Classification of neonatal epileptic seizures using support vector machine
by Velayutham Vimala; K. Ramar
International Journal of Advanced Intelligence Paradigms (IJAIP), Vol. 12, No. 1/2, 2019

Abstract: Neonates are infants who are in their first 28 days of life. The diagnoses of neonatal seizures have been advocated by the use of clinical observations and electroencephalography (EEG). The continuous monitoring of neonatal EEGs in neonatal intensive care units is tedious and involves experts' intervention. The use of clinical decision support systems into the neonatal intensive care units has proved to produce aid to neonatal staff. The neonatal seizures of epileptic origin are more common and we recommend an approach to aid in the classification of the same using EEG signals of the neonates. Daubechies wavelet transform is used for the task of separation of frequency bands and the extraction of features. The theta rhythm of EEG reflects rightly the occurrence of epileptic seizures in neonates. The features taken into consideration for the classification are mean, variance, skewness and kurtosis. The support vector machine (SVM)-based classification is adopted for the development of the system which detects the presence or absence of epileptic seizures. The performance of this diagnostic aid system has been studied and the system has a sensitivity of 94% and specificity of 96%. The receiver operating characteristic curve is also used in the performance assessment.

Online publication date: Fri, 14-Dec-2018

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Advanced Intelligence Paradigms (IJAIP):
Login with your Inderscience username and password:

    Username:        Password:         

Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email