Independent component analysis: algorithms and applications
Neural Networks
Computer Methods and Programs in Biomedicine
Hi-index | 0.00 |
In this paper, a method for characterizing cerebral infarction (CI) utilizing spontaneous electroencephalogram (EEG) is described. We obtained the time-frequency representations (TFRs) of EEG signals recorded from both normal subjects and CI patients. The corresponding characteristics were depicted by relative frequency band energy (RFBE) and Shannon entropy (SE) of TFR. Comparing with the normal subjects, the CI patients had some changes in EEG as follows: (1) delta and theta rhythms were attenuated while beta and gamma rhythms were enhanced, and the changes of delta and beta were more significant, (2) alpha was also blocked with eyes open, however the blocking action was less evident, (3) SE increase was pronounced. Consequently, the quantitative EEG methods are promising tools to provide helpful and sensitive information for the detection and diagnose of CI.