AUTOMATIC ABNORMAL WAVES DETECTION FROM THE ELECTROENCEPHALOGRAM OF EPILEPSY WITH DWT

Abstract

This paper proposes a feature extraction and recognition algorithm for interictal and ictal EEG signals using Discrit Wavelet Transform (DWT).
Patients seizure consist 4 Males, 6 Females ages 3-35 years. Clinical status epileptic without seizure 10 Males, 18 Female, ages 10-40 years. Clinical status non epileptic 8 Male, 5 Female ages 8-42 years. Numerical data were acquired with EEG system at Karyadi hospital Semarang 2008-2013. The categorization is confirmed by Fast Fourier Transform (FFT) analysis. The dataset includes waves such as sharp, spike through DWT ( For this a mother daubechies 7, coiflets 1 and coiflets 5) of EEG records.
The experimental results show that this algorithm can achieve the sensitivity of 94.00% and pecificity of 93.75% for interictal and ictal EEGs,and the sensitivity of 92.50% and specificity of 92.75%, total accuracy of 91.21% for normal and ictal EEGs on data sets.Besides,the experiment with interictal and ictal EEGs from karyadi Hospital data set also yields sensitivity of 90.05% . specificity of 95% and total accuracy of 94.63% .
Automatic seizure detection is very helpful to review prolonged EEGs.The research carried out so far was to find the prospect of this digital signal processing on EEG waves to support the doctors work in this field.