Automated sleep stage detection and classification of sleep disorders

Malaekah, E 2016, Automated sleep stage detection and classification of sleep disorders, Doctor of Philosophy (PhD), Electrical and Computer Engineering, RMIT University.


Document type: Thesis
Collection: Theses

Attached Files
Name Description MIMEType Size
Malaekah.pdf Thesis Click to show the corresponding preview/stream application/pdf;... 2.01MB
Title Automated sleep stage detection and classification of sleep disorders
Author(s) Malaekah, E
Year 2016
Abstract Studies have demonstrated that more than 1 million Australians experience some sort of sleep-related disorder in their lifetime [12]. In order to improve the diagnostic and clinical treatment of sleep disorders, the first important step is to identify or automatically detect the sleep stages. The most common method, known as the visual sleep stage scoring, can be a tedious and time-consuming process. Because of that, there is a need to create or develop an improved automatic sleep stage detection method to assist the sleep physician to efficiently and accurately evaluate the sleep stages of patients or non-patients.

This research project consisted of two parts. The first part focused on the automatic sleep stages detection based on two individual bio-signals, which made up an overnight polysomnography (PSG), such as the electroencephalogram (EEG), and electrooculogram (EOG). Several features were extracted from these two bio-signals in the time and frequency domains. The decision tree and classification methods were utilised for the classification of the sleep stages.

The second part of this project focused on the automatic classification of different sleep and psychiatric disorders, such as patients with periodic limb movements of sleep (PLMs), sleep apnea-hypopnea syndrome (SAHS), primary insomnia, schizophrenia and healthy sleep. Different PSG parameters were computed for the classification of sleep disorders, such as descriptive statistics of sleep architecture. In conclusion, the advantage of an automatic sleep stage detection method based on a single-channel EEG or EOG signal can be undertaken with portable sleep stage recording instead of full the PSG system, which includes multichannel bio-signals. An automatic classification method of sleep and psychiatric disorders based on the descriptive statistics of sleep architecture statistics was found to be an effective technique for screening sleep and psychiatric disorders. This classification method can assist physicians to quickly undertake a diagnostic procedure.
Degree Doctor of Philosophy (PhD)
Institution RMIT University
School, Department or Centre Electrical and Computer Engineering
Keyword(s) Sleep scoring
Polysomnography
Electrooculography
Feature extraction
Sleep quality
Versions
Version Filter Type
Access Statistics: 389 Abstract Views, 1035 File Downloads  -  Detailed Statistics
Created: Wed, 13 Apr 2016, 11:48:43 EST by Denise Paciocco
© 2014 RMIT Research Repository • Powered by Fez SoftwareContact us