Thesis defense Marianne Severens (Donders Series 135)
16 October 2013
Promotors: Prof.dr. P.W.M. Desain, Prof.dr. J. Duysens
Copromotor: dr. J.D.R. Farquhar
Towards clinical BCI applications: Assistive technology and gait rehabilitation
A brain-computer interface (BCI) allows for control of a device by detecting brain activity and translating this into actions, without involvement of peripheral muscle activity. BCIs can be used for different applications, depending on the target group. For example, patients with locked-in syndrome may use a BCI for communication or control of other assistive technology, whereas stroke patients could in the future benefit from BCI facilitated training for motor rehabilitation. In this thesis I investigate BCIs based on electroencephalographic (EEG) signals for these clinical applications.
In the first part of this thesis, the use of the somatosensory modality for BCI for assistive technology is investigated. Our results show that the somatosensory domain can be used in a gaze-independent BCI for communication or control. This was shown, not only in healthy controls, but also in a group of ALS patients with mild to moderate impairments. Nevertheless, with the current knowledge, a BCI with visual stimuli will likely perform better. However, patients who have strongly impaired vision or eye-gaze could possibly benefit from a BCI based on tactile stimulation.
In the second part of this thesis, the feasibility of using BCI for gait rehabilitation was investigated. The difficulty here is the influence of movement and muscle artifacts on the EEG recordings. We have shown that with special artifact removal methods, walking related brain signals can be detected and used in a BCI. This paves the way for using BCI training in gait rehabilitation.