GR Semicolon EN

Show simple item record

dc.contributor.author
Tentsoglidis, Iordanis
en
dc.date.accessioned
2019-03-21T10:06:03Z
dc.date.available
2019-03-22T01:00:21Z
dc.date.issued
2019-03-21
dc.identifier.uri
https://repository.ihu.edu.gr//xmlui/handle/11544/29268
dc.rights
Default License
dc.subject
Sleep Classification
en
dc.title
Exploration of algorithms of automatic detection of sleep microstages ( of morphology of cycling alternating pattern (CAP) on Non-REM sleep)
en
heal.type
masterThesis
en_US
heal.creatorID.email
iordanis.ten@gmail.com
heal.classification
Machine Learning
en
heal.language
en
en_US
heal.access
free
en_US
heal.license
http://creativecommons.org/licenses/by-nc/4.0
en_US
heal.recordProvider
School of Science and Technology, MSc in Data Science
en_US
heal.publicationDate
2019-03-18
heal.abstract
The last decades the explosion in diverse domains of IT technology so in hardware as in software set new horizons for further research in many fields or defined from the ground whole new scientific disciplines. Bioinformatics and biomedicine definitely took advantage of tech explosion. As a consequence, sleep medicine is one of those new research areas that arose. This study tries to reproduce, extend and optimize state of the art technics for automatic sleep microstructure analysis. Especially focuses on the Cycling Alternating Pattern (CAP) and the detection of a CAP’s prominent feature, the A-phase. An algorithm is reproduced according to state of the art techniques and experimental approaches are tested to classify records between Non A-phase and A-phase events achieving 91% accuracy, 75% sensitivity and 91% specificity. As the results are competitive compared to other studies employing the same dataset (Physionet CAP Sleep database), it is believed that the applied techniques could contribute to sleep research.
en
heal.advisorName
Diamantaras, Konstantinos
en
heal.committeeMemberName
Diamantaras, Konstantinos
en
heal.academicPublisher
IHU
en
heal.academicPublisherID
ihu
en_US


This item appears in the following Collection(s)

Show simple item record

Related Items