Volume 29, Issue 10 (Monthly-Jan 2019)
Studies in Medical Sciences 2019, 29(10): 707-715 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
ghasemi M, puryusef M. CLASSIFICATION OF EPILEPTIC SEIZURE IN EEG SIGNAL USING ADAPTIVE NEURO FUZZY INFERENCE SYSTEM. Studies in Medical Sciences 2019; 29 (10) :707-715
URL: http://umj.umsu.ac.ir/article-1-4517-en.html
URL: http://umj.umsu.ac.ir/article-1-4517-en.html
University of Neyshabur , m.ghasemi@neyshabur.ac.ir
Abstract: (5250 Views)
Background & Aims: Epilepsy is a brain disorder in which nerve cells receive abnormal inputs. This disease can lead to abnormal behaviors, feelings and symptoms such as loss of consciousness, which is called the seizure. Identification and classification of the epileptic seizure events in electroencephalographic signal against free seizure intervals plays an important role in clinical investigations.
Materials & Methods: We used five groups of 100 EEG signals recorded at Bon University. EEG time series recorded in surface EEG recordings from healthy volunteers and intracranial EEG from epilepsy patients during the seizure-free interval within and outside the seizure. In the first step, statistical features were extracted from the time-frequency characteristics of EEG signals in five main spectra. Reduced dimension of the statistical features was fed to adaptive neuro fuzzy inference system as a strong classifier.
Results: The results obtained in this study improved the accuracy of their pre-published researches. The first and second error in our method has reached zero and 0.02, respectively.
Conclusion: This research is an effective way for diagnostic seizure events, specifically once there are suspected clinical symptoms of epileptic such as occurred in newborns.
Materials & Methods: We used five groups of 100 EEG signals recorded at Bon University. EEG time series recorded in surface EEG recordings from healthy volunteers and intracranial EEG from epilepsy patients during the seizure-free interval within and outside the seizure. In the first step, statistical features were extracted from the time-frequency characteristics of EEG signals in five main spectra. Reduced dimension of the statistical features was fed to adaptive neuro fuzzy inference system as a strong classifier.
Results: The results obtained in this study improved the accuracy of their pre-published researches. The first and second error in our method has reached zero and 0.02, respectively.
Conclusion: This research is an effective way for diagnostic seizure events, specifically once there are suspected clinical symptoms of epileptic such as occurred in newborns.
Type of Study: Research |
Subject:
Neuroscience
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
