Advancing brain computer interface with multi-sensor integration
Background: A brain computer interface (BCI) system utilises neuro-physiological correlates of voluntary cognitive tasks to facilitate direct communication between human brain and computing devices without the involvement of neuromuscular pathways. One of the predominant voluntary cognitive tasks used in BCI is motor imagery, e.g. imagination of left and right hand movement. BCI is an emerging research area that can contribute significantly to enhancing the accessibility of information and communication technology (ICT) systems for the elderly and disabled. It is, in general, progressing in two areas: BCI for communication for improving independence & quality of life of severely disabled people such as sufferers of MND and spinal cord injury, and BCI for rehabilitation purposes, e.g. motor restoration in paralysis due to stroke.
The Research Problem and the Proposal: Current BCI systems, however, suffer from several limitations. They are not robust enough and performance variability among users is quite high. As a result, there is very little take-up for real-world applications. One of the critical limitations is because of the non-stationary brain waves obtained from EEG recordings. Due to this, it is difficult to obtain stable neurophysiological correlates of voluntary cognitive tasks to design the BCI. One way to counter this limitation is to augment the BCI with additional biosensor channels and thus enhance the degrees-of-freedom. Therefore, the project proposes to investigate additional physiological signals correlated with motor imagery tasks used in establishing a BCI. Recently a few studies have been reported confirming autonomic alterations in certain physiological signals during motor imagery [1-7]. These signals may include the skin resistance, blood flow, heart and respiratory rates. The proposal objective is therefore to investigate the most appropriate signal (or signal combination) that can provide features, which in combination with features obtained from EEG during motor imagery, enhances the robustness of the BCI system.
Expected Outcome: The proposed project is anticipated to facilitate designing of a much enhanced and practical BCI developed through the fusion of correlated autonomic signals with EEG signals obtained during motor imagery tasks.
RESOURCES NEEDED: All necessary bio-sensors and computing systems are available as part of a state-of-the-art BCI laboratory within ISRC.
References:
1. Pfurtscheller, G, Ortner, R, Bauernfeind, G, Linortner, P and Neuper, C. Does conscious intention to perform a motor act depend on slow cardiovascular rhythms? Neuroscience Letter, 2010, 468: 46-50.
2. Pfurtscheller, G, Leeb, R, Friedman, D and Slater, M. Centrally controlled heart rate changes during mental practice in immersive virtual environment: a case study with a tetraplegic, International Journal of Psychophysiology, 2008, 68: 1-5.
3. Scherer, R. Muller-Putz, GR and Pfurtscheller, G. Self-initiation of EEG-based brain-computer communication using the heart response, Journal of Neural Engineering, 2007, 4: L23-L29.
4. Pfurtscheller, G, Leeb, R and Slater, M. Cardiac responses induced during thought-based control of a virtual environment, International Journal of Psychophysiology, 2006, 62: 134-140.
5. Guillot, A and Collet, C. Contribution from neurophysiological and psychological methods to the study of motor imagery, Brain Research Reviews, 2005, 50: 387-397.
6. Decety, J. The neurophysiological basis of motor imagery, Behavioural Brain Research, 1996, 77: 45-52
7. Shahid, S., Prasad, G, and Sinha, R. K. On fusion of heart and brain signals for hybrid BCI. In: 5th Int. IEEE EMBS Conference on Neural Engineering, Cancun, Mexico. IEEE. 2011 , 5 pp.
Personnel Involved
First Supervisor: Prasad, G Dr
Second Supervisor: Shahid, S Dr
Collaborator: Dr Rakesh K. Sinha, Department of Biomedical Instrumentation, Birla Institute of Technology, Mesra, Ranchi, India.
Collaboration: This project does not involve collaboration with another establishment