Noise Detection Motor Imagery

About Brain Computer Interfaces, EEG, neurofeedback, related event announcements, jobs, etc.
Post Reply
Posts: 1
Joined: Thu Feb 05, 2015 6:49 pm

Noise Detection Motor Imagery

Post by AditiAnand »

I am working on creating scenario in Open Vibe for motor imagery left and right. I am using Emotive-epoch headset. I am running in to this problem:

The problem is: When we to data collection for BCI it works fine until online scenario.In this scenario what was happening was even when there is no one was wearing the device on their head(basically we turn the device on and put it on the table.) It still generates left and right signal randomly. Which is wrong since there are no signals. Its a behavior of classifier and I am guessing its force to make a choice. Is their any way I could make all the other signals(which is not left/right) noise?
Posts: 110
Joined: Sun Mar 14, 2010 12:58 pm

Re: Noise Detection Motor Imagery

Post by »

Hello, indeed, the motor imagery scenario classifies left or right motor imagery, i.e., it assumes the user is making one or the other during the instructed period. So if you don't have signal, it will provide a random detection, which will be random left or right indeed, during the feedback periods. You need to modify the scenarios to perform noise detection (and add your own noise detection pipeline since this is not something readily available). Note that the motor imagery scenarios are synchronous BCI, i.e., they require the user to perform left or right motor imagery during dedicated time periods. If you want your BCI to detect when the user does nothing (or when there is nothing, e.g., when you turned off your device), you should try to build an asychronous BCI. This is a much harder task though, you should look into the literature to do so, see, e.g.:

Leeb, R., Friedman, D., Müller-Putz, G. R., Scherer, R., Slater, M., & Pfurtscheller, G. (2007). Self-paced (asynchronous) BCI control of a wheelchair in virtual environments: a case study with a tetraplegic. Computational intelligence and neuroscience, 2007.

Scherer, R., Lee, F., Schlogl, A., Leeb, R., Bischof, H., & Pfurtscheller, G. (2008). Toward self-paced brain–computer communication: navigation through virtual worlds. Biomedical Engineering, IEEE Transactions on, 55(2), 675-682.

Scherer, R., Schloegl, A., Lee, F., Bischof, H., Janša, J., & Pfurtscheller, G. (2007). The self-paced Graz brain-computer interface: methods and applications. Computational intelligence and neuroscience, 2007.

Bashashati, A., Ward, R. K., & Birch, G. E. (2007). Towards development of a 3-state self-paced brain-computer interface. Computational Intelligence and Neuroscience, 2007.

Faradji, F., Ward, R. K., & Birch, G. E. (2009). Plausibility assessment of a 2-state self-paced mental task-based BCI using the no-control performance analysis. Journal of neuroscience methods, 180(2), 330-339.

Lotte, F., Mouchere, H., & Lécuyer, A. (2008, December). Pattern rejection strategies for the design of self-paced eeg-based brain-computer interfaces. In Pattern Recognition, 2008. ICPR 2008. 19th International Conference on (pp. 1-5). IEEE.

please also kindly note that the emotiv epoc is not really suitable for motor imagery given its channels locations (no channels over the motor cortex)

Post Reply