Computerized Clinical Electroencephalography in Perspective

Abstract

Recent developments in the field of computerized clinical electroencephalography (EEG) are surveyed, with particular reference to techniques of analysis of background (stationary) EEG activity, transient (nonstationary) activity, and to integrated systems for multichannel clinical EEG's. A variety of approaches have been used for the basic EEG analyses. For background activity, the fast Fourier transform (FFT) and autoregressive approaches have predominated. For fast transients, segmentation, double differentiation, and inverse filtering have been prevalent. Some integrated systems, which provide a summary, and in some cases an evaluation of the basic EEG analyses, are limited to processing of only the background activity, whereas others include both background and transient activity. Some systems have been designed primarily to be an aid to the clinical electroencephalographer in the preparation of his report. A very few systems have been designed to provide a printed report similar to the conventional clinical EEG report. Although not considered extensively in this review, artifact rejection and/or compensation will necessarily be a major aspect of any fully computerized system. Overall, the field of computerized clinical EEG is now a relatively rapidly evolving one in which further progress is likely to be accelerated by the utilization of microprocessors and high-speed arithmetic processing devices.