ARMA – Filtering of Evoked Potentials

Summary

A method of evoked potential data processing alternative to the averaging method will be discussed. It is to extract evoked potentials only from two segments. The theoretical basis of this procedure is the so-called ARMA model which gives a high frequency resolution of a periodic signal mixed with noise in a relatively low signal-to-noise ratio. The method is tested by simulation studies. The applicability will be demonstrated by clinical examples.

Eine Methode zur Erfassung evozierter Potentiale wird diskutiert, die im Gegensatz zum sog. Averaging eine Schätzung einzelner Potentiale aus zwei Segmenten gestattet. Grundlage des Verfahrens ist das sog. ARMA-Modell, das eine Frequenzanalyse hoher Genauigkeit eines periodischen Signals gestattet, das mit Rauschen in einem relativ niedrigen Signal – Rausch-Verhältnis additiv gemischt ist. Die Methode wird anhand von Simulationsbeispielen getestet. Ihre Anwendung wird mit klinischen Beispielen demonstriert.

Key-Words:

Evoked Potentials - ARMA Model - Stochastic Process - Selective Filtering

Schlüssel-Wörter:

Evozierte Potentiale - ARMA-Modell - stochastische Prozesse - selektive Filterung

• References

• 1 Basar E.. EEG-Brain Dynamics. Relation between EEG and brain-evoked potentials.. Amsterdam: Elsevier; 1980
• 2 Cadzow J. A.. High performance spectral estimation-A new ARMA method. IEEE Trans. Acoust., Speech, Signal Processing, ASSP-2S 1980; 524-529.
• 3 Chan Y. T., Langford R. T.. Spectral estimation via the high-order Yule-Walker equations. IEEE Trans. Acoust., Speech, Signal Processing, ASSP-1982-30 689-698.
• 4 Childers D. G.. Evoked responses: elec-trogenesis, models, methodology, and wavefront reconstruction and tracking analysis. Proceed. IEEE 1977; 65: 611-625.
• 5 Coppola R., Tabor R., Buchsbaum M. S.. Signal to noise ratio and response variability measurements in single trial evoked potentials. Electroenc. clin. Neurophysiol 1978; 44: 214-222.
• 6 Friedmann Y., Carmon A.. Analysis of single cerebral evoked responses by adaptive filtering. Transact. San Diego Biomed. Symp., pp 233-238. 1979;
• 7 Heinze H. J., Künkel H.. EMD effects on visual evoked potentials (in press).
• 8 Heinze H. J., Künkel H.. Extraction of single evoked potentials. Paper presented at the Academy of Sciences, Prague, Nov 1982
• 9 Heinze H. J., Künkel H., Massing W.. Selective filtering of single evoked potentials by high performance ARMA methods. In N. Yamaguchi, and K. Fujisa-wa. (Eds) Recent Advances in EEG and EMG Data Processing, pp. 29-46. Amsterdam: Elsevier/North-Holland Biomedical Press; 1982
• 10 Jenkins G. M., Watts D. G.. Spectral Analysis and its Application. San Francisco, CA: Holden-Day; 1968
• 11 Kay S. M.. A new ARMA spectral estimator. IEEE Transact. Acoust., Speech, Signal Processing, ASSP-2S 1980; 585-588.
• 12 Mayne D. Q.. On the calculation of pseudo inverses. IEEE Transact. Automat. Contr., AC 1969; 14: 204-205.
• 13 McGillem C. D., Aunon J. I.. Measurements of signal components in single visually evoked brain potentials. IEEE Transact. biomed. Engrg., BME 1977; 13: 232-241.
• 14 Stearns S. D.. Digitale Verarbeitung analoger Signale, pp. 154-156. München/ Wien; Oldenbourg: 1979