Utilizing High Gamma (HG) Band Power Changes as a Control Signal for Non-Invasive BCI
Current electroencephalography (EEG) Brain-Computer Interface (BCI) methods typically use control signals (P300, modulated slow cortical potentials, mu or beta rhythm) that suffer from a slow time scale, low signal to noise ratio, and/or low spatial resolution. High gamma oscillations (70–150 Hz; HG) are rapidly evolving, spatially localized signals and recent studies have shown that EEG can reliably detect task-related HG power changes. In this chapter, we discuss how we capitalize on EEG resolved HG as a control signal for BCI. We use functional magnetic resonance imaging (fMRI) to impose spatial constraints in an effort to improve the signal to noise ratio across the HG band. The overall combination lends itself to a fast-responding, dynamic BCI.
KeywordsMotor Imagery Blood Oxygenation Level Dependent Blood Oxygenation Level Dependent Response High Gamma Slow Cortical Potential
- G. Pfurtscheller, EEG event-related desynchronization (ERD) and event-related synchronization (ERS), in Electroencephalography: Basic Principles, Clinical Applications and Related Fields, 4th edn, eds. by E. Niedermeyer, F. H. Lopes da Silva (Williams and Wilkins, Baltimore, 1999), pp. 958–967Google Scholar
- R. Prückl, B.Z. Allison, C. Hintermüller, B. Großwindhager, C. Kapeller, C. Guger, G. Edlinger, Poor performance in SSVEP BCIs: are worse subjects just slower? in 34th Annual International IEEE EMBS Conference of the IEEE Engineering in Medicine and Biology Society (2012)Google Scholar
- J.R. Wolpaw, D. Flotzinger et al., Timing of EEG-based cursor control. J. Clin. Neurophysiol. 16, 529–538 (1997)Google Scholar