International Journal of Basic & Applied Sciences IJBAS-IJENS Vol: 11 No: 02 17 113002-7575 IJBAS-IJENS © April 2011 IJENS I J E N S Transformation of EEG Signals Into Image Form During Epileptic Seizure Muhammad Abdy and Tahir Ahmad Department of Mathematics, Faculty of Science & Theoretical and Computational Modeling for Complex System, Nanotechnology Research Alliance, University Teknologi Malaysia, 81310 Skudai, Johor, Malaysia Abstract-- Electroencephalogram (EEG) is a recording of electrical activity of the brain and it contains valuable information related to the different physiological states of the brain. A quantitative EEG analysis has been developed over the years that introduce objective measure, reflecting not only the characteristics of the brain activity itself but also giving clues concerning the underlying associated neural dynamics. In this paper, the EEG signals during epileptic seizures are transformed into image form. The transformation is conducted at flat EEG (fEEG). fEEG is a new method to map high dimensional EEG signal into low dimensional space. The transformation of the signals consists of three steps. Initially, fEEG is divided into pixels and each of the pixels is determined its membership value in a cluster center. Then, membership value of the pixels at fEEG is determined by using maximum operator of fuzzy set. Finally, the membership degree of the pixels is transformed into image data. Index Term— EEG signals, flat EEG, Fuzzy Neighborhood, Fuzzy Region, Image. 1. Introduction Epileptic seizures are manifestations of epilepsy caused a temporary electrical disturbance in a group of brain cells (neurons). Symptoms experienced patients during seizures depending on the location and extent of the affected brain tissue. Epileptic seizures may cause negative physical, psychological and social consequences, including loss of consciousness, injuries and sudden death (Guo at el, 2010). Unfortunately, the occurrence of an epileptic seizure seems unpredictable and its process is very little understood. Surgical treatment may be an option in patients having epileptic seizures refractory to medication (Engle at el., 1993). However, if surgery is needed, the surgeons have to find out where the seizures originates. The main problem in epilepsy surgery is to resolve the size and location of the epileptic foci. Therefore, it is crucial to determine an accurate technique which is capable to localize the epileptic focus in patient who is suffering from epilepsy. Electroencephalogram (EEG) is a recording of electrical activity of the brain and it contains valuable information related to the different physiological states of the brain. There are two types of EEG depending on where the signal is taken in the head: intracranial or scalp (Adeli at el, 2003). For intracranial EEG is obtained by special electrodes implanted in the brain during a surgery. Meanwhile, scalp EEG, in a non- invasive way, small metal discs, also known as electrodes, are placed on the scalp with good mechanical and electrical contact. EEG recording is considered to be critical in localizing the epileptogenic zone. A careful analysis of the first clinical signs and symptoms of a seizure and of the evolution of the seizure symptomatology can provide important localizing clues. Although surface EEG recordings are less sensitive than invasive studies, they provide the best overview and, therefore, the most efficient way to define the approximate localization of the epileptogenic zone (Noachtar and Rémi, 2009). The traditional way of analyzing brain electrical activity, on the basis of EEG records, still relies mostly on its visual inspection and years of training. Although it is quite useful, of course, one has to acknowledge its subjective nature that hardly allows for a systematic protocol. To overcome this undesirable feature, a quantitative EEG analysis has been developed over the years that introduce objective measure, reflecting not only the characteristics of the brain activity itself but also giving clues concerning the underlying associated neural dynamics. Automatic EEG processing based on a Fourier transform has been used in clinical EEG. This approach is based on earlier observations that the EEG spectrum contains some characteristic waveform that fall primarily within four frequency bands. However, the Fourier transform requires stationary of the signal as well, and EEGs are highly non-stationary (Rossa at el., 2002). Pardalos at el. (2003) proposed a statistical information approach to the adaptive analysis of the EEG. Robust Bayesian estimation and order approximation provides the optimal selection of a local stochastic model. In the work of Subasi and Ercelebi (2005), delta, theta, alpha and beta sub-frequency of the EEG signal were extracted by using lifting-based discrete wavelet (LBDW). The LBDW coefficients of the EEG signals were used as input to logistic regression and multilayer perceptron neural network (MLPNN) that could be used to detect epileptic seizures. Guo at el. (2010) described a method for automatic epileptic seizure detection, which uses approximate entropy features derived from multiwavelet transform and combines with an artificial neural network to classify the EEG signals regarding the existence or absence of seizure. Based on the literature review, all of the electrical activity in the brain during seizures epileptic use an EEG signal in the waveform. In this paper, the EEG signals during epileptic Corresponding author: Tahir Ahmad (e-mail: tahir@ibnusina.utm.my).