Title :
Non-Gaussian modeling of EEG data
Author :
Charles, Prophete J. ; Sclabassi, Robert J. ; Sun, Mingui
Author_Institution :
Lab. for Comput. Neurosci., Pittsburgh Univ., PA, USA
Abstract :
It is common when modeling EEG data to employ an assumption of normality. While this assumption usually provides a modest approximation for random variables, its use for EEG data is limited. In this paper we examine why general modeling of EEG data as normal is inadequate and provide an example of approximating various stages of a seizure with non-normal distributions
Keywords :
Gaussian distribution; brain models; electroencephalography; probability; EEG data; nonGaussian modeling; nonnormal distributions; normality; seizure; Brain modeling; Electroencephalography; Frequency; Gaussian distribution; Histograms; Maximum likelihood estimation; Random variables; Robustness; Surges; Testing;
Conference_Titel :
[Engineering in Medicine and Biology, 1999. 21st Annual Conference and the 1999 Annual Fall Meetring of the Biomedical Engineering Society] BMES/EMBS Conference, 1999. Proceedings of the First Joint
Conference_Location :
Atlanta, GA
Print_ISBN :
0-7803-5674-8
DOI :
10.1109/IEMBS.1999.804176
