A Novel Approach for the Analysis of Epileptic Spikes in EEG
ICCIMA '07 Proceedings of the International Conference on Computational Intelligence and Multimedia Applications (ICCIMA 2007) - Volume 01
Computers in Biology and Medicine
Optimizing the performance of an MLP classifier for the automatic detection of epileptic spikes
Expert Systems with Applications: An International Journal
Hi-index | 0.00 |
The present paper attempts to introduce a new approach for the automatic detection of epileptic spikes in EEG signal which plays a vital role in diagnosing epilepsy. In this approach, we have detected and classified epileptic spikes by using extracted features and Fuzzy ARTMAP neural network. The performance of classifying system is evaluated by three criteria of sensitivity, selectivity and specificity. For the classifying system applied in the current study, the obtained values of these three criteria are 88.24%, 93.75% and 90.9%, respectively, which compared with MLP classifier system, benefits from a higher speed and precision.