Automatic seizure detection based on support vector machines with genetic algorithms
SEAL'06 Proceedings of the 6th international conference on Simulated Evolution And Learning
Hi-index | 0.00 |
We propose an Alzheimer's disease (AD) recognition method combined the genetic algorithms (GA) and the artificial neural network (ANN). Spontaneous EEG and auditory ERP data recorded from a single site in 16 early AD patients and 16 age-matched normal subjects were used. We made a feature pool including 88 spectral, 28 statistical and 2 nonlinear characteristics of EEG and 10 features of ERP. The combined GA/ANN was applied to find the dominant features automatically from the feature pool, and the selected features were used as a network input. The recognition rate of the ANN fed by this input was 81.9% for the untrained data set. These results lead to the conclusion that the combined GA/ANN approach may be useful for an early detection of the AD. This approach could be extended to a reliable classification system using EEG recording that can discriminate between groups.