Sleep Stage Classification Using Dempster–Shafer Theory for Classifier Fusion

Mehdi Abdollahpour, Tohid Yousefi Rezaii, Ali Farzamnia, Saeed Meshgini

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

5 Citations (Scopus)

Abstract

The importance of the automatic sleep stage classification is increasing in order to study of sleep stage transitions and sleep health. The researches are introducing new methods to obtain the highest accuracy compared to the expert scored hypnograms through classification of the electroencephalogram (EEG). Because of limitations when using the single information source this article combines the outcomes of classifiers obtained from different sources. By extracting 13 features from two channels of EEG signal and using these features for learning and testing linear discriminant analysis classifier, we reached accuracy of 71.93% for the Fpz-Cz channel and 6S.33% for the Pz-Oz channel. In the last step, Dempster-Shafer theory of evidence is used for classifier fusion and combining the outputs derived from the classification of two EEG channels, the classification accuracy increased by 88.23.

Original languageEnglish
Title of host publicationProceedings - 2018 IEEE International Conference on Artificial Intelligence in Engineering and Technology, IICAIET 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages32-35
Number of pages4
ISBN (Electronic)9781538678138, 9781538678121
ISBN (Print)9781538678145
DOIs
Publication statusPublished - 10 Feb 2019
Externally publishedYes
Event2018 IEEE International Conference on Artificial Intelligence in Engineering and Technology - Kota Kinabalu, Sabah, Malaysia
Duration: 8 Nov 20188 Nov 2018

Conference

Conference2018 IEEE International Conference on Artificial Intelligence in Engineering and Technology
Abbreviated titleIICAIET 2018
Country/TerritoryMalaysia
CityKota Kinabalu, Sabah
Period8/11/188/11/18

Cite this