Nonstationarity Detection: The Use of the Cross Correlation Integral in ECG, and EEG Profile Analysis
Authors:
- Bunty B. E. KiremireBKBunty B. E. Kiremire
- Tshilidzi MarwalaTMTshilidzi Marwala
- 1 January 2008
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 5, 373-378
Abstract
This paper proposes the Stationarity Index, a measure of the similarities of the auto correlation integral of a section of a time series and the cross correlation of that section with others of the same time series. This measure of similarity is a measure of the stationarity of the time series and therefore can be used not only to detect nonstationarity but to also quantify it. The index is then successfully used in the analysis of electrocardiogram (ECG) and electroencephalogram (EEG) profiles to identify the changes in the dynamics of the signals as well as the occurrence of various events. The index displays sensitivity to changes in the dynamics exhibited in ECG signals that are the result of partial epileptic seizures, with the index going high in responseto the changes in the dynamics.Keywords
This publication has 17 references indexed in Scilit:
- The use of luminosity effects to calculate scaling regimes in the galaxy distributionAstronomy & Astrophysics, 2002
- Measuring Nonstationarity by Analyzing the Loss of Recurrence in Dynamical SystemsPhysical Review Letters, 2002
- Robust test for detecting nonstationarity in data from gravitational wave detectorsPhysical Review D, 2000
- Recurrence quantification analysis and principal components in the detection of short complex signalsPhysics Letters A, 1998
- Statistical test for dynamical nonstationarity in observed time-series dataPhysical Review E, 1997
- Detecting and Analyzing Nonstationarity in a Time Series Using Nonlinear Cross PredictionsPhysical Review Letters, 1997
- Stationarity and nonstationarity in time series analysisPhysica D: Nonlinear Phenomena, 1996
- Application of the correlation integral to respiratory data of infants during REM sleepBiological Cybernetics, 1995
- Distinguishing between low-dimensional dynamics and randomness in measured time seriesPhysica D: Nonlinear Phenomena, 1992
- Spurious dimension from correlation algorithms applied to limited time-series dataPhysical Review A, 1986