Automatic Sleep Staging using Support Vector Machines with Posterior Probability Estimates

Author

Gudmundsson, Steinn ; Runarsson, Thomas Philip ; Sigurdsson, Sven

Author_Institution

Dept. of Comput. Sci., Iceland Univ., Reykjavik

Volume

2

fYear

2005

fDate

28-30 Nov. 2005

Firstpage

366

Lastpage

372

Abstract

This paper describes attempts at constructing an automatic sleep stage classifier using EEG recordings. Three different feature extraction schemes were compared together with two different pattern classifiers, the recently introduced support vector machine and the well known k-nearest neighbor classifier. Using estimates of posterior probabilities for each of the sleep stages it was possible to devise a simple post-processing rule which leads to improved accuracy. Compared to a human expert the accuracy of the best classifier is 81%

Keywords

electroencephalography; estimation theory; feature extraction; medical signal processing; neurophysiology; pattern classification; probability; sleep; support vector machines; EEG recording; automatic sleep stage classifier; feature extraction; k-nearest neighbor classifier; post-processing rule; posterior probability estimate; support vector machine; Band pass filters; Computer science; Electroencephalography; Electromyography; Electrooculography; Feature extraction; Information filtering; Sleep; Support vector machine classification; Support vector machines;

fLanguage

English

Publisher

ieee

Conference_Titel

Computational Intelligence for Modelling, Control and Automation, 2005 and International Conference on Intelligent Agents, Web Technologies and Internet Commerce, International Conference on

Conference_Location

Vienna

Print_ISBN

0-7695-2504-0

Type

conf

DOI

10.1109/CIMCA.2005.1631496

Filename

1631496