Epilepsy Recognition by Higher Order Spectra Analysis of EEG Signals

Epilepsy Recognition by Higher Order Spectra Analysis of EEG Signals

Seyyed Abed Hosseini
DOI: 10.4018/978-1-4666-5888-2.ch546
(Individual Chapters)
No Current Special Offers

Chapter Preview



A lot of research has been undertaken in assessment of epilepsy over the last few years (Andrzejak, 2001; Andrzejak & Widman, 2001; Tzallas, 2007; Chua, 2007; Hosseini, 2013; Guo, 2010). Gotman (1982) presented a computerized system for recognizing a variety of seizures. Murro et al. (1991) developed a seizure recognition system based on the discriminant analysis of the EEG signal recorded from the intracranial electrodes. They used three features include relative amplitude, dominant frequency and rhythmicity. A recognition sensitivity of 91-100% was achieved, with false positive rates of 1.5-2.5 per hour. Kannathal et al. (2005) have shown the importance of various entropies for recognition of epilepsy. Subasi (2007) used a method for analysis of EEG signals using discrete wavelet transform (DWT) and classification using an adaptive neuro fuzzy inference system (ANFIS). They conclude that, the ANFIS model achieved accuracy rates which were higher than that of the stand-alone artificial neural network model. In most researches, choosing suitable features is important for epilepsy seizure recognition. Higher-order spectral (HOS) or polyspectra analysis is by now a well established signal analysis technique with many applications in science and engineering, especially biomedical signal processing (Shahid, 2005; Hosseini, 2009; Hosseini, 2010; Xiang, 2002; Zhou, 2008; Abootalebi, 2000).

This chapter studies features related to the third order statistics of the signal, namely the bispectrum and bicoherence with both quantitative and qualitative view. The rest of this chapter is as follows: the database, brief review of higher-order spectral features, Hinich test, normalization and classifier are explained. Then the results and performance is illustrated. Finally, the discussion is reported.

Key Terms in this Chapter

Bicoherence: Is a squared normalized version of the bispectrum.

Higher Order Spectra: ( HOS): Refers to functions which use the third or higher power of a sample, as opposed to more conventional techniques of lower-order statistics, which use constant, linear, and quadratic terms.

Bispectrum: The 3rd order polyspectrum is called the bispectrum. It is contain phase information and is a function of two independent frequencies, which could take on both positive and negative values.

Electroencephalography (EEG): Is a technique which contains much information about the patient’s psycho-physiological state. It can be recorded in two essential ways: The first and most common is non-invasive recording known as scalp recording. The second is invasive recording that often is known as inter-cranial EEG.

Epilepsy: Is a brain disorder that is characterized by sudden and recurrent seizures. It can cause abnormal electrical activity in the brain and may alter consciousness, perception, sensation, behavior and body movement.

Complete Chapter List

Search this Book: