Paper Info
Title
Auto Mutual Information Analysis with Order Patterns for Epileptic EEG
Abstract
In this study, we investigated auto mutual information (AMI), based on order patterns analysis, as a tool to evaluate the dynamical characteristics of electroencephalogram (EEG) during interictal, preictal and ictal phase, respectively. Permutation entropy quantifies regularity in time series, while AMI detects the mutual information (MI) between a time series and a delayed version of itself. The results show that AMI method was able to reveal that the highest entropy values were assigned to interictal EEG recordings and the lowest entropy values were assigned to ictal EEG recordings. The classification ability of the AMI measures is tested using ANFIS classifier. Test results confirm that AMI method has potential in classifying the epileptic EEG signals.
Year
DOI
Venue
2009
10.1109/FSKD.2009.33
FSKD (5)
Keywords
Field
DocType
auto mutual information analysis,time series,ami method,eeg recording,lowest entropy value,epileptic eeg,permutation entropy quantifies regularity,highest entropy value,order pattern,ami measure,ictal phase,order patterns,epileptic eeg signal,auto mutual information,classification,entropy,electroencephalography,time series analysis,pattern analysis,mutual information,data mining
Time series,Pattern recognition,Computer science,Permutation entropy,Speech recognition,Artificial intelligence,Mutual information,Adaptive neuro fuzzy inference system,Classifier (linguistics),Ictal,Electroencephalography,Machine learning
Conference
Citations 
PageRank 
References 
0
0.34
8
Authors
3
Name
Order
Citations
PageRank
Gaoxiang Ouyang19412.36
Wang Yao28511.45
Xiaoli Li3427.35