Anzeige
Erschienen in:
2021 | OriginalPaper | Buchkapitel
Epileptic Seizure Detection Using Tunable Q-Factor Wavelet Transform and Machine Learning
verfasst von : Ala Tokhmpash, Sarah Hadipour, Bahram Shafai
Erschienen in: Advances in Neuroergonomics and Cognitive Engineering
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Abstract
Epileptic seizures constitute an important group of neurological disorders in brain that affect many people globally each year. Complexity of EEG signals due to their high-dimensional nature, as well as artifacts in data due to equipment flaws, pose significant challenges to physicians in diagnosing epileptic seizures directly and manually from EEG signals. In this paper, a method is proposed to combine signal processing and machine learning for diagnosing epileptic seizures and tested on the Bonn University database. We used Tunable Q-Factor wavelet transform (TQWT) method to transform signals. Subsequently, various statistical properties, frequency, chaotic and fractional were extracted from the TQWT sub-bands. Subset selection techniques were used in order to reduce the features of the methods used and their results were compared. Finally, a SVM method with different kernels were tested, where the empirical results show the high efficiency of the proposed method in the diagnosis of epileptic seizures.