| | | | Course module | | SOW-BKI323 | Category | | - | Language of instruction | | English | Offered by | | Radboud University; Faculty of Social Sciences; Artificial Intelligence; | Lecturer(s) | | | | Academic year | | 2017 | | Period | | PER1-PER2 | (04/09/2017 to 04/02/2018) |
| Starting block | | PER1 | |
| Course mode | | full-time | |
| Remarks | | - | Registration using OSIRIS | | Yes | Course open to students from other faculties | | Yes | Pre-registration | | No | Waiting list | | No | Placement procedure | | - |
| | | | | |
Upon completion of the course, students will:
- Generally understand how a Brain-Computer-Interface (BCI) works, recognise the basic constituent parts, understand how each of these work and how all parts work together to create a completely functional system.
- Understand the difficulties associated with BCI and know where the current major challenges in BCI development lie. Students should be able to specifically identify the major challenges in each of the part of the BCI cycle.
- Understand the difference between passive and active BCIs and understand what the purpose of intentional encoding is in making active BCIs effective.
- Understand how interesting neuro/psycho phenomena are transformed into practical BCI systems.
- Possess scientific knowledge of the underlying design of the current common BCI systems (specifically, IM, p300-based, SSEP-based).
- Possess knowledge of the range of applications to which BCI techniques can be applied, including neurofeedback, rehabilitation, consciousness detection, lie-detection.
- Understand how BCI system's limitations can influence the design of the rest of the BCI system including shared-autonomy, predictive-models (e.g. language models), error-correction.
- Understand the wider implications of the use of a BCI on society as a whole.
- Have experience in working with a team.
|
|
|
A Brain Computer Interface (BCI) is used to extract information about a person’s mental state directly from their brain signals and uses this information to control external devices. The detected mental state may be actively generated when the user performs specific mental tasks, such as imaging movements, to encode their intentions, or passively generated as a consequence of other influences on the user, such as their current stress level. BCI is a highly interdisciplinary field which builds on ideas and techniques from many areas including Neuroscience, Psychology, Signal Processing, and Artificial Intelligence.
This course provides a detailed introduction to BCI and covers all the major steps involved in making a working Brain-Computer Interface. The course begins with an introduction to the BCI cycle which explains how the signal inside the user's head is transmitted via all the necessary sensors and computers to the outside world. The course then goes on to investigate each of the cycle's steps in more detail.
Broadly speaking, there are two sides to a BCI:
1. the 'human-focused' side where mental-tasks and user stimulus are selected to ensure that the user makes the strongest signals possible, and,
2. the 'computer-focused' side where signal-processing techniques are used to extract the user-generated signals from the background system noise.
This course covers both sides of the BCI. The later part of the course will concentrate on the 'human-focused' side and will use ideas from Psychology and Cognitive-Neuroscience with focus on maximising the strength of the user generated signals.
Students will also gain hands-on experience; they will see what EEG signals look like and see how a BCI works during demonstration sessions. During these sessions a member of each group will have their brain signals measured and displayed in real-time. They will then perform various tasks (such as watching flashing lights) whilst observing the effects on the measured brain signals to identify the signal changes used in BCIs. They will also control a working BCI using these same signals.
The final six weeks of the course will focus on the group assignment. Working in small groups, students will get to apply all the knowledge they have gained in order to design, implement, test and evaluate an EEG-based BCI design of their choice. The final assignment will take the form of a competition to develop the best BCI to complete a 'brain-race' based on that used in the international Cybathalon BCI contest.
|
|
|
|
|
Course Work: Working in small groups, students will: • Develop a design for a BCI , • Implement this design, and conduct EEG experiments to test its effectiveness. • Present the design and write a report on the BCI and the experiment. A three hour written, closed-book, examination.
Total course assessment: 40% course work, 60% examination mark. |
• Knowledge of introductory Psychology; • Knowledge of basic mathematics (linear algebra, fourier analysis); • Knowledge of basic machine learning. |
Dr.J.D.R.Farquhar, T:024-3611938, E: j.farquhar@donders.ru.nl |
| | | Required materialsCourse materialBCI is a new field there is, as such, no suitable textbook for this course. This means that most of the literature will be in the form of slides which are used for teaching the course, and current research papers, which form compulsory course reading |
| ArticlesGerven, 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" in Journal of Neural Engineering (2009) Vol. 6: p. 041001. |
|
|
Recommended materialsBookAlso, 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. |
|
|
Instructional modesLecture RemarkLectures and tutorials
| Tutorials
|
| TestsExamTest weight | | 1 |
Opportunities | | Block PER1, Block PER2 |
|
|
| | |
| |
| |