|Language of instruction||English|
|Offered by||Radboud University; Faculty of Social Sciences; Artificial Intelligence; |
|SEM2|| (04/02/2019 to 12/07/2019)|
|Registration using OSIRIS||Yes|
|Course open to students from other faculties||Yes|
- Understand what is needed to build a Brain Computer Interface.
- Be able to implement a simple BCI experiment using the hardware/software tools available at the DCC.
- Be able to run a small BCI pilot experiment, analyze results, and report results in the scientific literature.
A Brain Computer Interface (BCI) is a device for translating user intentions, encoded by performing specific mental tasks, into control signals which can be used to control external devices. Whilst BCI builds on ideas and techniques from multiple disciplines including Neuroscience, Psychology, Cognitive Neuroscience, Signal Processing, Artificial Intelligence, it is fundamentally an experimental subject. All ideas/designs must be validated against real subjects generating real signals to prove their worth.
This course consist of three components:
- Lectures on Advanced Topics in Brain Computer Interfacing, focusing on advanced signal analysis and machine elarning techniques.
- Tutorials on how to develop a modern advanced BCI and conduct EEG BCI experiments using our in-house developed BCI framework.
- Assignment where students work in small teams (3-4 people) to apply the knowledge gained in the Lecture and Tutorials to design, implement and evaluate their own EEG based BCI system. Consisting of stimulus type/mental task, signal detection & processing and feedback/output generation.
During the assignment phase, teams will have access to all the facilities available in the department, including EEG-equipment, stimulus-generation software, signal-analysis software, and output-devices (e.g. video, audio, Lego-robots, internet-BCI-games, virtual-reality environments). Finally, they will test their design by conducting a set of experiments and report the results.
After developing their BCI students give a demonstration to prove it works correctly, conduct a set of experiments to validate this further and report on these experiments in the form of a scientific article.
Note: this course includes a significant practical component using a mix of EEG hardware and Python based analysis software.
- Individual Assingments
- Practical Assignments
- Poster project
- Individual Assignments 20%
- Written Examination 20%
- Group Project Report 30%
- Software product quality 30%
|Basic Python (or MATLAB) programming experience. |
Basic understanding of linear algebra and signal analysis, e.g. fourier decomposition of a signal -- as provided by taking "Mathematics for Artificial Intelligence" (BKI316) or equivalent.
Basic background in Brain Computer Interfaces -- as provided by taking "Introduction to Brain Computer Interfaces" (BKI323) or equivalent.
|Dr. J. Farquhar, T: ++00 24 3611938, E: J.email@example.com|
|BCI is a new field, thus there is no good text-book for this course. As such, most of the literature will be from the slides used to teach the course, the tutorial documentation, and from current research papers.|
|Gerven, M.V., J. Farquhar, R. Schaefer, R. Vlek, J. Geuze, A. Nijholt, N. Ramsey, P. Haselager, L. Vuurpijl, S. Gielen, and P. Desain, "The Brain-Computer Interface cycle," Journal of Neural Engineering (2009). Vol. 6: p. 041001.|
|Also, whilst not compulsory, the following book provides a good overview of current BCI technologies:
Wolpaw, J.R. & Wolpaw E.W., Eds., Brain-Computer Interfaces: Principles and Practice. Oxford; New York: Oxford University Press, 2012.|
|BCI Design Tutorials|
GeneralBCI Experiment design, implementation and testing project
General4 Advanced topics lectures, 2 hours per week
5 BCI tutorial sessions, 2x2 hours per week
General20 group workshops, 2x2 hours per week
|Opportunities||Block SEM2, Block SEM2|