Discrete Wavelet Transform based Statistical features for the Diagnosis of Epilepsy

Epilepsy is one of the most unpredictable chronic neurological disorders characterized by the recurrent unprovoked seizure. Following Stroke and Alzheimer's Syndrome, Epilepsy stands third in neurological disorders and is active in children over 5years. Electroencephalography (EEG) being a prominent component in the evaluation of epilepsy, provides information about epileptic form discharges, which are being utilised in the diagnosis of electrochemical syndromes. Enumerous techniques are being implemented in EEG signal for detecting an epileptic seizure. In the present work, Principle of Disc rete Wavelet Transform (DWT) is applied to identify the seizure from the background signal. Alongside parameters like Approximate Entropy (ApEn), Energy and standard deviation of a normal and epileptic person are being compared. Also to study any significant changes in these parameters during pre-ictal, ictal and post-ictal epileptic discharges. Significant differences have been found between the ApEn, Energy and standard deviation values of an Epileptic person. These par a meters also showed variations between pre-ictal, ictal and post-ictal phases. Results depict that EEG acquires lower ApEn values compared to the pre-ictal activity and standard EEG. And energy reveals that Epileptic EEG has a higher amplitude than normal.