Epileptic Seizure Forecasting Using AR Model Coefficients and Machine Learning Techniques

Celebi M., GÜLLÜ M. K.

Medical Technologies Congress (TIPTEKNO), İzmir, Turkey, 3 - 05 October 2019, pp.66-69, (Full Text) identifier identifier

  • Publication Type: Conference Paper / Full Text
  • Volume:
  • Doi Number: 10.1109/tiptekno.2019.8894960
  • City: İzmir
  • Country: Turkey
  • Page Numbers: pp.66-69
  • Kocaeli University Affiliated: Yes

Abstract

Epilepsy is a neurological disorder that affects human life negatively. In this study, epileptic seizure prediction study on EEG signals is performed by using AR model coefficients and machine learning techniques in order to estimate the epileptic seizure. For this purpose, AR model coefficients are extracted from the EEG signals and used as features. Then, these features are applied to three different classifiers and the performance results are obtained. Experiments are conducted on the American Epilepsy Society dataset. The results show that AR model coefficients are suitable features for epileptic seizure prediction studies.