Accepted:
20 Dec 2003
| |
/vol10/das02/ | © Copyright 2002 | |||
| Volume 10 | Received: Accepted: |
09 May 2002 20 Dec 2003 |
|||
|
Nonlinearity in EEG: investigation by surrogate data analysis
Das, A. & Das, P. |
|
| Abstract | |
| Nonlinearity has to be tested before applying nonlinear time series methods to quantify the complex dynamical structure of electroencephalogram [EEG]. Such an attempt is made in this work with three different EEG signals and a synthetic time series of known nature. We shall employ the surrogate data analysis technique to remove any nonlinearity present in the four datasets and then employ nonlinear dynamical tools to estimate complexity of the datasets. A comparison will be made for each dataset with its surrogate counterpart to find difference in calculated measures, meaning loss of information due to the removal of nonlinearity. Any significant difference will characterize that dataset as nonlinear. We find that EEG data cannot be ascribed as always being linear or nonlinear. Test for nonlinearity should be made using method described above. | |
|
|
Multimedia Links (none) Reference Links Citation Reference Das, A. & Das, P. (2002), Nonlinearity in EEG: investigation by surrogate data analysis, Complexity International, Volume 10, Paper ID: das02, URL: http://www.complexity.org.au/ci/vol10/das02/ |
Get viewers for PS & PDF   |
