presentation slides are that of the organizer, not of the IJCNN conference or IEEE, or any other entity.

Last update: February 23, 2015.