Boosting BCI technology: Towards a multi-dimensional fully implanted BCI

Abstract

For individuals with Locked-In Syndrome (LIS), assistive technology (AT) can be the last resource to reinstate communication and to allow meaningful interaction with their environment. Typically, AT relies on residual voluntary movement to control an adapted keyboard, a computer mouse, an eye-tracker, or another device. However, for many of these individuals, AT based on muscle activity is not an option, either because their muscle control is very weak or too erratic to produce volitional control. In these cases, neurotechnologies controlled without any residual muscle activity present a promising alternative tool to restore communication. An example of such communication tools is a Brain-Computer Interface or BCI. A BCI records signals from the brain, extracts features from the signal, translates them into a control signal, and provides feedback to the user, ideally without significant delay. In the last four years, we implemented and tested for the first time a fully implanted BCI in a patient with late-stage amyotrophic lateral sclerosis. This proof-of-concept study allowed us to demonstrate that electrocorticography (ECoG) is a usable technique for the stable recording of neuronal signals in people with LIS, and that attempted hand movement is an adequate strategy for reliable control of a BCI. While the results of this study are remarkable and provide unique data, the system still presents some limitations, the most important being that it only allows the detection of one hand movement, which is translated into a one degree-of-freedom control signal. A possible strategy to increase the number of degrees-of-freedom is to increase the number of implanted channels, by using high-density (HD) ECoG grids. Therefore, in the last four years we worked on extending the number of degrees-of-freedom for control, and thereby the speed and versatility of the system. For that, we addressed several issues in the design of a multi-dimensional fully implantable communication BCI based on HD-ECoG grids. We focused on several aspects of all parts of the BCI system, providing not only tools and methods that will facilitate the development of multi-dimensional ECoG-based BCIs, but also neuroscientific insight into which paradigms and brain areas may be used for multi-dimensional control.