Analogue Evolutionary Brain Computer Interfaces
Lead Research Organisation:
University of Essex
Department Name: Computer Sci and Electronic Engineering
Abstract
The keyboard is a piece of plastic with lots of switches, which provides us with a reliable, but very unnatural form of input. The mouse is slightly less primitive. Still, it is only an electro-mechanical transducer of musculoskeletal movement. Both have been with us for many years and are still the best computer interfaces we have at the moment, yet they are unusable for people with severe musculoskeletal disorders and are themselves known causes of work-related upper-limb and back disorders: both hugely widespread problems for the UK's workforce.Wouldn't it be nice some day to be able to dispose of them and replace them with Brain-Computer Interfaces (BCIs) capable of directly interpreting the desires and intentions of computer users?This adventurous proposal aims to carry out an innovative and ambitious interdisciplinary research programme at the edge of Computer Science, Biomedical Engineering, Neuroscience and Psychology aimed at developing precisely these devices with a novel powerful BCI approach recently developed by the applicants.BCI has been a dream of researchers for many years, but developments have been slow and, with rare exceptions, BCI is still effectively a curiosity that can be used only in the lab. So, what's different about this project?In very recent research, we were able to develop a prototype BCI mouse capable of full 2-D motion control (a rarity in the BCI world), which uniquely can be used by any person without any prior training within minutes. This was possible thanks to our taking a completely innovative approach to BCI. Previous BCI designs were based on the paradigm of observing EEG signals looking for specific features or waves, manipulating them and then making a yes-or-no decision as to whether such features or waves were present. Contrary to this design wisdom, we completely dispose of the decision step and allow brain waves to directly control the computer via simple analogue transformations. Furthermore, we only partially design the system, leaving the completion and customisation of the design for each specific user to an evolutionary algorithm. This, thanks to an artificial form of Darwinian evolution inside the computer, performs the difficult tasks of selecting the best EEG channels, waves and analogue manipulations. Using these same two ingredients (analogue approach and evolutionary design) and starting from our successful experimental BCI mouse, this project specifically aims at developing brain-computer interfaces which are sufficiently robust, flexible and cheap to leave the lab and that can start making a serious impact in the real world.To maximise performance, preliminary work will determine the optimal visual presentation conditions that minimise cognitive load, perceptual errors and target-distractors interference.
Organisations
Publications
Citi L
(2010)
Documenting, modelling and exploiting P300 amplitude changes due to variable target delays in Donchin's speller.
in Journal of neural engineering
Poli R
(2009)
Analogue evolutionary brain computer interfaces [Application Notes
in IEEE Computational Intelligence Magazine
Poli R
(2012)
A genetic programming approach to the evolution of brain-computer interfaces for 2-D mouse-pointer control
in Genetic Programming and Evolvable Machines
Poli R
(2011)
Applications of Evolutionary Computation
Poli R
(2010)
Eigenbrains: The free vibrational modes of the brain as a new representation for EEG.
in Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference
Poli R
(2010)
Reaction-time binning: a simple method for increasing the resolving power of ERP averages.
in Psychophysiology
Poli R
(2011)
Genetic Programming
Salvaris M
(2011)
Novel sequential protocols for a ERP based BCI mouse
| Description | Since no design techniques existed for a brain-computer interface mouse, in the project we combined knowledge from signal processing and psychology with machine-learning techniques, including innovative evolutionary algorithms and support-vector machines. However, this was not enough to overcome the extremely difficult challenges involving in eliciting and interpreting brain signals. We also had do carry out fundamental research to understand how the shape and amplitude of P300 waves varies depending on the type and timing of the visual stimuli used. In turn, this required developing better ways of recording and averaging such waves. Results with this holistic approach were extremely promising. In particular, in the exploration of visual stimuli, we found a totally innovative protocol, which uses periodic, and thus predictable, sequences of stimuli instead of the traditional random sequences. This goes against all the psychophysiology literature on the generation of P300s, yet, with an appropriate mental task for users, we found that periodic sequences produced much stronger and more widely distributed P300s than traditional sequences. Correspondingly, this produced a marked increase in performance in our mouse. In using evolutionary algorithms to aid the design of our BCI mice, we discovered that they devoted much attention to minimising the effects of muscular artifacts (such as eye blinks or swallowing) on the mouse trajectories. When we combined our new periodic stimulation patterns with the best machine learning technology and the best evolved system for artifact rejection, we obtained excellent results. Subjects were able to use the mouse in both controlled conditions and in a standard Windows environment with very good accuracy only minutes after wearing the electrode cap and with no previous training. Our analogue mouse is currently the best P300-based BCI mouse in the literature by far, performing one movement every 100ms. In addition, we were able to exploit the newly acquired knowledge on P300 variability beyond our BCI mouse, building a matrix speller that provides a significant improvement in accuracy over the top-performing algorithm in the literature to date. Finally, while developing our analogue systems, we stumbled onto a couple of other useful results. While further research is required to make our BCI mouse and speller competitive for able-bodied people, performance and reliability appear to be now sufficient for people suffering from severe muscular disorder or locked-in syndrome. |
| Exploitation Route | We have taken this research forward in a number of ways in EP/K004638/1 "Global engagement with NASA JPL and ESA in Robotics, Brain Computer Interfaces, and Secure Adaptive Systems for Space Applications" |
| Sectors | Healthcare |