Brain Recording in Real-World Envifonments using Cognionics Dry-Wireless EEG
The objective of this reseach is to test the ability of the Cognionics Dry-Wireless EEG to be able to record brain activity in real-world settings in the presence of extensive environmental and physiological noise. For practical application of EEG in real-world settings it is required that the system does not require gel and is highly portable.
The Cognionics EEG System is evaluated using an auditory evoked potential experiment in an open cockpit biplane with the engine and avionics off and with the engine and avionics on. To determine offline whether real-time analysis of the dry-wireless EEG system is possible given the considerable environmental and physiological artifacts brain activity was recorded and compared between cruise flight and an aerobatic spin maneuver. Independent component analysis was used to separate artifacts and task related activity.
Experimental Environment: Open-Cockpit Biplane
Clear auditory evoked potentials AEP could be seen for both the avionics and engine off as well as the avionics and engine on conditions while recording in the open cockpit biplane. Single trials could be decoded for presence or absence of an auditory signal using least squares probabilistic classification. The features used for decoding included amplitude measures centered around the time of the peak of the AEP. Performance for the avionics and engine off condition was 72% and for the avionics and engine on condition was 69%.
AEP for 2 ICs Avionics and Engine Off
AEP for 2 ICs Avionics and Engine On
Brain activity of task related independent components in different frequency bands was found that significantly differed between cruise and aerobatic spin flight. Dipole analysis was used to source localize the component activity in the brain.
Please install the Flash Plugin
Clear auditory evoked potentials can be recorded in an open-cockpit biplane with the engine and avionics on using the Cognionics dry-wireless EEG systems.
Band-pass filtering and independent component analysis are necessary to remove environmental and physiological artifacts.
Real-time extraction of brain activity during extreme conditons is possible but requires individual specific independent component analysis models for artifact removal.