FPGA IMPLEMENTATION OF A WAVELET NEURAL NETWORK WITH PARTICLE SWARM OPTIMIZATION LEARNING FOR EPILEPTIC SEIZURE DETECTION

The main objective of this paper is to model and implementation of Field Programmable Gate Array (FPGA) for a Wavelet Neural Network (WNN) with Particle Swarm Optimization (PSO) learning ability to detect epileptic seizure. The electroencephalogram (EEG) signals were first pre-processed using Discrete Wavelet Transforms (DWTs) haar, dB2, sym8, and dB4. This was followed by the feature selection stage, where a set of four representative statistics were computed. The features obtained were fed into the input layer of WNNs. A more suitable PSO method is selected for is a population-based learning algorithm for WNN. In the approximation of a nonlinear activation function, we use a Taylor series and a look-up table (LUT) to achieve a more accurate approximation. A group of twenty epileptic patients were studied in this research. The accuracy of 98.27% is obtained for dB4 wavelet with Morlet activation function