Use of Databases in QEEG Evaluation

Abstract

Background. Quantitative electroencephalography (qEEG) analysis incorporating the use of normative or reference database comparison has developed from being primarily a research tool into an increasingly widely used method for clinical neurophysiological evaluation. Method. A survey of several of the most widely used qEEG databases as well as issues surrounding the construction and use of these databases is presented, comparing and contrasting the various features of these databases, followed by a discussion of critical issues in this developing technology. Results. This review considers the concept of normalcy, norming of qEEG features, and validation of clinical findings. Technical issues such as methods for recording and analysis, filter use, broad bands versus single Hz finer frequency resolution, the number of variables relative to the number of cases, and the problem of multiple statistical testing are addressed. The importance of the recording electrode and montage reformatting for normative EEG data is emphasized. The use of multiple references is suggested. Discussion. A brief review of the characteristics of several major databases will exploit new technological developments and increasing sophistication in statistical analysis of EEG data. Implementing new measures such as variability over time and extraction of features such as event-related desynchronization (see Pfurtscheller, Maresch, & Schuy, 1985) and gamma synchrony (Rennie, Wright, & Robinson, 2000) are likely to have important clinical impact. Caution is urged in the use of automated classification by discriminant analysis.