EEG monitoring based on fuzzy classification
W. Armin Kittel, Cooper Epstein, Monson H. Hayes III · 2003
The problem of automatic monitoring of electroencephalogram (EEG) recordings is addressed. A new approach based on fuzzy classification of spike events in the EEG is used in a monitoring system to reduce the number of false positive classifications. The overall monitoring system is divided into three phases of analysis: the transformation of the monitored signals into a symbolic representation; the syntactic classification of potential spikes; and the semantic verification of these spike events for the 16-channel EEG. This system is described along with results from recognition experiments.>