Fuzzy techniques for classification of epilepsy risk level from EEG signals

The aim of this paper is to develop a fuzzy classification model for epilepsy risk level analysis from EEG signals. The parametric values such as energy, positive and negative peaks, sharp spikes and events are derived in each epoch of the EEG signal channels. Fuzzy techniques are used to classify the risk level in each epoch for all the channels; the risk level patterns obtained are found to have a low percentage of performance and quality value. In order to increase the classification rate, an optimization technique is used and a quality value of 11.9 is achieved when compared to the value 6.25 achieved in the previous case. A comparison of fuzzy techniques with and without optimization is studied. The focal epilepsy problem in normal fuzzy classification is solved using this new approach. A group of six patients with known epilepsy findings are used in this study. Further research work can be carried out in the classification of epilepsy risk level of a diabetic neuropathy patient. An analytic study is to be undertaken for classification, based on fuzzy aggregation operators and fuzzy discriminators. The number of samples may be increased to improve the classification rate.