Comparison of Time-Frequency Feature Extraction Methods for EEG Signals Classification

Autor:	Grzegorz Rutkowski, Krzysztof Patan, Paweł Leśniak
Rok vydání:	2013
Předmět:	Discrete wavelet transform Signal processing medicine.diagnostic_test Computer science business.industry Feature extraction Pattern recognition Electroencephalography Signal Time–frequency analysis Time–frequency representation medicine Artificial intelligence business S transform
Zdroj:	Artificial Intelligence and Soft Computing ISBN: 9783642386091 ICAISC (2)
DOI:	10.1007/978-3-642-38610-7_30
Popis:	Biomedical signal analysis can assist the diagnostic process for each neurological dysfunction. At that time the diagnostic tool continues to evolve, which is related to the development of both technological and innovative solutions proposed by scientists. One of possible solutions for automatic EEG analysis is to use modern signal processing tools, which are able to give a time-frequency representation of a signal. This paper presents a comparison of methods such as the discrete wavelet transform, the matching pursuit algorithm, and the S transform for feature extraction and then classification of neurological disorders. The research was carried out using real EEG recordings of epileptic patients as well as healthy subjects.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::e090ede581a194dcc71a3a0020461565 https://doi.org/10.1007/978-3-642-38610-7_30 Zobrazit plný text záznamu