Feature extraction in epilepsy using a cellular neural network based device - first results
Ronald Tetzlaff, Christian Niederhofer, P. Fischer · 2003
In this paper, the bioelectrical activity of a human brain in epilepsy is analyzed using a cellular neural network-universal machine (CNN-UM) proposed by T. Roska and L.O. Chua (IEEE Trans. Circuits and Systems II, vol. 40, pp. 163-173, 1993). Therefore a feature extraction method based on binary input-output patterns and Boolean CNN with linear weight functions called pattern detection algorithm (R. Tetzlaff et al, IEEE Int. Workshop on Cellular Neural Networks and Their Appl., pp. 259-266, 2002) is used. First results of a hardware application with a CNN-UM realized as a mixed-mode array processor (M. Laiho et al, Int. J. Circuit Theory and Appl., pp. 165-180, 2002) are presented.