ABSTRACT
A series of 63 clinical EEGs showing a variety of normal and abnormal patterns was analysed by computer with particular reference to the different types of pattern within the same EEG. Boundaries between different patterns were established by means of adaptive segmentation, so that the duration of the resulting segments was determined by the particular EEG itself (thus the term 'adaptive'). Four channels from each EEG were analysed, paired (left and right) channels were simultaneously segmented and analysed interactively. Similar segments were then clustered without supervision by estimating a probability density function in a 2-dimensional 'feature space' having dimensions of mean frequency and mean power. Individual clusters emerged as well-defined peaks of the surface, individual segments or small groups of duration insufficient to constitute a separate cluster, being identified as 'singular events' (e.g., rare sharp waves, artifacts). The autocorrelation function was used to characterize the EEG both for the segmentation and for the subsequent clustering of the resulting segments. In confirmation of our previous work, adaptive segmentation based on the autocorrelation function of the EEG was found to be quite satisfactory. Unsupervised clustering by estimation of the probability density function in feature space was found to give the correct number of clusters (usually less than 5) in a majority of the records (65%), but in the remaining minority of cases (35%), either overclustering or underclustering occurred. Further, the 'singular events' were occasionally partly included in a formal cluster. Comparison of these results of EEG clustering by unsupervised probability density function estimation with earlier results obtained by supervised hierarchical clustering suggests that there may be subtle cues used by the electroencephalographer in the classification of EEG patterns which have not been adequately approximated by the computer algorithms thus far used in this work. Hence at least some minimal degree of supervision in the clustering process may be necessary, at least for the present. On the other hand, the method recommends itself for the representation of illustrative EEG summaries which, in conjunction with a short written report, would provide the clinical neurologist with a sufficient picture of the real EEG without, in most cases, the need to inspect the original record.
Subject(s)
Electroencephalography , Action Potentials/classification , Arousal/physiology , Brain Diseases/physiopathology , Cerebral Cortex/physiology , Epilepsy/physiopathology , HumansABSTRACT
A phenomenological model for the representation of clinical EEGs is proposed. It assumes each individual record to consist of a few repetitive patterns which are described sufficiently by their power spectra. An algorithm for automatic EEG evaluation is described. It consists of two steps, a segmentation process which isolates the elementary patterns, and a clustering procedure which groups similar patterns with each other. Results are represented in graphical form. Diagnostic classification is not attempted. An appendix highlights the advantages of autoregressive modelling for EEG spectral analysis and, in particular, the estimation of the power contained in the various "rhythms".
Subject(s)
Computers , Electroencephalography , Microcomputers , Biometry , Humans , Pattern Recognition, AutomatedABSTRACT
The first step in a procedure for automatic EEG analysis is to compress the incoming data into a manageable format while preserving the essential diagnostic information. In our approach we mimic the visual procedure of looking through the record for segments and events of particular interest. We assume that the EEG is composed of roughly stationary segments of variable length, possibly superposed by sharp transients. By using an autoregressive model we have developed a procedure to detect the segment boundaries and locate transients, and to represent the information in the segments in terms of a set of parameters specifying their power spectra. In this way, the time structure as well as the frequency content of the signal is preserved. Examples of segmentation and transient detection are shown for several EEG signals, and the quality of the representation is demonstrated by simulating the original signal from the parameters. Possible applications to practical EEG analysis are discussed.