Cortically-coupled computer vision for rapid image search”, Neural Systems and Rehabilitation Engineering

Abstract We describe a real-time EEG-based brain-computer interface (BCI) system for triaging imagery presented using rapid serial visual presentation (RSVP). A target image in a sequence of non-target distractor images elicits in the EEG a stereotypical spatio-temporal response, which can be detected. A pattern classier uses this response to re-prioritize the image sequence, placing detected targets in the front of an image stack. We use single-trial analysis based on linear discrimination to recover spatial components that reect differences in EEG activity evoked by target vs. non-target images. We nd an optimal set of spatial weights for 59 EEG sensors within a sliding 50 ms time window. Using this simple classier allows us to process EEG in real-time. The detection accuracy across ve subjects is on average 92%, i.e. in a sequence of 2500 images, resorting images based on detector output results in 92 % of target images being moved from a random position in the sequence to one of the rst 250 images (rst 10 % of the sequence). The approach leverages the highly robust and invariant object recognition capabilities of the human visual system, using single-trial EEG analysis to efciently detect neural signatures correlated with the recognition event. Index Terms electroencephalography, braincomputer inter-face, corticallycoupled computer vision, rapid serial visua