Automatic Detection of Epileptic Seizures in EEG Using Machine Learning Methods

Authors: ,

Abstract: Epilepsy is a common neurological disorder in which the sudden onset of seizures can cause delirium, body twitching, foaming at the mouth and other symptoms requiring immediate treatment. Electroencephalogram (EEG) is the primary method for objectively detecting epilepsy in patients and must be conducted by a trained physician or specialist. Therefore, the development of methods for the rapid and accurate diagnosis of epilepsy could potentially save considerable time and cost. The This study uses Discrete Wavelet Transform to analyze physiological EEG parameters and retrieve a plurality of sub-bands. Multiple classification methods are used to compare the diverse data to select the most suitable method for classifying epilepsy EEG data and to develop a diagnostic mode. The results found that C4.5 in the sub-D2 produced diagnoses with an accuracy of up to 90%. Rule-based of decision trees can be used to quickly analyze large amounts of data, thus accelerating epilepsy diagnosis.

WSEAS Transactions on Biology and Biomedicine, ISSN / E-ISSN: 1109-9518 / 2224-2902, Volume 14, 2017, Art. #10

PDF
XML
Cite
×
Citation Tools
WSEAS AMA
Ying-Fang Lai, Hsiu-Sen Chiang, "Automatic Detection of Epileptic Seizures in EEG Using Machine Learning Methods," WSEAS Transactions on Biology and Biomedicine, vol. 14, pp. 69-74, 2017, DOI:
Ying-Fang Lai, Hsiu-Sen Chiang. Automatic Detection of Epileptic Seizures in EEG Using Machine Learning Methods. WSEAS Transactions on Biology and Biomedicine. 2017;14:69-74.
Copy to Clipboard
Citation copied to Clipboard

Certification

---