Processing of the electroencephalogram in cardiac surgery.

The objective of this investigation was to find a method to automatically detect several types of abnormal EEG patterns which occurred during cardiac surgery. The EEGs were analyzed by means of several EEG processing and pattern recognition methods. It was found that a good classification into normal and abnormal EEG patterns was generally obtained if only two EEG features were used. The results of this investigation are of importance for implementing into a computer-based monitoring system for use during cardiac surgery.