A Patient Specific Neural Networks (MLP) for Optimization of Fuzzy Outputs in Classification of Epilepsy Risk Levels from EEG Si

In recent years Neural Networks have been widely used as pattern and statistical classifiers in bio medical engineering. Most research to date using hybrid systems (Fuzzy-Neuro) focused on the Multi-Layer Perceptron (MLP). Here we focus on MLP network as an optimizer for classification of epilepsy risk levels obtained from the fuzzy techniques using the EEG signal parameters. The obtained risk level patterns from fuzzy techniques are found to have low values of Performance Index (PI) and Quality Value (QV). The neural networks are trained and tested with 480 patterns extracted from three epochs of sixteen channel EEG signals of ten known epilepsy patients. Different architectures of MLP network was compared based on the minimum Mean Square Error (MSE), the better MLP network (2-4-2) were selected. The MLP network out performs the fuzzy techniques with high Quality Value (QV) of 25 when compared to low QV of 6.25.

• Type: Journal
• Year: 2006
• Where: ENGL
• Authors: R. Sukanesh, R. Harikumar
• Keywords: ENGL 2006, Fuzzy Techniques, MLP Network, Neural Networks