YISSUM RESEARCH DEVELOPMENT COMPANY OF THE HEBREW UNIVERSITY OF JERUSALEM LTD.;B.G. NEGEV TECHNOLOGIES & APPLICATIONS LTD., AT BEN-GURION UNIVERSITY
发明人:
DEOUELL, LEON Y.,GEVA, AMIR B.,FUHRMANN ALPERT, GALIT,MANOR, RAN EL,SHALGI, SHANI
申请号：
CA2909017
公开号：
CA2909017A1
申请日:
2014.04.13
申请国别(地区):
CA
年份:
2014
代理人:
摘要：
Systems and method for conduction of single trial classification of EEG signals of a human subject generated responsive to a series of images containing target images and non-target images, the method comprising: obtaining said EEG signals in a spatio-temporal representation comprising time points and respective spatial distribution of said EEG signals； classifying said time points independently, using a linear discriminant classifier, to compute spatio-temporal discriminating weights； using said spatio-temporal discriminating weights to amplify said spatio-temporal representation by said spatio-temporal discriminating weights at tempo-spatial points respectively, to create a spatially-weighted representation； using Principal Component Analysis (PCA) on a temporal domain for dimensionality reduction, separately for each spatial channel of said EEG signals, to create a PCA projection； applying said PCA projection to said spatially-weighted representation onto a first plurality of principal components, to create a temporally approximated spatially weighted representation containing for each spatial channel, PCA coefficients for said plurality of principal temporal projections； and classifying said temporally approximated spatially weighted representation, over said number of channels, using said linear discriminant classifier, to yield a binary decisions series indicative of each image of the images series as either belonging to said target image or to said non-target image.
