Performance Evaluation of Different Neural Networks Used for Seizure Detection

  IJCOT-book-cover
 
International Journal of Computer & Organization Trends (IJCOT)          
 
© 2011 by IJCOT Journal
Volume-1 Issue-3                          
Year of Publication : 2011
Authors : Miss Ashwini D. Bhople, Prof. P. A. Tijare

Citation

Miss Ashwini D. Bhople, Prof. P. A. Tijare. "Performance Evaluation of Different Neural Networks Used for Seizure Detection". International Journal of Computer & organization Trends (IJCOT), V1(3):35-39 Nov - Dec 2011, ISSN 2249-2593, www.ijcotjournal.org. Published by Seventh Sense Research Group.

Abstract

Brain is one of the most important and complicated organs of humans. It is also susceptible to degenerative disorder, such as epilepsy. A disease caused due to temporary alternation in brain function due to abnormal electrical activity of a gro up of brain cells and is termed as epileptic seizure. It is a common chronic ne urological disorder characterized by seizures . In epilepsy, the normal pattern of neuronal activity becomes disturbed, causing strange sensations, emotions, and behavior or sometimes convuls ions, muscle spasms, and loss of consciousness. The term ‘Epilepsy’ is derived from the Greek word epilambanein or epilepsía which means ‘to seize or attack’. Anything that disturbs the normal pattern of brain cells (neuron) activity from illness to brain damage to abnormal brain development can lead to seizures. Epilepsy may develop because of an abnormality in brain wiring, an imbalance of nerve signaling chemicals called neurotransmitters, or some combination of these factors. Having a seizure does not n ecessarily mean that a person has epilepsy. Only when a person has two or more seizures is he or she considered to have epilepsy. The seizure occurs at random to impair the normal function of the brain. Seizures can be classified into two main categories d epending on the extent of involvement of various brain regions focal (or partial) and generalized. Generalized seizures involve from a circumscribed region of the brain, often called epileptic foci. EEGs and brain scans are most common and cost effective d iagnostic test for epilepsy. Worldwide, epilepsy affects 50 million people. In this paper different neural networks are studied and compared, for the detection of seizure

References

[1]. R. H arikumar and B. Sabarikumar narayanan ” Fuzzy techniques for class ification of epilepsy risk from EEG signal”, IEEE Conference on convergent technology for Asia - Pacific region TENCON 2003.
[2]. R. H arikumar and Dr. S. Raghavan and Dr . ( Mrs. ) R . Sukanesh “Genetic Algorithm for classification of Epilepsy risk level from EEG sig nals” , IEEE Conference on signals, Systems and Computers 2004 .
[3]. J. Gotman, “Automatic recognization of epileptic seizures in the EEG, Electroencephalogram”. Clin. Neurophysiol, vol. 54, pp. 530 – 540, 1982.
[4]. Nic olaos B. Karayiannis, Amit Mukherjee, “Detection of Pseu do sinusoidal Epileptic Seizure Segments in the Neonatal EEG by Cascading a Rule - Based Algorithm with a Neural Network” IEEE Transactions on Biomedical Engineering, vol. 53, no. 4, April 2006.
[5]. Vairav an Srinivasan, Chikkannan Eswaran, and Natarajan Sriraam, “Approximate Entropy - Based Epileptic EEG Detection Using Arti ficial Neural Networks” , IEEE Transactions on Information Technology in Biomedicine, vol. 11, no. 3, May 2007 .
[6]. Thasneem Fathima and M. Bedeeuzzaman , “Wavelet Based Features for Epileptic Seizure Detection” , MES Jour nal of Technology and Management .
[7]. N. Sivasankari. And Dr. K. Thanushkodi , ,”Automated Epileptic Seizure Detection in EEG Signals Using FastICA and Neural Network” Int. J. Advance. Soft Comput er . Appl., Vol. 1, No. 2, November 2009 .
[8]. N. McGrogen,”Neural Net works detection of Epileptic Seizures in the Electroencephalogram” , Probationary Research Transfer Report Oxford University infebruary 1999 .
[9]. Dr. R. Shantha Selva Kumari, J.Prabin Jose “ Seizure Detection in EEG Using Time Frequency Analysis and SVM ” , IEEE 20 11.

Keywords

Support Vector machine (SVM), Elman network (EN) , probabilistic Neural Ne twork (PNN), Generalized Feed Forward (GFFNN), Back Propagation Neural Network (BPNN) and Multilayer perceptron (MLP)