Los Angeles, CA
March 31, 2009 to April 2, 2009
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/CSIE.2009.652
Epilepsy is a common chronic neurological disorder that is characterized by recurrent unprovoked seizures. About 50 million people worldwide have epilepsy at any one time. This paper presents an Intelligent Diagnostic System for Epilepsy using Artificial Neural Networks (ANNs) and Neuro-Fuzzy technique. In this approach the feed-forward neural network has been trained using Back propagation algorithm (BPA) and by Adaptive Neuro Fuzzy Inference System (ANFIS). First, all the data (from UCI machine learning repository) has been normalized so that the value of every attribute is between 0 and 1. Out of 265 instances, 200 instances have been used for training the system and 65 have been used for testing purposes. The simulator has been developed using MATLAB and performance is compared by considering the metrics like accuracy of diagnosis, training time, number of neurons, number of epochs etc. The results obtained clearly shows that the presented methods have improved the inference procedures and are advantageous over the conventional architectures on both efficiency and accuracy.
Intelligent systems, artificial neural networks, neuro-fuzzy systems, epilepsy, diagnosis
Ritu Tiwari, Anupam Shukla, "Intelligent System for the Diagnosis of Epilepsy", CSIE, 2009, 2009 WRI World Congress on Computer Science and Information Engineering, CSIE, 2009 WRI World Congress on Computer Science and Information Engineering, CSIE 2009, pp. 755-758, doi:10.1109/CSIE.2009.652