Evaluation of a syntactic pattern recognition approach to quantitative electroencephalographic analysis.

Abstract

This paper reports an experiment designed to evaluate the usefulness of a syntactic pattern recognition approach to quantitative EEG analysis. This approach provides a systematic way of organizing and interpreting spatially and temporally distributed activity in an EEG record. Many automated scoring systems evaluate only background activity, ignoring the significant transient events that may occur in an EEG. In contrast, syntactic methods allow evaluation of background activity in combination with transient events. In this paper, 454 EEGs recorded from a population of renal patients are evaluated using (1) the syntactic approach, and (2) a method based on discriminant analysis. The main finding is that the syntactic approach outperforms the discriminant analysis method when each technique is compared to visual scoring.