Abstract
Epileptic patients are investigated in the phase of preoperative diagnostics of epilepsy by long-term monitoring (EEG, ECoG) to locate epileptic seizures. The introduced method extracts clinical important EEG-segments from recordings of several hours duration by calculating position and class symbol of the segments and emphasizing dependence between these passages. The analysis is divided into the steps segmentation, feature extraction, classification, syntactic analysis in time domain and clustering of the recording channels. The search for given sequences of features was written down as a context-free grammar and was realized by LR(0)-parser. The results of the method show good correlation to the visual EEG-results of experts, thus the time expensive work for manual EEG-examination can be reduced by concentrating to the relevant regions which are extracted by this automated analysis.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Ferber, G.: Verfahren zur automatischen Beschreibung des EEG im Rahmen der klinischen Diagnostik. Dissertation, Universität Hannover (1983)
Gaillard, J.M., Tissof, R.: Principles of automatic analysis of sleep records with a hybrid system. Comput. Biomed. Res. 6 (1973)
Gotman, J., Gloor, P.: Automatic recognition and quantification of interictal epileptic activity in the human scalp EEG. EEG & clin. Neurophysiology 44 (1976), pp. 513–529
Gotman, J., Qu, H.: Improvements in seizure detection performance by automatic adaption to the EEG of each patient. EEG & clin. Neurophysiology 86 (1992), pp. 79–87
Hellmann, G., Kuhn, M., Prosch, M., Spreng, M.: Extensible biosignal (EBS) file format: simple method for EEG data exchange. EEG & clin. Neurophysiology 99 (1996), pp. 426–431
Hofmann, W.: Ein Algorithmus zur Erkennung Epilepsietypischer Muster im Elektroenzephalogramm des Menschen. Diplomarbeit, Physikalisches Institut, Universität Bonn (1989)
Hofmann, W.: Untersuchung elektrocorticographischer Aufzeichnungen mit unüberwacht lernenden Klassifikatoren für die Epilepsiediagnostik. Dissertation, IPB, Friedrich-Alexander-Universität Erlangen-Nürnberg (1996)
Laible, T.: Digitale Filter: Ein Algorithmus zur Erkennung von Spike-WaveMustern in EEG-Daten. Hospitantenarbeit, IPB, Universität Erlangen-Nürnberg (1992)
Niemann, H.: Klassifikation von Mustern. Springer Verlag, Heidelberg (1983)
Schickentanz, T.: Implementierung und Bewertung von Verfahren zur Klassifikation von EEG-Registrierungen mittels Grammatiken. Diplomarbeit, IPB, Universität Erlangen-Nürnberg (1997)
Spreng, M., Hellmann, G., Kuhn, M., Reinhartz, K.-D., Stefan, H.: Unsupervised EEG/ECoG/MEG classification in epilepsy using array processors. in: Quantitative and topological EEG and MEG analysis. Universitätsverlag Druckhaus Mayer, Jena (1995)
Stefan, H.: Epilepsien: Diagnose und Behandlung. Chapman & Hall, New York (1995)
Walthers, R., Principe, J.C., Park, S.H.: Spike detection using a syntactic pattern recognition approach. Proc. 11th Ann. Int. Conf. IEEE Eng. in Medicine Biol. Soc. (1989)
Wendling, F., Bellanger, J.-J., Badier, J.-M., Coatrieux, J.-L.: Extraction of Spatio-Temporal Signatures From Depth EEG Seizure Signals Based on Objective Matching in Warped Vectorial Observations. IEEE Transactions on biomedical engineering 43 (1996), pp. 990–1000
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Schickentanz, T., Hellmann, G., Niemann, H., Spreng, M. (1997). Classification and Clustering of Electroencephalographical Recordings by Grammars in Epilepsy Diagnostic. In: Paulus, E., Wahl, F.M. (eds) Mustererkennung 1997. Informatik aktuell. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-60893-3_59
Download citation
DOI: https://doi.org/10.1007/978-3-642-60893-3_59
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63426-3
Online ISBN: 978-3-642-60893-3
eBook Packages: Springer Book Archive