MU Digital Repository
Logo

Signalprocessing techniques for the Detection of epilepsy

Nayak, Prabhakar K (2008) Signalprocessing techniques for the Detection of epilepsy. Phd. Thesis thesis, Manipal Institute of Technology, Manipal.

[img] PDF
Prabhakar nayak.pdf - Submitted Version
Restricted to Registered users only

Download (62MB) | Request a copy

Abstract

Electroencephalogram (EEG) signals are the potentials in the cortex or on the surface of the scalp caused by the physiological activities of the brain. The electroencephalogram obtained fromthe scalp electrodes is a superposition of a large number of electrical potentials arising from several neurons. The frequency content of EEG plays an important role in its analysis and unveilsinformation about the various functions of the brain. Since the discovery ofEEG in 1929, many signal processing techniques have been applied to its analysis. EEG signals are nonstationary and get corrupted with noise while recording. The non-stationarities are due to the associatedtransient phenomena and frequencies evolving with time. Artifacts are noise added to anEEG signal by patient's movement and sources of electric field outside the patient's body. Epilepsy is one of the most common neurological disorders with a prevalence of 0.6 - 0.8% of the world's population. The epilepsy is characterized by a sudden and recurrent malfunction of the brain, which is termed as "seizure". The detrimental effects of epileptic seizures vary from momentary lapse of muscle control to death. Epileptic seizures reflect the clinical signs of an excessive and hyper-synchronous activity of neurons in the brain. The presence of epileptiform activity in the electroencephalogram confiims the diagnosis of epilepsy, Epilepsy does not refer to a specific disease, but rather to a group of symptoms due to varied reasons. It is the chronic brain disorder of various aetiologies characterized by recurrent seizures due to excessive discharge of cerebral neurons. During seizures, the scalp EEG of patients with epilepsy is characterized by high amplitude synchronized periodic waveforms, reflecting abnormal discharge of a large group of neurons. Between seizures, epileptiform transient waveforms, which include spikes and sharp waves, are observed in the EEG. A spike is a transient that is clearly distinguished from the background activity, with pointed peak and duration from 20 to 70 milliseconds and variable amplitude. The epileptic waves have various morphologies with main ingredients like spikes (13.5-50 Hz), sharp (5-12.5 . . Hz) and slow (1-2.5 Hz) waves. Usually spikes and sharp waves (SSW) or sharp and slow waves (SWW) appear simultaneously. In the EEG signals of normal persons, the alpha rhythm with the frequency 8-12 Hz is present, in addition to beta wave (14-25 Hz) and a little theta and delta waveat lower frequency. The EEG recorded from the surface of the brain (the "cortical EEG") gets altered as it passes through the skull and scalp. By the time the EEG signal arrives at the scalp surface to be transduced by an electrode, it is at a microvolt level and prone to environmental electrical noise and artifacts. The EEG signals are inherently complicated due to their non-stationary, and often nonlinear nature. Computerized handling of EEG data has led to improvements in data storage, data reduction, and data display. Suspected seizures are evaluated using a routine electroencephalogram (EEG), typically with a 20 to 25 minutes recoding of patient's brain waves. Routine EEGs record inter-ictal hallmarks of epilepsy, including spikes, sharp waves, or spike-and-wave complexes. Clinical neuro-physiologists periodically review the EEG recordings and analyze the seizures that may have occurred during the monitoring session. In many cases, patients are not aware of the occurrence of their own seizures. From the pattern recognition point of view, the problem of detection of epilepsy is extremely complex. Moreover, an automated seizure detection system is thus of great interest in identifying EEG sections that need to be reviewed

Item Type: Thesis (Phd. Thesis)
Subjects: Engineering > MIT Manipal > Electronics and Communication
Depositing User: MIT Library
Date Deposited: 09 Feb 2015 11:53
Last Modified: 09 Feb 2015 11:53
URI: http://eprints.manipal.edu/id/eprint/141846

Actions (login required)

View Item View Item