SEIZURE DETECTION USING ARTIFICIAL NEURAL NETWORK IN VLSI ON-CHIPS

For epilepsy patients, the portable automated seizure identification device is very easy to transport. This paper proposes a very large scale integration (VLSI) architecture based on a nonlinear artificial neural network to make the device workable with high performance and a high detection precision. The architecture proposed consists primarily of an extraction and classification module. The classifier combines with a gaussian kernel, the modified sequence minimal optimization algorithm to make on-chip learning efficient. Results from the experiments show that the VLSI device designed increases precision in identification and education.